Related Experiment Video
Updated: Aug 12, 2025
13:07
Optical Coherence Tomography Based Biomechanical Fluid-Structure Interaction Analysis of Coronary Atherosclerosis Progression
Published on: January 15, 2022
4.0K
An artificial intelligence-based platform for automatically estimating time-averaged wall shear stress in the
Lei Lv1, Haotian Li1, Zonglv Wu1,2
1Department of Cardio-Vascular Surgery, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, No. 107 Yan Jiang West Road, 510120 Guangzhou, China.
European Heart Journal. Digital Health
|January 30, 2023
Summary
A new AI platform accurately estimates time-averaged wall shear stress (TAWSS) in the ascending aorta, overcoming computational challenges of traditional methods like CFD for aortopathy risk assessment.
Area of Science:
- Cardiovascular Imaging
- Medical Artificial Intelligence
- Computational Fluid Dynamics
Background:
- Aortopathies require multiple indicators for risk assessment.
- Time-averaged wall shear stress (TAWSS) is a key indicator of aortopathy progression.
- Computational Fluid Dynamics (CFD) is complex and computationally expensive for TAWSS calculation.
Purpose of the Study:
- To develop a deep learning platform for accurate TAWSS estimation in the ascending aorta.
- To provide a computationally efficient alternative to CFD for TAWSS analysis.
Main Methods:
- Trained an AI model on thoracic computed tomography angiography data from 154 patients.
- Validated the AI platform against manual segmentation and CFD calculations.
- Assessed model performance using Dice Coefficient (DC), Normalized Mean Absolute Error (NMAE), and Root Mean Square Error (RMSE).
Main Results:
- The AI platform achieved a Dice Coefficient (DC) of 0.86 with manual segmentation.
- NMAE was 7.8773% ± 4.7144% and RMSE was 0.0098 ± 0.0097 compared to CFD findings.
- The AI approach resulted in a 12000-fold reduction in computational cost.
Conclusions:
- A high-efficiency, robust AI platform can automatically estimate TAWSS in the ascending aorta.
- This AI platform shows potential for clinical applications in aortopathy management.
- The study offers insights for solving CFD-related challenges through AI.
