University of Ghana  http://ugspace.ug.edu.gh
   
ADAPTATION OF COMPUTED RADIOGRAPHIC SYSTEM FOR 
TREATMENT SETUP VERIFICATION IN EXTERNAL BEAM 
RADIOTHERAPY. 
 
BY 
 
FRANCIS SANDY 
 
THIS THESIS IS SUBMITTED TO THE GRADUATE SCHOOL OF NUCLEAR 
AND ALLIED SCIENCES, UNIVERSITY OF GHANA, LEGON  
 
 
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF 
MPHIL DEGREE 
 
 
IN 
 
 
MEDICAL PHYSICS  
 
 
 
JULY, 2018 
University of Ghana  http://ugspace.ug.edu.gh
   
DECLARATION 
This thesis is a result of a work undertaken by Francis Sandy in the Medical Physics 
department, School of Nuclear and Allied Science, University of Ghana under the 
supervision of Mr. Samuel Nii Adu Tagoe, Dr. Stephen Inkoom and Professor John H. 
Amuasi. 
I, Francis Sandy, declare that except for references cited in this work, the contents of this 
research was done by me and that this thesis work has not been submitted by any one for   
academic examination towards any qualification in the University of Ghana or elsewhere. 
 
Sign---------------------------                                                   Sign------------------------------- 
Francis Sandy                                                                       Dr. S. Inkoom 
(Student)                                                                                (Principal Supervisor) 
Date----------------------------                                                   Date ----------------------------- 
 
Sign-----------------------------                                                  Sign----------------------------- 
Prof. J. H. Amuasi                                                                  Mr. S. N. Tagoe 
(Co-Supervisor)                                                                       (Co-Supervisor) 
Date-----------------------------                                                   Date----------------------------                                                                                                                            
 
ii 
 
University of Ghana  http://ugspace.ug.edu.gh
   
ACKNOWLEDGEMENTS 
 The glory goes to God almighty for his strength, wisdom, protection, guidance and 
divine favour over me throughout this research study, and taking me through another 
educational level in my life. 
My second thanks and appreciation goes to my wonderful supervisors Dr. S. Inkoom, 
Prof. John Humphrey Amuasi and Mr. Samuel Nii Adu Tagoe who have followed every 
detailed progress of this research work to a successful completion, and have used their 
knowledge and experience to constructively criticise and correct my mistakes.  
With great appreciation to the staff of medical physics at the Korle-Bu Teaching Hospital 
for their support throughout this study. 
Much thanks to my sponsorship body International Atomic Energy Agency (IAEA) for 
their financial support throughout this programme in Ghana.  
To all my lecturers at SNAS a big thanks and appreciation for their knowledge imparted 
in me during the course of this programme, and I want to assure them that their good 
work will bring a change in my life and my country Sierra Leone. 
To all my friends and colleagues, I am grateful for the love shown during the period of 
this programme. 
To my relatives, family thanks for your support. Your support has been my strength 
throughout this programme and I wish you all God blessing. 
iii 
 
University of Ghana  http://ugspace.ug.edu.gh
   
TABLE OF CONTENTS 
 
DECLARATION ................................................................................................................ ii 
ACKNOWLEDGEMENTS ............................................................................................... iii 
TABLE OF CONTENTS ................................................................................................... iv 
LIST OF FIGURES ......................................................................................................... viii 
LIST OF TABLE ................................................................................................................ x 
LIST OF ABRREVIATIONS ........................................................................................... xii 
ABSTRACT ...................................................................................................................... xv 
CHAPTER ONE ................................................................................................................. 1 
INTRODUCTION .............................................................................................................. 1 
1.1 Background ......................................................................................................... 1 
1.2 Statement of Research Problem and Justification .......................................... 9 
1.3 Objectives ............................................................................................................ 9 
1.4 Specific Objectives.............................................................................................. 9 
1.5 Scope and Delimitation .................................................................................... 10 
1.6 Thesis Organization ......................................................................................... 10 
CHAPTER TWO .............................................................................................................. 12 
LITERATURE REVIEW ................................................................................................. 12 
2.0 Introduction ........................................................................................................... 12 
iv 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.1. The physics of computed radiography ............................................................... 12 
2.2 Historical background .......................................................................................... 15 
2.3 Megavoltage imaging in radiotherapy ................................................................ 15 
2.3.1 Cone Beam Computed Tomography (CBCT) ............................................. 16 
2.3.2 Kilovoltage Cone Beam Computed Tomography (KVCBCT) .................. 16 
2.3.3 Megavoltage Cone Beam Computed Tomography (MVCBCT) ............... 17 
2.3.4 Radio Frequency Beacons (RFB) ................................................................. 18 
2.3.5 Magnetic Resonance Device (MRD) ............................................................. 18 
2.3.6 Video Surface Imaging (VSI) ........................................................................ 18 
2.3.7 Electronic Portal Imaging Device (EPID) .................................................... 19 
2.3.8 Video-Camera Based EPIDs ......................................................................... 19 
2.3.9   Matrix- Ion Chamber Array ....................................................................... 21 
2.3.10 Scanning Linear Arrays (Solid state system). ........................................... 25 
2.3.11 Flat-Panel Device ......................................................................................... 26 
2.4 Radiographic Films ............................................................................................... 26 
2.4.1 The Base .......................................................................................................... 27 
2.4.2 Emulsion ......................................................................................................... 28 
2.5 Image Quality Assessment with Image J software............................................. 29 
2.5.1 Calculation of Signal-to Noise ratio (SNR) .................................................. 32 
2.5.2 Calculation of contrast-to-Noise ratio .......................................................... 33 
v 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.5.3 Calculation of Relative Modulation Transfer Function (RMTF).............. 36 
2.6. Quality Assurance of External Beam Radiotherapy ........................................ 39 
2.6.1 Quality Assurance .......................................................................................... 39 
2.6.2 Quality Control .............................................................................................. 40 
2.6.3. Need for quality assurance in radiotherapy ............................................... 40 
2.6.4 Verification of radiation field shape in radiotherapy ................................. 41 
2.6.5 Verification of patient setup .......................................................................... 42 
2.7 Accidents in radiotherapy .................................................................................... 44 
CHAPTER THREE .......................................................................................................... 47 
MATERIALS AND METHOD ........................................................................................ 47 
3.2. Theratron Equinox 100 Cobalt 60 teletherapy Machine. ............................ 47 
3.3. Computed Radiography Cassette ................................................................... 50 
3.4. Computed Radiography Reader ..................................................................... 51 
3.5. Radiotherapy Film Cassette. ........................................................................... 52 
3.6. Scanner (Microtek Scan Maker 9800XL) ...................................................... 54 
3.7. Image J Software.............................................................................................. 55 
3.8. Film Processor .................................................................................................. 56 
3.9. Phantom ............................................................................................................ 57 
3.10. Image Quality Assessment ............................................................................ 59 
CHAPTER FOUR ............................................................................................................. 61 
vi 
 
University of Ghana  http://ugspace.ug.edu.gh
   
RESULTS AND DISCUSSION ....................................................................................... 61 
4.1 Signal-to-Noise ratio for both CR and Film. .................................................. 61 
4.2 Comparing SNR of the two detectors (CR and Film) of irradiation time 0.1 
min. ........................................................................................................................... 66 
4.3 Contrast to Noise Ratio for both CR and Radiographic Film at irradiation 
time 0.2min .............................................................................................................. 68 
4.4 Comparing CNR of the two detectors (CR and Radiographic Film) of 0.2 
min ............................................................................................................................ 72 
4.5 Spatial Resolution ............................................................................................. 75 
4.6 Comparing the RMTF for the two detectors (CR and Film) at 0.1 Min ..... 78 
4.7 summary of findings ......................................................................................... 80 
1.8 Contribution to knowledge. ......................................................................... 81 
CHAPTER FIVE .............................................................................................................. 82 
CONCLUSION AND RECONMENDATION ................................................................ 82 
5.1 CONCLUSION ................................................................................................. 82 
5.2 RECONMENDATION ..................................................................................... 83 
References ......................................................................................................................... 85 
 
  
vii 
 
University of Ghana  http://ugspace.ug.edu.gh
   
LIST OF FIGURES 
Figure 1. 1: Diagram showing the chart of steps in radiation therapy process. .................. 5 
 
Figure 2. 1: Picture depicting sequence of events leading to the formation of images in 
CR ..................................................................................................................................... 13 
Figure 2. 2: Picture of a video-based EPID (From Boyer et al.1992) .............................. 21 
Figure 2. 3: Matrix-ion chamber array EPID (From Van Herk, 1987) ............................. 23 
Figure 2. 4: Cross section of radiographic film ................................................................ 26 
Figure 2.5:/Step-wedge……………………………………………………………          33 
Figure 2.6: Exposed image of the step-wedge……………………………………           33 
Figure 2.7: Comparing CNR for two system…………………………………..        33 
Figure 2.8: Comparing contrast of two systems……………………………….        33 
Figure 3. 1: Theraton equinox-100 cobalt-60 unit at Korle-Bu Teaching Hospital.......... 50 
Figure 3. 2: Fujifilm IP Cassette type CC ......................................................................... 51 
Figure 3. 3: CR reader at Korle-Bu Teaching Hospital .................................................... 52 
Figure 3. 4: Picture of the radiographic film cassette ....................................................... 54 
Figure 3. 5: Scanner (Microtek scan maker 9800XL) ...................................................... 55 
Figure 3. 6: Open view of image J software interface ...................................................... 56 
Figure 3. 7: Picture of film processor at Korle-Bu Teaching Hospital ............................. 57 
Figure 3. 8: Experimental setup showing the constructed phantom ................................. 59 
 Figure 3.9: Scan image of phantom on the film…………………………………………59 
Figure 3.10: Scan image of phantom on CR……………………………………………..59 
 
Figure 4. 1: Graph of SNR for CR of irradiation time of 0.1 min .................................... 63 
viii 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Figure 4. 2: Graph of SNR for Film of irradiation time of 0.1 min .................................. 65 
Figure 4. 3. Graph showing the two detectors SNR (CR and film) .................................. 67 
Figure 4. 4: Graph of CNR for CR of 0.2 min irradiation time. ....................................... 69 
Figure 4. 5: Graph of CNR for Film of 0.2 min ................................................................ 71 
Figure 4. 6: A graph showing combine values of CNR for both Film and CR ................ 73 
Figure 4. 7: CR detector showing the phantom image at 0.2min treatment time ............. 73 
Figure 4. 8: Film detector showing phantom image at 0.2min treatment time ................. 73 
Figure 4. 9: Graph of RMTF for CR of 0.1 min ............................................................... 73 
Figure 4. 10: Graph of RMTF for Film detector ............................................................... 73 
Figure 4. 11: Graph of RMTF for both CR and Film detectors ........................................ 78 
 
 
 
 
 
 
 
 
 
 
ix 
 
University of Ghana  http://ugspace.ug.edu.gh
   
LIST OF TABLE 
Table 2. 1: Causes of misadministration in external beam therapy in the 2000 IAEA 
Report (Thwaites et al., 2003). ........................................................................................ 45 
 
Table 4. 1: Result of scanned data of phantom using image J software (CR) of 0.1 
min .................................................................................................................................... 61 
Table 4. 2: Calculated results of SNR for CR obtained from scanner data of 
irradiation time of 0.1 min ............................................................................................. 62 
Table 4. 3: Result of scanned data using image J software (FILM) of 0.1 min ......... 64 
Table 4. 4: Calculated results of SNR for Film obtained from scanner data of 
irradiation time of 0.1 min ............................................................................................. 64 
Table 4. 5: SNR values for both detectors (CR and FILM) ........................................ 66 
Table 4. 6: Result of CNR from scanned data using image J (CR) 0.2 min .............. 68 
Table 4. 7: Calculated result of CNR for CR acquire from scanner data of 
irradiation time of 0.2 min ............................................................................................. 68 
Table 4. 8: Result of CNR from scanned phantom for film of 0.2 min ...................... 70 
Table 4. 9: Calculated result of CNR for Film acquire from scanner data of 
irradiation time of 0.2 min ............................................................................................. 70 
Table 4. 10: Showing CNR values for CR digital detector and Film. ........................ 72 
Table 4. 11: showing RMTF for CR digital detector of 0.1 min irradiation time ..... 75 
Table 4. 12: Showing RMTF for Film of 0.1 min ......................................................... 76 
Table 4. 13: Shows RMTF values for CR digital detector and Film .......................... 78 
 
x 
 
University of Ghana  http://ugspace.ug.edu.gh
   
xi 
 
University of Ghana  http://ugspace.ug.edu.gh
   
LIST OF ABRREVIATIONS 
ALARA                  As low as reasonable achievable 
CR                          Computed Radiography 
CNR                        Contrast-to-noise ratio 
CT                           Computed Tomography 
CBCT                      Cone-beam computed tomography 
CRI                          Computed Radiographic Imaging 
DRRs                        Digitally Reconstructed Radiographs 
EBRT                        External Beam Radiotherapy 
EPID                          Electronic Portal Imaging Device 
ESF                             Edge Spread Function 
HV                              High Voltage 
ICRU                          International Commission on Radiation Units and Measurement 
IGRT                           Image Guidance Radiation Therapy 
IP                                 Imaging Plate 
IEC                              International Electrotechnical Commission 
IAEA                           International Atomic Energy Agencies 
KVCBCT                    Kilovoltage Cone-Beam Computed Tomography 
xii 
 
University of Ghana  http://ugspace.ug.edu.gh
   
KBTH                        Korle-Bu Teaching Hospital 
LED                           Light-Emitting Diode  
MTF                          Modulation Transfer Function 
MVCBCT                 Megavoltage Cone-Beam Computed Tomography 
MeV                          Mega-electron Volts 
MR                           Magnetic Resonance 
MLC                         Multileaf Collimator 
MW                           Motorized Wedge 
NTCP                         Normal Tissues Complication Probability 
PACS                          Picture Archival Communication System 
PSP                             Photostimulable Phosphor Plate 
PMT                            Photomultiplier Tube 
PSL                              Photostimulated Luminescence 
PV                              Pixel Values 
PTV                            Planning Target Volume 
PVC                         Polyvinyl Chloride 
PDDs                         Percentage Depth Dose 
PMMA                      Poly-Methyl Methacrylate 
xiii 
 
