In Silico Development in the Medical Device Industry
In Silico Medical Device Regulation Reaches Global Inflection Point
The regulatory landscape for computational modeling in medical device development has undergone significant transformation as of January 2026. The FDA's November 2023 credibility framework remains the global gold standard, but the European Commission's December 2025 MDR simplification proposal explicitly legitimizing in silico evidence and Japan's aggressive digital health reforms signal a fundamental shift toward computational evidence acceptance worldwide.
The most consequential development is the EU's COM(2025) 1023 proposal, which for the first time formally recognizes "in silico testing, computational modeling or simulation" as acceptable non-clinical evidence under Article 61—a significant departure from the EU's historically conservative stance. Meanwhile, the FDA's MDDT program qualified its first mechanical testing computational model in May 2025, demonstrating practical regulatory acceptance of physics-based simulations as replacements for physical bench testing.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🇺🇸 FDA 🇪🇺 EU 🇯🇵 JAPAN 🌍 IMDRF ║
║ ════════ ════════ ════════ ════════ ║
║ ║
║ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ ║
║ │▓▓▓▓▓▓▓▓▓▓│ │▒▒▒▒▒▒▒▒▒▒│ │░░░░░░░░░░│ │██████████│ ║
║ │▓▓▓▓▓▓▓▓▓▓│ │▒▒▒▒▒▒▒▒▒▒│ │░░░░░░░░░░│ │██████████│ ║
║ │▓▓ GOLD ▓▓│ │▒PARADIGM▒│ │░ DIGITAL░│ │█HARMONI-█│ ║
║ │▓STANDARD▓│ │▒▒ SHIFT▒▒│ │░ ACCEL ░│ │█ ZATION █│ ║
║ │▓▓▓▓▓▓▓▓▓▓│ │▒▒▒▒▒▒▒▒▒▒│ │░░░░░░░░░░│ │██████████│ ║
║ │▓▓▓▓▓▓▓▓▓▓│ │▒▒▒▒▒▒▒▒▒▒│ │░░░░░░░░░░│ │██████████│ ║
║ └──────────┘ └──────────┘ └──────────┘ └──────────┘ ║
║ ║
║ Nov 2023 Dec 2025 Jan 2024 Jan 2025 ║
║ ✅ FINAL ⏳ PENDING ✅ ACTIVE ✅ FINAL ║
║ ║
║ ║ ║ ║ ║ ║
║ ╚═══════════════════╩═══════════════════╩═══════════════════╝ ║
║ │ ║
║ ▼ ║
║ ┌────────────────────────────────────────┐ ║
║ │ 🔗 CONVERGENCE TOWARD GLOBAL ACCEPTANCE │ ║
║ │ OF COMPUTATIONAL EVIDENCE │ ║
║ └────────────────────────────────────────┘ ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
███████╗██████╗ █████╗
██╔════╝██╔══██╗██╔══██╗
█████╗ ██║ ██║███████║ 🇺🇸 UNITED STATES
██╔══╝ ██║ ██║██╔══██║
██║ ██████╔╝██║ ██║
╚═╝ ╚═════╝ ╚═╝ ╚═╝
FDA Maintains Leadership with Stable Framework
The FDA's computational modeling regulatory framework centers on the November 16, 2023 final guidance "Assessing the Credibility of Computational Modeling and Simulation in Medical Device Submissions" (Document GUI01500056, Federal Register 88 FR 80314, Docket FDA-2021-D-0980). No revisions or new CM&S-specific guidance documents have been published in 2025-2026, reflecting regulatory maturity rather than stagnation.
The guidance establishes a nine-step credibility assessment framework applicable to first principles-based, physics-based, and mechanistic computational models. Critically, it explicitly excludes AI/ML models—these fall under the separate January 6, 2025 draft guidance on AI-enabled device software functions. The framework introduces model risk as the product of model influence (contribution relative to other evidence) and decision consequence (significance of adverse outcomes).
