MedShapeNet – a large-scale dataset of 3D medical shapes for computer vision

Biomedical Engineering / Biomedizinische Technik, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 29, 2024

Abstract Objectives The shape is commonly used to describe the objects. State-of-the-art algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models used. This seen growing popularity of ShapeNet (51,300 models) Princeton ModelNet (127,915 models). However, a large collection anatomical shapes (e.g., bones, organs, vessels) 3D surgical instruments missing. Methods We present MedShapeNet translate data-driven vision applications adapt state-of-the-art problems. As unique feature, we directly model majority on data real patients. use cases classifying brain tumors, skull reconstructions, multi-class anatomy completion, education, printing. Results By now, includes 23 datasets with more than 100,000 that paired annotations (ground truth). Our freely accessible via web interface Python application programming can be for discriminative, reconstructive, variational benchmarks as well various virtual, augmented, or mixed reality, Conclusions contains will continue collect applications. project page is: https://medshapenet.ikim.nrw/ .

Language: Английский

HIT: Estimating Internal Human Implicit Tissues from the Body Surface

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Journal Year: 2024, Volume and Issue: unknown, P. 3480 - 3490

Published: June 16, 2024

Language: Английский

On Predicting 3D Bone Locations Inside the Human Body

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 336 - 346

Published: Jan. 1, 2024

Language: Английский