University of Ghana  http://ugspace.ug.edu.gh
   
QC                           Quality Control 
ROI                             Region of Interest 
SNR                            Signal-To-Noise Ratio 
SF                               Screen Film  
TRS                            Technical Series Report 
TCP                            Tumour Control Probability 
VEPIDs                      Video-Camera Electronic Portal Imaging Device 
 
  
xiv 
 
University of Ghana  http://ugspace.ug.edu.gh
   
ABSTRACT 
In radiation therapy, the traditional way of providing accuracy of megavoltage radiation 
treatment fields, is by the use of portal radiographs. In these radiographs the film detector 
is positioned at the exist point of the beam beside the patient and details of its anatomical 
parts located within the field borders are taken. There is poor visualization of landmarks 
in portal films due to the high energy of photons used for treatment. Due to this draw-
back, an electronic portal imaging device (EPID) is now being assembled with modern 
therapy machine. This device uses kilovoltage energies for treatment setup verification, 
which provides clear anatomical landmarks of patient for treatment. The importance for 
field edges verification is to: 
 Verify the shape of the radiation beam. 
 To establish a coordinate system common to both reference and portal image in 
which to account for deviation in patient setup. 
The aim of this study was to assess the possibility of using computed radiography (CR) 
system instead of the conventional film in the use for treatment setup verification with the 
telecobalt machine at the radiotherapy department of the Korle-Bu Teaching Hospital. In 
diagnostic radiology, the use of CR has been long in existence to obtain digital 
radiographic images which help in providing picture archival communication system 
(PACS). The film is substituted with a photostimulable phosphor plate (PSP) in CR 
system. Energies deposited by the X-ray beam on the PSP are read by the scanner laser 
beam in the CR reader. After the PSP has been scanned by the reader, it can also be ready 
for reuse by exposing it to light for about forty seconds for the erasure of any remaining 
xv 
 
University of Ghana  http://ugspace.ug.edu.gh
   
image on the plate. A phantom was constructed and fabricated from a perspex which 
compensates for the human tissue. The phantom was irradiated using cobalt-60 source in 
which both the CR and the radiographic film were used as detectors to detect the images 
of the built phantom. These images were scanned using the scan maker 9800 XL. This 
thesis defines the use of the current image quality matrices such as Modulation Transfer 
Function (MTF), Contrast to noise ratio (CNR) and Signal to noise ratio (SNR) which 
show a practical easy-to-use software based program Image J for measurable assessment 
of digital (CR) and conventional detector system (film). 
From the results obtained, the CR demonstrated superior values in CNR and SNR than 
that of the radiographic film, which shows better quality of images on CR. 
Notwithstanding these qualities in CR, the film also showed a better resolution than that 
of the CR. 
Therefore digital treatment portal and simulation images, can easily be obtained with     
CR system which offers a base for a picture archival communication system in radiation 
oncology. Simulation and digital portal images will ease verification of treatment setup. 
 
 
  
 
 
xvi 
 
University of Ghana  http://ugspace.ug.edu.gh
   
CHAPTER ONE 
 INTRODUCTION 
This section of the study deals with background, statement of research problems and 
justification, objectives of the study, scope and delimitation and the organization of 
the thesis. 
1.1   Background 
The major aim of radiation therapy is to deliver accurate prescribed radiation dose to a 
defined target volume in order to damage the life intimidating tumor. The contradictory 
goal is the avoidance of damage to the patient’s normal, tissue within the irradiated 
region. (Taylor & Francis, 2007). Over exposing normal tissues to radiation dose beyond 
their thresholds may result in the loss of functionality of the tissues involved or affliction 
of secondary cancer for an individual. It is therefore imperative to choose irradiation 
geometries that will maximize radiation dose to the intended target volume, while 
concurrently reducing radiation doses to normal tissues in close proximity to the target 
volume. This is done in other to achieve favorable treatment outcome during external 
beam radiotherapy (Buzdar et al., 2013). The determining factor is how well both goals 
can be achieved.  In external beam radiotherapy (EBRT), the total dose which is to be 
administered to a patient is divided into smaller segments (or fractions) and delivered 
over many days to weeks. This process is called fractionation. This is done to give the 
normal tissues time to repair sublethal damage, since normal tissues are faster in repairing 
sublethal damage than a tumour. Thus prolonging the radiation dose delivery and 
1 
 
University of Ghana  http://ugspace.ug.edu.gh
   
choosing the optimal prescribed dose per fraction will improve tumour control 
probability (TCP) and reduce normal tissues complication probability (NTCP). The time-
dose constraints are considered a lot by clinicians when prescribing radiation dose to a 
patient undergoing EBRT. Also, the radiation dose deposited at any point within the 
irradiated region of a patient must be accurately known. Dose distribution within the 
patient is calculated based on dosimetric functions and or data obtained in a full scatter 
water phantom and mathematical algorithms that can be used to explain the interaction of 
radiation with a medium. Owing to the complexities of the dose calculation procedures, 
determination of the dose distribution within a patient has now become the preserve of 
specialized computers known as treatment planning systems (TPS). The process of 
simulating the treatment delivery with a treatment planning system is to realize the intent 
of treatment prior to the real treatment delivered, (known as treatment planning). 
Treatment planning procedures have to be in agreement with the International 
Commission on Radiation Units and Measurement(Monti et al., 1995) (Stroom & 
Heijmen, 2002). Treatment planning involves the most appropriate way to irradiate the 
patient. Treatment planning is the first and one of the most important parts of the 
radiation delivery process. It outlines course of action for the weeks of individual 
treatment session.  It begins immediately after the radiotherapy decision is made and 
chosen as the treatment modality.  Unlike pharmaceuticals, radiation is not metabolized. 
Once dose of radiation has been delivered, it cannot be removed.  Therefore, it is most 
important that the initial planning should be done accurately. Radiation kills cancer cells 
by depositing energy within the cell. Malignant cells that are not struck by radiation will 
not be affected by treatment, thus resulting in spreading and eventual death of the patient. 
2 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Similarly, when a large number of normal tissue cells are hit by high levels of radiation, 
they are most likely to die, which result in complication of the patient. Therefore patient 
treatment designed to treat tumor must be accurate to avoid unnecessary irradiation of 
normal tissue. Treatment planning procedures are as follows: 
 Selecting the right location and immobilization technique so that treatment 
will be the same throughout the treatment phase. 
 Classifying the position, shape of the tumor, and the neighboring organs 
surrounding the tumour. 
 Choosing an appropriate beam arrangement 
 Assessing the distribution of dose around the planned target. 
 Calculations are done based on  the treatment machine settings to deliver 
the required total dose.(Khan, 2007) 
Figure1.1 below gives details of steps in treatment planning. From the figure,                                  
treatment verification is one of the key component in treatment planning system. 
Treatment verification is a way of assessing localized treatment beam relative to patient’s 
anatomy. In EBRT, dosimetric verification and geometric localization are done to 
enhance accuracy and confidence in dose delivery. Dosimetric verifications are normally 
done by measuring the dose at certain points inside (natural cavity) or outside the patient. 
These measured doses are then compared with the calculated and predicted doses at the 
same points by the treatment planning system. Geometric localization is enhanced by 
acquiring a portal image of patient within the course of treatment. Treatment fields 
earmarked the treatment of patient by creating radiographic images with teletherapy 
machine. This is done to ascertain the relationship between the treatment radiation fields 
3 
 
University of Ghana  http://ugspace.ug.edu.gh
   
and the anatomical landmarks within the irradiated region of the patient. The 
radiographic images acquired with teletherapy machine are compared with those obtained 
during treatment simulation. These procedures help in establishing uncertainties 
associated with patient setup and chance to rectify these uncertainties prior to treatment 
delivery. The process of verifying the localization of an intended target volume being 
irradiated is known as treatment verification. Treatment verification plays a key part in 
administration of patients getting radiation treatment. Verification of patients, as a 
component of quality assurance program is planned to identify mistakes and errors that 
happen during the course of treatment.   
4 
 
University of Ghana  http://ugspace.ug.edu.gh
   
University of Ghana  http://ugspace.ug.edu.gh
   
than predicted and are bound to take place in all stages of planning and treatment. These 
faults are shared in two classes: those related to prescription, computation, and 
specification of target doses and those that happen during the delivery stage. This takes 
account of treatment machine or patient alignment relative to the treatment machine. 
Portal imaging, in which radiographs are generated with transmitted megavoltage beams 
through patient are received on image receptors. The receptors are placed vertically to the 
direction of propagation beam and is used to verify errors (Shalev, 1995), such as; patient 
motion, proper alignment of treatment fields on the patient. Radiotherapy treatment beam 
is mostly used for portal imaging in therapy to ensure that prescribed irradiation 
treatment within anatomical regions and surrounding normal tissues are spared. The 
commonly used image receptor in portal imaging is a radiographic detector, positioned in 
lead covered cassette.  Verification of field arrangement, using portal imaging can 
increase accuracy by identifying localization errors. It has also been reported that a 
significant reduction in localization error could be achieved with increasing frequency of 
portal imaging. Identifying localization errors with portal imaging is greatly influenced 
by our ability to see clearly certain anatomical landmarks (such as bones) and their spatial 
resolutions to a radiation fields on a portal image. Most portal images suffer from 
extremely poor contrast.(Roehrig, 1990) This poor contrast is due to the megavoltage 
energy range beams which are used to acquire the images. As the energy increases from 
the diagnostic range i.e. from 30 – 150 kV, to therapeutic range i.e. 1 – 10 MV the mode 
of interaction of photons with biological material changes from photoelectric absorption 
to Compton scattering, which leads to the decrease in differential absorption of the 
tissues. Additionally, the size of the irradiation source (geometrical status) and voluntary 
6 
 
University of Ghana  http://ugspace.ug.edu.gh
   
or involuntary movement on the part of the patient are all causes of image degradation. 
The traditional way of proving the accuracy of megavoltage radiation treatment fields is 
by the use of portal radiography.  The film detector is positioned at the exit point of the 
beam beside the patient and details of its anatomical parts located within the field borders 
are taken. Portal radiographs are divided into three types (i) a localization radiograph is a 
portal radiograph produced by a short exposure and then comparing it to treatment time 
on daily basis required for that field of treatment. This portal film is used to regulate 
patient positioning and field boundaries prior to the delivery of the daily treatment. (ii) a 
verification radiography is formed when the image receptors are exposed to the field 
during the whole treatment sessions. Insensitive detectors such as slow films are used in 
this portal radiographs. (iii) the last one is the double exposure radiography, in which 
small exposure is done with an open field followed by another exposure in the real 
treatment field. It has the benefit of viewing anatomical structures beyond the field 
borders, which is significant in classifying patient setup, but most often  use is restricted 
by the radiation acceptance of exposed normal tissues (Shalev, 1995).  Portal films are 
emulsion-type film placed in a light-tight enclosure. Intensifying screens are not used in 
radiotherapy films as compared to those used in diagnostic radiology. Instead, a thin 
single layer of copper or aluminum metal is placed on the top surface of the film (beam 
entry side) which offers an electronic buildup that helps to raise the efficiency of the film. 
Due to the draw-back of poor visualization of landmarks in portal film systems, an 
electronic portal imaging device is now being assembled with modem radiotherapy 
machines (linear accelerators) for treatment. Electronic portal imaging device (EPID) is 
used to measure the transmitted X-ray intensity that goes out through a patient from a 
7 
 
University of Ghana  http://ugspace.ug.edu.gh
   
radiation port during a treatment phase. The radiation signal is converted electronically 
into a two dimensional (2D) digital radiographic image to verify the true beam location in 
relation to patient’s anatomy.  Advantages of EPID are that images can be acquired and 
viewed prior to treatment delivery. This research work was undertaken at the Korle-Bu 
Teaching Hospital in Ghana which has both linear accelerator and cobalt teletherapy units 
for radiotherapy treatment. Unfortunately during this research work the linear accelerator 
was installed but not commissioned and which had the EPI device as a package.  
Therefore, at Korle-Bu Teaching Hospital where this study was done, radiographic film 
system was used to verify radiotherapy treatment. Fuji Corporation was the first body to 
introduce computed radiography (CR) which gives the opportunity for the first practical 
digital X-ray system to be seen by the radiologist. The CR system was explored to obtain 
digital portal images for treatment verification at the hospital. The aim of the CR system, 
was to digitize X-ray images from storage phosphor plates. CR system consists of a 
cassette containing imaging detector, image reader/ digitizer, a workstation, software, 
monitors, and a printer.  There are two types of CR readers :(i) single-plate readers. In 
this reader, cassettes are loaded manually and read separately. (ii) the other reader is the 
multiple-plate readers, which holds multiple plates-up to eight, and these plates are 
stacked and loaded automatically. Imaging detectors are placed in a radiographic cassette 
and images are obtained using the X-ray system. Therefore, the use of the CR system for 
portal image verification at the Korle-Bu-Teaching Hospital would be of importance 
since they are planning to go digital, and this work will emphasize more on the use of CR 
system to verify patient treatment. This intervention would help to obtain better image 
quality, and enhance accurate treatment of patients which is the goal of radiation therapy. 
8 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
1.2 Statement of Research Problem and Justification 
Computed radiographic system (CR) has been used in diagnostic X-ray facilities as a 
substitute for radiographic films to minimize cost and reduce exposure to hazardous 
chemicals used for film processing, which is also labour intensive. With the CR system, 
images are able to be stored in a digital format making it possible to address issues 
relating to long term storage stability and retrieval of films. The radiotherapy department 
of the Korle-Bu-Teaching Hospital is planning to digitize its patient records, and has 
proposed the use of CR system for treatment verification with the telecobalt machine. 
The proposal was also as a result of the deplorable condition of the film processor within 
the facility, and also issues relating to procurement of films. Since the beam energy of the 
therapy level is by far higher than that of the diagnostic level, there is need to assess the 
possibility of using the CR system and how to enhance the quality of image obtained with 
the CR system. 
 
1.3 Objectives 
The main aim of this study was to assess the possibility of using CR system instead of the 
conventional screen films in use for treatment verification with the telecobalt machine at 
the radiotherapy department of the Korle-Bu-Teaching Hospital. 
1.4 Specific Objectives 
The specific objective of the study was: 
9 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 To construct a phantom from which CR (digital detector) and conventional film 
detectors images would be assessed. 
 To quantitatively determine the quality of images obtained from CR and film 
using the image quality parameters such as: signal to noise ratio (SNR), contrast 
to noise ratio (CNR) and Relative Modulation Transfer Function (RMTF). 
    
1.5 Scope and Delimitation 
This project work was carried out at the Korle-Bu-Teaching Hospital in Accra, Ghana. 
This work covered the use of a constructed phantom which replicated patient geometries. 
The CR cassettes would be used for treatment verification with the phantom after being 
irradiated with telecobalt machine. Test objects having densities similar to tissue of 
human body will be placed in the phantom to determine their contrast level. The CR 
reader would be used to read the image on the CR cassette. Quantitative analysis would 
be done on the images acquired on both CR and radiographic film system using Image J 
software. 
 
