Anatomical Shape

Anatomical shape analysis focuses on quantitatively characterizing and comparing the shapes of anatomical structures, primarily using statistical shape modeling (SSM). Current research emphasizes developing robust and efficient deep learning methods, such as those employing point clouds, implicit shape representations, and variational information bottleneck approaches, to automate SSM construction directly from medical images, reducing reliance on manual segmentation. This work is significantly impacting medical image analysis by enabling more accurate and efficient diagnosis, treatment planning, and population-level studies of anatomical variation, particularly in areas like surgical navigation and disease progression monitoring.