エピソード

  • #462: Implementing an eQMS: The Ultimate Move-In Guide for MedTech Leaders

    #462: Implementing an eQMS: The Ultimate Move-In Guide for MedTech Leaders
    2026/06/29
    In this episode, host Etienne Nichols sits down with Michaela Kivett, a seasoned medical device consultant at Greenlight Guru, to break down the complexities of implementing an electronic Quality Management System (eQMS). Drawing from her background in orthopedic implant contract manufacturing and pharmaceutical process engineering, Michaela shares firsthand accounts of the critical inefficiencies that plague traditional paper and generic electronic repositories like SharePoint or Google Drive.The conversation centers around the strategic planning required to transition between quality management states. Michaela introduces a powerful moving house analogy, illustrating that simply dragging and dropping messy, legacy records into a new digital environment will not solve underlying organizational issues. Instead, a successful migration requires an intentional internal self-evaluation, a culture of quality, and a structured, room-by-room approach to data and process transfer.Additionally, the episode highlights how forward-thinking MedTech companies are leveraging advanced tools, including artificial intelligence, to streamline their eQMS implementation. By using AI to scan documents for compliance deficiencies against standards like ISO 13485, categorize sprawling folders, and map out workflow updates, manufacturers can dramatically mitigate the transitional efficiency dip and establish a mature, robust foundation for future scale.Key Timestamps00:42 – Michaela Kivett’s background: Transitioning from orthopedic quality engineering to pharma process engineering, and finding a passion for MedTech consulting.03:15 – Operational friction: Real-world pain points of on-site communication, tracking down physical signatures across 100-acre facilities, and booking conference rooms.04:32 – Version control nightmares: The consequences of multiple departments making parallel redlines without localized system notifications.06:12 – Defining the eQMS: Distinguishing between a basic electronic file repository (SharePoint/Google Drive) and a specialized, medical device-focused quality platform.08:58 – The universal MedTech pain point: Systemic organizational complexity and the hidden administrative burden of manual document referencing.10:43 – The Rube Goldberg illustration: How disconnected spreadsheets, Word files, and manual trackers create fragile operational systems.13:02 – The three legs of the medical device stool: Balancing ethical, legal, and monetary drivers to build organizational maturity.16:04 – The "moving house" migration framework: Why dragging and dropping cluttered records fails and how to evaluate a legacy data landscape before a move.19:25 – Operational entropy: Managing legacy supplier history and updating training matrices during a system overhaul.21:10 – Leveraging AI in eQMS implementation: Using automated tools to scan documents for ISO 13485 gaps and auto-categorize large file volumes.Quotes"Organization is the most common pain point. And it's a very simple pain point. I think every industry probably feels that... but you underestimate exactly how many different documents and records you're going to be producing and how many different places they tie into each other." — Michaela KivettTakeawaysAudit Before You Migrate: Treat an eQMS implementation as an internal audit. Do not lift and shift messy legacy files; instead, use the transition to purge obsolete records and refine active procedures.Mitigate the Efficiency Dip: Anticipate a temporary slowdown during a software transition. Minimize this area under the curve by building a sequential plan that prioritizes core procedures and training matrices before migrating complex design or risk data.Design for Future Scale: Choose and configure your digital quality architecture not just for the team you have today, but for the corporate milestones of tomorrow—whether that involves clinical trials, an international 510(k) submission, or M&A.Deploy AI for Compliance Mapping: Utilize AI tools to systematically scan old documentation folders for standard gaps (such as ISO 13485 or ISO 14971 compliance) and to automate the heavy lifting of categorizing thousands of uncategorized records.ReferencesISO 13485: The international standard outlining quality management system requirements specific to the medical device industry.Greenlight Guru: Purpose-built medical device software platform offering specialized QMS and EDC solutions to accelerate commercialization and ensure lifecycle compliance.Connect with the host, Etienne Nichols on LinkedIn.MedTech 101 SectionWhat is the difference between a QMS and an eQMS?Think of your QMS (Quality Management System) as the blueprint for an entire house. It represents the actual words, rules, regulations, and standard operating procedures (SOPs) that dictate how your company builds safe medical hardware.The eQMS (electronic Quality Management System) is the physical structure and construction ...
