ARPA-H launches program to create high-quality, diverse, representative medical imaging data exchange platform

Program aims to revolutionize medical AI tools development

The Advanced Research Projects Agency for Health (ARPA-H), an agency within the U.S. Department of Health and Human Services (HHS), today announced a new funding opportunity through its ImagiNg Data EXchange (INDEX) program. INDEX intends to develop an exchange platform to connect medical imaging data providers, users, and services to create robust and trustworthy artificial intelligence (AI) tools for radiology, pathology, and surgical imaging.  

Medical imaging is a critical step in diagnosing and treating most diseases. However, the demand for imaging exams exceeds the number of trained health care providers needed to interpret them. AI and machine learning (ML) have the potential to resolve this ongoing challenge. Large amounts of imaging data, such as digital pathology, surgical video, or radiological scans - MRI, CT, PET, and SPECT – are needed to train AI/ML applications. Currently, medical software developers and innovators lack access to sufficient quantity of diverse, representative, high-quality medical imaging data.  

“For AI-enabled medical diagnostic tools to become commonplace in medical settings, we must work to enable consistent performance, which requires large volumes of diverse and representative data for training, testing, and validation,”​ said ARPA-H Director Renee Wegrzyn, Ph.D. “What if high-quality, diverse, and representative imaging data were easily available for medical AI development? INDEX intends to answer this question with a platform that does not currently exist.”

If successful, the INDEX platform will provide much greater access to medical imaging data, resources for image processing, and ML algorithm development. It will include tools for algorithm assessment at multiple stages of its lifecycle, thus simplifying and accelerating imaging AI algorithm development, testing, and deployment. INDEX has the potential to help health care providers process more medical images quickly and with higher confidence, reducing provider burnout and giving people faster and more accurate diagnoses.    

“Medical imaging data is scarce, expensive, siloed, and not under a practical quality system,” said INDEX Program Manager Ileana Hancu, Ph.D. “INDEX is not another database; instead, it seeks to be a single, affordable, and sustainable exchange platform with built-in tools to develop regulatory-ready algorithms. The platform intends to benefit the entire imaging ecosystem, from data providers to data users and patients.”  

Through a forthcoming Innovative Solutions Opening (ISO) solicitation, INDEX invites proposals across three technical areas: business model and business plan, system architecture design and implementation, and enrollment. Proposals are expected to address all technical areas and involve teams from a wide ecosystem, including software developers, data aggregators, clinician experts, and more, solving technical challenges that enable faster, more accurate medical AI algorithm development. While initial INDEX offerings will include whole slide pathology imaging, one radiology focus area, and one surgical video focus area, the platform is intended to expand to accommodate other types of data.

ARPA-H and the U.S. Food and Drug Administration (FDA) are collaborating to accelerate medical AI/ML innovation by removing barriers to obtaining data that are intended to represent the U.S. population appropriately. As part of the INDEX program, FDA’s Office of Science and Engineering Laboratories (OSEL) in the Center for Devices and Radiological Health (CDRH) will develop regulatory science toolkits that will assess data quality, diversity, and usage.  

Multiple awards under this ISO are anticipated. Awards will depend on the quality of the proposals received and the availability of funds. Learn more about INDEX on its program page, including information about the solicitation, Proposers’ Day registration, and how to state interest to form an applicant team.