Evaluation of Disability Progression in Multiple Sclerosis via Magnetic-Resonance-Based Deep Learning Techniques

Short-term disability progression was predicted from a baseline evaluation in patientswith multiple sclerosis (MS) using their three-dimensional T1-weighted (3DT1) magnetic resonanceimages (MRI). One-hundred-and-eighty-one subjects diagnosed with MS underwent 3T-MRI andwere followed up for two to six years at two sites, with disability progression defined according tothe expanded-disability-status-scale (EDSS) increment at the follow-up. The patients' 3DT1 imageswere bias-corrected, brain-extracted, registered onto MNI space, and divided into slices alongcoronal, sagittal, and axial projections. Deep learning image classification models were applied onslices and devised as ResNet50 fine-tuned adaptations at first on a large independent dataset andsecondly on the study sample. The final classifiers' performance was evaluated via the area underthe curve (AUC) of the false versus true positive diagram. Each model was also tested against itsnull model, obtained by reshuffling patients' labels in the training set. Informative areas were foundby intersecting slices corresponding to models fulfilling the disability progression predictioncriteria. At follow-up, 34% of patients had disability progression. Five coronal and five sagittal sliceshad one classifier surviving the AUC evaluation and null test and predicted disability progression(AUC > 0.72 and AUC > 0.81, respectively). Likewise, fifteen combinations of classifiers and axialslices predicted disability progression in patients (AUC > 0.69). Informative areas were the frontalareas, mainly within the grey matter. Briefly, 3DT1 images may give hints on disability progressionin MS patients, exploiting the information hidden in the MRI of specific areas of the brain.