    続きを読む 一部表示
    39 分
  • #463: Finding MedTechs Leading Voices with Sean Smith

    #463: Finding MedTechs Leading Voices with Sean Smith
    2026/07/06
    The medical device industry functions as a highly complex ecosystem where diverse niches—including regulatory affairs, quality assurance, marketing, and reimbursement—must seamlessly interconnect to bring life-saving technologies to life. In this episode, host Etienne Nichols sits down with Sean Smith, a 25-year B2B marketing veteran, journalist, and founder of the weekly LinkedIn newsletter MedTech Leading Voices. Together, they peel back the curtain on why traditional, corporate-centric marketing strategies often fail within the specialized medical device space, exploring instead how service providers, consultants, and experts can effectively communicate their value without losing their human touch.As generative AI tools begin to saturate digital platforms with automated content, the landscape of B2B marketing has grown increasingly crowded and noisy. Sean discusses the critical paradigm shift required to cut through this digital "slop," emphasizing that true marketing success is rooted in foundational human behavior rather than algorithm hacking or perfect websites. He challenges the standard corporate playbook, pointing out the futility of over-polished taglines and unread online case studies, and explains how authentic storytelling and real-world problem-solving serve as the primary mechanisms for earning industry trust.Looking toward the future of professional expertise, the conversation addresses the long-game nature of organic visibility and the rising importance of specialized professional networks. Sean outlines actionable strategies for medical device experts to transition their buried, day-to-day insights into public thought leadership by focusing on what truly drives human behavior: helping peers make money or keeping them out of trouble. Ultimately, the episode serves as a powerful reminder that robust, human-centric communities are the ultimate safeguard for safeguarding one's livelihood and career longevity in an automated world.Key Timestamps00:05 – Introduction of guest Sean Smith and the interconnected MedTech ecosystem.01:52 – The origin story of MedTech Leading Voices and the challenges of navigating LinkedIn's closed platform.03:11 – How the B2B marketing landscape has changed since 2022 with the rise of AI-generated content.04:15 – Core philosophical pillars of modern marketing: Being human first and showing genuine interest in others.06:22 – Debunking the B2C vs. B2B crossover myth and the trap of corporate website redesigns.07:54 – How MedTech experts can unlock hidden value by documenting their day-to-day problem-solving.09:41 – Repackaging technical content (webinars, articles, infographics) for different audience attention spans.11:15 – The ultimate human motivators in professional spaces: Making money and staying out of trouble.12:55 – Journalism principles in marketing: The critical value of expert attribution over anonymous AI output.14:38 – Tactful visibility tricks for LinkedIn and the hidden power of a postscript (P.S.) in email communications.15:52 – Etienne's "accidental" conference moderation strategy for gaining immediate executive access.17:15 – Tactical advice for vendors: Building advisory boards and dropping the constant hard pitch.18:50 – Leveraging free platforms like Substack to establish niche authority through consistency.21:04 – Redefining "community" as an existential shield against automated AI displacement.22:20 – Benchmarking success on social platforms: Shifting from vanity metrics to aggregate long-term trends.23:45 – Case Study: How focusing on a niche topic like CAPA can generate a massive, passionate industry following.Quotes"First be human. First be a human being... I think that it's very difficult to humanize the kind of in-depth regulatory, quality, risk discussions that we have... and so we're trying to make this information accessible, understandable, and keep it at a human scale." - Sean Smith"The European audience wants the same thing that everybody wants. They want to know how to make more money and how to stay out of trouble. Those are the only two things that motivate human beings. So if you can tether the story that you're telling to one of those two things... people don't like making mistakes." - Sean SmithTakeawaysRegulatory & Quality AssuranceCommoditization of "How-To" Content: Baseline regulatory explanations (e.g., standard steps to file a 510(k)) are rapidly becoming digitized and automated by LLMs. True value lies in sharing your unique perspective, industry misconceptions, and nuanced edge cases rather than boilerplate text.Medical Device R&DDocument the Daily Defenses: Engineers and developers provide value continuously through emails, internal PowerPoint presentations, and custom proposals. Capture these unique problem-solving workflows systematically to build a library of narrative proof points.Marketing & SalesDrop the Pitch-Slap: Avoid aggressive, transactional sales messaging on ...