1.6 Thesis Organization 
This research write up covers five chapters. The first chapter focuses on the introduction 
of the study. Second chapter reviews existing literature relating to research problem of 
your study. Third chapter focuses on materials and methodology of your study. The 
fourth chapter looks at the results and discussions, while the fifth chapter provides the 
10 
 
University of Ghana  http://ugspace.ug.edu.gh
   
conclusion and recommendations from the overall findings. The reference section 
contains all the relevant works cited throughout the project work. 
 
 
 
 
 
 
 
 
 
 
 
 
11 
 
University of Ghana  http://ugspace.ug.edu.gh
   
CHAPTER TWO 
 LITERATURE REVIEW 
2.0 Introduction 
In this section relevant literature on the subject is reviewed. These include the physics of 
computed radiography, historical background of CR system, megavoltage imaging, 
radiographic films, electronic portal imaging device, and papers relating to the study. 
2.1. The physics of computed radiography 
The use of  computed radiography (CR) is presently built on photostimulable phosphors, 
BaFBr: Eu2+ commonly identified as storage phosphor (Sonoda et al., 1983), (Rowlands, 
2002). Phosphors are mostly in powder form and are obtained from  barium fluorohalide 
family (Rowlands, 2002), which are deposited on the substrate to form an imaging plate. 
Imaging plate (IP) used in the CR system is positioned in a light-tight enclosure. This 
plates are exposed to an x-ray source and are read out using the raster scanner which has 
a laser that releases the photostimulated luminescence (PSL). The light guide collects the 
blue PSL light and the light is detected using photomultiplier tube (PMT). The signal 
collected from the PMT is digitized and  images are formed (Fujita et al., 1989). The 
huge dynamic range, digital nature, easy convenience and unique image quality of CR 
system has led to an extensive approval in diagnostic radiology. The sequence of events 
engaged in producing a PSL signal begins as shown in figure 2.1 
 
 
12 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
 
 
 
 
 
 
 
 
 
Figure 2. 1: Picture depicting sequence of events leading to the formation of images in 
CR 
 
 Upon exposure of the PSP to an X-ray source, energy is transferred to the electron which 
results in the excitation of electrons in to a metastable condition. Mostly, half of these 
electrons go immediately to their ground state resulting in the quick release of light.  
Over time the 50% of the metastable electrons will go back to their ground state. This 
sometimes causes fading in the latent image which therefore requires the IP to be read 
immediately after exposure. The signal formed on the imaging plate (IP) are significantly 
degraded after eight hours of exposure of the CR cassette  (Bushong, 2013) . 
The other stage in the process is stimulation, which requires an infrared light with 
diameter 50 to 100 µm to be focused on the PSP.  The beam of infrared light penetrates, 
and spreads on the PSP. The amount of spread increases with the PSP thickness, thus 
stimulating a lot of metastable electrons to return to their ground state thus emitting a lot 
13 
 
University of Ghana  http://ugspace.ug.edu.gh
   
of light for the formation of latent image. The third step is the detection (reading) of the 
stimulated emission. The laser beam makes the metastable electrons in the PSP to go 
back to their original state thus leading to the emission of a shorter wavelength light.  
Results leading to the loss of signal are as follows:  
 Scattered emitted light 
 
 The collection proficiency of the photo detector. 
 
 
The signal acquisition at the stimulation stage of PSL does not totally change all the 
metastable electrons to their original state, some still remain in that higher state. Ghosting 
normally appears on subsequent use of imaging plate, if residual latent images are not 
removed. Residual latent images are removed by exposing the phosphor to high strong 
white light from the storage of a designed fluorescent lamps. At the stimulation cycle of 
PSP, result of no latent image would be formed when the dwell time of the laser beam is 
much longer at each position on the PSP. Background radiation usually fogged images 
due to the high sensitivity of the PSP. The laser light used to stimulate the PSP is 
monochromatic.  HeNe gas laser was used as stimulating source, but has been replaced 
with a solid state laser. Solid-state laser produces longer wavelength light and therefore is 
less likely to interfere with emitted light. Even so, optical filters are necessary to allow 
only emitted light that reaches the photo detector while blocking the intense stimulated 
light. 
 
 
 
14 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.2 Historical background 
The original innovation of CR system was done by Kodak (Luckey & Rochester, 1975) 
who considered keeping an X-ray image in a phosphor screen. The conceptualization of 
application was done by Fuji (“Thermoluminescence of solids,” 1980) to manufacture the 
leading imagers in the medical settings. Fuji, being the leading designer of CR in the 
early years, used BaFBr: Eu phosphor and a cassette-based method. At the same time, 
Agfa and Kodak undertook a work based on x-ray imaging on phosphor screen and 
intended to make an improvement on the same method done by Fuji, but unfortunately 
they were inhibited from publicizing their work based on patent right issues and fear of 
destroying the foundation they had on screen-film.  At that time, effect of storage on 
screen-film applications were observed when prints not needed were done over, i.e. a 
ghost image of previous exposure on the screen that appears on a succeeding film shown 
in the same cassette. The outcome of storage was based on the principle of thermal 
induced luminescence of exposed materials, i.e. thermoluminescence.  There are histories 
on photoluminescence and thermoluminescence that can be traced back  (Mckeever, 
1985) and forwarded to medicine in current day applications e.g. radiation dosimetry, and  
archaeological dating. 
 
2.3 Megavoltage imaging in radiotherapy 
Megavoltage imaging was initially developed to make image guided radiotherapy (IGRT) 
possible, and made use of imaging practices on daily basis.  
 
15 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.3.1 Cone Beam Computed Tomography (CBCT) 
 Conventionally, CT scanners have round ring detectors, moving opposite an x-ray tube. 
Moreover, CT scans also have detectors imbedded circular ring of the flat panel. The 
geometry used by this type of CT scan is called cone-beam computed tomography 
(CBCT). Multiple planer projection images are obtained using cone-beam CT, as the 
source moves about the patients through 180 degree and more. Patient anatomical 
structures in three dimensions, sagittal, cross-sectional and coronal planes are 
reconstructed based on the data provided by the multidirectional images of the CBCT. 
Volumetric images are reconstructed using an algorithm called filtered back-projection 
(Murphy et al., 2007). CBCT systems are commercially available as accessories to linear 
accelerators. The CBCT is connected to the gantry of the accelerator with which 
volumetric image data are acquired under normal treatment conditions, thus enabling 
localization of intended target volume and critical structures before each treatment. 
Application of the system is done using kilovoltage x-ray source or megavoltage beams. 
 
2.3.2 Kilovoltage Cone Beam Computed Tomography (KVCBCT) 
Kilovoltage cone-beam CT (KVCBCT) systems are made by connecting a conventional 
X-ray tube to a retractable arm at 90 degrees to the therapy beam position.  Flat panel of 
X-ray detectors are attached to the X-ray tube, thus providing very good images which 
are suitable to be used in cone-beam CT and also in 2-D radiography and fluoroscopy. 
The advantages of a KVCBCT are its capacity to: 
16 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 Have quality contrast and submillimeter spatial resolution due to the volumetric 
CT images produce by the system. 
 Obtain images in therapy room conditions. 
 Also uses two dimensional methods of radiographic and fluoroscopic procedure to 
confirm the accuracy in portal imaging, patient movement, and ensuring position 
and dosimetric adjustment before and during treatments. 
 
2.3.3 Megavoltage Cone Beam Computed Tomography (MVCBCT) 
In megavoltage cone- beam CT (MVCBCT) system, X-ray beam and EPID of the linear 
accelerator are mounted in opposite direction to the source. EPIDs with Si flat panel 
detectors are highly sensitive to permit speedy acquisition of multiple, low dose images 
as the gantry is revolving through 180 degrees or more. After these, multidirectional two 
dimensional images, volumetric CT images are reconstructed (Gray et al., 2005). 
The MVCBCT produces quality images for bony anatomical structures and, in some 
cases, soft tissue targets as well. MVCBCT is an important instrument for on-line 
verification of patient, and identification of implanted metal markers if used for patient’s 
setup. 
Although KVCBCT has better image quality (resolution and contrast), MVCBCT has the 
following potential advantage over KVCBCT: 
 Less exposure to artifacts due to high-Z objects such as metallic markers in the 
target, metallic hip implants, and dental fillings. 
17 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 In MVCBCT, extrapolation of attenuation coefficients from kilovoltage to 
megavoltage energies is of no importance in term of dosimetric corrections. 
 2.3.4 Radio Frequency Beacons (RFB) 
In RF beacons system, small RF transponders can be implanted in the patient instead of 
x-ray fiducials. Location of these beacons can be determined electrically by the system in 
real time (occurring ten times in every seconds).  Real-time monitoring of location of 
target is allowed by the system during therapy, and it also gives the chance to stop the 
exposure if the target beacon moves beyond the predetermined acceptance. 
 
2.3.5 Magnetic Resonance Device (MRD) 
Few new equipment using MRI have been established for IGRT treatment. Scanning of 
patients are done in the treatment location, patients are then moved straight beneath the 
accelerator for treatment. Another device is also developed (view ray), that incorporate a 
rotating gantry system fixed with three cobalt -60 sources equally spread about the ring. 
Each of these rings has its own multi leaf collimator (MLC) system which is connected to 
a 0.3 T permanent magnet of MRI imaging system. Several  MR- Linac units have 
presently been established, each with the objective of delivering sophisticated photon 
treatment using 6 MV photons while at the same time executing 1.5 T MR imaging. 
 
 2.3.6 Video Surface Imaging (VSI) 
 This system uses video camera methods for projection onto patients skin so that the 3D 
shape of the patient can be determined in real time. Images acquired can be compared to 
18 
 
University of Ghana  http://ugspace.ug.edu.gh
   
the expected model of the patient surface as a way of identifying positional or motional 
differences between the current patient surface and the model. 
2.3.7 Electronic Portal Imaging Device (EPID) 
If the portal images are in digital form then they can also be processed to improve image 
quality. The devices that produce digital portal images are called Electronic Portal 
Imaging Devices (EPIDs). Due to the delays caused in film processing, took advantage of 
real time portal imaging for treatment. In the past years, great effort has been devoted to 
developing new equipment, i.e., electronic portal imaging devices. These include 
scanning linear arrays of diodes (Taborsky et al., 1981) and scintillating crystals (Morton 
et al., 1991), video-camera based systems(Munro et al., 1990) (Baily et al., 1980), matrix 
ion-chamber systems(Meertens et al., 1985), an Flat- panel arrays (Zhao & Rowlands, 
1992) (Antonuk et al., 1990) 
 
2.3.8 Video-Camera Based EPIDs 
Video-camera electronic portal imaging devices (VEPIDs) consist of a metal plate with 
the underside coated with a fluorescent phosphor screen to produce visible photons.   A 
450 mirror deflects some of the light onto a video camera. The major advantages of 
VEPIDs are that: 
 Radiations exiting the patient have the ability to produce a signal in the EPID, i.e. 
the detector views the entire area of the irradiated screen 
 The spatial resolution can be high, depending upon the thickness of the screen. 
 
19 
 
University of Ghana  http://ugspace.ug.edu.gh
   
The system can generate an image quickly. Images can also be acquired rapidly, at up to 
30 frames per second. However a major limitation to the design is that, only a very small 
fraction of light is captured by the lens (0.01-0.1%) and is focussed onto the 
camera(Munro et al., 1990). Because of this, the image quality can be reduced in two 
ways: 
 When none of the light produced by the interaction of the photon with X-ray 
detectors reaches the TV camera, then no measurable signal is produced. 
 If the noise produced by the preamplifier and the electronics of the system is 
much greater than the signal produced on the TV camera.  
 
However from figure 2.2, the development of this device has concentrated on improving 
the light collection efficiency by increasing the output of the phosphor, increasing the 
aperture of the lens and finally increasing the detection efficiency of the camera. 
Moreover, the thicker screen will increase the quality of images, but suffers from the loss 
of spatial resolution and blemishes. Also increasing the lens size to increase the light 
collection efficiency will decrease spatial resolution causing non- uniform brightness and 
image distortion. 
 
20 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
 
Figure 2. 2: Picture of a video-based EPID (From Boyer et al.1992) 
 
2.3.9   Matrix- Ion Chamber Array 
Figure 2.3 shows a type of EPID system that  was first developed by (Meertens et al., 
1985) and (Meertens et al., 1988) . The system employs ion transport in a liquid. The 
EPID consists of: (i) camera cassette and (ii) control unit. 
The camera cassette consists of a 256- channel electrometer system, control electronics 
and a 256-channel high voltage switch system. The detector consists of two sets of 256 
strip electrodes perpendicular to each other. The electrodes lie on the inner surface of two 
printed circuit boards, parted by a 0.8 mm gap filled with liquid film (Iso-octane, 
spectroscopically pure Merck), which serve as the ionization medium. A set of the 
electrodes is linked to a high-voltage supply, and the next is connected to a sensitive 
electrometer. A small ionisation chamber is positioned at the cross-point of the 
electrodes. There is a 1mm steel plate in front of the upper board which acts as the main 
radiation build-up material. The liquid is ionised when it is irradiated. In other to acquire 
an image, the ionization matrix should be scanned row by row, by successively switching 
the high voltage (HV) to different voltage electrodes and measuring the currents in all the 
21 
 
University of Ghana  http://ugspace.ug.edu.gh
   
256 column electrodes. Since the analog to digital A/D conversion time is ~ 4µs, a single 
row can be read-out in 1ms. The outputs of the electrometers are multiplexed to a single 
amplifier and digitized in the control unit. 
When the HV is switched to a row, a transient pulse is induced in the electrometers. Thus 
each row cycle consists of a wait-time for this pulse followed by readout of the 256 
electrometers. The HV is the latest faster model of this detector which is 500V .The HV 
cycle time is based on two cycles of the accelerator 360    timing pules is 5.6 ms, 
comprising of a waiting time of 3ms, and the remainder is signal readout time. The 
current in each ionization chamber is measured ten times, and the values are averaged. 
The total frame acquisition time is therefore ~1.4 s (which is a result of multiplication of 
256 electrometers by 5.6 ms pulse time). For the older models the total frame acquisition 
time is ~3s (Boyer et al., 1992). The signal readout is synchronised with the pulse timing 
of the accelerator to obtain smooth images without artifacts due to radiation pulse 
variation. Several frames can be averaged depending on the time available, and the dose 
rate delivered. Based on the long read out time of the device, it is only possible to 
measure the dose rate instead of measuring the integrated dose directly. The relation 
between output pixel values PV ,that is  ionization current, and dose rate is given by 
(Essers et al., 1995) as follows: 
  PV = G (D) =a* D1/2 + b* D, ----------------2.1 
Where a and b are parameters with units (min/cGy) 1/2 and min/cGy, respectively. G is 
the gray level and is a function of dose rate. 
22 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
                     Figure 2. 3: Matrix-ion chamber array EPID (From Van Herk, 1987) 
                             
