Back

Detection of Region-specific Fiber Damage within Injured Spinal Cord Using Advanced Diffusion MRI

Wang, F.; Xu, J.; Wu, T.-L.; Yang, P.-F.; Chen, L. M.; Gore, J. C.

2025-07-03 neuroscience
10.1101/2025.07.02.662840 bioRxiv
Show abstract

This study aimed to evaluate diffusion parameters derived from diffusion tensor imaging (DTI) and spherical mean technique (SMT) for detecting region-specific, fine-grained tissue damage and white matter (WM) tract disruptions following spinal cord injury (SCI). Diffusion MRI data were acquired from the cervical spinal cord of monkeys before and after a unilateral dorsal column lesion at the C5 level, using a 9.4T scanner. Parametric maps derived from DTI and SMT effectively detected regional fiber damage around 16 weeks post-injury. Post-mortem silver staining served as the ground truth for assessing region-specific fiber damage. Diffusion MRI maps aligned well with histological measures and captured the severity of WM damage at the lesioned segment (in an order of dorsal > ventral > lateral WM tracts) and along the dorsal column tract across segments (in an order of lesion center > rostral > caudal). Among the diffusion parameters, fractional anisotropy (FA), axonal volume fraction (Vax), radial diffusivity (RD), and extra axonal transverse diffusivity (Dex) showed most significant changes at and around the lesion site where severe tissue damage occurred. FA, Vax, and axial diffusivity (AD) exhibited marked changes in dorsal column proximal to the lesion center, where moderate axonal damage occurred. Additionally, AD and FA showed the greatest sensitivity (true positive rate) and specificity (true negative rate) to mild fiber disruption and demyelination in regions distal to the lesion. Overall, FA provided the highest sensitivity and specificity for detecting fiber degeneration and demyelination, while Vax demonstrated the strongest spatial correlation with histologic markers of regional fiber damage. The combination of DTI and SMT thus offers reliable biomarkers for assessing SCI.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.