    続きを読む 一部表示
    45 分
  • #465: Why Good Medical Devices Fail: Reimbursement Strategy with Ali Samiian

    #465: Why Good Medical Devices Fail: Reimbursement Strategy with Ali Samiian
    2026/07/20
    In this episode, host Etienne Nichols sits down with Ali Samiian, founder and managing principal of Popular Access Advisors, to demystify the critical and often misunderstood world of MedTech reimbursement. Far too many early-stage medical device companies treat reimbursement as a secondary, post-launch paperwork exercise, only to find that their brilliant, FDA-cleared technology fails because no one has figured out who will pay for it. Ali draws on his 20-plus years of experience in market access, health economics, and executive leadership to explain why reimbursement must be treated as a core product strategy long before submission.The conversation explores how commercialization pathways are inherently dictated by the site of care—whether inpatient, outpatient, or home use. Ali highlights common and costly pitfalls, such as designing clinical trials solely for FDA clearance while neglecting the specific evidence endpoints that insurance payers demand. Payers do not just look at safety and efficacy; they look at long-term clinical value, accessibility, durability, and standard-of-care comparisons. Using practical, real-world examples, Ali demonstrates how simple adjustments to product design and clinical study lengths can proactively align a device with existing or novel code requirements.Finally, the episode highlights the shifting regulatory landscape and new initiatives designed to accelerate market access for breakthrough innovations. Etienne and Ali discuss the FDA’s Total Lifecycle Product Advisory (TAP) program and the emerging CMS Regulatory Alignment for Predictable and Immediate Device (RAPID) program. By understanding these frameworks and embedding payer-relevant outcomes into early-stage research, innovators can significantly compress their revenue cycles, avoid product redesigns, and successfully deliver life-changing technologies into the hands of patients.Key Timestamps00:03 - Introduction to MedTech reimbursement and guest Ali Samiian.01:46 - The difference between clearing the FDA bar and achieving commercial success.02:16 - Case Study: How product design and classification categories impact commercialization.03:45 - The risk of ignoring durable medical equipment (DME) requirements during design.05:22 - Categorizing sites of care: Inpatient, outpatient, ASCs, and home use (DMEPOS).06:15 - Mistake #1: Postponing reimbursement strategy until after FDA approval.07:35 - Mistake #2: Designing clinical studies for the FDA without considering payer-relevant endpoints.09:02 - Understanding standard of care, durability data, and minimizing study bias for payers.10:30 - Exploring the FDA's TAP program and the new CMS RAPID program for breakthrough devices.12:15 - Mistake #3: Rushing regulatory pathways without assessing commercial and price-point implications.13:50 - Identifying stakeholders and understanding the oblique nature of CMS and payer regulations.14:38 - Deconstructing how to build new codes and establish premium pricing from scratch.Quotes"Clearing the FDA bar is not really what gets us to commercialization. We need to have product and clinical differentiation... A lot of reimbursement is actually more of a strategy exercise." - Ali Samiian"The FDA basically looks at is the product safe and effective? Payers look at is there value and is there accessibility for the product?" - Ali SamiianTakeawaysIncorporate Payer Endpoints Early: MedTech innovators should involve a reimbursement advisor during clinical trial design to incorporate payer-relevant endpoints (like durability and standard-of-care comparisons), avoiding the need for an expensive second study.Align Design with Code Descriptors: Ensure product features and testing durations match the strict regulatory definitions of your target site of care (e.g., verifying a home-use device meets the three-year durability testing threshold for DME classification).Evaluate Pathways Holistically: Assess regulatory pathways (510(k) vs. De Novo vs. PMA) not just by upfront cost or speed to market, but by their long-term implications on pricing, coding, and time-to-reimbursement.Leverage Breakthrough Programs: Companies with breakthrough device designation should actively follow and align with collaborative initiatives like the FDA's TAP and CMS's RAPID programs to secure immediate coverage upon clearance.ReferencesFDA TAP Program: The Total Lifecycle Product Advisory program designed to provide early, strategic communication with senior FDA leadership for breakthrough devices.CMS RAPID Program: Regulatory Alignment for Predictable and Immediate Device program, an initiative aimed at accelerating coverage pathways for breakthrough innovations.Etienne Nichols: Connect with the host on LinkedIn.MedTech 101 SectionReimbursement vs. FDA Clearance: Think of FDA clearance as getting a driver's license—it proves your device is safe to be on the road. Reimbursement is like getting a toll pass; it determines who is actually going to pay for the ...
