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

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

Authors

Institutions

Publication Details

Journal
Open Access CRIS of the University of Bern
Published
2024-12-30
DOI
https://doi.org/10.48620/93828
Primary Topic
Medical Imaging and Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

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

Fin Hendrik Bahnsen, Claudia M. Krebs, João Pedrosa, Hamza Mekhzoum et al.
Open Access CRIS of the University of Bern
Medical Imaging and Analysis
article

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

Fin Hendrik Bahnsen, Claudia M. Krebs, João Pedrosa, Hamza Mekhzoum, Alain Lalande, Thomas Küstner, Yannick Suter, Letícia Rittner, Hans Liebl, Amr U. Abourayya, Konstantinos Tserpes, Vicky Vandenbossche, Thomas J. J. Maal, Ramtin Gharleghi, Lars Erik Podleska, Rajni V. Patel, Yannik Hanusrichter, Afaque Rafique Memon, Evie Vereecke, Patrick Ferdinand Christ, Martin Urschler, Jana Fragemann, Moon Kim, Guilherme Aresta, Daniel Haehn, Stefan Cornelissen, Paul Friedrich, Mariana Lindo, António Sá Cunha, Zdravko Marinov, Timo van Leeuwen, Fabian Horst, Jakob Wasserthal, Heinrich Mächler, Max J. H. Witjes, Roy Eagleson, Felix Nensa, Hongwei Bran, Yuanfeng Ji, Chongyu Qu, Zongwei Zhou, Enrico Nasca, Richard Frayne, Ahmed Rekik, Miriam Balzer, Kangxian Xie, Jan Egger, Naida Solak, Seyed-Ahmad Ahmadi, Arcot Sowmya, Gregor Schiele, Gijs Luijten, Lukas Heine, Julius Keyl, Rainer Röhrig, Marcel Dudda, Dženan Zukić, Constantin Seibold, Costantino Grana, Peter F. Hoyer, Gian Marco Melito, Xiaojun Chen, Ti-chiun Chang, Alexander Brehmer, Jianxu Chen, Oliver Basu, Christoph John, Ping Luo, Christopher Schlachta, Pascal Fua, Jiancheng Yang, Patrick Langenhuizen, Christos Davatzikos, Andreas Nürnberger, Jens Kleesiek, Michele R. Aizenberg, Tom Vercauteren, Alan L. Yuille, Moritz Rempe, Aurélio Campilho, Soumick Chatterjee, Frank Hölzle, Victor Alves, Vahid Badeli, Sergi Pujades, Stanislav Malorodov, André Ferreira, Vincent Andrearczyk, Behrus Puladi, Sergios Gatidis, Bjoern Menze, Emmanuel Audenaert, Christina Gsaxner, Alexander Jaus, Hauke Heidemeyer, Hans Lamecker, Yuan Jin, Carlos Ferreira, Mauricio Reyes, Hendrik Mattern, Nicholas Heller, Nadya Shusharina, Ana Sofia Santos, Xiaoxi Chen, Edmond Boyer, Anjany Sekuboyina, Vincenzo Ferrari, Thomas M. Deserno, Aline Van Oevelen, David G. Ellis, Matthias Gunzer, Narmada Ambigapathy, Jef Vandemeulebroucke, Adrien Depeursinge, Hamidreza Salehi, Achraf Ben-Hamadou, Florian Dubost, Maximilian T. Löffler, Shireen Elhabian, Reuben Dorent, Eduardo A. Garza-Villarreal, Timo van Meegdenburg, Matthias A. Fink, Frederic Jonske, Amin Dada, Aaron Kujawa, Mathieu Hatt, Martin Weßling, Jianning Li, Rainer Stiefelhagen, Jun Ma, Wenxuan Li, Susann Beier, Tiezheng Zhang, Stefanie Schreiber, Kate Duquesne, Kathrin Krieger, Jose Garcia, Luca Lumetti, Luc Duong, Roberto Souza, Jan Stefan Kirschke, Paul F. Jaeger, Yao Zhang
article en

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

Open Access CRIS of the University of Bern
Laboratoire d’Imagerie Biomédicale (FR)
Reduced inequalities
Openalex Percentile: Top 59%
Medical Imaging and Analysis
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.