The generation of synthetic 3D cardiovascular anatomy is a crucial aspect of biomedical engineering and computational anatomy, especially for advancing in-silico trials of device deployment. A fundamental challenge in this domain is ensuring that generative models produce topologically accurate anatomical representations. This involves maintaining the correct number and type of topological components, such as connected components, loops, and voids, and ensuring accurate topological interactions between different tissue types, including containment, adjacency, and exclusion. These aspects are vital for the accurate simulation of cardiovascular physics, as minor topological errors can lead to significant unphysiological effects. This project aims to develop and implement topological regularization methods within generative models to penalize and correct topological defects, thereby producing anatomically viable configurations.
**Prerequisites**
• Substantial experience in deep learning methodologies.
• Proficiency in computational geometry and mesh processing.
• A strong background in image processing.
**Good to have**
• Experience in persistent homology.
• Experience in skeletonization techniques.
• Familiarity with medical imaging technologies and data.
The generation of synthetic 3D cardiovascular anatomy is a crucial aspect of biomedical engineering and computational anatomy, especially for advancing in-silico trials of device deployment. A fundamental challenge in this domain is ensuring that generative models produce topologically accurate anatomical representations. This involves maintaining the correct number and type of topological components, such as connected components, loops, and voids, and ensuring accurate topological interactions between different tissue types, including containment, adjacency, and exclusion. These aspects are vital for the accurate simulation of cardiovascular physics, as minor topological errors can lead to significant unphysiological effects. This project aims to develop and implement topological regularization methods within generative models to penalize and correct topological defects, thereby producing anatomically viable configurations.
**Prerequisites**
• Substantial experience in deep learning methodologies.
• Proficiency in computational geometry and mesh processing.
• A strong background in image processing.
**Good to have**
• Experience in persistent homology.
• Experience in skeletonization techniques.
• Familiarity with medical imaging technologies and data.
• Development of a framework for the detection and quantification of topological defects in cardiovascular anatomy.
• Development of deep-learning-based approaches for regularizing the topological quality of these models.
• Benchmarking the effectiveness of these approaches against existing state-of-the-art algorithms in the field.
• Development of a framework for the detection and quantification of topological defects in cardiovascular anatomy.
• Development of deep-learning-based approaches for regularizing the topological quality of these models.
• Benchmarking the effectiveness of these approaches against existing state-of-the-art algorithms in the field.
Candidates interested in this project should submit their CV and cover letter to kkadry@mit.edu
Candidates interested in this project should submit their CV and cover letter to kkadry@mit.edu