    続きを読む 一部表示
    39 分
  • #466: Leaving the Ivory Tower - How Notified Body Engagement Unlocks Startup Growth

    #466: Leaving the Ivory Tower - How Notified Body Engagement Unlocks Startup Growth
    2026/07/27
    Many early-stage medical device founders face a major dilemma: how do you prove regulatory maturity to investors and partners before you actually hold a final CE mark or FDA approval? Waiting until the end of a long development cycle creates significant commercial risk. By treating regulatory readiness as a continuous maturity process rather than an all-or-nothing milestone, startups can build trust early and avoid costly late-stage surprises.In this episode, host Etienne Nichols sits down with Malte Knowles Schmidt, Global Portfolio Lead for Medical Device Software, AI, and Cybersecurity at TÜV SÜD. Drawing from his background as an ICU nurse, an R&D product leader for Class III cardiac implantables at Biotronik, and now a notified body leader, Malte breaks down how startups can step out of their silos. He emphasizes that notified bodies and medical device companies must both "leave the ivory tower" to establish practical, real-world communication long before a formal audit occurs.The conversation explores concrete strategies for demonstrating regulatory maturity during development, including standing up an enterprise Quality Management System (eQMS), securing partial ISO 13485 certification for core design processes, and leveraging external testing as strategic "breadcrumbs." Malte also cautions founders against chasing "Pyrrhic certifications"—winning regulatory approval at the cost of commercial viability—and shares practical guidance on when and how to initiate early structured dialogues with notified bodies.Key Timestamps00:00 - 02:15 | Introduction to the "Regulatory Ivory Tower"Etienne introduces guest Malte Knowles Schmidt and sets up the challenge of proving regulatory maturity early in the startup lifecycle.02:16 - 05:04 | Why Communication Gap Exists Between Startups & Notified BodiesMalte discusses why structured dialogues often feel too abstract for founders and why concrete examples are needed to make early engagement approachable.05:05 - 08:30 | Defining Regulatory Maturity & The Agile QMSExploring how investors view regulatory progress, why build-measure-learn mindsets belong inside a QMS, and how an eQMS acts as a foundational framework.08:31 - 12:10 | Unlocking Value Through Partial ISO 13485 CertificationA breakdown of how certifying core design and development processes early builds commercial credibility and attracts investor funding before full scope audits.12:11 - 16:45 | Leveraging External Testing & Avoiding "Pyrrhic Certifications"How penetration testing and biocompatibility act as evidence breadcrumbs, plus the trap of sacrificing commercial viability just to get a certificate.16:46 - 20:30 | When and How to Initiate Contact with a Notified BodyPractical advice for founders on overcoming the fear of reaching out, preparing essential homework (intended purpose, risk classification), and taking the first step.Quotes"A Pyrrhic certification comes from a Pyrrhic victory, where you win the battle, but the losses are so great that the victory is basically meaningless... startups make the certification their core goal and not the commercial success." - Malte Knowles Schmidt"If you're asking yourself the question, 'Should I be talking to a notified body?'—then stop right there, because the answer is yes." - Malte Knowles SchmidtKey TakeawaysEstablish Early Regulatory Breadcrumbs: Investors want to see continuous progression. Utilizing an eQMS, conducting third-party testing (e.g., penetration or biocompatibility testing), and mapping regulatory roadmaps provide tangible proof of maturity long before final approval.Consider Partial ISO 13485 Certification: Startups do not need to wait for full scope certification. Certifying core design and development processes first demonstrates organizational discipline and can unlock major funding rounds.Avoid the Pyrrhic Certification Trap: Do not sacrifice your core business model or reduce critical product capabilities solely to make certification easier. Always align regulatory strategy with ultimate commercial viability.Treat Your QMS as an Agile System: Quality management is not a static set of restrictive rules; it is an iterative framework that should evolve alongside your product and team processes.Initiate Dialogue Early: Notified bodies are accessible for preliminary discussions. Reaching out early helps validate your intended purpose, risk classification, and submission assumptions before sinking capital into the wrong pathway.ReferencesWhite Paper: Leaving the Regulatory Ivory Tower: How Early Notified Body Dialogues Reduce Business Risk by Malte Knowles Schmidt (TÜV SÜD).ISO 13485 Standard: Quality management systems requirements for regulatory purposes in the medical device sector.Host LinkedIn: Etienne Nichols on LinkedInMedTech 101 SectionPyrrhic Certification: Named after King Pyrrhus of Epirus, whose army won a battle against the Romans but suffered devastating losses in the process. In MedTech, a Pyrrhic ...
