Effect of segmentation from different diffusive metric maps on diffusion tensor imaging analysis of the cervical spinal cord

Deep Learning-Based Auto-Segmentation of Spinal Cord Internal Structure of Diffusion Tensor Imaging in Cervical Spondylotic Myelopathy

Cervical spondylotic myelopathy (CSM) is a chronic disorder of the spinal cord. ROI-based features on diffusion tensor imaging (DTI) provide additional information about spinal cord status, which would benefit the diagnosis and prognosis of CSM. However, the manual extraction of the DTI-related features on multiple ROIs is time-consuming and laborious. In total, 1159 slices at cervical levels from 89 CSM patients were analyzed, and corresponding fractional anisotropy (FA) maps were calculated. Eight ROIs were drawn, covering both sides of lateral, dorsal, ventral, and gray matter. The UNet model was trained with the proposed heatmap distance loss for auto-segmentation. Mean Dice coefficients on the test dataset for dorsal, lateral, and ventral column and gray matter were 0.69, 0.67, 0.57, 0.54 on the left side and 0.68, 0.67, 0.59, 0.55 on the right side. The ROI-based mean FA value based on segmentation model strongly correlated with the value based on manual drawing. The percentages of the mean absolute error between the two values of multiple ROIs were 0.07, 0.07, 0.11, and 0.08 on the left side and 0.07, 0.1, 0.1, 0.11, and 0.07 on the right side. The proposed segmentation model has the potential to offer a more detailed spinal cord segmentation and would be beneficial for quantifying a more detailed status of the cervical spinal cord.

The Correlation between Functional Connectivity of the Primary Somatosensory Cortex and Cervical Spinal Cord Microstructural Injury in Patients with Cervical Spondylotic Myelopathy

Objectives

To explore functional connectivity reorganization of the primary somatosensory cortex, the chronic microstructure damage of the cervical spinal cord, and their relationship in cervical spondylotic myelopathy (CSM) patients.

Methods

Thirty-three patients with CSM and 23 healthy controls (HCs) were recruited for rs-fMRI and cervical spinal cord diffusion tensor imaging (DTI) scans. Six subregions (including leg, back, chest, hand, finger and face) of bilateral primary somatosensory cortex (S1) were selected for seed-based whole-brain functional connectivity (FC). Then, we calculated the apparent diffusion coefficient (ADC) and fractional anisotropy (FA) values of the cervical spinal cord. Correlation analysis was conducted between FC values of brain regions and DTI parameters of cervical spinal cord (ADC, FA), and their relationship with each other and clinical parameters.

Results

Compared with the HC group, the CSM group showed decreased FC between areas of the left S1_hand, the left S1_leg, the right S1_chest, and the right S1_leg with brain regions. The mean FA values of the cervical spinal cord in CSM patients were positively correlated with JOA scores. Especially, the FA_pos values of bilateral posterior funiculus were positively correlated with JOA scores. The ADC and FA values of bilateral posterior funiculus in the cervical spinal cord were also positively correlated with the FC values.

Conclusions

There was synchronization between chronic cervical spinal cord microstructural injury and cerebral cortex sensory function compensatory recombination. DTI parameters of the posterior cervical spinal cord could objectively reflect the degree of cerebral cortex sensory function impairment to a certain extent.