Abstract #0373

Pathology-Guided AI System for Accurate Segmentation and Diagnosis of Cervical Spondylosis in MR T2 Images

Qi Zhang¹, Xiuyuan Chen², Lianming Wu³, Kun Wang², Jianqi Sun^1,4,5, and Hongxing Shen²

¹School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ²Department of Spine Surgery, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China, ³Department of Radiology, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China, ⁴National Engineering Research Center of Advanced Magnetic Resonance Technologies for Diagnosis and Therapy (NERC-AMRT), Shanghai, China, ⁵Med-X Research Institute, Shanghai Jiao Tong University, Shanghai, China

Synopsis

Keywords: Diagnosis/Prediction, Segmentation

Motivation: Cervical spondylosis requires precise and automated diagnostic methods to handle the complexity of MRI-based T2-weighted assessments.

Goal(s): To develop an expert-driven AI system for automated segmentation and diagnosis of cervical spondylosis using MR T2-weighted images.

Approach: We introduce the PG-nnUNet model, which incorporates pathology-guided segmentation and edge loss for enhanced segmentation accuracy, and an expert-based framework for key clinical indicators.

Results: The system achieved high accuracy in both segmentation and diagnostic tasks on MR T2 images, outperforming existing methods and showing reliability for clinical use.

Impact: This AI-based framework enhances precision and efficiency in cervical spondylosis diagnosis, reducing clinician workload and improving diagnostic accuracy, with PG-nnUNet supporting consistent, automated decision-making. Future efforts will focus on multimodal imaging for broader applicability.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.