On March 15, 2024, the US Food and Drug Administration (FDA) stepped into the future of medical innovation with the release of a significant white paper, titled “Artificial Intelligence & Medical Products: How CBER, CDER, CDRH, and OCP are Working Together”. This landmark document reveals FDA’s broad perspective on the deployment and regulation of artificial intelligence (AI) in the life cycle of medical products.
Unveiling the FDA’s vision in response to the Executive Order 14110 issued in October 2023, this paper is an early insight into the potential policy framework that will guide the application of AI in healthcare and human services. Including drugs and devices safety, and public health among its areas of focus, the AI White paper represents a progressive pivot towards technology in the realm of health and human services.
“Medical products,” in the scope of this influential document, encompasses biological products, drugs, devices, and combination products. No stone seems to be left unturned as the FDA reveals its broad and inclusive approach to AI.
As mentioned in the white paper’s title, the Center for Biologics Evaluation and Research (CBER), the Center for Drug Evaluation and Research (CDER), the Center for Devices and Radiological Health (CDRH) and the Office of Combination Products (OCP) are working together.
These entities have all come together to deliver this comprehensive blueprint, putting forth a strategic collaborative work plan that looks at fostering innovation, developing harmonized standards, and supporting research in the realm of AI. This introductory look at the FDA’s new direction provides a snapshot of the potential changes and advancements on the horizon for AI in medical products.
Amplifying collaboration
Front and center of these progressive plans lies a robust endeavor to foster vital collaboration between a variety of key stakeholders. Different players such as AI developers, patient groups, and international regulators all have a part to play in shaping a future where AI is integral to healthcare. To implement a comprehensive, patient-centered regulatory approach, the FDA will be fostering discourse on specific topics like cybersecurity and quality assurance.
While the actual process of seeking input may vary, you can expect the FDA to utilize familiar strategies that involve public workshops, draft guidance documents, or proposed rules. All these avenues are designed to garner insights and feedback from a diverse range of perspectives and expertise. This approach does more than just gathering feedback; it helps to create an inclusive environment that welcomes all players in the AI arena.
Boosting medical AI advancements
The white paper also underlines the need to fuel innovation by providing clarity and predictability in regulatory policies. To keep up with the rapid pace of AI development, the FDA will monitor emerging trends and make timely adaptations to the evaluation of premarket regulatory submissions. This dynamic approach is designed to preemptively address potential challenges and ensure that new medical AI advancements can seamlessly integrate within regulatory frameworks.
What does this mean for AI innovators or medical product manufacturers? It signals a future where data and observation will shape regulation rather than staunch policies. Assume an adaptable stance on AI policy in medical products as the FDA embarks on this journey of breaking barriers for advancing AI adoption in medicine.
Refining good ML practices
Despite the advancements thus far, the FDA’s white paper signifies that there is more work to be done, particularly in refining and developing Good Machine Learning Practices (GMLP) for medical device development. International regulatory bodies, such as the International Medical Device Regulators Forum (IMDRF), can play a significant role in promoting harmonized standards and guidelines for AI integration into medical product development and post-market safety.
Envisioning a future where AI plays an integral role in health services, the FDA underlines the importance of leveraging multidisciplinary expertise. This collaboration aims to understand the desired benefits and potential patient risks tied to AI technology. Implementing good software engineering and security practices, as well as ensuring the integrity and security of data in AI and machine learning applications, are vital aspects of this vision. Polycentric efforts from companies like USDM Life Sciences, which offer training and expertise in establishing data governance frameworks, could prove paramount in achieving this goal.
The FDA channels collaboration
International collaboration can catalyze the mutual development and acceptance of standards, guidelines, and best practices. To this end, the FDA is engaging with global collaborators, such as Health Canada and the UK’s Medicines and Healthcare products Regulatory Agency (MHRA), and encouraging similar collaborations with other countries. Bodies like the IMDRF could be instrumental in combining resources and insights to advance GMLP.
As the FDA commits itself to promoting the responsible and ethical development and deployment of AI, it places focus on the performance of the human-AI team. Ensuring safety and efficacy through testing during clinically relevant conditions is of utmost importance. Moreover, ensuring that clinical study participants and data sets are representative of the intended patient population will help the industry develop solutions that are inclusive and wide-ranging.