Investigating the role of aortic mechanics in Marfan patients
Supervisor: Dr Ankush Aggarwal
School: Engineering
Industry Partner: Great Ormond Street Hospital
Description:
Aortic diseases cause 6000 deaths annually in the UK, highlighting the importance of timely intervention. This is even more important for patients with Marfan syndrome, a genetic abnormality, who are at a higher risk of aortic diseases. Currently, the decision to intervene is only based on aortic diameter. However, this measure is inadequate, as half of the reported deaths have diameter below the established threshold. Thus, there is an urgent need for a more accurate risk predictor. Emerging research suggests that aortic mechanics in the aortic disease progression.
In this project, we will investigate this relationship based on a unique data from our collaborators at Great Ormond Street Hospital (GOSH). We will extract the aortic stiffness and longitudinal stretch from the images using automated deep learning models and then quantify their correlations. We will also build a finite element model and find the tissue properties (stiffness and stresses). The novel insight will take us closer to refining the clinical guidelines that are more detailed than just the aortic diameter. Eventually, the results will help save thousands of patient lives who are currently being misdiagnosed.
Retrospective data from GOSH (we already have access to this) will be used, which contains MR images from 87 Marfan patients. The 3D static images will be segmented to delineate the aorta, and a finite element suitable mesh will be created. The mesh will be imported into FEBio (a finite element software), and the loading over a cardiac cycle will be simulated. The resulting deformation will be compared to the 2D cine MR images, to determine the stiffness parameters of the aortic tissue. The entire framework has been implemented and tested by a PhD student, which will be extended from six patients to 87 patients in this internship project. The results will be summarized in terms of mean, and quartile values, and correlations will be found with respect to longitudinal data about patients. The resulting insight will establish the role of mechanics in aortic diseases, and how these differ specifically in Marfan patients.