Artificial intelligence (AI) computing company NVIDIA and the American College of Radiology (ACR) are collaborating to make an AI-based toolkit available for radiologists to build, share, locally adapt and validate AI algorithms.
Following a three-month pilot program by both parties, ACR is integrating the NVIDIA Clara AI toolkit into the newly announced ACR Data Science Institute ACR AI-LAB, which is a free software platform that will be made available to more than 38,000 ACR members and other radiology professionals to build, share, locally adapt and validate AI algorithms, while also ensuring patient data stays protected at the local institution, according to officials who made the announcement this week at the World Medical Innovation Forum 2019 in Boston.
The pilot with the Ohio State University (OSU) and the Massachusetts General Hospital and Brigham and Women’s Hospital’s Center for Clinical Data Science (CCDS) helped NVIDIA and ACR define the assets and pathways necessary to enable facilities to work together and with industry to refine AI algorithms without sharing potentially sensitive patient data, officials stated. “Bringing an AI model to the patient data, instead of patient data to the model, can help increase diversity in algorithm training, facilitate validation of the algorithms and enable radiologists to learn the steps needed to adapt algorithms to their institutions’ clinical needs,” they said.
Specifically using the NVIDIA Clara AI toolkit, OSU professionals were able to quickly import a pre-trained model developed by CCDS. This model was customized to local variables and successfully labeled OSU data for further testing and improvement of the algorithm, all of which took place behind their own firewall. It resulted in a highly accurate and enhanced cardiac computed tomography angiography model, and the shared approach reduced algorithm training, validation and testing times by days, officials attested.
“This software will offer radiologists, without computer programming experience, the ability to build and improve AI algorithms without the need to share their data,” said Keith Dreyer, D.O., Ph.D., chief data science officer at Partners HealthCare and associate professor of radiology at Harvard Medical School. “Algorithms typically work best within the sites where they were trained, but those limited training sets are not always representative of the population at large. Training AI models on data from diverse sites helps ensure resiliency while reducing algorithm bias, resulting in improved inference across broader populations.”
Partners HealthCare also has announced that CCDS is planning broad roll-out to the Partners system over the next 12 months, and is already offering AI capabilities and support service. “The truth is, you don’t have to be a computer scientist or data scientist to participate in the creation of AI—we are just starting to see increasing availability of tools to enable on-premises development of AI models by clinicians,” said Dreyer.
The ACR AI-LAB already has the support of top industry companies such as GE Healthcare and Nuance, along with a vast network of healthcare startups and leading research institutes, according to officials. “This collaboration marks a significant milestone in an extraordinary ACR Data Science Institute project, helping enable the launch of the ACR AI-LAB, giving radiologists in any practice environment an opportunity to become involved in AI development at their own institutions, using their own patient data to meet their own clinical needs,” said Bibb Allen Jr., M.D., chief medical officer of the Data Science Institute at the American College of Radiology.
The initial version of ACR AI-LAB will be shown at the 2019 ACR Annual Meeting in Washington, from May 18 to 22. Attendees will be able to explore and experiment with the AI tools necessary to modify and refine AI models. Soon after, ACR plans to provide online access and sample data from publicly available patient datasets.
“Enabling a network of artificial intelligence between hospitals will create more robust algorithms, greater efficiencies and likely lead to better patient outcomes,” said Richard White, M.D., chair of the department of Radiology and Medical Imaging Informatics at the Ohio State University Wexner Medical Center.
