Unsupervised Detection of Post-Stroke Brain Abnormalities
This addresses the challenge of capturing secondary structural changes in post-stroke patients for better imaging biomarkers, but it is incremental as it builds on existing unsupervised methods with specific data improvements.
The paper tackled the problem of detecting post-stroke brain abnormalities, including focal lesions and non-lesional changes, using an unsupervised flow-based generative model called REFLECT, and found that training on healthy controls improved lesion segmentation (Dice = 0.37 vs. 0.27) and sensitivity to non-lesional abnormalities (FROC = 0.62 vs. 0.43).
Post-stroke MRI not only delineates focal lesions but also reveals secondary structural changes, such as atrophy and ventricular enlargement. These abnormalities, increasingly recognised as imaging biomarkers of recovery and outcome, remain poorly captured by supervised segmentation methods. We evaluate REFLECT, a flow-based generative model, for unsupervised detection of both focal and non-lesional abnormalities in post-stroke patients. Using dual-expert central-slice annotations on ATLAS data, performance was assessed at the object level with Free-Response ROC analysis for anomaly maps. Two models were trained on lesion-free slices from stroke patients (ATLAS) and on healthy controls (IXI) to test the effect of training data. On ATLAS test subjects, the IXI-trained model achieved higher lesion segmentation (Dice = 0.37 vs 0.27) and improved sensitivity to non-lesional abnormalities (FROC = 0.62 vs 0.43). Training on fully healthy anatomy improves the modelling of normal variability, enabling broader and more reliable detection of structural abnormalities.