News | Stories
Aortic Stent Graft Demonstrator
Animations of a representative unsuccessful (i.e., proximal Type IA endoleak) thoracic stent graft deployment (left) and a representative successful (i.e., no endoleak) thoracic stent graft deployment (right).
With the considerable amount of time, resources and risk involved in performing clinical assessments, the biomedical industry is increasingly looking to innovative methods, such as in silico trials, as a viable alternative to help meet urgent demand. Combining machine learning with our extensive engineering simulation expertise, our Biomechanics & Life Sciences team creates virtual models of living systems that can serve as a substitute to physical testing.
In a recent R&D project, the team produced an in silico workflow to simulate a surgical implantation of a generic thoracic stent graft for a virtual patient population of thoracic aortic aneurysm (TAA) anatomies. They applied machine learning to analyze how the patients’ anatomical features and attributes of the device influenced therapeutic outcomes. The analysis showed quantitatively which therapy inputs have the most influence on the likely chronic proximal seal zone length of the stent after deployment.