Related Experiment Video
Updated: May 3, 2026
Assessing Cortical Cerebral Microinfarcts on High Resolution MR Images
Published on: November 20, 2015
Sexual Dimorphism of Radiomic Features in the Brain: An Exploratory Study Using 700 μm MP2RAGE MRI at 7 T.
Marius E Mayerhoefer1, Timothy M Shepherd, Michael Weber
1From the Department of Radiology, NYU Grossman School of Medicine, New York, NY (M.E.M., T.M.S., D.L., S.W.); Division of General and Pediatric Radiology, Department of Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria (M.E.M., M.W.); Florey Institute of Neuroscience and Mental Health, Melbourne, Australia (H.R.P.); Comprehensive Epilepsy Center, Department of Neurology, NYU Grossman School of Medicine, New York, NY (H.R.P.); and Department of Radiology, University of Missouri Columbia, Columbia, MO (J.W.P.).
High-resolution MRI reveals significant differences in brain radiomic features between women and men. These sex-based brain variations, particularly in key structures, may require ultrahigh resolution for accurate detection and further study.
Area of Science:
- Neuroimaging
- Radiomics
- Medical Physics
Background:
- Radiomic features can quantify tissue characteristics from medical images.
- Understanding sex-based differences in brain structure is crucial for personalized medicine.
- Image resolution can influence the detection of subtle imaging biomarkers.
Purpose of the Study:
- To investigate sex-based differences in radiomic features of key cerebral structures.
- To determine the impact of MRI image resolution on detecting these sex-based differences.
Main Methods:
- Retrospective analysis of ultrahigh resolution (UHR) 3D MP2RAGE T1-weighted MRI scans from 30 healthy subjects (18 women, 12 men).
- Convolutional neural network used for segmenting seven key cerebral structures.
- Extraction of 87 radiomic features at UHR (0.7 mm 3 voxel size) and standard clinical resolution (1.4 mm 3 voxel size).
- Principal component analysis (PCA) and independent samples t tests used to compare radiomic features between sexes.
Main Results:
- At UHR, significant differences in principal components (PCs) of radiomic features were found between sexes in 6 out of 7 analyzed cerebral structures.
- At standard clinical resolution, only one PC from the corpus callosum showed a significant sex-based difference.
- The findings highlight that higher image resolution is more sensitive in detecting sex-based variations in brain radiomic features.
Conclusions:
- Substantial sex-based differences exist in the radiomic signatures of several key brain structures.
- Ultrahigh spatial resolution MRI is likely necessary to fully uncover and study the sexual dimorphism of brain structures.
- These findings underscore the importance of considering sex as a biological variable in neuroimaging research and clinical applications.
