In a move to streamline the regulatory process for Machine Learning-Enabled Medical Devices (MLMDs), regulators have introduced five guiding principles. These principles are designed to alleviate the regulatory reassessment burden on manufacturers when they make certain updates or changes to their devices.
The five guiding principles
Derived from the 10 principles of Good Machine Learning Practice, the new guiding principles for MLMD manufacturers are as follows:
Focused and bounded: The principles emphasize that any changes a manufacturer plans to implement should be specific and well-defined.
Risk-based: The creation and execution of a Predictive Change Control Plan (PCCP) should be rooted in a risk-based approach, ensuring adherence to risk management principles.
Evidence-based: Manufacturers must consistently demonstrate that the benefits of their device outweigh potential risks throughout its lifecycle.
Transparent: It’s imperative for manufacturers to offer clear, detailed information about their devices and maintain transparency with all stakeholders, ranging from patients to healthcare professionals.
Total product lifecycle perspective: Manufacturers should always consider the viewpoints of all stakeholders to enhance the quality and integrity of a PCCP.
Current UK regulatory landscape
At present, any significant modifications or updates to a medical device in the UK necessitate manufacturers to inform their conformity assessment body. Subsequently, the device may undergo reassessment to ensure that the changes haven’t compromised its safety or performance.
Given that AI and machine learning-driven medical devices often require frequent updates, this could lead to a cumbersome reassessment process after every change. This poses a considerable regulatory challenge for both developers and assessors.
The role of PCCPs
PCCPs come into play by allowing MLMD manufacturers to specify the changes and updates they intend to make. They can also detail how they plan to maintain the safety and effectiveness of their device, eliminating the need for constant regulatory oversight.
The guidelines highlight the areas where the Medicines and Healthcare Products Regulatory Agency (MHRA), the US Food and Drug Administration (FDA), and Health Canada share expectations regarding an acceptable PCCP. This alignment aims to minimize or even eliminate the need for reassessment.
National guidance and international collaboration
While these guiding principles will serve as a foundation for PCCP development across the UK, US, and Canada, each country will have its specific national guidelines that manufacturers must adhere to. For instance, the MHRA is set to release its guidance in 2024.
Dr. Paul Campbell, the MHRA’s head of software and AI, emphasized the growing prevalence of AI and MLMDs. He stated, “Regulators must adapt their processes to support innovations for patients while continuing to ensure their safety.”
He further highlighted the importance of international collaboration, noting, “By collaborating with the FDA and Health Canada on these guiding principles, we can clearly outline where we align on our expectations for a successful change control plan and help reduce the regulatory burden for manufacturers.”
The introduction of the five guiding principles for MLMDs marks a significant stride in the evolving landscape of medical device regulation. By fostering international collaboration and providing clear guidelines, regulators aim to support innovation while ensuring patient safety. As AI and machine learning continue to reshape the medical field, these principles will play a crucial role in balancing innovation with regulation.
A Step-By-Step System To Launching Your Web3 Career and Landing High-Paying Crypto Jobs in 90 Days.