    続きを読む 一部表示
    32 分
  • #467: Combination Product Compliance: PMOA, 21 CFR Part 4 & QMS Alignment

    #467: Combination Product Compliance: PMOA, 21 CFR Part 4 & QMS Alignment
    2026/08/03
    Navigating the regulatory landscape for combination products requires understanding how primary modes of action (PMOA) dictate oversight pathways. In this episode, host Etienne Nichols sits down with Jim Fentress, Director of Regulatory Affairs at Galero (a Santa Group company), to unpack the structural differences and hidden pitfalls when medical device and pharmaceutical worlds collide. They discuss how the FDA handles lead agency designation across CDRH and CDER using interagency agreements and official Requests for Designation (RFD).A central theme of the discussion is managing quality management systems under 21 CFR Part 4. The pair explore the friction that occurs when pharmaceutical companies act as lead applicants for drug-led combination products, requiring them to incorporate device design controls (ISO 13485 / QMSR) and CAPA systems into their existing CGMP framework. Jim explains the practical realities of integrating Part 210/211 elements—such as calculation of yield and stability testing—into a single, operational QMS without overcomplicating procedures.Finally, the conversation delves into critical execution details: risk management under AAMI TIR105 (ISO 14971 vs. ICH Q9), labeling classifications (single entity, co-packaged, and cross-labeled), and strict change control protocols. Jim highlights how post-market design changes to a device constituent part can impact the pharmaceutical partner's NDA or baseline regulatory filings, underscoring the necessity of transparent cross-industry communication from initial development through full commercial release.Key Timestamps00:00 – Introduction to combination products and guest Jim Fentress.00:48 – Understanding Primary Mode of Action (PMOA) and regulatory pathways (FDA vs. European authorities).01:57 – FDA interagency agreements (CDRH and CDER) and Requests for Designation (RFD).03:00 – 21 CFR Part 4 quality system integration (CGMP Part 210/211 and QMSR/Part 820).05:22 – Navigating the communication gap between pharma companies and device manufacturers.07:44 – Calculation of yield in drug manufacturing vs. medical device production.09:05 – Risk management for combination products (AAMI TIR105: evaluating device-on-drug and drug-on-device risks).12:15 – Bridging ISO 14971 and ICH Q9 framework structures in registration files.13:16 – Design controls, user needs, and human factors validation (Module 5 / Section 3.2.R ECTD filings).15:06 – Labeling pathways: Single Entity (Integral), Co-Packaged, and Cross-Labeled products.18:18 – Change control risks: How minor device modifications affect drug application filings (NDAs, CBER/CDER supplements).20:41 – Advice for device manufacturers partnering with pharma: Alignment on risk, documentation depth, and cleanroom requirements.Standout Quotes"There's four aspects of risk that you need to take into account: what is the risk of the drug alone, the risk of the delivery system alone, the risk of the drug on the device, and the risk of the device on the drug." — Jim Fentress"Before you even think about making a change, you need to talk to your pharmaceutical partners because now what's represented as the co-packaged device constituent element is changing, and they need to inform the FDA." — Jim FentressActionable TakeawaysEstablish Cross-Disciplinary Risk Management Early: Adopt frameworks like AAMI TIR105 to integrate traditional device risk protocols (ISO 14971) with pharmaceutical risk management (ICH Q9). Ensure assessment of cross-interaction hazards (e.g., drug interactions with delivery plastics, viscous drug effects on ejection times).Define Clear Part 4 QMS Interfaces: If operating primarily under device rules (QMSR/ISO 13485), build project-specific addenda to account for drug CGMP requirements such as stability testing, container-closure assessments, and calculation of yield limits.Align Post-Market Change Control Protocols: Establish explicit notification procedures between the device supplier and the NDA holder. Simple component material updates or geometry changes to a constituent part may require formal NDA supplements or changes-being-effected (CBE) filings with CDER.Scope Document Deliverables Upfront: Clarify whether the pharmaceutical partner requires high-level summary reports or the complete device master record (DMR) and design history file (DHF) to populate Section 3.2.R of their eCTD submission.Validate Cleanroom and Sterility Assumptions: Discuss cleanroom requirements early to avoid unnecessary cost structures; verify if an ISO 8 or ISO 7 environment is scientifically required for the assembly of non-sterile device constituents before adopting conservative pharma-grade aseptic norms (ISO 5).Essential References21 CFR Part 4: Regulation governing current good manufacturing practice (CGMP) requirements for combination products.AAMI TIR105: Technical Information Report providing guidance on the application of risk management to combination ...
