Automated quantification of axonal and myelin changes in contusion, dislocation, and distraction spinal cord injuries: Insights into targeted remyelination and axonal regeneration

Abstract

Quantifying axons and myelin is essential for understanding spinal cord injury (SCI) mechanisms and developing targeted therapies. This study proposes and validates an automated method to measure axons and myelin, applied to compare contusion, dislocation, and distraction SCIs in a rat model. Spinal cords were processed and stained for neurofilament, tubulin, and myelin basic protein, with histology images segmented into dorsal, lateral, and ventral white matter regions. Custom MATLAB scripts identified axons and myelin through brightness-based object detection and shape analysis, followed by an iterative dilation process to differentiate myelinated from unmyelinated axons. Validation showed a high correlation with manual counts of total and myelinated axons, with no significant differences between methods. Application of this method revealed distinct injury-specific changes: dislocation caused the greatest axonal loss, while distraction led to the lowest myelin-to-axon-area ratio, indicating preserved axons but severe demyelination. All injuries resulted in increased axon diameter and a decreased myelin-sheath-thickness-to-axon-diameter ratio, suggesting disrupted myelination. These results indicate that remyelination therapies may be most effective for distraction injuries, where preserved axons make remyelination crucial, while axonal regeneration therapies are likely better suited for dislocation injuries with extensive axonal loss. Contusion injuries, involving both axonal and myelin damage, may benefit from a combination of neuroprotective and remyelination strategies. These findings highlight the importance of tailoring treatments to the distinct pathophysiological features of each SCI type to optimize recovery outcomes.