# Probabilistic mapping and automated segmentation of human brainstem white matter bundles

## TL;DR

This paper introduces a new automated method to map brainstem white matter bundles using MRI, which could improve understanding of neurological disorders.

## Contribution

The novel contribution is a convolutional neural network-based tool for automatic segmentation of eight brainstem white matter bundles.

## Key findings

- The BrainStem Bundle Tool (BSBT) successfully segments brainstem white matter bundles across different MRI protocols.
- BSBT detects microstructural changes in brainstem bundles associated with Alzheimer’s, Parkinson’s, multiple sclerosis, and traumatic brain injury.
- BSBT shows prognostic utility in predicting recovery from traumatic coma through longitudinal analysis.

## Abstract

Vital brainstem functions are relayed through clustered myelin-coated axons termed white matter (WM) bundles. There is presently no reliable method for delineating these brainstem structures, largely due to their morphological complexity. We map WM contrast and create a neural network model to automatically trace eight brainstem WM bundles in diffusion MRI. We validate this methodology with in vivo and ex vivo diffusion MRI data and demonstrate that microstructural and morphologic changes in distinct subsets of these bundles are associated with Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, and traumatic brain injury. Our method establishes a foundation for fully automated brainstem connectivity mapping, which will enhance our understanding of brainstem contributions to multiple neurological disorders.

Brainstem white matter (WM) bundles are essential conduits for neural signals that modulate homeostasis and consciousness. Their architecture forms the anatomic basis for brainstem connectomics, subcortical circuit models, and deep brain navigation tools. However, their small size and complex morphology, compared to cerebral WM, makes mapping and segmentation challenging in neuroimaging. As a result, fundamental questions about brainstem modulation of human homeostasis and consciousness remain unanswered. We leverage diffusion MRI tractography to create BrainStem Bundle Tool (BSBT), which automatically segments eight WM bundles in the rostral brainstem. BSBT performs segmentation on a custom probabilistic fiber map using a convolutional neural network architecture tailored to detect small anatomic structures. We demonstrate BSBT’s robustness across diffusion MRI acquisition protocols with in vivo scans of healthy subjects and ex vivo scans of human brain specimens with corresponding histology. BSBT also detected distinct brainstem bundle alterations in patients with Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, and traumatic brain injury through tract-based analysis and classification tasks. Finally, we provide proof-of-principle evidence for the prognostic utility of BSBT in a longitudinal analysis of traumatic coma recovery. BSBT creates opportunities for scalable mapping of brainstem WM bundles and investigation of their role in a broad spectrum of neurological disorders.