Optimization of Gradient Echo Pulse Sequence Parameters of the Brain Protocol Using Magnetic Resonance Imaging for Clinical Applications

Abstract

Magnetic Resonance Imaging (MRI) using Gradient Echo (GRE) pulse sequences is highly sensitive to magnetic field inhomogeneities, which can degrade image quality and compromise the accuracy of brain tissue segmentation and volumetric measurements. This research aimed to optimize two GRE pulse sequences on a 1.5 T Philips MRI scanner to establish reliable imaging protocols for clinical and research applications. The GRE sequences include the Fast Field Echo (FFE) and Three Dimensional T1-weighted Turbo Field Echo (3D T1W-TFE). A cross-sectional study design was employed. Optimization of the FFE was achieved through a quality control test using the American Association of Physicists in Medicine (AAPM) Report 100 and American College of Radiology (ACR) 2015 guidelines. The 3D T1W-TFE optimization was achieved using images acquired from five healthy volunteers who underwent brain MRI examinations with varying turbo factors 50–225. Magnetic field homogeneity (MFH), signal-to-noise ratio (SNR), geometrical distortion (GD), percentage image uniformity (PIU), and ghosting (GS) were assessed in the phantom study, while brain image quality metrics including SNR, contrast-to-noise ratios between gray matter and white matter (GM/WM), gray matter and cerebrospinal fluid (GM/CSF), and Euler number, were evaluated using the FreeSurfer suite. Key findings showed that MFH ranged between 0.13 and 0.60 ppm with optimal homogeneity (0.10 ppm) and maximum SNR achieved at the isocenter using a 400 mm field of view (FOV) with shimming. All FFE derived image quality metrics met ACR and AAPM acceptance criteria. In the in vivo study, turbo factor 200 produced the highest SNR and CNR values. Although variations in turbo factor influenced SNR and CNR trends, no statistically significant differences (p > 0.05) were observed in brain tissue volumes across turbo factors, indicating high reproducibility of volumetric estimates on the 1.5 T system. The findings demonstrate that optimization strategies including shimming, pulse sequence adjustment, and appropriate FOV selection substantially enhance image quality in GRE sequences. The optimized protocols, especially the 3D T1W-TFE turbo factor 200, offered improved efficiency and reproducibility for brain imaging. These results contribute valuable reference standards for future MRI quality control assessments and support the need for periodic magnetic field homogeneity evaluations, particularly following major system maintenance or upgrades.