    続きを読む 一部表示
    36 分
  • #461: Why Manufacturing is Part of Product Development with Mike Dolphin

    #461: Why Manufacturing is Part of Product Development with Mike Dolphin
    2026/06/22
    The traditional approach to medical device commercialization often treats manufacturing as a distinct, isolated step executed after the design phase is completed. In this episode, Mike Dolphin, CEO of GuideStar Medical Devices, challenges this linear mindset by arguing that manufacturing process development is fundamentally an extension of product development itself. Drawing from his unique background spanning aerospace engineering at JPL, scientific research, and medical device ventures, Dolphin shares how upfront constraints shape a more predictable path to market.The conversation centers heavily around the engineering and clinical challenges of epidural anesthesia delivery, a high-consequence procedure historically reliant entirely on a physician's tactile sense. Dolphin details how his company approached this clinical risk profile by designing a closed-loop system capable of automatically stopping a needle upon sensing the epidural space. By establishing critical manufacturing constraints—such as choosing injection-molded plastics and radiation sterilization from day one—the design team avoided the common trap of engineering a prototype that cannot be scaled.Additionally, the episode dives into the practical friction between tight physical tolerances and production realities, showcasing a creative approach to mold development that bypasses typical vendor limitations. Dolphin also shares his perspective on balancing rigorous documentation with early-stage agility, warning founders against premature lock-down of design controls within a Quality Management System (QMS). Ultimately, the discussion underscores that true commercial readiness requires a unified view where the final product and the manufacturing pipeline are developed in parallel.Key Timestamps00:41 – Guest introduction: Mike Dolphin’s transition from aerospace engineering at JPL to MedTech leadership.02:02 – Cross-industry lessons: How regulatory oversight, documentation, and system thinking in aerospace translate directly to medical device design.03:02 – The clinical problem: Demystifying the high-consequence risks of epidural anesthesia, including accidental dural puncture and nerve damage.05:14 – Engineering an actuator: Shifting from the clinical request for "better sensors" to building a closed-loop mechanical system.07:34 – Epidural procedure metrics: The market scale of labor, delivery, and chronic pain injections in the US and globally.09:47 – Integrating manufacturing early: Why sterilization and material choices must be established during initial requirements gathering.12:02 – Common founder pitfalls: The danger of designing a product looking for a problem versus evaluating cost, market size, and manufacturability from the start.13:58 – The documentation vs. QMS overhead balance: Knowing when to record choices and when to formally lock down design controls to preserve startup capital.16:47 – Overcoming injection molding tolerance limitations: A case study on utilizing first principles physics and progressive mold variations to achieve a 10-micron output consistency.21:04 – Managing manufacturing consistency: Dealing with brittle plastic runs, operator variances, and securing lines against unauthorized process shortcuts.22:25 – Impact on the 510(k) pathway: Defining commercial readiness as manufacturing readiness for final finished product submissions.Quotes"Having worked in aerospace and in medical device, I can say that this is harder than launching rockets." — Mike Dolphin"Manufacturing is part of development in medical devices. You develop your product, you develop a prototype that works. Now you need to develop your manufacturing process. That takes time, that takes real engineering and real know-how." — Mike DolphinTakeawaysIntegrate Manufacturing Into R&D: Do not treat manufacturing as a post-development handoff. Developing the manufacturing pipeline is a core engineering activity required to establish a fully validated, commercial-ready device.Establish Production Constraints Early: Define your sterilization methods, primary materials, and fabrication methods (e.g., injection molding) during initial requirement generation to restrict the design space and eliminate unproducable prototypes.Leverage First Principles for Tolerances: When manufacturing vendors claim tight tolerances are impossible due to material shrinkage, analyze the underlying physical limitations. Strategies like building progressive progressive molds can deliver highly consistent micro-level outputs.Audit Process Consistency: Component quality depends entirely on process parameters. Even with identical raw materials, minor adjustments to cycle times or cooling rates by different operators can alter material properties like brittleness.De-risk the 510(k) With Finished Production Runs: Because a 510(k) submission requires testing on the final finished product, achieving manufacturing readiness is the critical path to ...
