Researchers from MIT, Harvard University, and Massachusetts General Hospital have developed an artificial intelligence tool that can, for the first time, reliably identify and track key white matter fiber bundles in the human brainstem using diffusion MRI. The brainstem controls vital functions such as breathing, heart rate, sleep, and consciousness, yet its small size and constant motion have made it difficult to image in detail.
The new software, called the BrainStem Bundle Tool (BSBT), was trained on manually labeled MRI scans and validated against post-mortem brain dissections. It automatically segments eight distinct brainstem fiber bundles and consistently identifies them across repeated scans. BSBT combines probabilistic fiber mapping with a convolutional neural network to distinguish these bundles, overcoming noise from fluid flow and physiological motion.
When applied to patient datasets, BSBT revealed disease-specific patterns of structural damage. Alzheimer’s disease showed changes in one bundle, Parkinson’s disease in several, and multiple sclerosis in even more, while traumatic brain injury primarily affected fiber integrity rather than volume. The tool also tracked bundle healing in a coma patient, aligning with clinical recovery.
Overall, BSBT provides a powerful new imaging biomarker for detecting, monitoring, and understanding brainstem damage in neurological disease and injury.
Source: MIT News | Original Article: https://www.pnas.org/doi/10.1073/pnas.2509321123
