NIH Launches AI Program to Advance Biomedical, Behavioral Research
The National Institutes of Health launched the Bridge2AI program to accelerate the widespread use of artificial intelligence in biomedical and behavioral research.
Source: Getty Images
By - The National Institutes of Health (NIH) is investing $130 million over the next four years through the NIH Common Fund’s Bridge to Artificial Intelligence (Bridge2AI) program to advance the use of AI by the biomedical and behavioral research communities.
The Bridge2AI program is an NIH-wide effort to generate data, resources, and tools responsive to AI approaches. These efforts include collaborations aimed at creating standards and guidance for the development of ethically sourced datasets and tools that do not perpetuate disparities or inequities, according to the press release.
“Generating high-quality ethically sourced data sets is crucial for enabling the use of next-generation AI technologies that transform how we do research,” said Lawrence A. Tabak, DDS, PhD, performing the duties of the director of NIH, in the press release. “The solutions to long-standing challenges in human health are at our fingertips, and now is the time to connect researchers and AI technologies to tackle our most difficult research questions and ultimately help improve human health.”
The program also aims to bolster the formation of diverse research teams with varying perspectives, backgrounds, and disciplines to support equity goals. Through these collaborations, Bridge2AI hopes to develop ethical practices related to data generation and use, such as data trustworthiness, bias reduction, and privacy.
To date, NIH has awarded funding to four data generation projects from the University of South Florida (USF), University of California, San Diego (UCSD), University of Washington (UW), and Massachusetts General Hospital (Mass General).
USF’s project will focus on voice as a biomarker of health. It aims to create a voice database that researchers focused on machine learning can use to train models to detect conditions known to have associations with voice alterations. The project is a collaboration between USF, Weill Cornell Medicine, Washington University School of Medicine in St. Louis, and nine other stakeholders, according to the institutions’ press releases.
UCSD’s project aims to create maps of the structure and function of entire human cells, which will be used to develop precision medicine AI algorithms to analyze patients’ genomes and detect disease. The project brings together researchers from Yale and other institutions across the country, according to a Yale School of Medicine press release.
UW’s project will develop a dataset focused on disease trajectory that is AI-friendly and hypothesis-agnostic, which will be used to create a disease model that researchers can replicate with other diseases for future studies. Participants in this project include Johns Hopkins University and Stanford University, among others.
Mass General’s project is concerned with creating a dataset using 100,000 patient EHRs, social determinants of health (SDOH), and waveform physiology information.
NIH also awarded funding to three additional projects related to creating a Bridge Center for integration, dissemination, and evaluation activities, which “will be responsible for integrating activities and knowledge across data generation projects, and disseminating products, best-practices, and training materials,” the press release states.
