One of the important goals of the project is to build a unique data repository that combines images of bruises and other injuries, with measurements, clinical and demographic information about the victim, and information inferred by artificial intelligence.
The repository is an integral part of the developed platform, which integrates data collected in main arms of the project:
- Controlled lab data: Data collection efforts are underway. Data are collected from volunteers with existing injuries and those who agree to have bruises induced in a controlled environment. See the “Engage” section for details.
- Clinical data from partner organizations: We are partnering with forensic medical units and hospitals to retrospectively collect existing imaging and clinical data on injuries. We are also prospectively collecting data from current patients.
- Contributed data from the community: We’d like everyone who wants to contribute their data to the project to do so. We are setting up the platform and mobile data collection tools to allow for massive data collection in the community.
This diversity of data and data sources is key to the success of this project and the building of artificial intelligence tools. Please see the Engage section to see how you can contribute.
The repository is designed to ensure that equity is a foundational component of any developed AI. Currently, the platform includes about 32,000 images of bruises collected using visible and alternative light sources and is being constantly expanded. To make the impact and build high-quality tools we plan to collect tens of thousands of bruise images with corresponding clinical data.