Meeting Banner
Abstract #0379

Multi-Stage Deep Learning Architecture for Carotid Artery Segmentation and Stenosis Degree Evaluation: A Comparative Study with DSA

Zhiji Zheng1, Xin Cao2, Qingluan Yang3, Wanchen Liu2, and Daoying Geng4
1Academy for Engineering and Technology, Fudan University, Shanghai, China, 2Department of Radiology, Huashan Hospital, Fudan University, Shanghai, China, 3Department of Infectious Diseases, Huashan Hospital, Fudan University, Shanghai, China, 4Huashan Hospital, Fudan University, Shanghai, China

Synopsis

Keywords: Diagnosis/Prediction, AI/ML Image Reconstruction

Motivation: This paper aims to mitigate the labor-intensive, time-consuming processes and interobserver variability that currently limit diagnostic efficiency in managing atherosclerotic diseases.

Goal(s): To develop a deep learning-enhanced architecture for automated segmentation of extracranial carotid artery and an intelligent quantitative diagnosis of the degree of stenosis, in comparison with DSA.

Approach: This two-stage architecture comprises modules for artery localization, automatic segmentation, and quantitative stenosis evaluation. It localizes extracranial carotid arteries within an ROI, subsequently segmenting and classifying stenosis from 3D reconstructions.

Results: The model achieved DSC of 0.9737 and AUC of 0.89, validating its effectiveness in enhancing segmentation performance and diagnostic efficiency.

Impact: This pipeline demonstrates high concordance with DSA and could significantly enhance cardiovascular risk assessment and atherosclerotic disease diagnosis in a non-invasive, radiation-free manner. Its clinical implementation may streamline diagnostic workflows and aid in the management of carotid artery disease.

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.

Keywords