When I was working on my Masters in data science, one of the projects I did was to create an algorithm that would take an intended use statement for a medical device and predict whether FDA would require a clinical trial. It worked fairly well, with accuracy of about 95%.
Since that’s a dynamic algorithm in which the user inputs an intended use statement and gets a prediction of FDA’s decision, I wanted to go about a similar task this month: create a static word cloud to show what words are most associated with intended use statements where FDA has required a clinical trial. At least in theory, this static representation might give you a sense of words in an intended use statement that are more likely to push your device toward a clinical trial.
Introduction
Hardly a day goes by when we don’t see some media report of health care providers experimenting with machine learning, and more recently with generative AI, in the context of patient care. The allure is obvious. But the question is, to what extent do health care providers need to worry about FDA requirements as they use AI?
Blog Editors
Recent Updates
- Proposed Modernization of the HIPAA Security Rules
- In Confirmation Hearings, AG Nominee Pledges to Defend the Constitutionality of the False Claims Act
- A Primer on Executive Orders and a Preview of the Road Ahead
- At Long Last, DEA’s Remote Prescribing Rules 2.0 Are (Really) Here! (Pending Further Consideration by the Incoming Administration . . .)
- Massachusetts District Court Applies “But-For Causation” Standard, Dismisses AKS-Based FCA Case After Evaluating Facts and Circumstances of Independent Contractor Arrangements