The electronic portal imaging device (EPID), has several possible advantages over 
conventional X-ray film for portal imaging.  Images acquired using EPID are available 
instantly and can be easily used for interactive adjustment of patient field positioning 
throughout radiotherapy. The images obtained are digital, which helps image processing, 
contrast enhancement and image matching. Moreover, because of the digital archiving 
nature of the EPID, a lot of space is available which permits recollection of images over a 
network system. The disadvantage of the EPID system is that they have huge or bulky 
devices that are of limited practicality with disappointing image quality. There is a 
tendency of changes with the introduction of more modern technology, such as 
23 
 
University of Ghana  http://ugspace.ug.edu.gh
   
amorphous silicon-based devices, but there still remain issues with image quality 
.(Langmack, 2001)  
Lastly we would also made mention of the computed radiography (CR) as an imaging 
detector that is used in the therapy department for patient verification. The CR system is 
designed to acquire and digitize X-ray images from image storage phosphor plates. The 
CR system consists of several components such as: an image reader/digitizer, cassettes 
containing imaging receptors (photostimulable-phosphor plates), a computer console, 
software, monitors, and a printer. The system has two plate readers. The first one is a 
single-plate readers (each cassette is loaded manually and read separately) and the next is 
the multiple-plate readers (multiple plate –up to -10- can be stacked and loaded 
automatically). In CR systems, images are obtained using X-rays in which imaging plates 
are inserted in a radiographic table cassette holder. When the imaging plate is exposed to 
an X-rays, electrons in the phosphor plate are excited to higher energy state, thus forming 
a latent image. The laser spot found within the reader scan the phosphor. Energy is 
absorbed by the trapped electrons and light is emitted as they go back to their original 
state. The light guide collects the light and transmits it to a photomultiplier tube, which 
produces an analog electrical signal that is amplified, and converted to digital signal, and 
stored digitally. After use the plate can be reused after it has been exposed to an erasing 
light that removes residual radiation. Images captured on the imaging plate by the X-ray 
source are extracted by the image reader. In the reader the cassette is slotted manually or 
automatically. Digital images acquired are downloaded unto the image-processing 
system, usually a computer workstation, for display and manipulation of images within 
30-120 seconds. The plate is erased for reuse. Mostly damages to imaging Plates are due 
24 
 
University of Ghana  http://ugspace.ug.edu.gh
   
to reckless handling and are expensive to replace. Modern CR readers have brushes or 
fans that instantly clear dust off plates, which helps to prevent scratches. The 
performance of the imaging plates can be deteriorated by cracks or scratches which thus 
reduce the quality of the plate for imaging. The light guide should always be cleaned as 
part of routine maintenance (quality of images are affected by dirt accumulation). 
 
2.3.10 Scanning Linear Arrays (Solid state system). 
This system uses a linear array of 255 diodes with a centre-to-center spacing of 2 mm that 
scans the image in 2 mm increments using a stepping motor. A 1.1 mm thick lead plate 
covering the diode array acts as build up layer. As the array only covers a small portion 
of the field the doses required to form an image are large. Spatial resolution is lower, due 
to large diode spacing than a finer detail system such as fluoroscopy camera devices. 
(Morton et al., 1991) produced a linear array of scintillating crystals (ZnWO4), each 
5x5x25 mm in size. The crystals are arranged in a double row, 64 crystals per row with 
each row offset by half the crystal width to reduce the sampling interval. The X-rays 
interact and create high-energy electrons that pass through the volume of these crystals 
resulting in the creation of visible photons that are detected by photodiodes. As the 
crystals are optically transparent they can be long (25 mm) and hence the detection 
efficiency for the X-ray photons is very high, ~ 50% for 6 MV. The high quantum 
efficiency of the system allows the image to be produced with high subject contrast. 
However the spatial resolution is low. 
 
25 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.3.11 Flat-Panel Device 
Flat-panel based X-ray imaging is a new technology used to meaningfully increase the 
quality of on-line portal imagers. There are two main types of flat-panel solid state 
imaging device being developed for megavoltage imaging. These are amorphous silicon 
photodiode array (Street et al., 1992), and amorphous selenium photoconductor arrays 
(Zhao & Rowlands, 1992).  
 
2.4 Radiographic Films 
Figure 2.4 shows a cross section of a radiographic film system. During the early 1960s, at 
the height of the nuclear weapons testing, X-ray film manufacturers took extraordinary 
precautions to ensure that contamination from radioactive fallout did not invade their 
manufacturing environment, which could seriously fog the film. A radiographic film has 
two parts: (figure 2.4) the base and the emulsion 
 
      Figure 2. 4: Cross section of radiographic film 
 
26 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 Emulsion coated on both sides in an X-ray film is called double emulsion film. The 
adhesive layer is made of a thin coating material found between the emulsion and the 
base, and its function is to provide uniform adhesion between the base and the emulsion. 
The adhesive layer maintains proper contact between the emulsion and the base during 
use. The emulsion is protected by a protecting covering of galatin called overcoat. The 
overcoat protects the emulsion from pressure, scratches, and contamination during 
handling, processing, and storage. 
 
2.4.1 The Base 
The foundation of a radiographic film is the base. The function of the base is to provide a 
hard structure onto which the emulsion can be coated. The radiographic film can easily 
be handled due to the flexibility and fracture resistant nature of the base. The base of a 
radiographic film keeps the size and shape of the film during processing and use thus 
preventing image distortion. Originally, radiographic film base was made of a glass plate 
which was referred to as X-ray plate.  High-quality glass in the World War 1 became 
extremely unavailable while medical application of X-rays, by the military, was 
increasing every day. The next substitute material used at that time was cellulose nitrate, 
which became the standard base. But this cellulose nitrate had one major deficiency: it 
was highly flammable. During the 1920s and early 1930s, due to improper and handling 
of some X-ray film files resulted in severe hospital fires. In the-1920s, film with a safety 
base cellulose triacetate, was introduced. Cellulose triacetate has a property similar to that 
of cellulose nitrate but not as flammable as cellulose nitrate. In the early 1960s, a 
27 
 
University of Ghana  http://ugspace.ug.edu.gh
   
polyester base was introduced. Polyester is more resistant and stronger than cellulose 
triacetate, allowing easier transport through automatic processors. Polyester base are 
more used as film base. 
 
2.4.2 Emulsion 
The heart of the radiographic film is the emulsion.  The emulsion is made of material 
with which the light photon, X-ray of the radiographic intensifying screen interacts. The 
emulsion consist of a homogeneous mixture of galatin and silver halide crystals. The 
galatin is clear, so it transmits lights, and it is highly porous for processing chemicals that 
penetrate the crystals of silver halide. The main function of the galatin is to give 
mechanical support by holding silver halide crystals uniformly together in place.  
The most active component of the radiographic emulsion is the silver halide crystal. 
Ninety eight percent of halides in the emulsion is silver bromide and two percent is silver 
iodide. The atomic numbers of these halides in the emulsion are relatively higher (ZBr = 
35, ZAg=47, ZI=53) compared to that of galatin and base (for both, Z~7). The formation 
of latent image on the radiographs rapidly increases when materials with high-Z atoms 
interact with X-ray and light photons. The silver halide crystals of the emulsion have 
different shapes, of which some are tabular, cubic, octahedral, polyhedral, or irregular in 
shape. The crystals are formed by dissolving metallic silver (Ag) in nitric acid (HNO3). 
Light-sensitive silver bromide (AgBr) crystals are made by mixing silver nitrate with 
potassium bromide (KBr). The reaction is as fellow: 
AgNO3 +KBr   AgBr + KNO3. ----------------2.2 
28 
 
University of Ghana  http://ugspace.ug.edu.gh
   
From equation (2.2) silver bromide is precipitated while the potassium nitrate, which is 
soluble, is washed away. Radiographic film is manufactured in total darkness. From the 
moment the emulsion ingredients are brought together until final packaging, no light is 
present. 
This radiographic film is still used for verification of patients in radiotherapy department. 
Nevertheless, their wide use is declining every day due to the facts of fixed dose latitude, 
fixed non- linear grey scale response, and limited potential for reducing dose to patient. 
The image contrast of these films remains unchanged once they have been processed. In 
spite of this, films are expensive, labour intensive, uses dangerous chemicals for 
processing, and finally the retrieval of films after been kept long is difficult. The 
compatibility of picture archiving and communication systems (PACS) in radiographic 
film system is poor. 
 
2.5 Image Quality Assessment with Image J software. 
 
Image quality assessment remains challenging in the field of image processing.  
According to the International Atomic Energy Agency (IAEA: TRS-457, 2007), 
optimization processes which involve balancing radiation dose and image quality do not 
always lead to reduction in radiation dose and it is essential to emphasize that image 
quality must always be sufficient to meet clinical requirements. Assessment of image 
quality plays a major role in the various imaging processing applications. It is meant to 
quantify a visual quality or anatomical amount of distortion or degradation in a given 
image. These degradations (artifacts) are inevitably part of any digital image processing 
29 
 
University of Ghana  http://ugspace.ug.edu.gh
   
chain from acquisition, processing and transmission of images (Kud, 2012). Efforts have 
been made which seek to develop various image quality matrices that correlate very well 
with the perceived quality measurement, but only limited success has been 
achieved.(Zhou & Bovik, 2002)  Assessment of image quality subjectively or by utilizing 
physically-based matrices is still in practice in modern radiation therapy. 
Notwithstanding that, a method for objective task-based image quality (IQ) assessment in 
radiation therapy was proposed by (Barrett & Kupinski, 2013). However details 
regarding its implementation remain to be investigated and the methodology has yet to 
find widespread application in the field of radiation oncology.  This study shows a 
comprehensive implementation and evaluation of the matrices (SNR, CNR and RMTF) 
image quality assessment methodology. According to (Mohammadi & Ebrahimi-
Moghadam, 2014) suggested that the need for evaluating the reliability has increased due 
to the increasing demand for image-based applications. According to these authors, this is 
based on the fact that assessing image quality is of importance for several image 
processing applications where the aim of image quality assessment methods is to 
instantly evaluate the quality of images in accordance with human judgements. In the 
field of radiation therapy, oncologist depends greatly on imaging modalities for quality 
image. From initial consultation to treatment completion, images are used to direct nearly 
every step of the patient encounter. Image guidance radiation therapy (IGRT) has become 
a standard of care for many treatment situations in radiation therapy (Thapa & Molloy, 
2015). During simulation in IGRT, the CT scan obtained provides virtually full details 
regarding patient-specifics, localized subject contrast. This information helps to establish 
an imaging goal and to select patient-specific imaging acquisition parameters that can 
30 
 
University of Ghana  http://ugspace.ug.edu.gh
   
improve the similarity between reference and daily set-up images.(Thapa & Molloy, 
2015).  Generation of digitally reconstructed radiographs (DRRs) is one of the vital steps 
in treatment planning processes. These radiographs are used as reference images and 
processing tools which give full anatomical detail of the patients (Thapa & Molloy, 
2015). The quality of a medical image is determined by the parameters such as: the 
characteristics of the equipment, and the imaging methods selected by the operator. The 
assessment of image quality is a composite of at least five factors: contrast, blur, noise, 
artifacts, and distortion(Sprawls, 1996). However, quality image is important for patient 
treatment in radiation therapy. Generally image quality can be evaluated mostly by using 
two approaches: objective and subjective methods (Kud, 2012). Subjective image quality 
assessment is by visual inspection of 2-D images by human observers. Objective image 
quality assessment uses mathematical models to calculate the quality of an image 
accurately. 
 Image J software is an imaging analysis programme created by the National Institute of 
Health, UAS, that performs image quality assessment objectively. The programme 
calculates the different parameters (area, standard deviation etc.) depending on the set 
measurements needed. Signal-to-noise ratio, (SNR) Contrast to noise ratio,(CNR)  and 
Relative Modulation Transfer Function (RMTF) can be calculated using image J 
software, by selecting one or more specific regions of interest (ROI) in the image.  
 
31 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.5.1 Calculation of Signal-to Noise ratio (SNR) 
The signal to noise ratio (SNR) expresses the ratio between signal and noise in a large 
scale object. Using image J software, region of interest (ROI) is drawn on each of the step 
to determine their mean, standard deviation and other parameters. A study was carried 
out by (Roehrig, 1990) to determine the SNR of the CR and conventional film. The SNR 
was determined by taking the difference in the mean gray value of the objects and its 
background by the quadratic sum of the standard deviation of the background as a 
measure of noise.  
SNR=       =            (   )                 (   ) ---------2.3 
     √                         
Where (std object) is the standard deviation of the object, (std background) is the 
standard deviation of the background, (mean object) is the mean gray value of the object 
within the region of interest and (mean background) is the mean gray value of the 
background with the ROI. According to the author, the result obtained from this study 
shows that SNR for CR system was meaningfully more effective than conventional films. 
Irradiation at different dose levels i.e. (9, 6, 3 and 1 rad) were done on both detectors. The 
SNR value for film at dose of 9 rad used was 0.766. Most of the SNR for the computed 
radiography system were significantly higher. At a dose of 9 rads the value was 1.13, at a 
dose 6 rad the value was1.11and at a dose of 3 rad the value was 1.03. Only at a dose of 1 
rad was the SNR smaller than that of the film, which shows value of 0.732. 
 
32 
 
University of Ghana  http://ugspace.ug.edu.gh
   
University of Ghana  http://ugspace.ug.edu.gh
   
University of Ghana  http://ugspace.ug.edu.gh
   
 
Figure 2.9: Portal image QC phantom obtained with BEAMVIEW Portal imaging 
system. The ROIs used for QC test marked on the image 
 
The CNR for the QC phantom was calculated by using the formula 
 
CNR =     -------------------------2.5   
 
 Where    and    are the average gray values of the brightest and darkest ROI and   is the 
random image noise. In their study, the authors determined CNR by comparing the 
energy of linac 6 MV and 23 MV.  
35 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
Figure 2.10: Plot of CNR for the BEAMVIEW portal image system 
 
Figure 2.10 shows a plot of CNR values recorded on a daily basis for BEAMVIEWplus 
portal imaging system. During the 25 day calibration period, the mean values of CNR 
were 15.9   1.2 and 26.4   3.2 for 6 and 23 MV beams respectively. The dotted and 
dashed lines represent   3 standard deviations determined during the calibration period 
of 25 days. According to the author it shows that the CNR for 23 MV is higher than that 
of 6 MV, which is due to the higher dose rate (300 cGy/min) at 23 MV compared to that 
of 6 MV (200 cGy/min). 
 
