Radiology
Julián N. Acosta, Guido J. Falcone, Pranav Rajpurkar
Apr 19 2022
Abstract
What is the context?
- The state of medical artificial intelligence–enabled devices would benefit from the curation of data sets and algorithm development, allowing for comparison to prior examinations, which is crucial for several key imaging interpretation tasks in everyday clinical practice.
- The use of artificial intelligence (AI) has grown dramatically in the past few years in the United States and worldwide, with more than 300 AI-enabled devices approved by the U.S. Food and Drug Administration (FDA).
What is the problem?
- Most of these AI-enabled applications focus on helping radiologists with detection, triage, and prioritization of tasks by using data from a single point, …
- …but clinical practice often encompasses a dynamic scenario wherein physicians make decisions on the basis of longitudinal information.
- Unfortunately, benchmark data sets incorporating clinical and radiologic data from several points are scarce, and, therefore, the machine learning community has not focused on developing methods and architectures suitable for these tasks.
What is the solution?
- Current AI algorithms are not suited to tackle key image interpretation tasks that require comparisons to previous examinations.
- Focusing on the curation of data sets and algorithm development that allow for comparisons at different points will be required to advance the range of relevant tasks covered by future AI-enabled FDA-cleared devices.
Originally published at https://pubs.rsna.org
About the authors & authors affiliations
From the Department of Neurology,
Yale School of Medicine, New Haven, Conn (J.N.A., G.J.F.); and
Department of Biomedical Informatics,
Harvard Medical School, 10 Shattuck St, Boston, MA 02115 (P.R.).
