Meeting Banner
Abstract #1851

Deep Learning-Driven Architecture for Automated Segmentation and High-Risk Carotid Plaque Identification in High-Resolution MRI

Xin Cao1, Zhiji Zheng2, and Qingluan Yang3
1Department of Radiology, Huashan Hospital, Fudan University, Shanghai, China, 2Academy for Engineering and Technology, Fudan University, Shanghai, China, 3Department of Infectious Diseases, Huashan Hospital, Fudan University, Shanghai, China

Synopsis

Keywords: Diagnosis/Prediction, Machine Learning/Artificial Intelligence

Motivation: Carotid atherosclerosis significantly contributes to cardiovascular diseases, necessitating improved diagnostic methods for plaque characterization in HR-MRI.

Goal(s): To develop a deep learning framework that automates the segmentation and classification of high-risk carotid plaques, enhancing diagnostic accuracy and risk assessment.

Approach: Leveraging a two-stage deep learning model, the study employed a modified 3D U-net for accurate plaque segmentation and a ResNet-based architecture for classification. Advanced image augmentation techniques were applied to enhance data robustness before training.

Results: Achieved a DSC of 0.8316, demonstrating high accuracy (89.74%) and AUC (0.9035) in differentiating symptomatic and asymptomatic plaques, confirming the model’s efficacy in clinical diagnostics.

Impact: This study significantly enhances the diagnostics of carotid atherosclerosis by leveraging high-resolution MRI and deep learning to accurately identify high-risk plaques, improving early intervention and potentially transforming outcomes in cardiovascular patient care.

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