2.5.3 Calculation of Relative Modulation Transfer Function (RMTF) 
A study was carried out by (Rajapakshe et al., 1996) to determine the relative modulation 
transfer function. The square wave modulation transfer function (SWMTF) of the 
36 
 
University of Ghana  http://ugspace.ug.edu.gh
   
imaging system was defined as shown in equation 2.6, since the modulation of the bar 
pattern is in a form of a wave. 
SWMTF =  ( )----------------------------2.6 
   
Where           ( ) are the modulation of input to and output of the system 
According to this work they were not interested in an absolute measurement of the 
SWMTF, but only in the day-to-day variation in the system resolution, which use a 
relative measure (RMTF) of the SWMTF by calculating 
RMTF (f) =   ( ) ----------------------------2.7 
  (  )
Where   (  ) is the output modulation for the lowest frequency. Based on this study the 
output modulation is usually difficult to be obtain from noisy image, therefore,(Droege & 
Morin, 2011)  suggested using the relationship between a signal amplitude and its 
variance (M)2 within an ROI containing the bar pattern, and the above relation can be 
rewritten as: 
RMTF (f) =  ( ) ----------------------------2.8 
 (  )
In the presence of random image noise, M (f) can be obtained by 
M2 (f) =     (f) -    (f) ---------------------2.9 
Where =     (f) and   (f) are the measured total variance and the variance due to random 
noise respectively.  The total variance     ( ) is obtained by measuring the pixels of the 
37 
 
University of Ghana  http://ugspace.ug.edu.gh
   
ROI corresponding to frequency f. In order to measure the random noise in an image 
according to the author a pair of similar images are subtracted, and the standard deviation 
is obtained from the difference, thus avoiding contributions from fixed pattern noise. The 
variance of the subtracted ROI (      ) will be 
       =      +   ------------------------------2.10 
Where     and     are the random noise variances of the ROIs for each image. The author 
therefore assume that these variances are equal and hence 
   =   =      / √   ---------------------------2.11 
 This work was comparing the resolution of a linac system with dual energies 6 MV and 
23 MV. From the figure 2.11 below, the dotted and dashed lines represents the   3 
standard deviations determined during the calibration period and extrapolated to the test 
period for 6 and 23 MV respectively. Based on the author the system resolution of 6 MV 
is superior to that at 23 MV, this result is due to the larger physical beam penumbra at 
higher energies  (Munro et al., 1990) and increase transmission through the bar patterns 
by the higher energy photons. 
38 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
Figure 2.11: Plot of f50 recorded on daily basis for the BEAMVIEW portal imaging 
system at 6 and 23 MV 
 
2.6. Quality Assurance of External Beam Radiotherapy  
2.6.1 Quality Assurance 
In radiotherapy, quality assurance is all measures that ensure reliability of medical 
prescription, and safe completion of that prescription, as regards the dose to the target 
volume, together with negligible dose to normal tissue, minimal exposure of personal and 
adequate patient monitoring aimed at determining the end result of the treatment. It must 
be stressed again that quality assurance in radiotherapy is connected to all aspects of 
radiotherapy process and all groups of staff should be involved in proceedings, since 
quality activities are interdependent. 
 
39 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.6.2 Quality Control 
Quality control is a tool used to regulate a process through which quality performance is 
measured with existing standards. Actions necessary to maintain the performance of 
quality are kept within the speculated standards. Quality control is one part of overall 
quality assurance, and is concerned with operational techniques and activities used: 
 To ensure that quality requirements are in place. 
 
 To adjust and correct performance if the requirements are not to standard. 
 
 
2.6.3. Need for quality assurance in radiotherapy 
 High accuracy assessment in clinical setting in therapy is important in yielding a desired 
result of tumour control rates that is consistent in keeping complication rates within 
standard levels. Quality assurance measures in radiotherapy can be characterized as 
follows: 
 Uncertainties and errors in dosimetry, treatment planning, equipment 
performance, and treatment delivery are minimized when quality assurance are in 
place and enforced. Quality assurance thus helps to improve dosimetric, 
geometric accuracy and the precision of dose delivery. With this, radiotherapy 
results will be improved (treatment outcomes), thus increasing tumour control 
rates and reducing complication and recurrence rates. 
 Quality assurance as a tool does not only reduce the likelihood of errors and 
accident occurring, but it also provides the opportunity for recognizing and 
rectifying problems that tend to occur during patient treatment. It is not only for 
40 
 
University of Ghana  http://ugspace.ug.edu.gh
   
incidents that are big but also for those that have higher possibility of minor 
incidents. 
 Quality assurance also provides the opportunity of comparing true results 
obtained at different radiotherapy centres, thus ensuring quality and accurate 
dosimetry and treatment delivery.  
 In modern radiotherapy technology, complex treatments can only be achieved 
when there is a high level of consistency and accuracy in the treatment. 
The objective of radiotherapy is to ensure that patient safety regarding exposure to 
normal tissue is kept as low as reasonably achievable (ALARA) in accordance with dose 
delivered to the planning target volume (PTV). The aim of providing quality in 
radiotherapy treatment is to ensure patient safety and evasion of accidental exposures. 
The safety of patient is therefore key in quality assurance of radiotherapy treatments. 
 
2.6.4 Verification of radiation field shape in radiotherapy 
Detection of radiation field edges is required for two purposes: (i). to verify the shape of 
the radiation beam, (ii). to establish a coordinate system common to both reference and 
portal image in which to account for deviation in patient setup. Several image processing 
methods have been reported which extract the edges of radiation field from portal images, 
e. g (Bijhold et al., 1991b).(Eilertsen et al., 1994). (Petrascu et al., 2000). With the 
advance of MLC technology, methods are being developed to accurately detect the leaf 
positions, e.g. (Zhou & Verhey, 1994)  (Eilertsen, 1997)    Although MLC systems 
generally have their own feed-back control systems, there is an increasing interest in 
41 
 
University of Ghana  http://ugspace.ug.edu.gh
   
independent online verification of proper leaf positioning using portal imaging, 
e.g.(James et al., 2000). Other methods for verification of beam shaping elements are 
based on geometric moments (Bijhold et al., 1992) chamfer matching (Gilhuijs & van 
Herk, 1993). The increasing use of small beam intensity modulated fields (e.g. by use of 
MLC technology) provides an interesting challenge for the verification of patient setup 
from portal images. The modulation process produces portal images with limited feed-
back mechanism for gated radiation therapy in order to cope effectively with organ 
motion during treatment. The largest challenge, however, is to narrow the existing gap 
between image processing research and the daily clinical routine. 
 
2.6.5 Verification of patient setup 
Inaccuracies in any steps, tolerances in mechanical components, and anatomical changes 
in the patient, movement of internal organs, incorrect immobilization, and human 
mistakes can lead to  differences between the prescribed volumetric dose  and the dose 
that is delivered over a period(Gilhuij et al., 1996),(Bansal et al., 1999). Instances of 
curve features are edges or ridges in the projection of bony anatomy or generalized gray 
value media axes. Template features are usually rectangular regions of pixel values. 
42 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Figure 2.12: Picture illustrating systematic and random error in field placement 
 
The methods described figure 2.12 above are based on image arrangement in two 
dimensions (2D). Since the position and orientation of a patients are three-dimensional 
concepts, the comparison of portal with reference images in 2-D sometimes yields 
insufficient information to establish a complete correction of the setup in 3-D. For 
instance, patients rotating along the axes that are not perpendicular to the imaging plane 
cannot be quantified. Moreover, analysis in 2-D only provides accurate results of patient 
translations when the errors in rotational are small. For example out-of-plane rotations of 
the pelvis larger than 2 degrees in prostrate treatment may cause dosimetrically 
significant misinterpretation of the position of the isocenter inside the patient(Hanley et 
al., 1995),(Bijhold et al., 1991a) (Remeijer et al., 2000). Thirdly, geometrical degeneracy 
in the alignment of rotation symmetrical objects like the femoral heads may cause 
ambiguous alignment in 2-D from a single view. 
43 
 
University of Ghana  http://ugspace.ug.edu.gh
   
2.7 Accidents in radiotherapy 
Patient with disease treated using therapy are faced with twofold of risk: 
 Firstly, there is a possibility of failure to control the original disease, which, when 
it becomes malignant, is finally lethal to the patient 
 There is also risk from increased radiation exposure to normal tissue. 
 
 In radiotherapy wrong administration or an accident is serious if results are due to 
overdose or an underdose. Overdoses are of general concern in conventional radiation 
protection legislation and protocols. There is an accident or wrong administration when 
the difference between prescribed and delivered dose are not in line as proposed. From 
the general aim of an accuracy approaching 5% (95% CL), about twice this seems to be 
an accepted limit for the definition of an accidental exposure (i.e. a 10% difference). 
Moreover dose applied accidentally outside the planned target volume may lead to bigger 
complications. In 2000 the IAEA investigated series of accidental exposures in 
radiotherapy, (Thwaites et al., 2003) and tended to draw lessons for the prevention of 
such occurrences. Criteria for classifying radiological accidents are: 
 Direct causes of wrong administration 
 Classification of possible dangers 
 Preventability of misadministration. 
 
 
 
44 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 Table 2.1 below shows some examples of direct causes of misadministration in external 
beam radiotherapy cataloged and analyzed in the IAEA report. 
Table 2. 1: Causes of misadministration in external beam therapy in the 2000 IAEA 
Report (Thwaites et al., 2003). 
Cause Number of accidents 
Calculation error of exposure time or dose 15 
Inadequate review of patient chart 9 
Error in anatomical areas to be treated 8 
Error in identifying the correct patient 4 
Error involving lack of or misuse of a wedge 4 
Error in calibration of Co-60 source 3 
Transcription error of prescribed dose 3 
Decommissioning of teletherapy source error 2 
Human error during simulation 2 
Error in commissioning of TPS 2 
Technologist misread the treatment time or MU 2 
Malfunction of accelerator 1 
Treatment unit mechanical failure 1 
Accelerator control software error 1 
Wrong repair followed by human error 1 
 
 
45 
 
University of Ghana  http://ugspace.ug.edu.gh
   
At every stage of treatment, errors can be done by every staff member involved in the 
process. Most accidental exposures in therapy are due to the failure in the application of 
adequate quality assurance programme. Mostly errors caused by human beings are 
leading to lack of knowledge, complacency and overconfidence. 
 Errors relating to human beings will always occur in any organization and in any 
activity. Therefore one of the aims of applied quality assurance programme is to reduce 
the amount of incidence and to classify them as early as possible in other to reduce the 
consequences. 
 
 
 
 
 
 
 
 
 
 
 
46 
 
University of Ghana  http://ugspace.ug.edu.gh
   
CHAPTER THREE 
MATERIALS AND METHOD 
 
This chapter outlines in details the experimental method and materials employed in this 
study. 
 3.1. Study Area 
This research was conducted in two major departments of the Korle-Bu Teaching 
Hospital in Ghana. These units were the radiotherapy and radiology departments of the 
hospital. This hospital was chosen because of the availability of CR system and the only 
teaching hospital using Cobalt machine for cancer treatment within Accra. 
 
3.2. Theratron Equinox 100 Cobalt 60 teletherapy Machine. 
Portal image quality assessments were performed for an Equinox 100 cobalt 60 
teletherapy machine (Best Theratronics, Canada). The telecobalt unit was manufactured 
in April, 2013. The treatment head of the telecobalt machine is mounted isocentrically 
with source axial distance of 100 cm. In the treatment head is contained a double 
encapsulated cobalt 60 source having initial total source activity of 399.0 TBq measured 
on August 1st , 2013, by source manufacturer, Nordion Inc., Canada. This gives the 
teletherapy machine a reference beam output in water at a maximum depth dose (0.5 cm) 
of 189.49 cGy/min on December 12, 2013), measured after installation of the telecobalt 
47 
 
University of Ghana  http://ugspace.ug.edu.gh
   
machine, based on International Atomic Energy Agency (IAEA) technical series report 
(TRS) 398 protocol.(Andreo et al., 2006). Percentage depth dose for the reference field 
size of 10 cm x 10 cm for a depth of 10 cm in water (PDD10), which was used as a beam 
quality specifier for megavoltage beam (Tagoe et al., 2017), is 58.36 % for the telecobalt 
machine. High activity encapsulated cobalt 60 source within the treatment head of the 
teletherapy machine has a diameter 2 cm and length of 4 cm, and is classified as C-146 
teletherapy source capsules by Canadian Nuclear Safety Commission (Effects, 2016). 
The cobalt 60 source was formed by exposing stable cobalt 59 with neutrons in a reactor. 
The cobalt 60 source decays to nickel 60 with the emission of beta particles (Emax = 0.32 
MeV) and two photons energies 1.17 MeV and 1.33 MeV, given the cobalt 60 source an 
average energy of 1.25 MeV. The source is embedded in a source drawer mechanism 
which uses a pneumatic system to bring the source in and out of treatment position. 
Within the treatment head are asymmetrical collimators that allow the jaws, which define 
the shape of beam to move independently of each other, providing more freedom in 
treatment planning. Attached to the collimator system is an accessory holder with block 
tray code interlock to prevent the use of wrong accessory for treatment. The distance of 
radiation source to the accessory holder (or block tray) is 59.3 cm. The system has two 
timers: one is activated to sense the source when fully exposed and the other is activated 
to sense when the source when returns to its safe position. A door interlock device is also 
linked to the machine in a way that:  
 It is possible to return the beam when the door is open. 
 
 Also if the door is open when the beam is on, the source will automatically returns 
to its safe position 
48 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
When the unit is not in use, the source will be in the off position thus preventing radiation 
dose from exiting the treatment head. The source is slid or rotated into the treatment 
position when dose delivery is required. Dose delivered, is measured by treatment time as 
opposed to monitor unit in a linac machine. The source exposure system is designed in a 
way that under suitable faulty circumstance, such as mains power failure or any control 
system failure, the source exposure mechanism will return the source to its safe state.  
According to the International Electrotechnical Commission (IEC) standard a manual 
beam control system is available that directly operates on the source exposure device that 
is used in an emergency to return system to its beam-off position. 
The telecobalt machine is configured to have features of a modern medical linear 
accelerator with the exception of a multileaf collimator system and an electronic portal 
imager. The field size that can be set on the machine ranges from 1x1 to 43 x 43 cm2 
(defines at the machine isocenter). The machine features a new motorized wedge (MW) 
system, which allows one to treat with any wedge angle ranging from 0 to 60 degrees. 
This is made possible with a fixed 60 degrees physical wedge permanently positioned in 
the treatment head of the machine, which can be brought automatically in and out of the 
path of the radiation beam during treatment delivery, such that combinations of time 
weighted beams with and without the wedge filter create the dosimetric effects of the 
required wedge. A picture of the telecobalt unit is shown in Figure 3.1 
49 
 
University of Ghana  http://ugspace.ug.edu.gh
   
GANTRY 
COLLIMATOR 
GANTRY ANGLE 
TREATMENT 
 
Figure 3. 1: Theraton equinox-100 cobalt-60 unit at Korle-Bu Teaching Hospital 
 
3.3. Computed Radiography Cassette  
 The CR cassette used for this study was available at the Korle-Bu Teaching Hospital in 
the radiology department. The model of the CR cassette was Fuji film with dimension 
35.4 x 43.0 cm, serial number (CCWSL 17-189081) and was clinically accepted for use 
in 2012.  The CR cassette used has rectangular plastic container which holds the imaging 
plate (IP). The function of the cassette is to: 
 hold the IP and protect it from damage 
 
 Exclude dusts and dirt from the sensitive imaging plate. 
 