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🔬 9-STEP CREDIBILITY ASSESSMENT FRAMEWORK 🔬 ║
║ ║
║ ┌─────┐ ┌─────┐ ┌─────┐ ┌─────┐ ┌─────┐ ║
║ │ 🟦 │───▶│ 🟦 │───▶│ 🟦 │───▶│ 🟦 │───▶│ 🟦 │ ║
║ │ 1 │ │ 2 │ │ 3 │ │ 4 │ │ 5 │ ║
║ │ QOI │ │ COU │ │ REQ │ │V&V │ │ UQ │ ║
║ └─────┘ └─────┘ └─────┘ └─────┘ └─────┘ ║
║ │ │ │ │ │ ║
║ │ │ │ │ │ ║
║ ┌─────┐ ┌─────┐ ┌─────┐ ┌─────┐ ║
║ │ 🟩 │───▶│ 🟩 │───▶│ 🟩 │───▶│ 🟩 │ ║
║ │ 6 │ │ 7 │ │ 8 │ │ 9 │ ║
║ │ DOC │ │RISK │ │ REV │ │ DEC │ ║
║ └─────┘ └─────┘ └─────┘ └─────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ 1. QOI = Quantity of Interest 6. DOC = Documentation │ ║
║ │ 2. COU = Context of Use 7. RISK = Risk Assessment │ ║
║ │ 3. REQ = Requirements 8. REV = Review Process │ ║
║ │ 4. V&V = Verification/Validation 9. DEC = Decision │ ║
║ │ 5. UQ = Uncertainty Quantification │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ╭─────────────────────────────────────────────────╮ ║
║ │ ⚖️ MODEL RISK EQUATION ⚖️ │ ║
║ │ │ ║
║ │ ┌─────────────────┐ ┌──────────────────┐ │ ║
║ │ │ MODEL INFLUENCE │ × │ DECISION │ │ ║
║ │ │ (% of total │ │ CONSEQUENCE │ │ ║
║ │ │ evidence) │ │ (patient impact) │ │ ║
║ │ └─────────────────┘ └──────────────────┘ │ ║
║ │ │ │ │ ║
║ │ └────────┬───────────┘ │ ║
║ │ ▼ │ ║
║ │ ┌─────────────────┐ │ ║
║ │ │ MODEL RISK │ │ ║
║ │ └─────────────────┘ │ ║
║ ╰─────────────────────────────────────────────────╯ ║
║ ║
║ 📌 SCOPE: Physics-based, mechanistic, first-principles models ║
║ 🚫 EXCLUDES: AI/ML models → separate Jan 2025 guidance ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
MDDT Qualification Milestone: May 2025
A landmark MDDT qualification occurred in May 2025: the ENDPOINT numaScrew Virtual Pullout Test (Submission U240247), developed by Sawbones and Numalogics, became the first FDA-qualified mechanical test for an orthopedic device. This finite element analysis tool simulates screw pullout behavior conforming to ASTM F543 standards, validating computational predictions as surrogates for physical testing.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🏆 FIRST FDA-QUALIFIED MECHANICAL TEST FOR ORTHOPEDICS 🏆 ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 📋 ENDPOINT - numaScrew Virtual Pullout Test │ ║
║ │ ════════════════════════════════════════════ │ ║
║ │ │ ║
║ │ Submission: U240247 │ ║
║ │ Qualified: May 2025 │ ║
║ │ Developers: Sawbones + Numalogics Inc. │ ║
║ │ Standard: ASTM F543 (screw pullout testing) │ ║
║ │ │ ║
║ │ ┌───────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ ░░░░░░░░░░░░░ │ │ ║
║ │ │ ░░░ ░░░ 🦴 BONE FOAM BLOCK │ │ ║
║ │ │ ░░░ ╔══════╗ ░░░ (20 PCF polyurethane) │ │ ║
║ │ │ ░░░ ║██████║ ░░░ │ │ ║
║ │ │ ░░░ ║██████║ ░░░ 🔩 METALLIC BONE SCREW │ │ ║
║ │ │ ░░░ ║██████║ ░░░ (FEA mesh model) │ │ ║
║ │ │ ░░░ ║██████║ ░░░ │ │ ║
║ │ │ ░░░ ╚══════╝ ░░░ ▲ │ │ ║
║ │ │ ░░░ ░░░ │ │ │ ║
║ │ │ ░░░░░░░░░░░░░ │ PULLOUT │ │ ║
║ │ │ │ FORCE │ │ ║
║ │ │ │ │ ║
║ │ └───────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ │ 📊 PERFORMANCE METRICS │ ║
