ACR’s Allen on why AI use cases are so important to radiology
On Oct. 26, the American College of Radiology Data Science Institute (ACR DSI) announced the release of standardized artificial intelligence (AI) use cases designed to improve AI adoption in radiology. Why, exactly, are these use cases so vital to the specialty? Bibb Allen, MD, ACR DSI chief medical officer, provided some important context in a new analysis published by the Journal of the American College of Radiology.
“A structured AI use case includes a narrative description and flowcharts that define exactly how an AI algorithm takes in images and/or other information from the clinical workflow and provides specific output to end users,” Allen, a radiologist at Grandview Medical Center in Birmingham, Alabama, wrote in his analysis. “Structured AI use cases also include parameters for how algorithms are trained, tested, and validated for regulatory approval and clinical use; how they are deployed into clinical workflows; and how their effectiveness can be monitored in clinical practice.”
Creating these use cases also helps radiologists play a “leading role” in the development of these algorithms, Allen added, and highlights the specialty’s overall value.
These use cases also help provide radiology with consistency when it comes to its AI algorithms. A majority of AI developers are currently working with individual radiologists right now, which means progress is being made, but it could prove difficult to take “single-site” applications and generalize them for patient populations in different areas. The ACR’s aim in releasing its standardized AI use cases is to avoid such confusion and help keep radiologists throughout the United States on the same page. The ACR developed standardized process for this, Technology Oriented Use Cases in Healthcare AI (TOUCH-AI), which is an “open framework authorizing system for defining clinical and operational AI use cases” in radiology, Allen explained.
“The TOUCH-AI framework provides detailed descriptions of the goals the algorithm should meet, the required clinical inputs, how the algorithm should integrate into the clinical workflow, and how it should interface with both human end users and an array of electronic resources, such as reporting software, PACS, and electronic health records,” Allen wrote.
The 50 AI use cases released by the ACR DSI include cases in radiation oncology, interventional radiology and all diagnostic radiology subspecialties. They are all freely available to developers. The FDA worked with the ACR to ensure each case was defined as part of the FDA Medical Device Development Tool program.
“The ACR DSI structured use-case development program is a cornerstone in the creation of an AI ecosystem for radiology,” Allen concluded. “Structured AI use cases for the radiologic sciences can convene multiple stakeholders, ensure patient safety, promote diversity in algorithm development, and foster collaborations with federal regulatory agencies and even Congress to facilitate the introduction of AI algorithms into the market, which will enhance the care radiology professionals provide for their patients.”
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.