    続きを読む 一部表示
    47 分
  • #464: Why LLM Unpredictability is a Liability in MedTech

    #464: Why LLM Unpredictability is a Liability in MedTech
    2026/07/13
    Artificial intelligence has officially entered the mainstream cultural zeitgeist, creating a wave of excitement—and a fair share of fatigue—across the medical device industry. In this episode, host Etienne Nichols sits down with Tyler Harmon, biomedical engineer and CEO of Iaso Automated Medical Systems, to cut through the marketing buzzwords. Together, they explore the technical realities behind the technology stack, shifting the conversation away from generic AI toward specific, actionable engineering frameworks.The discussion highlights a critical distinction between traditional machine learning models and consumer-oriented Large Language Models (LLMs). Harmon explains that while technologies like convolutional neural networks (CNNs) have successfully processed medical imaging for years, modern LLMs introduce an intentional element of randomness to mimic human conversation. This lack of predictability presents unique challenges for medical device developers who operate in a deterministic, safety-critical environment where reproducibility is paramount.Looking toward practical deployment, the episode addresses how companies can responsibly govern these tools both within their software architectures and their internal Quality Management Systems (QMS). From classifying external AI models as Software of Unknown Provenance (SOUP) under IEC 62304 to leveraging machine learning for early detection of Acute Respiratory Distress Syndrome (ARDS) in the ICU, this conversation serves as an essential guide for innovators looking to build the next generation of safe, compliant, and effective medical technologies.Key Timestamps00:05 – Introduction to the dual nature of AI in MedTech: embedded clinical algorithms versus internal process optimization.02:14 – Demystifying the math: Breaking down artificial intelligence into linear and non-linear algorithmic transformations.04:30 – The Turing Test, Markov chains, and why consumer LLMs are mathematically designed to be unpredictable.07:15 – Real-world success stories: How convolutional neural networks (CNNs) revolutionized emergency stroke triage.09:42 – Inside Iaso Automated Medical Systems: Using non-LLM machine learning to identify Acute Respiratory Distress Syndrome (ARDS) in critical care.12:10 – AI Governance in the QMS: Designing specialized Standard Operating Procedures (SOPs) and Machine Learning Management Systems (AIMS).15:35 – Evaluating recent FDA 510(k) clearances for LLM-adjacent technologies and managing third-party stacks as SOUP.Quotes"If we as innovators can't explain things to a more general audience, we generally don't understand them ourselves. And if you can't do that, it's probably not the best idea to be implementing it into your products." - Tyler Harmon"I am probably going to be the biggest advocate you'll ever talk to about 'doctors need enablement, not replacement.' We need to give them the tools, the force multipliers to tackle the challenges they're going to face this century." - Tyler HarmonTakeawaysClassify External AI as SOUP: Treat third-party language models and external tech stacks as Software of Unknown Provenance (SOUP) under IEC 62304 frameworks, implementing rigorous risk management boundaries to isolate the core medical device logic.Engineer Out Randomness: Recognize that consumer LLMs purposefully integrate randomness layers to maximize user engagement. For clinical safety, developers must utilize architectural harnesses or alternative machine learning methods (like CNNs or random forests) to force more deterministic outcomes.Establish an AI Management System: Expand your organizational compliance beyond standard Quality Management Systems (QMS) and Information Security Management Systems (ISMS). Implement specific AI standard operating procedures and work instructions to govern internal token usage and data handling.Prioritize Clinical Enablement Over Automation: Focus clinical software engineering on clearing workflow bottlenecks and flagging early-stage critical conditions (such as ARDS) to allow bedside clinicians to deploy their hands-on expertise faster.ReferencesBerlin Criteria: The formal, quantitative medical classification standard used by clinicians to diagnose and grade the severity of Acute Respiratory Distress Syndrome.IEC 62304: The international standard governing medical device software lifecycle processes, specifically detailing the management of Software of Unknown Provenance (SOUP).Connect with Etienne Nichols on LinkedIn to stay updated on the latest episodes and industry insights.MedTech 101 SectionUnderstanding Non-Linear Math and LLMsThink of a traditional medical device software algorithm like a standard thermometer tracking a fever. It follows a straight, predictable line: if the temperature input increases by one degree, the reading on the screen changes by exactly one degree. This is a linear system.Modern AI, like Large Language Models (LLMs), works more like a seasoned doctor ...
