Setayesh Abiazi Shalmani, “Integrating Computational Fluid Dynamics and Machine Learning for Predicting Cerebrovascular Aneurysm Growth”
Mentor: Mahsa Dabagh, Biomedical Engineering, Engineering & Applied Science (College of)
Poster #64
Cerebrovascular aneurysms pose a severe health risk, as their rupture can lead to life-threatening consequences such as stroke or death. Research has shown that aneurysm growth is closely linked to hemodynamic conditions, including pressure, velocity, and wall shear stress. Computational fluid dynamics (CFD) is widely used to simulate and analyze these factors; however, traditional CFD simulations are computationally expensive and time-consuming, making real-time clinical assessments challenging. This study aims to develop a predictive framework that integrates CFD with machine learning (ML) to improve efficiency while maintaining accuracy. Currently, our research is focused on the CFD phase, where we preprocess computed tomography angiography (CTA) images to reconstruct three-dimensional patient-specific vascular geometries. We then simulate blood flow within these geometries using CFD to extract key hemodynamic parameters associated with aneurysm growth. These simulations will generate the dataset required for the subsequent ML phase, where models will be trained to predict hemodynamic features more rapidly and accurately. By reducing reliance on time-intensive CFD simulations, this approach has the potential to significantly enhance aneurysm risk assessment. Although our work is still in progress, we anticipate that integrating ML into CFD analysis will enable more efficient predictions, ultimately aiding neurosurgeons in identifying high-risk aneurysms and making timely treatment decisions. This research not only advances computational modeling techniques but also has real-world clinical applications in improving cerebrovascular health care. By streamlining the assessment process, our method could provide a faster, more accessible tool for evaluating aneurysm growth, reducing rupture risks, and improving patient outcomes.