The plate is coated with europium-activated Barium Fluoro-Halide (BaFX: Eu2+). The 
data are kept in the PSP imaging plate as electrons, in a semi-stable higher state. Image 
50 
 
University of Ghana  http://ugspace.ug.edu.gh
   
details are obtained by scanning the plate with laser beam. The CR cassettes are 
extremely sensitive to background radiation and scatter. If the CR cassette is out of use 
for days, it should be inserted into the reader for re-erasure to prevent fogging on the 
image. The CR cassettes get deteriorated with usage over a period of two years especially 
when not properly handled. This effect of damage on the cassette for not closing tightly 
on the imaging plate (IP) will lead to poor quality image. Figure 3.2 shows samples of 
different dimensions of the IP cassette used in CR systems. 
 
Figure 3. 2: Fujifilm IP Cassette type CC 
 
3.4. Computed Radiography Reader 
Figure 3.3 shows the CR reader that was employed for this study and was found at the 
radiology department of the Korle-Bu Teaching Hospital. The model is CR-IR 359, with 
serial No. 26126981. The reader was manufactured by Shimadzu Corporation, Japan in 
January 2012, and installed for clinical use in 2012. Averagely, this CR reader serves 
fifty (50) patients per day.  It has two insertion openings for cassette of different sizes. 
The reader is used to extract the image that has been stored on the imaging plate. The 
cassettes were inserted manually into the reader. The reader scans the latent image that 
51 
 
University of Ghana  http://ugspace.ug.edu.gh
   
has been stored on the phosphor plate with a laser spot. Energy is absorbed by the 
electrons upon interaction with the laser spot and lights are emitted as they go back to 
their original level. The light guide collects the light and transfer it to the photomultiplier 
tube, which produces an analog electrical signal that is amplified, and converted to a 
digital signal which is kept digitally. After 45 seconds digital image were acquired and 
downloaded to a computer workstation for display and manipulation. The images 
obtained were then printed on a film for examination. The plate was erased for reuse after 
exposure to white light. 
 
                             Figure 3. 3: CR reader at Korle-Bu Teaching Hospital 
 
3.5. Radiotherapy Film Cassette. 
Figure 3.4 show the radiographic film cassette used by the radiotherapy department of the 
Hospital for setup verification.  The film cassette specification used for this study was 
52 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Curix Universal AGFA-GEVAERT of size 34 × 43 cm in dimension. The film cassette is 
a rectangular rigid holder that contains the film. The top cover facing the beam entrance 
is made of material with low atomic number such as plastic which helps to minimize the 
attenuation of the X-ray beam. The back cover is made of metal to minimize back scatter. 
Its functions are:  
 Hold the film and protect it from damaged. 
 Exclude all light from entering the cassette and thus fogging the film. 
 Maintain close and uniform contact of the film. 
 Prevent the film from accumulating dust and dirt 
 
The features of the film cassette used are as follow: 
 Lightweight to facilitate easy handling and carrying. 
 Strong and rigid to withstand physical damage from daily wear and tear. 
 The internal surface have adequate layer of lead foil attached to reduce the risk of 
back scatter 
 
The cassette containing the film is irradiated with an X-ray beam. The film is removed 
from the cassette and loaded manually into the automatic processing system at the 
hospital. A lot of steps are involved in the processing. This include wetting, development, 
fixing, washing and drying. From this processes, the film is then ready to be used. 
 
 
53 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
Figure 3. 4: Picture of the radiographic film cassette 
 
3.6. Scanner (Microtek Scan Maker 9800XL) 
There are many scanners with high quality, which are used to scan coloured films. Figure 
3.5 shows Microtek Scan Maker 9800XL which was used to scan the films of various 
sizes. It is flat bed with a large format A3 scanner with super and versatile function 
specifically designed for proper image quality. It weighs 12 kg with dimensions 62.7 × 
37.6 × 13 cm (W× D× H). 
With the worm-up free energy saving Light-Emitting Diode (LED) source, the scanner 
echoes to go green. The scanner has an optical dynamic range of 3.7 Dmax which deliver 
accurate colour, superb shadows and sharp images. It has a smart-touch button (scan) 
located at the front panel of the scanner which provides a quick and easy way to capture 
image that can automatically save as files or sent to another application. The tabloid size 
scan bed of this scanner allow oversize scanning of original contact sheets, large pieces 
of art, mechanical blue prints and X-rays.  
54 
 
University of Ghana  http://ugspace.ug.edu.gh
   
  With a true optical resolution of 1600dpi scanning, the Scan Maker 9800XL can capture 
a greater amount of details from the format A3 scanner areas, allowing scan to be printed 
with high image quality. 
 
Figure 3. 5: Scanner (Microtek scan maker 9800XL) 
 
3.7. Image J Software 
The Image J is a Java-based image processing program created at the National Institute of 
Health (NIH). This software was designed with an open architecture that provides 
extensibility via java plugins and records macros. User written plugins make it possible to 
solve many image processing and analysis problems. Figure 3.6 shown below shows an 
open view of the software interface that was used for the study. 
 
55 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
    Figure 3. 6: Open view of image J software interface 
                    
3.8. Film Processor 
The film processor that was used for this study is shown in figure 3.7, which is Kodak 
Medical X-ray Processor 102, with serial No. 117510-0903-7107 and was made in 
Germany for CARESTREAM HEALTH .INC. Rochester. NY 14608. This processor was 
installed for clinical use in 2008 at the Korle-Bu Teaching Hospital. The amount of film 
processed by the system is approximately 35 per day depending on the cases available for 
treatment. The processor is placed in a dark room at the hospital. Image acquisition 
involves several steps: The film is manually removed from the cassette and inserted in the 
processor. In the processor the processing sequence involves the wetting of the film to 
swell the emulsion, so that subsequent chemical baths can reach all the part of the 
emulsion uniformly. The developing stage is very short and highly critical. The film is 
rinsed in an acid solution designed to stop the developing process and remove excess 
developer chemicals from the emulsion. The next stage is fixing which removes 
remaining silver halide from emulsion and hardens galatin. The final stage is washing and 
drying in other to removes excess chemicals and water for viewing of radiograph. The 
time duration of the process is approximately two minutes.  
56 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
Figure 3. 7: Picture of film processor at Korle-Bu Teaching Hospital 
 
3.9. Phantom 
The constructed phantom was made of plexiglas of thickness 1.8 cm with dimension 40 x 
40 x17 cm3. Various objects such as steel, cerrobend, plexiglas   having densities similar 
to tissues (bone and soft tissue) found within the human body were placed in the phantom 
to give contrast level. 
Two step wedges made with cerrobend and plexiglas materials were placed at the two 
opposing side of the phantom which were used to determine signal to noise ratio and 
contrast level as well. The cerrobend is eutectic alloy 50% bismuth, 27% lead, 13.3% tin, 
and 10% cardium by weight.  Bar patterns made of steel of different thickness ranging 
from 4, 3, 2, and 1 mm with spacer made with plexiglas material were used to determine 
the resolution of the imaging system. 
Patient irradiation geometries were replicated by the use of the constructed phantom. The 
thickness of this phantom replicated the anatomical structure of the abdomen of human. 
57 
 
University of Ghana  http://ugspace.ug.edu.gh
   
The phantom was irradiated with Theraton equinox-100 cobalt 60 unit at the hospital. 
The experimental setup was done by placing the phantom on the couch of the treatment 
machine employing source to surface distance (SSD) technique of 100 cm. Both the CR 
and the radiographic film cassette were mounted on the cassette holder one at a time for 
every irradiation. The cassette on the holder was level accurately using a sprit or alcohol 
level in other to maintain equal balance of the cassette on the holder. The cassette was 
then placed under the couch with detector to couch distance 14.5 cm, and source to 
detector distance 131.6 cm.  
The phantom was irradiated with CR as detector with beam having a field size of  30 x 30 
cm and at various treatment time ranging from 0.02 to 0.4 min. Immediately after each 
irradiation the  CR cassette obtained   from the radiological department of the Korle-Bu 
Teaching hospital was taken back to be read with the CR reader. 
The CR cassette with the imaging plate was manually inserted into the reader for image 
acquisition. After 45 seconds, the reader which was connected to a computer, displayed 
an image on the screen. The digitized image was printed on a film for scanning.  
 A similar procedure was repeated using a radiographic film as a detector. The film 
cassette was then taken to the dark room for processing using an automatic film 
processor, in order to acquire the image. Both the CR and the radiographic images were 
acquired and scanned using the scan maker 9800XL model at the teaching hospital. 
Images were finally analyzed using Image J software. Figure 3.8 shows the experimental 
setup with the constructed phantom. 
58 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Collimator 
Constructed Phantom 
 
Figure 3. 8: Experimental setup showing the constructed phantom 
 
3.10. Image Quality Assessment 
A sample of images were acquired at treatment time ranging from 0.02, 0.2, 0.3, 0.4 
minutes for examination done on each CR and radiographic film system employed. 
Image quality was assessed using Image J software (version 1.48). These images were 
assessed by measuring the SNR, CNR and RMTF which determine the spatial resolution 
of the imaging system. 
In calculating the contrast to noise ratio (CNR), regions of interest (ROI) were marked on 
each of the step wedge design from the image of the phantom constructed, and a 
background ROI was also marked.  The mean gray values for each of the steps and the 
59 
 
University of Ghana  http://ugspace.ug.edu.gh
   
background results were obtained by the image J software. The formula used to estimate 
CNR is found in the literature review with equation (2.4). 
The results obtained from this formula were then used to compare the CNR for CR 
detector and the screen film used at the therapy department for portal imaging. 
The SNR was determined by drawing ROIs on each of the six steps of the phantom. SNR 
was determined by taking the difference in the mean gray value of the object and its 
background by the quadratic sum of the standard deviation of the background as a 
measured noise. The SNR was then calculated by using the formula found in chapter two 
with equation (2.3)  
The relative modulation transfer function (RMTF) which determines the resolution of an 
imaging system was determined using equation (2.8), (2.9), (2.10), (2.11). 
The images for both CR and radiographic film from which these quality assessments 
were made are show in figures 3.9 and 3.10 below. 
60 
 
University of Ghana  http://ugspace.ug.edu.gh
   
University of Ghana  http://ugspace.ug.edu.gh
   
 
No Area of Mean Gray Standard Min Gray Max Gray 
ROI Values of ROI Deviation (std) Values Values 
1 10000 152.66 3.29 143 161 
2 10000 142.95 2.89 131 151 
3 10000 114.59 3.12 105 122 
4 10000 62.27 3.36 51 76 
5 10000 15.57 1.91 10 20 
6 3480  10000 8.08 0.28 8 9 
 
 
Table 4. 2: Calculated results of SNR for CR obtained from scanner data of irradiation 
time of 0.1 min 
Std object + std   Step Transmission 
Mean(M) -
of Background                 √                       SNR=  Thickness Values  
√               
Background 
(cm) 
(B) 
144.58 3.57 1.89 76.42 6.18 0.04 
134.86 3.17 1.78 75.65 5.07 0.07 
106.51 3.40 1.84 57.71 3.82 0.14 
54.18 3.64 1.90 28.37 2.63 0.26 
7.48 2.19 1.48 5.04 1.23 0.54 
 
62 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Table 4.1 shows measured values obtained from the scanned phantom using image J.  A 
region of interest (ROI) was drawn on each of the step found in the phantom and 
measured values such as the area, standard deviation, mean, minimum and maximum 
gray values seen in the table were generated by the software.  
Table 4.2 was deducted form table 4.1 which helps in the calculation of signal to noise 
ratio of CR. In this table, the SNR, transmission values for each of the steps with various 
thickness were calculated as shown in the table. The calculated SNR is high (76.42) when 
the step thickness is high (6.18 cm) thus showing a very low transmission value (0.04). 
Also form table 4.2 its clear that the SNR value decreases as the step thickness decreases 
while the transmission value increases with the decrease in both the step and SNR. 
 
SNR For CR (0.1min) 
90
80
70
60
50
40
30
20
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Transmision Value 
 
Figure 4. 1: Graph of SNR for CR of irradiation time of 0.1 min 
 
 
63 
 
SNR Value 
University of Ghana  http://ugspace.ug.edu.gh
   
Table 4. 3: Result of scanned data using image J software (FILM) of 0.1 min 
No Area of Standard Mean Values Min Gray Max Gray 
ROI deviation (std) of ROI Values Values 
1 10000 1.38 199.59 183 203 
2 10000 0.97 198.29 180 203 
3 10000 1.52 192.28 172 197 
4 10000 1.43 181.69 167 186 
5 10000 1.84 166.14 153 173 
6 10000 1.63 144.03 129 150 
 
 
Table 4. 4: Calculated results of SNR for Film obtained from scanner data of 
irradiation time of 0.1 min 
Std object + std   Step Transmission 
Mean(M) -
of Background  SNR=                √                       Thickness Values  
√               
Background (B) 
(cm) 
55.56 3.02 1.73 31.96 6.18 0.04 
54.26 2.61 1.61 33.56 5.07 0.07 
48.24 1.52 1.23 39.06 3.82 0.14 
37.66 3.07 1.75 21.46 2.63 0.26 
22.11 3.48 1.86 11.85 1.23 0.54 
 
64 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Table 4.3 shows measured values obtained from scanned phantom using image J.  A 
region of interest (ROI) was drawn on each of the steps found in the phantom and 
measured values such as the area, standard deviation, mean, minimum and maximum 
gray values seen in the table were generated by the software.  
Table 4.4 was deducted form table 4.3 which helps in the calculation of signal to noise 
ratio of film. In this table, the SNR, transmission values for each of the steps with various 
thicknesses were calculated as shown in the table. The calculated values for SNR are not 
consistent with the step thickness for film as that shown for CR in table 4.2.The 
transmission values increase as the step thickness decreases.  
 
