Current Status of FDA-Approved Medical Devices and Algorithms Based on Artificial Intelligence

Since we are still in the beginning stages of this technological revolution, a lot of people are holding out hope that artificial intelligence and machine learning will improve medical diagnosis, management, and treatment when they are initially released. On the other hand, there are a lot of challenges that might prevent the use of machine learning and artificial intelligence in clinical settings, most notably in the field of AI regulation. The Food and Drug Administration (FDA) of the United States has given its approval to a number of AI/ML-based medical devices and algorithms that are now available. As a result, we provide an overview of these products. By making it abundantly evident if a medical equipment is based on artificial intelligence or machine learning, our objective was to bring attention to the significance of regulatory agencies. After doing a thorough cross-check and validation of all approvals, we were able to identify sixty-four AI/ML-based medical devices and algorithms that have been authorized by the FDA. In the official notice made by the FDA, just 32 of those individuals included any terms relating to artificial intelligence or machine learning. Sixteen (17.56%) got de novo route clearance, while one (2.4%) gained premarket approval (PMA) clearance. The bulk of the substances were authorized by the FDA with a 623(k) clearance, which accounted for 85.9% of the total. These technologies were created for the fields of Radiology, Cardiology, and Internal Medicine/General Practice, respectively, with 42 (53.7%), 32 (34.2%), and 18 (22.5%) being the most notable examples. With the introduction of the first complete and open access database of medical solutions that are solely based on artificial intelligence and machine learning, we have received approval from the Food and Drug Administration. Ongoing updates will be made to the database whenever possible.