║ │ ┌────────────────────────┬────────────────────────────────────┐ │ ║
║ │ │ Predictive ICC │ 0.979 (excellent reliability) │ │ ║
║ │ ├────────────────────────┼────────────────────────────────────┤ │ ║
║ │ │ Mean Absolute Error │ 8.0% │ │ ║
║ │ ├────────────────────────┼────────────────────────────────────┤ │ ║
║ │ │ vs Chapman Formula │ 23.2% MAE (computational wins) │ │ ║
║ │ ├────────────────────────┼────────────────────────────────────┤ │ ║
║ │ │ Spearman Correlation │ 0.985 │ │ ║
║ │ └────────────────────────┴────────────────────────────────────┘ │ ║
║ │ │ ║
║ │ ✅ SIGNIFICANCE: First computational tool qualified to REPLACE │ ║
║ │ (not just supplement) traditional physical bench testing │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
Other Qualified Computational MDDTs
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 📋 FDA-QUALIFIED COMPUTATIONAL MDDTS (As of January 2026) ║
║ ║
║ ┌──────────────┬─────────────────────────────────────┬──────────────┐ ║
║ │ DATE │ TOOL │ CATEGORY │ ║
║ ├──────────────┼─────────────────────────────────────┼──────────────┤ ║
║ │ May 2021 │ IMAnalytics with MRIxViP │ MRI Safety │ ║
║ │ │ (active implanted devices) │ │ ║
║ ├──────────────┼─────────────────────────────────────┼──────────────┤ ║
║ │ Nov 2021 │ Virtual MRI Safety Evaluations │ MRI Safety │ ║
║ │ │ (MED Institute) │ │ ║
║ ├──────────────┼─────────────────────────────────────┼──────────────┤ ║
║ │ Mar 2023 │ MRI Temperature Rise Prediction │ Thermal │ ║
║ │ │ (orthopedic implants) │ │ ║
║ ├──────────────┼─────────────────────────────────────┼──────────────┤ ║
║ │ May 2025 │ numaScrew Virtual Pullout Test │ Mechanical │ ║
║ │ │ (Sawbones/Numalogics) 🆕 │ │ ║
║ └──────────────┴─────────────────────────────────────┴──────────────┘ ║
║ ║
║ 🔔 HeartFlow K250902 (Sep 2025): CFD + AI-powered 3D coronary modeling ║
║ for fractional flow reserve assessment ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
███████╗██╗ ██╗██████╗ ██████╗ ██████╗ ███████╗
██╔════╝██║ ██║██╔══██╗██╔═══██╗██╔══██╗██╔════╝
█████╗ ██║ ██║██████╔╝██║ ██║██████╔╝█████╗ 🇪🇺 EUROPEAN UNION
██╔══╝ ██║ ██║██╔══██╗██║ ██║██╔═══╝ ██╔══╝
███████╗╚██████╔╝██║ ██║╚██████╔╝██║ ███████╗
╚══════╝ ╚═════╝ ╚═╝ ╚═╝ ╚═════╝ ╚═╝ ╚══════╝
EU December 2025 Proposal: A Paradigm Shift
The European Commission published COM(2025) 1023 final on December 16, 2025, proposing amendments to Regulations (EU) 2017/745 (MDR) and 2017/746 (IVDR) that fundamentally alter the regulatory treatment of computational evidence.
The proposal explicitly amends Article 61 (Clinical Evaluation) to "allow additional non-clinical sources of data, in particular 'in vitro, ex vivo, in silico testing, computational modeling or simulation.'" This represents the first explicit EU-level recognition that manufacturers may rely on these non-clinical testing methods where duly justified, without necessarily requiring clinical investigation data.