 
SNR FOR FILM (0.1min) 
45
40
35
30
25
20
15
10
5
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Transmision Values 
 
Figure 4. 2: Graph of SNR for Film of irradiation time of 0.1 min 
 
65 
 
SNR Values 
University of Ghana  http://ugspace.ug.edu.gh
   
Figure 4.1 shows a graph of signal to noise ratio versus transmission values of the CR 
detector when using a step wedge found in the phantom of 0.1 minute treatment time.  
The graph shows a decrease in SNR value from 76.42 to 5.04 exhibiting a decrease from 
31.4% to 2.1%,  with an increase in the transmission values from 0.04 to 0.54 as the step 
thickness  decreases from (6.18 cm) to (1.23 cm). On the other hand, the graph of Figure 
4.2 shows a conventional film detector with an increase in SNR value from 31.96 to 
39.06 exhibiting an increment from 23.2% to 28.3%. The graph shows a decrease in SNR 
from 39.06 to 11.85 exhibiting a decrease of 28.3% to 8.6%. 
 
 
4.2 Comparing SNR of the two detectors (CR and Film) of irradiation time 0.1 min. 
Table 4. 5: SNR values for both detectors (CR and FILM) 
SNR (CR) SNR (FILM) STEP THICKNES TRANSMISSION 
(cm) VALUES 
76.42 31.96 6.18 0.04 
75.65 33.56 5.07 0.07 
57.71 39.06 3.82 0.14 
28.37 21.46 2.63 0.26 
5.04 11.85 1.23 0.54 
 
Table 4.5 shows calculated values of signal to noise ratio, step thickness and transmission 
for the two detectors (CR and film) used in the experiments. 
66 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
SNR FOR THE DETECTORS (CR AND FILM) OF 0.1 min  
90
80
70
60
50
40
30
20
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Transmision Values 
SNR (CR) SNR (FILM)
 
Figure 4. 3. Graph showing the two detectors SNR (CR and film) 
In figure 4.3 at step five where the step thickness is high (6.18 cm), with the lowest 
transmission value of 0.04, CR shows a superior SNR value (76.42) exhibiting 31.4% 
more than that of the film (31.96) exhibiting 23.2%. From the graph, a decrease in steps 
in CR, the SNR values decreases from (76.42) to (5.04) with an increase in the 
transmission values from (0.04) to (0.54). With the film on the other hand at step 
thickness of 6.18 cm it shows a SNR value of (31.96), a decrease in steps, it shows a 
gradual increase of (39.06), immediately after this point there is a decrease in SNR to 
(11.85). At the lower steps (1.23 cm) of the two detectors the film shows a higher value 
in SNR of 11.85 (8.6%) to that of CR of value 5.04 (2.1%).  
 
 
67 
 
Axis Title 
University of Ghana  http://ugspace.ug.edu.gh
   
 
4.3 Contrast to Noise Ratio for both CR and Radiographic Film at irradiation time 0.2min 
 
Table 4. 6: Result of CNR from scanned data using image J (CR) 0.2 min  
No Area of Standard Mean Values Min Gray Max Gray 
ROI deviation (std) of ROI  Values Values 
1 10000 5.30 126.49 114 149 
2 10000 4.96 108.16 94 120 
3 10000 4.50 83.29 63 95 
4 10000 2.81 29.16 21 39 
5 10000 0.79 9.87 8 12 
6 10000 0.50 8.50 7 9 
 
 
Table 4. 7: Calculated result of CNR for CR acquire from scanner data of irradiation 
time of 0.2 min 
Signal in A- Signal in B    Step Transmission 
(SA-SB)      thickness Values 
  
                 
(cm) 
18.33 36.59 6.18 0.04 
24.87 49.64 5.07 0.07 
68 
 
University of Ghana  http://ugspace.ug.edu.gh
   
54.12 108.03 3.82 0.14 
19.29 38.51 2.63 0.26 
 
Table 4.6 shows measured values obtained from scanned phantom when evaluating for 
CNR in CR using the software. Regions of interest (ROI) on each of the steps give 
measured values for each parameters: area, standard deviation, mean, maximum and 
minimum gray values seen in the table were generated by the image J software. 
Table 4.7 was obtained from table 4.6 which helps in the calculation of CNR of the CR. 
The transmission values for each of the step with various thickness were calculated as 
shown in the table. From the table the CNR values for CR shows some fluctuations as the 
step thicknes decreases.  
 
CNR FOR CR OF 0.2min 
120
100
80
60
40
20
0
0 0.05 0.1 0.15 0.2 0.25 0.3
Transmision Values 
 
Figure 4. 4: Graph of CNR for CR of 0.2 min irradiation time. 
 
69 
 
CNR Values 
University of Ghana  http://ugspace.ug.edu.gh
   
 
 
Table 4. 8: Result of CNR from scanned phantom for film of 0.2 min 
No Area of Standard Mean Values of Min Gray Max Gray 
ROI deviation (std) ROI Values Values 
1 10000 2.23 179.87 157 188 
2 10000 1.51 179.11 167 184 
3 10000 1.56 170.41 156 176 
4 10000 1.58 158.67 149 165 
5 10000 2.61 132.05 111 141 
6 10000 2.64 102.85 82 113 
 
 
Table 4. 9: Calculated result of CNR for Film acquire from scanner data of irradiation 
time of 0.2 min 
Signal in A- Signal in B    Step Transmission 
(SA-SB)      thickness Values 
  
                 
(cm) 
0.75 0.28 6.18 0.04 
8.70 3.29 5.07 0.07 
11.73 4.44 3.82 0.14 
26.62 10.08 2.63 0.26 
 
70 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Table 4.8 shows measured values obtained from scanned phantom when evaluating for 
CNR in film using the software. Region of interest (ROI) on each of the steps gives 
measured values for each parameters: area, standard deviation, mean, maximum and 
minimum gray values seen in the table were generated by the image J software. 
Table 4.9 was obtained from table 4.8 which helps in the calculation of CNR of the Film. 
In the table the CNR, transmission values for each of the steps with various thicknesses 
were calculated and results shown in the table. From the table the CNR values for Film 
shows an increase as the step thickness decreases from 6.18 cm to 2.63 cm while the 
transmission values increases. 
CNR FOR FILM OF 0.2 min 
12
10
8
6
4
2
0
0 0.05 0.1 0.15 0.2 0.25 0.3
Transmision Values 
 
Figure 4. 5: Graph of CNR for Film of 0.2 min 
 
Figure 4.4 shows a graph of contrast to noise ratio versus transmission values of the CR 
detector when using a step wedge found in the phantom of 0.2 minutes treatment time.  
The graph therefore shows an increase in CNR value from 36.59 (15.7 %) to 108.03 (46.4 
71 
 
CNR Values 
University of Ghana  http://ugspace.ug.edu.gh
   
%) with an increase in transmission value from 0.04 to 0.14, with a decrease in step 
thicknes from 6.18 cm to 3.82 cm. After this pick there is sharp fallout in the CNR from 
108.03 exhibiting (46.4%) to 38.51 exhibiting (16.5%). On the other hand, the graph of 
Figure 4.5 shows a conventional Film detector. In this graph the CNR value increases 
from 0.28 to 10.08 exhibiting an increase from (1.6%) to (55.6%) with an increase also in 
the transmission value from 0.04 to 0.268 while the step thickness decreases from 6.18 
cm to 2.63 cm. 
 
4.4 Comparing CNR of the two detectors (CR and Radiographic Film) of 0.2 min 
Table 4. 10: Showing CNR values for CR digital detector and Film. 
Step thickness (cm) Transmission CNR for CR CNR for Film 
Values 
6.18 0.04 36.59 0.28 
5.07 0.07 49.64 3.29 
3.82 0.14 108.03 4.44 
2.63 0.26 38.51 10.08 
 
Table 4.10 shows compared calculated values for the two detectors (CR and Film). From 
the table the calculated values of CNR in the CR are far much higher than that for the 
film with the same transmission and step thicknes values. 
 
72 
 
University of Ghana  http://ugspace.ug.edu.gh
   
University of Ghana  http://ugspace.ug.edu.gh
   
University of Ghana  http://ugspace.ug.edu.gh
   
4.5 Spatial Resolution 
Table 4. 11: showing RMTF for CR digital detector of 0.1 min irradiation time 
No Area   Mean  Standard Frequency    =   /√  Variance RMTF=   -    /124.9     
(ROI) (ROI) deviation (M) (1lp/mm) ( )  
1 10000 96.57 11.17 0.12 16.34 124.94       0.8 
2 10000 50.74 16.35 0.16 16.34 267.35       2.0 
3 10000 17.52 6.59 0.25 16.34 43.44       0.2 
4 10000 13.29 1.78 0.5 16.34 3.18       0.1 
  
Table 4.11 shows the parameters: standard deviation, mean, variance and frequencies 
values on the bar pattern in the phantom which were used to calculate the relative 
modulation function, of the CR system to determine the resolution of the imaging unit 
used as a function of the frequency. 
 
75 
 
University of Ghana  http://ugspace.ug.edu.gh
   
RMTF FOR CR AT 0.1 min 
2.5
2
1.5
1
0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Frequency(1lp/mm) 
 
Figure 4. 9: Graph of RMTF for CR of 0.1 min 
 
Table 4. 12: Showing RMTF for Film of 0.1 min 
No Area   Mean  Standard Frequency    =   /√  Variance RMTF=   -        
(ROI) (ROI) deviation (M) (1lp/mm) ( )  /124.9 
1 10000 191.77 3.251 0.12   7.04 10.56       0.3 
2 10000 180.10 5.023 0.16   7.04 25.23       1.7 
3 10000 170.58 4.71 0.25   7.04 22.18       1.4 
4 10000 167.84 1.554 0.5   7.04 2.41       0.4 
 
Table 4.12 shows the parameters as: standard deviation, mean, variance and frequencies 
values on the bar pattern in the phantom that were used to calculate the relative 
modulation function, of the film system to determine the resolution of the imaging unit 
used as a function of the frequency 
 
76 
 
RMTF 
University of Ghana  http://ugspace.ug.edu.gh
   
RMTF FOR FILM AT 0.1 min 
2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Frequency (1lp/mm) 
 
Figure 4. 10: Graph of RMTF for Film detector 
 
Figure 4.11 and 4.12 shows the graphical representation of RMTF for CR and Film. From 
the graph in figure 4.11 the CR shows a higher RMTF value of 2.0 at a lower frequency 
of 0.12 lp/mm, and a fall in RMTF at higher frequencies. Also in fig 4.12, the film shows 
a higher RMTF value of 1.7 at a frequency of 0.16 1lp/mm and also a fall in RMTF at 
higher frequencies. 
 
 
 
 
 
77 
 
RMTF 
University of Ghana  http://ugspace.ug.edu.gh
   
4.6 Comparing the RMTF for the two detectors (CR and Film) at 0.1 Min 
 
Table 4. 13: Shows RMTF values for CR digital detector and Film 
Frequency  
(1lp/mm) RMTF (CR) RMTF (FILM) 
 0.12  0.8 0.3  
 0.16  2 1.7 
 0.25 0.2 1.4  
 0.5 0.1 0.4 
 
 
RMTF FOR CR AND FILM 
2.5
2
1.5
1
0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Frequency ( 1lp/mm) 
CR FILM
 
Figure 4. 11: Graph of RMTF for both CR and Film detectors 
 
Figure 4.13 shows the RMTF of both the CR and Film drawn on the same axis. From this 
graph at lower frequency of 0.12 lp/mm to 0.16 lp/mm the RMTF of the CR is higher 
than that of the film from 25.8% to 7.8% and 64.5% to 44.7%. As the frequency increases 
78 
 
RMTF 
University of Ghana  http://ugspace.ug.edu.gh
   
from 0.25 lp/mm to 0.5 lp/mm, there was a drastic reduction in the RMTF of the CR to 
that of the film from 6.4% to 36.8%. The resolution of the CR at 50% of RMTF was 0.2 
lp/mm and at 50% of the film RMTF the resolution was 0.4 lp/mm at higher frequency. 
At 50% RMTF the frequency (lp/mm) of the CR is 0.2lp/mm while that of the Film is 0.4 
lp/mm indicating better resolution.  This result is in accordance with work done by 
(Konstantinidis, 2011) (Williams et al., 2007) , (Lu, Nickoloff, So, & Dutta, 2003), and 
(Bansal, 2006). 
 
 
 
 
 
 
 
 
 
 
 
79 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 4.7 summary of findings 
Table 4.14: Shows the findings of the research in accordance with the specific 
objectives 
Specific Objectives Findings based on the detectors 
(CR and Film) 
To quantitatively determine the quality of  
images obtained from CR and Films 
using the image quality parameters  
Signal to noise ratio (SNR)  The calculated values of SNR for both 
detectors shows that the CR system have 
a higher SNR than the film. 
Contrast to noise ratio (CNR) It also shows that CNR value is higher for 
CR system than film 
Relative modulation transfer The spatial resolution is however 
function (RMTF) marginally higher for film system than 
CR system. 
Visibility response of images on Another very important finding of the 
detectors (CR and Film) at low work was that at low doses anatomical 
and high doses or treatment time. landmarks are clearly visible on the CR 
system, while nothing is seen on the film. 
At higher doses, images on the CR 
system become darker while images on 
the film now become clear. This implies 
that high doses are needed for the film to 
be useful while low doses are enough for 
the CR system thereby sparing patients of 
the deleterious effects of high doses. 
 
 
80 
 
University of Ghana  http://ugspace.ug.edu.gh
   
1.8 Contribution to knowledge. 
            The study 
 Establishes the superiority of image quality with the CR to that acquired with film 
as portal imager. 
 Establishes the suitability of the CR system for routine use for patient positioning 
verification at radiotherapy department of the Korle-Bu Teaching Hospital 
 Involved the construction of a phantom for studying radiation dose distribution 
and assessing the positioning of patients. This is a phantom that the resident 
Medical Physicist could be using on daily basis for verifying the patient 
positioning in telecobalttherapy set ups 
 
 
 
 
 
 
 
 
81 
 
University of Ghana  http://ugspace.ug.edu.gh
   
CHAPTER FIVE 
 CONCLUSION AND RECONMENDATION 
This chapter provides the conclusion drawn from this study and recommendations on the 
use of CR detector for treatment setup verification at the Korle-Bu Teaching Hospital in 
Ghana. 
 
