AI Improves Consistency in Regional Brain Volumes Measured in Ultra-Low Field MRI and 3T MRI

Kh Tohidul ISLAM¹Shenjun Zhong¹Parisa Zakavi¹Helen Kavnoudias²Shawna Farquharson³Gail Durbridge⁴Markus Barth⁵Andrew Dwyer^6,7Katie McMahon⁸Paul M. Parizel^10,9Richard McIntyre¹Gary F Egan¹Meng Law²Zhaolin Chen^1,11*

• ¹Monash Biomedical Imaging, Monash University, Melbourne, Australia
• ²Department of Neuroscience, Central Clinical School, Faculty of Medicine, Nursing & Health Sciences, Monash University, Melbourne, Victoria, Australia
• ³National Imaging Facility, Brisbane, Australia
• ⁴Herston Imaging Research Facility, University of Queensland, Queensland, Australia
• ⁵School of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, Australia
• ⁶South Australian Health and Medical Research Institute (SAHMRI), Adelaide, South Australia, Australia
• ⁷SA Medical Imaging, SA Health, South Australia, Australia
• ⁸School of Computer Science, Faculty of Science, Queensland University of Technology, Brisbane, Queensland, Australia
• ⁹David Hartley Chair of Radiology, Royal Perth Hospital, Western Australia, Australia
• ¹⁰Medical School, University of Western Australia, Western Australia, Australia
• ¹¹Department of Data Science and AI, Monash University, Clayton, Victoria, Australia

This study compares volumetric measurements of various brain regions using different magnetic resonance imaging (MRI) modalities and deep learning models, specifically 3T MRI, ultra-low field (ULF) MRI at 64mT, and AI-enhanced ULF MRI using SynthSR and HiLoResGAN. The aim is to evaluate the alignment and agreement among field strengths and ULF MRI with and without AI. Descriptive statistics, paired t-tests, effect size analyses, and regression analyses are employed to assess the relationships and differences between modalities. The results indicate that volumetric measurements derived from 64mT MRI deviate significantly from those obtained using 3T MRI. By leveraging SynthSR and LoHiResGAN models, these deviations are reduced, bringing the volumetric estimates closer to those obtained from 3T MRI, which serves as the reference standard for brain volume quantification. These findings highlight that deep learning models can reduce systematic differences in brain volume measurements across field strengths, providing potential solutions to minimize bias in imaging studies.