While recent draft guidance from the Food and Drug Administration on artificial intelligence-enabled medical devices is non-binding, the document indicates the agency is intensifying its regulatory scrutiny on these technologies, said Dr. Scott Schell, chief medical officer of IT consulting firm Cognizant.
“I think the timeliness of this draft guidance is very important because it signals that the agency is now ready to embrace some of the unique characteristics about AI that differ from other software-based health tech,” Schell said in an interview with Information Security Media Group.
The FDA’s draft guidance, issued in the final days of the Biden administration, covered a wide range of pre-market and life cycle management considerations that apply to developers and manufacturers of medical devices with one or more software functions that incorporate machine learning, deep learning and neural networks, as well as other types of AI (see: FDA Warns of Cyber Risks in Guidance for AI-Enabled Devices).
“This is a good opportunity to focus on decision-making versus decision-support capabilities, because clinicians are not ready to let the AI make the decision to basically invoke the next care action for their patients,” he said.
“If you look at what the FDA has done historically, draft guidance ends up being a pretty useful signal as to what to expect the full and permanent guidance to be,” he said.
“What’s the takeaway? If you’re in this space and you have IP or innovation around this kind of med tech, then you really need to be thinking about every stage of design data systems, integration, testing and validation,” he said. “And all of this should be conducted in the context of this draft guidance. I don’t actually think it’s a large lift because it really represents an extension of so much that we’re already doing.”
In this audio interview with Information Security Media Group (see audio link below photo), Schell also discussed:
- The potential of multi-agentic AI use for healthcare;
- HIPAA-related concerns involving AI-enabled medical devices and health technology;
- Why open-source AI technology from Chinese firm DeepSeek is not likely to be applied in the U.S. healthcare sector, at least not imminently.
Schell is a senior executive, surgeon and healthcare futurist with more than 30 years of experience in the healthcare, biotech and technology sectors both in the United States and globally. During his career, he has led development and implementation of large-scale population health and predictive analytics platforms at organizations including Alere, Cleveland Clinic and UPMC. He has founded and exited five healthcare startups, as well as served as managing partner for a private equity firm with a portfolio of digital health assets.