5.1 CONCLUSION   
Based on the results obtained from the study using CNR image quality evaluator shows 
that at irradiation time of 0.2 minutes, the CR shows an average value of 58.2 while that 
of the conventional film exhibit a value of 4.52 at the same irradiation time. The above 
result indicates that CR digital detector shows a superior image quality than that of the 
film. 
The SNR average values at irradiation time of 0.1 min was also calculated as a means of 
quantifying the level of noise in the images acquired from both conventional film and CR 
digital detector used. The average value for CR was 48.6, while the film value was 27.6. 
This result clearly shows that images acquired on the CR show less noise to that of the 
film, which thus exhibit superior image quality to that of the film.  
 Notwithstanding this, the RMTF values for both the film and CR detectors were 
evaluated to determine their spatial resolution. The value of RMTF at 0.1 min irradiation 
82 
 
University of Ghana  http://ugspace.ug.edu.gh
   
time for film was 0.95 while the CR was 0.78. From this results, the film shows a better 
spatial resolution ability than that of the CR system.  
Furthermore, irradiation of treatment time of 0.02 minutes was delivered on the 
constructed phantom, with which the two detectors were used to acquire images. At this 
treatment time, image of the phantom was visible on the CR digital detector while no 
image of the phantom was seen on the film detector. This result clearly shows that at 
lower dose, anatomical land marks images of the phantom can be clearly displayed on the 
CR detector.   Also with the CR, images acquired becomes darker as the irradiation time 
start increasing from 0.2 min and above, indicating that quality images are acquired with 
low doses whilst with the film good images are acquired when the treatment time is high. 
 The research indicated that patient going through radiotherapy treatment using 
conventional portal film for imaging are been induced with more dose than using CR 
system  for portal imaging. 
This clearly shows that using CR to acquire portal images would allow the radiation 
oncologist to take advantage of computerized image-processing techniques and Picture 
Archival Communication (PAC) technology with minimal expenditure of time and 
money. 
 
5.2 RECONMENDATION 
At the Korle-Bu Teaching Hospital were this research was undertaken, conventional film 
for portal imaging is still in use for patient setup verification. From the results of this 
83 
 
University of Ghana  http://ugspace.ug.edu.gh
   
study, it is recommended that the use of CR system be incooperated into the radiotherapy 
department of the Korle-Bu Teaching Hospital for portal imaging in the verification of 
patient setup for treatment. The use of CR system for patient verification at the therapy 
department would help to minimize patient dose during setup verification. Since the 
hospital is also planning to go digital, the use of CR at the radiotherapy department would 
help to improve the quality of patient treatment.  
Also for further research, it would be better if the findings of this work could be applied 
to some of the patients visiting the center for treatment.  
It would be better also for policy makers, regulators of ionizing radiation to look into 
these recommendations seriously in other to ensure the safety of patient’s undertaken 
treatment at the center. 
 
 
 
 
 
 
 
 
 
84 
 
University of Ghana  http://ugspace.ug.edu.gh
   
 
 
 
REFERENCES 
Andreo, P., Burns, D. T., Hohlfeld, K., Huq, M. S., Kanai, T., Laitano, F., … Vynckier, 
S. (2006). Absorbed Dose Determination in External Beam Radiotheraphy: An 
International Code of Practice for Dosimetry based on Standards of Absorbed Dose 
to Water. Iaea Trs-398, 2006(June), 183. https://doi.org/10.1097/00004032-
200111000-00017 
Antonuk, L. E., Boudry, J., Yorkston, J., Wild, C. F., Longo, M. J., & Street, R. A. 
(1990). Radiation damage studies of amorphous-silicon photodiode sensors for 
applications in radiotherapy X-ray imaging. Nuclear Inst. and Methods in Physics 
Research, A, 299(1–3), 143–146. https://doi.org/10.1016/0168-9002(90)90764-W 
Baily et al. (1980). Fluoroscopic visualization of megavoltage therapeutic x-ray beams, 
6(March), 935–939. 
Bansal, G. J. (2006). Digital radiography. A comparison with modern conventional 
imaging. Postgraduate Medical Journal, 82(969), 425–428. 
https://doi.org/10.1136/pgmj.2005.038448 
Bansal, Staib, Chen, Rangarajan, Knisely, Nath, & Duncan. (1999). A minimax entropy 
registration framework for patient setup verification in radiotherapy. Computer 
85 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Aided Surgery, 4(6), 287–304. https://doi.org/10.1002/(SICI)1097-
0150(1999)4:6<287::AID-IGS1>3.0.CO;2-9 
Barrett, H. H., & Kupinski, M. A. (2013). objective assessment of image quality VI: 
Imaging in radiation therapy. Physics in Medicine and Biology, 58(22), 8197–8213. 
Bijhold et al. (1991a). aligning features in portal and simulator images. Medical Physics, 
36, 1665–1679. 
Bijhold et al. (1991b). Radiation field edge detection in portal images. Physics in 
Medicine and Biology, 36(12), 1705–1710. https://doi.org/10.1088/0031-
9155/36/12/015 
Bijhold et al. (1992). Maximizing setup accuracy using portal images as applied to a 
conformal boost technique for prostatic cancer. Radiotherapy and Oncology, 24(4), 
261–271. https://doi.org/10.1016/0167-8140(92)90233-K 
Boyer, A. L., Antonuk, L., Fenster, A., Van Herk, M., Meertens, H., Munro, P., … 
Wong, J. (1992). A review of electronic portal imaging devices (EPIDs). Med Phys. 
https://doi.org/10.1118/1.596878 
Bushong. (2013). No Title Radiologic science for Technologists-physics, Biology and 
protection, 10 Edition. 
Buzdar et al. (2013). Accuracy requirements in radiotherapy treatment planning. Journal 
of the College of Physicians and Surgeons Pakistan, 23(6), 418–423. 
https://doi.org/06.2013/JCPSP.418423 
Desai et al. (2010). Practical Evaluation of Image Quality in Computed 
86 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Radiographic(CR) Imaging System. Proc. SPIE7622, Medical Imaging: Physics of 
Medical Imaging, 1(c), 76224Q–76224Q–10. https://doi.org/10.1117/12.844640 
Droege, & Morin. (2011). A practical method to measure the MTF of CT scanners, 
758(1982). https://doi.org/10.1118/1.595124 
Effects, R. H. (2016). Canadian Nuclear Safety Commission, 1–5. 
Eilertsen, K. (1997). Automatic detection of single MLC leaf positions with corrections 
for penumbral effects and portal imager dose rate characteristics Automatic 
detection of single MLC leaf positions with corrections for penumbral effects and 
portal imager dose rate charact. Medical Physics, 313. 
Eilertsen et al. (1994). Methods for fully automated verification of patient set-up in 
external beam radiotherapy with polygon shaped fields. Medical Physics, 993(39), 
993–1012. 
Essers et al. (1995). Dosimetric characteristics of a liquid-filled electronic portal imaging 
device. International Journal of Radiation Oncology, Biology, Physics, 33(5), 1265–
1272. https://doi.org/10.1016/0360-3016(95)00108-5 
Fujita, H., Ueda, K., Morishita, J., Fujikawa, T., Ohtsuka, A., & Sai, T. (1989). Basic 
imaging properties of a computed radiographic system with photostimulable 
phosphors. Medical Physics, 16(1), 52–59. https://doi.org/10.1118/1.596402 
Gilhuij et al. (1996). Interactive Three Dimensional Inspection of Patsent Setup in. 
Medical Physics, 34(4), 873–885. 
Gilhuijs, K. G. A., & van Herk, M. (1993). Automatic on-line inspection of patient setup 
87 
 
University of Ghana  http://ugspace.ug.edu.gh
   
in radiation therapy using digital portal images. Medical Physics, 20(3), 667–677. 
https://doi.org/10.1118/1.597016 
Gray, J. E., Archer, B. R., Butler, P. F., Hobbs, B. B., Mettler, F. A., Pizzutiello, R. J., … 
Yaffe, M. J. (2005). Reference Values for Diagnostic Radiology: Application and 
Impact. Radiology, 235(2), 354–358. https://doi.org/10.1148/radiol.2352020016 
Hanley et al. (1995). The effects of out-of-plane rotations on two dimensional portal 
image registration in conformal radiotherapy of the prostate. International Journal 
of Radiation Oncology, Biology, Physics, 33(5), 1331–1343. 
https://doi.org/10.1016/0360-3016(95)02062-4 
James et al. (2000). Verification of dynamic multileaf collimation using an electronic 
portal imaging device Verification of dynamic multileaf collimation using an 
electronic portal imaging device. Physics in Medicine and Biology, 495. 
Khan, F. M. (Ed.). (2007). Treatment Planning in Radiation Oncology, 2nd Edition 
(Second Edi). Lippincott Williams & Wilkins, 2007. 
Konstantinidis, A. (2011). Evaluation of digital x-ray detectors for medical imaging 
application, 9(1), 165–168. 
Kud, M. (2012). Image Quality Assessment, 94–99. 
Langmack, K. A. (2001). Review article Portal imaging, 74, 789–804. 
Lu, Z. F., Nickoloff, E. L., So, J. C., & Dutta, A. K. (2003). Comparison of computed 
radi, 41(2), 1–42. Retrieved from 
http://www.ndt.net/article/imagingNDE2007/Zscherpel1.pdf 
88 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Luckey, & Rochester. (1975). Apparatus and methods for producing images 
corresponding to patterns of high energy radiation, 54(19). 
Mckeever. (1985). Thermoluminescence of solids, 94–95. 
Meertens et al. (1985). A liquid ionisation detector for digital radiography of therapeutic 
mega. Plasma Sources Science and Technology, 30(4), 687–702. 
https://doi.org/10.1088/0031-9155/30/4/004 
Meertens et al. (1988). An inverse filter for digital restoration of portal images. Physics in 
Medicine and Biology, 33(6), 687–702. 
Mohammadi, P., & Ebrahimi-Moghadam, A. (2014). subjective and objective Quality 
Assessment of image:A survey. arXiv, (june), 1–50. https://doi.org/10.1088/0031-
9155/58/22/8197 
Monti, A. F., Ostinelli, A., Frigerioa, M., Cosentinob, D., Bossib, A., Cazzanigab, L. F., 
… Vallib, M. C. (1995). An ICRU 50 radiotherapy treatment chart, 35, 145–150. 
Morton et al. (1991). A linear array, scintillation crystal-photodiode detector for 
megavoltage imaging. Medical Physics, 18(4), 681–691. 
https://doi.org/10.1118/1.596661 
Munro et al. (1990). Therapy imaging: A signal-to-noise analysis of a fluoroscopic 
imaging system for radiotherapy localization. Medical Physics, 17(5), 763–772. 
https://doi.org/10.1118/1.596474 
Murphy, M. J., Balter, J., Balter, S., Bencomo, J. A., Das, I. J., Jiang, S. B., … Yin, F. F. 
(2007). The management of imaging dose during image-guided radiotherapy: Report 
89 
 
University of Ghana  http://ugspace.ug.edu.gh
   
of the AAPM Task Group 75. Medical Physics, 34(10), 4041–4063. 
https://doi.org/10.1118/1.2775667 
Petrascu, O., Bel, A., Linthout, N., Verellen, D., Soete, G., & Storme, G. (2000). 
Automatic on-line electronic portal image analysis with a wavelet-based edge 
detector. Medical Physics, 27(2), 321–329. https://doi.org/10.1118/1.598834 
Rajapakshe et al. (1996). A quality control test for electronic portal imaging devices. 
Medical Physics, 23(7), 1237–1244. https://doi.org/10.1118/1.597866 
Remeijer, P., Geerlof, E., Ploeger, L., Gilhuijs, K., van Herk, M., & Lebesque, J. V. 
(2000). 3-D portal image analysis in clinical practice: an evaluation of 2-D and 3-D 
analysis techniques as applied to 30 prostate cancer patients. International Journal 
of Radiation Oncology*Biology*Physics, 46(5), 1281–1290. 
https://doi.org/10.1016/S0360-3016(99)00468-X 
Roehrig, H. (1990). Use of computed radiography for portal imaging. Proceedings of 
SPIE, 1231(Jul), 492–497. https://doi.org/10.1117/12.18830 
Rowlands, J. A. (2002). The pysics of computed radiography. Physics in Medicine and 
Biology, 47, R123–R166. 
Shalev, S. (1995). Treatment verification using digital imaging. Radiation Therapy 
Physics, 155–173. 
Sonoda et al. (1983). Computed Radiography Scanning Utilizing Laser Stimulated 
Luminescence. Radiology, 148(3), 833–838. 
Sprawls, P. (1996). Review ______________________ Physical Principles of medical 
90 
 
University of Ghana  http://ugspace.ug.edu.gh
   
imaging. 2nd ed, (August), 1996. 
Street, R. A., Boudry, J., Huana, W., McShan, D. L., Morton, E. J., & Yorkston, I. J. 
(1992). Demonstration of megavoltage and diagnostic x-ray imaging with 
hydrogenated amorphous silicon arrays. Medical Physics, 19(6), 1455–1466. 
https://doi.org/10.1118/1.596802 
Stroom, J. C., & Heijmen, B. J. M. (2002). Geometrical uncertainties , radiotherapy 
planning margins , and the ICRU-62 report, 64, 75–83. 
Taborsky et al. (1981). Digital imaging for radiation therapy verification. Proc. SPIE, 
314(5), 164–171. 
Tagoe et al. (2017). Assessment of Dosimetric Functions of An Equinox 100 Telecobalt 
Machine, 3(3), 592–596. 
Taylor, & Francis. (2007). HANDBOOK OF RADIOTHERAPY PHYSICS: THEORY 
AND PRACTICE. (P. Mayles, A. Nahum, & J.-C. Rosenwald, Eds.). New York: 
CRC Press. 
Thapa, B., & Molloy, J. (2015). Physcica Medica Prospective image planning in radiation 
therapy for optimization of image of patient dose. Lancet Oncology, 7(10), 848–858. 
https://doi.org/10.1016/S1470-2045(06)70904-4 
Thermoluminescence of solids. (1980), (19). 
Thwaites, D. I., Centre, E. C., Hospital, W. G., Kingdom, U., Mijnheer, B. J., & Mills, J. 
a. (2003). Chapter 12 QUALITY ASSURANCE OF EXTERNAL BEAM 
RADIOTHERAPY. Quality Assurance, 407–450. 
91 
 
University of Ghana  http://ugspace.ug.edu.gh
   
Williams, M. B., Krupinski, E. A., Strauss, K. J., Breeden, W. K., Rzeszotarski, M. S., 
Applegate, K., … Seibert, J. A. (2007). Digital Radiography Image Quality: Image 
Acquisition. Journal of the American College of Radiology, 4(6), 371–388. 
https://doi.org/10.1016/j.jacr.2007.02.002 
Zhao, W., & Rowlands, J. A. (1992). A large area solid-state detector for radiology, 
1651, 134–143. 
Zhou, S. M., & Verhey, L. J. (1994). A robust method of multileaf collimator (MLC) 
leaf-configuration verification. Physics in Medicine and Biology, 39(11), 1929–
1947. https://doi.org/10.1088/0031-9155/39/11/009 
Zhou, & Bovik. (2002). A universal image quality index. Signal Processing Letters, 
IEEE, 9(3), 81–84. https://doi.org/10.1109/97.995823 
 
92