A deep learning–based segmentation and CT–MRI fusion framework for preoperative spatial assessment of cervical ossification of the posterior longitudinal ligament
This study proposes a deep learning framework utilizing an optimized nnUNetv2 model for automatic segmentation of cervical OPLL structures and a point cloud-based registration method for CT–MRI fusion, thereby enabling accurate, efficient, and integrated three-dimensional preoperative spatial assessment of ossified lesions relative to the spinal cord.