The internship focuses on improving 3D medical image segmentation accuracy by incorporating statistical prior knowledge into deep learning architectures. The project aims to develop novel methodologies that leverage shape statistics and anatomical constraints to enhance the robustness and reliability of cerebral aneurysm detection and segmentation.
Main Objectives:

• Develop and implement statistical shape models for intracranial aneurysm
• Design and integrate shape priors into deep learning architectures
• Evaluate and validate the proposed methods on medical imaging datasets

Bibliography:

1. Raju, Ashwin et al. “Deep Implicit Statistical Shape Models for 3D Medical Image Delineation.” AAAI Conference on Artificial Intelligence (2021).
2. Wickramasinghe, Udaranga et al. “Voxel2Mesh: 3D Mesh Model Generation from Volumetric Data.” International Conference on Medical Image Computing and Computer-Assisted Intervention (2019).

• Master’s student or final year engineering student in Applied Mathematics, Computer Science, or related field
• Strong mathematical background, particularly in statistics
• Demonstrated experience in deep learning and computer vision
• Practical knowledge of PyTorch
• Programming skills in Python and experience with scientific computing libraries
• Interest in medical imaging and healthcare applications
• Previous experience with medical image processing is a plus