James Pink’s Post

Practical, consensus guidance on how to get the best out of technology in healthcare is always welcome. User buy-in is critical for all devices, no matter how sophisticated, to be accepted and scaled to where they can help the most patients.

'SOUP' can ruin your AI-Enabled medical device development? "The process of fine tuning foundation models for more specific medical purposes may introduce significant quantities of Software of Unknown Provenance (SOUP), as the manufacturer performing the fine tuning may only have limited documentation for the base foundation model.' The "Good Machine Learning Practice for Medical Device Development: Guiding Principles" document outlines essential guidelines for developing safe AI-enabled medical devices. Guiding Principles: 1- Understanding Device Purpose and Multidisciplinary Expertise An understanding of a device's intended use within clinical workflows and leveraging multidisciplinary expertise throughout the product lifecycle is crucial.(AKA involve your clinicians!) 2-Good Software Engineering and Security Practices: This includes robust software engineering, usability, data quality assurance, data management, and cybersecurity. 3- Representative Clinical Study Participants and Datasets: Prevent bias and promote generalizability by ensuring datasets and study participants reflect the intended patient population (eg demographics and clinical conditions). 4- Independence of Training and Test Datasets: To improve validity, the 'training' and 'test' datasets should be independent.

The MDR, which came into full application in May 2021, has been a game-changer for Med Tech industry, bringing both challenges and opportunities. Key points to consider: 1. Stricter clinical evidence requirements: The bar for demonstrating device safety and performance has been raised significantly. 2. Enhanced post-market surveillance: Manufacturers must now implement more robust systems to monitor their devices in real-world use. 3. Unique Device Identification (UDI): This system is improving traceability and recall management, but implementation has been complex. 4. Notified Body capacity: The increased workload on Notified Bodies has led to bottlenecks in the certification process. 5. Software as a Medical Device (SaMD): The MDR has brought clearer guidelines for software classification and requirements. As we navigate these changes, it's crucial for manufacturers to stay informed and adapt their processes accordingly. The MDR's emphasis on patient safety and device effectiveness is pushing our industry towards greater innovation and accountability.

In today’s blog, Matt Walker discusses the guiding principles released by the U.S. FDA, UK MHRA, and Health Canada regarding transparency in machine learning-enabled medical devices (MLMDs). Click here for background on the guiding principles and how they are likely to inform international standards for Good Machine Learning Practice (GMLP) Principles https://bit.ly/3zBCgNh

MHRA, FDA, and Health Canada have collaborated to set guiding principles on transparency for machine-learning medical devices. These principles, in alignment with previous guidelines, provide essential transparency practices for all medical devices. Stay informed on these crucial standards: Published: 13 June 2024 #MHRA #FDA #HealthCanada #MedicalDevices #Medtech #AI #machinelearning

So proud to see Ketryx disrupting the SaMD industry as one of the six predictions for MedTech in 2024. We appreciate the kind words from the Rook Quality Systems team and are happy we can work together to bring life-changing technology to patients faster! "In short, Ketryx accelerates SaMD development, improving quality, safety, and compliance in the regulated field of medical devices" Some more details on our product: "Key features include DevTools integration for faster development while maintaining FDA compliance, real-time traceability for visibility and regulatory alignment, automated documentation to free up time for critical risk areas, and embedding Quality Management System (QMS) procedures in tools like Jira for enhanced quality and compliance. Ketryx also facilitates faster releases through CI/CD pipeline integration, saving time with automatic documentation and traceability" #samd #FDA #fdaregulations #medicaldevice #medicalsoftware #validation #validatedAI #medtech #fdaapproved #fdacleared #fdaapproval #fdacompliance #softwarerequirements #AI #medicalAI #regulatedAI #risk #riskmitigation #risk #riskassessment https://lnkd.in/eH-Gp2s8

🌐 Demystifying Global SaMD Regulations - Discover Our Latest Insights 🌐 I’m thrilled to share our latest blog post on “Software as a Medical Device: Guidelines Worldwide.” This comprehensive piece explores the intricate landscape of SaMD regulations across different regions, essential for developers and stakeholders in the medical software field. Learn about the varying regulatory frameworks and their impact on SaMD development. Dive into the details here: https://lnkd.in/dRuVW3sk #SaMd #MedicalDevices #HealthTech #DigitalHealth #RegulatoryCompliance #SoftwareDevelopment

It's #TransparencyTuesday, with a great set of guiding principles for #AITransparency. The FDA teamed up with Health Canada and the United Kingdom's Medicines and Healthcare products Regulatory Agency to provide definitions and fundamentals for #ML transparency. Even if your #AI tool has nothing to do with a medical device, they offer a great framework focused on the basics: who, why, what, where, when, and how. Reach out if you need help putting these into practice. #ExplainableAI #ResponsibleAI

Is each egg alike the other in the box? The guidance MDCG 2020-5 “Clinical Evaluation – Equivalence: A guide for manufacturers and notified bodies”, which has been available in its original version since April 2020, addresses the requirements for and the actual drafting of the equivalence assessment when preparing clinical evaluations. Generally speaking, the equivalence assessment should be carried out according to the valid rules of the MDR. The specifications and recommendations of MEDDEV 2.7/1 Rev. 4 continue to have their justification even three years after the MDR came into force; nevertheless, there are deviations from MDR which are discussed in the MDCG 2020-5 document without introducing new aspects and specifications. Only the MDR remains legally binding, but the guidelines are to be understood and taken into account as “state of the art”. In the same wording as the MDR, MEDDEV 2.7/1 Rev. 4 requires that technical, biological and clinical characteristics must be taken into account when assessing equivalence when preparing a clinical evaluation based on the equivalence principle. The differences between the MDR criteria and the MEDDEV 2.7/1 Rev. 4 criteria regarding the evaluation of these three characteristics are addressed in the MDCG 2020-5 document. #clinical #evaluation #equivalence #MDR #regulatoryaffairs #Blog #consulting #MEDDEV #seleon

The recent MHRA press release on regulatory direction for medical devices has some helpful guidance and insights about the future of GB medical device regulation and innovation. The roadmap appears to be intended to balance two goals that are in tension with each other - making the UK an appealing market for innovation, alongside greater worldwide regulatory alignment. My takeaways from the associated guidance document "Implementation of the Future Regulations" are: (1) The opportunity for releasing new self-certified medical devices recently shrank, as new devices must now self-certify under the EU Medical Device Regulation which has stricter classification rules. These rules apply under regulatory transition arrangements until 2030. (2) New post-market surveillance regulations will be prioritised for 2024, hopefully clarifying expectations but potentially requiring operational rework and possibly increasing overheads (consultation pending). (3) Regulatory changes are planned in the future to: - Change the classification of several types of devices, specifically increasing the class of certain software as a medical device. - Strengthen the requirements for quality management systems and technical documentation. - Bring the essential requirements for medical devices in the GB market into greater alignment with the EU. This will include cybersecurity requirements for software as a medical device, including AI. - introduce a requirement to have a Person Qualified in Regulatory Compliance. https://lnkd.in/d-dQEzt6