Sheet of ZHOU You

From 2024 : Université Bourgogne Europe, PhD Candidate, Dijon, France

2021 – 2023 : Université Bourgogne Franche-Comté, Computer Vision, Le Creusot, France

2020 – 2021 : Nanjing Tech University, Computer Vision, Nanjing, China

2016 – 2020 : Hebei University of Technology, Automation and Control Engineering, Tianjin, China

Background: Thoracic Aortic Aneurysm (TAA) refers to an abnormal widening of the aorta’s diameter in the chest region. This condition carries the risk of potentially fatal events such as aortic rupture or dissection. Current clinical decisions for intervention primarily rely on measuring the maximum aneurysm diameter. However, this single criterion is insufficient for accurately assessing the true risk for each individual patient. Therefore, a deeper understanding of the biomechanical properties of the aortic wall is crucial for improving risk assessment and patient management.

Objective: This project aims to leverage advanced 4D Flow Magnetic Resonance Imaging (4D Flow MRI) technology to develop novel methods for evaluating the biomechanical state of thoracic aortic aneurysms, thereby enabling more accurate identification of high-risk patients. 4D Flow MRI allows for the quantification of 3D blood flow dynamics and the calculation of key parameters like Wall Shear Stress (WSS). A significant bottleneck in its application, however, is the need for precise segmentation of the aorta throughout the entire cardiac cycle.

Core Tasks:

1. Develop Automatic Aorta Segmentation: Utilize deep learning methods, based on 4D Flow MRI data, to achieve fully automatic and accurate 3D segmentation of the aorta for each phase within the cardiac cycle.
2. Extract and Validate Biomechanical Parameters: Automatically extract biomechanical parameters, such as WSS, from the segmentation results and flow data, and visualize them on the patient’s 3D aortic model. Concurrently, validate these imaging-derived parameters by comparing them with data from ex-vivo mechanical testing (e.g., biaxial tensile tests) and histological analysis performed on tissue samples surgically resected from the same patients.
3. Define Risk Indicators and Visualization: Combine insights from imaging analysis and ex-vivo experiments to explore and define new indicators, or combinations thereof, that can better predict the risk of aneurysm rupture. Map potential high-risk zones onto the 3D aortic model.

Technical Implementation and Integration:
The algorithm development, data processing, and visualization functionalities involved in this project will be primarily implemented using the C++ language and the Qt development framework. These newly developed tools will be integrated into the QIR 4D software, which is developed and maintained by the CASIS company, to facilitate subsequent research applications