Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies
  1. Home
  3. Research Domains

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

Jianning Li1,2,3, Zongwei Zhou4, Jiancheng Yang5

  • 1Institute for Artificial Intelligence in Medicine (IKIM), University Hospital Essen (AöR), Essen, Germany.

Biomedizinische Technik. Biomedical Engineering
|December 29, 2024

Related Concept Videos

  • Information And Computing Sciences
  • Graphics, Augmented Reality And Games
  • Computer Aided Design
  • Medshapenet - A Large-scale Dataset Of 3d Medical Shapes For Computer Vision

    • Related Experiment Videos

    Three-Dimensional Shape Modeling and Analysis of Brain Structures
    05:33

    Three-Dimensional Shape Modeling and Analysis of Brain Structures

    Published on: November 14, 2019

    7.0K
    Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands
    10:59

    Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

    Published on: July 26, 2014

    14.4K
    3D Printing Model of a Patient's Specific Lumbar Vertebra
    07:30

    3D Printing Model of a Patient's Specific Lumbar Vertebra

    Published on: April 14, 2023

    1.8K

    View abstract on PubMed

    Summary
    This summary is machine-generated.

    MedShapeNet provides a large, accessible collection of 3D medical shapes for research and applications. This dataset aids in advancing computer vision for medical imaging, including tumor classification and surgical planning.

    Area of Science:

    • Medical Imaging
    • Computer Vision
    • 3D Shape Analysis

    Background:

    • State-of-the-art computer vision algorithms utilize various 3D shape representations like voxel grids, meshes, and point clouds.
    • Existing large-scale 3D shape datasets (e.g., ShapeNet, ModelNet) lack comprehensive medical anatomical and surgical instrument models.

    Purpose of the Study:

    • To introduce MedShapeNet, a novel dataset for 3D medical shapes.
    • To bridge the gap between computer vision and medical imaging by adapting data-driven algorithms for medical applications.
    • To provide a resource for developing and evaluating algorithms on real patient imaging data.

    Main Methods:

    • MedShapeNet directly models shapes from medical imaging data of real patients.
    • The dataset includes anatomical structures (bones, organs, vessels) and 3D models of surgical instruments.
    Keywords:
    3D medical shapesanatomy educationaugmented realitybenchmarkshapeomicsvirtual reality

    Related Experiment Videos

    Three-Dimensional Shape Modeling and Analysis of Brain Structures
    05:33

    Three-Dimensional Shape Modeling and Analysis of Brain Structures

    Published on: November 14, 2019

    7.0K
    Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands
    10:59

    Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

    Published on: July 26, 2014

    14.4K
    3D Printing Model of a Patient's Specific Lumbar Vertebra
    07:30

    3D Printing Model of a Patient's Specific Lumbar Vertebra

    Published on: April 14, 2023

    1.8K
  • Use cases demonstrate applications in brain tumor classification, skull reconstruction, anatomy completion, education, and 3D printing.

    • Main Results:

    • MedShapeNet currently comprises 23 datasets with over 100,000 annotated 3D shapes.
    • The data is publicly accessible through a web interface and a Python API.
    • The dataset supports discriminative, reconstructive, and variational benchmarks.

    Conclusions:

    • MedShapeNet establishes a valuable resource for medical shape analysis and computer vision applications.
    • The collection will continue to expand, supporting benchmarks and applications in virtual/augmented reality and 3D printing.
    • MedShapeNet is available at https://medshapenet.ikim.nrw/.