CNeuroGleam: Illuminating Small Vessel Disease Detection Through Deep Learning Based Segmentation of Brain MRI White Matter Hyperintensities

Abstract:

NeuroGleam is a deep-learning (DL) pipeline for fully automated segmentation of white-matter hyperintensities (WMH) on single-modal, low-resolution T2-FLAIR which is one of the most common MRI series in clinical care. Our preliminary investigation involved benchmarking of six DL architectures under four loss objectives. We found HRNet and its variants to be the best performing architecture and introduced a hyper-parameterized HRNet which was tuned using Bayesian optimization. Experiments span the Medical Image Computing and Computer Assisted Intervention (MICCAI) WMH Challenge dataset, and a 69-subject in-house clinical cohort called Assessing Population-based Radiological Brain Health in Stroke Epidemiology (APRISE). Metrics include Dice, Hausdorff95, lesion-wise sensitivity/F1, and average volume difference. The best HRNet achieves Dice 0.742 (MICCAI) and 0.651 (APRISE) with Hausdorff95 6.24-7.75 mm. Cross-dataset testing drops to Dice 0.523, underscoring domain-shift limits. We discuss practical design choices and outline ongoing work including transfer learning, domain adaptation and out-of-distribution detection to close the generalization gap and enable robust deployment.

Reference:

B. Iyer, B. Williamson, V. B. S. Prasath, B. J. Aronow, P. Khatri, H. Sucharew, V. Khandwala, J. LaPorta, L. Wang, R. Cornelius, M. Gaskill-Shipley, T. Tomsick, D. Wang, T. Maloney, P. S. Horn, J. Carrozzella, B. M. Kissela, A. Vagal. NeuroGleam: Illuminating small vessel disease detection through deep learning based segmentation of brain MRI white matter hyperintensities. 53rd Annual Applied Imagery Pattern Recognition (AIPR) Workshop, October 2025. Proc. Springer LNCS16446. doi:10.1007/978-3-032-18474-0_35

Back to MRI Projects Main Page. Back to Research Page.