The virtual skeleton database: an open access repository for biomedical research and collaboration

Area of Science:

• Biomedical Engineering
• Medical Informatics
• Computer-Aided Surgery

Background:

• Statistical shape models are crucial for medical image analysis but require large, often lost, datasets.
• Acquiring and managing these datasets is time-consuming, leading to duplicated efforts in research.

Purpose of the Study:

• To introduce the Virtual Skeleton Database (VSD) as a centralized storage and sharing system for medical data.
• To address challenges in data acquisition, management, and collaboration for statistical shape model development.

Main Methods:

• The VSD provides an online repository with tools for processing common image formats.
• It includes a statistical shape model framework and ontology-based search for data retrieval.
• The system supports collaborative research and data accessibility for the medical community.

Main Results:

• Exemplary projects demonstrate the VSD's utility in cochlear surgery optimization and femoral fracture risk analysis.
• An application for evaluating brain tumor segmentation was developed using the VSD.
• The repository supports international collaborations and population-based studies.

Conclusions:

• The VSD is a novel system enhancing scientific collaboration in medical imaging through a data-centric approach.
• Its semantically driven search facilitates anatomical structure retrieval.
• The VSD proves valuable for collaborative model building, biomechanical studies, and segmentation algorithm enhancement.