    続きを読む 一部表示
    53 分
  • #460: FDA AI Regulations: Master the QA/RA Skills to Stay Ahead

    #460: FDA AI Regulations: Master the QA/RA Skills to Stay Ahead
    2026/05/18
    The FDA is actively shaping the regulatory landscape for Artificial Intelligence (AI) and Machine Learning (ML) in real time. As the agency expands its internal expertise through the Digital Health Center of Excellence, FDA reviewers are becoming highly sophisticated. The era of submitting vague algorithm descriptions is over, paving the way for a more level playing field that rewards companies executing documentation correctly.Navigating this evolving space requires a dual-front approach for global medical device companies. Manufacturers must balance the FDA's framework with the EU AI Act, which classifies AI medical devices as high-risk systems demanding rigorous conformity assessments and human oversight. Fortunately, a robust quality management system designed around proactive frameworks, such as the Predetermined Change Control Plan (PCCP), can bridge the gap between US and international expectations.For Quality Assurance and Regulatory Affairs (QA/RA) professionals, this shift represents an unprecedented career opportunity. The future belongs to those who combine regulatory fluency with AI literacy. Success in the MedTech industry will not belong solely to the most complex algorithm, but to the companies and professionals who build compliant, disciplined systems around their AI technologies.Key Timestamps00:19 – Introduction to the current state of FDA AI regulation and leadership transitions.01:34 – The role of the FDA Digital Health Center of Excellence and shifting reviewer expectations.02:08 – Navigating global regulations: Balancing the EU AI Act and EU MDR.02:46 – The 5 guiding principles for AI/ML-based Software as a Medical Device (SaMD).03:41 – Analyzing FDA warning letters: Why documentation takes precedence over algorithm performance.04:19 – Bridging the language barrier between AI engineers and FDA reviewers in submissions.05:27 – The future of QA/RA careers: The rising demand for AI-literate regulatory professionals.06:21 – Actionable strategies to stay ahead: Implementing PCCPs early and training quality teams.07:23 – Treating post-market surveillance for AI products as an evolving product lifecycle.Quotes"The companies getting in trouble aren't the ones with bad AI, they're the ones with incomplete quality systems." - Etienne Nichols"Your job in a regulatory submission is not to demonstrate that your AI is sophisticated. Your job is to demonstrate that it's safe and effective in its intended use." - Etienne NicholsTakeawaysBuild Your PCCP First: Develop your Predetermined Change Control Plan (PCCP) concurrently with or prior to algorithm development to ensure post-clearance modifications match your design process.Close the Team Knowledge Gap: Educate quality engineering teams on fundamental AI concepts like training data, validation datasets, and demographic representation before facing regulatory audits.Proactively Audit Your DHF: Review your existing Design History File (DHF) against current FDA AI guidance documents well ahead of submission deadlines to eliminate documentation gaps without timeline pressure.Evolve Post-Market Surveillance: Treat your AI post-market surveillance plan as a living product by implementing version control, clear ownership, and defined thresholds to detect algorithm drift.Achieve Dual Literacy for Career Growth: QA/RA professionals who master both regulatory frameworks and basic AI literacy will position themselves at the top of an uncrowded talent pool.ReferencesFDA, Health Canada, & UK MHRA Joint Statement (2022): The five joint guiding principles established for machine learning medical device development.FDA AI/ML Action Plan (2021) & PCCP Guidance (2023): Core foundational reading material for understanding regulatory expectations.International Medical Device Regulators Forum (IMDRF) Guidance: Global harmonized guidelines concerning AI/ML-based SaMD.EU AI Act: High-risk classification rules and conformity requirements affecting medical software in Europe.Connect with the Host: Follow Etienne Nichols on LinkedIn for more MedTech insights and discussion.MedTech 101 SectionOverfittingThink of overfitting like a student who memorizes the exact questions and answers on a practice exam instead of learning the underlying concepts. When they take the real test with slightly altered questions, they fail. In AI, overfitting happens when an algorithm learns the training data too perfectly, making it excellent at analyzing that specific dataset but unable to make accurate predictions on new patient data.Algorithm DriftImagine a GPS map app that was programmed perfectly five years ago. Over time, new roads are built, traffic patterns change, and old exits close. If the app is never updated, its navigation becomes less accurate. Algorithm drift occurs when an AI medical device becomes less effective over time because the real-world clinical environment or patient demographics shift away from the original data it was trained on.SponsorsThis ...
    続きを読む 一部表示
    15 分