A key conceptual addition is "New Approach Methodologies" (NAMs), introduced in Recital 38. NAMs are not limited to replacing animal testing but may be extended to clinical evaluations, enabling AI-powered computational approaches to contribute to traditionally clinical evidence requirements.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🇪🇺 COM(2025) 1023 - MDR/IVDR SIMPLIFICATION PROPOSAL 🇪🇺 ║
║ ║
║ Published: December 16, 2025 ║
║ Procedure: 2025/0404 (COD) - Ordinary Legislative Procedure ║
║ Expected Adoption: Q2 2027 ║
║ Feedback Period: Open until March 4, 2026 ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 📜 KEY AMENDMENTS │ ║
║ │ │ ║
║ │ ┌─────────────────────────────────────────────────────────────────┐ │ ║
║ │ │ ARTICLE 61 (Clinical Evaluation) │ │ ║
║ │ │ ═══════════════════════════════ │ │ ║
║ │ │ │ │ ║
║ │ │ "...allow additional non-clinical sources of data, │ │ ║
║ │ │ in particular: │ │ ║
║ │ │ │ │ ║
║ │ │ • in vitro testing │ │ ║
║ │ │ • ex vivo testing │ │ ║
║ │ │ • 🆕 in silico testing │ │ ║
║ │ │ • 🆕 computational modeling or simulation" │ │ ║
║ │ │ │ │ ║
║ │ └─────────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 🧪 NEW APPROACH METHODOLOGIES (NAMs) - Recital 38 │ ║
║ │ │ ║
║ │ ┌─────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ NAMs NOT limited to replacing animal testing │ │ ║
║ │ │ │ │ │ ║
║ │ │ ▼ │ │ ║
║ │ │ May be EXTENDED to clinical evaluations │ │ ║
║ │ │ │ │ │ ║
║ │ │ ▼ │ │ ║
║ │ │ AI-powered computational approaches can │ │ ║
║ │ │ contribute to clinical evidence requirements │ │ ║
║ │ │ │ │ ║
║ │ └─────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 📊 SUMMARY OF MDR/IVDR PROPOSAL IMPACTS ║
║ ║
║ ┌───────────────────────┬──────────────────────────────────────────────┐ ║
║ │ PROPOSAL ELEMENT │ REGULATORY IMPACT │ ║
║ ├───────────────────────┼──────────────────────────────────────────────┤ ║
║ │ Article 61 │ Computational modeling explicitly accepted │ ║
║ │ amendments │ as non-clinical evidence │ ║
║ ├───────────────────────┼──────────────────────────────────────────────┤ ║
║ │ NAMs framework │ In silico methods elevated to clinical │ ║
║ │ (Recital 38) │ evaluation contribution │ ║
║ ├───────────────────────┼──────────────────────────────────────────────┤ ║
║ │ Rule 11 changes │ Software starts as Class I, │ ║
║ │ │ up-classified based on intended use │ ║
║ ├───────────────────────┼──────────────────────────────────────────────┤ ║
║ │ Regulatory sandboxes │ New Articles 59b, 59c enable emerging │ ║
║ │ │ technology testing environments │ ║
║ ├───────────────────────┼──────────────────────────────────────────────┤ ║
║ │ Fee reductions │ 50%+ reduction for micro manufacturers │ ║
║ │ │ 25%+ reduction for small enterprises │ ║
║ └───────────────────────┴──────────────────────────────────────────────┘ ║
║ ║
║ 💰 Expected cost savings: €3.3 billion/year ║
║ ║
║ ⚠️ STATUS: PENDING - No dedicated MDCG guidance on computational ║
║ modeling credibility yet; practical implementation details undefined ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
██╗ █████╗ ██████╗ █████╗ ███╗ ██╗
██║██╔══██╗██╔══██╗██╔══██╗████╗ ██║
██║███████║██████╔╝███████║██╔██╗ ██║ 🇯🇵 JAPAN
██║██╔══██║██╔═══╝ ██╔══██║██║╚██╗██║
██║██║ ██║██║ ██║ ██║██║ ╚████║
╚═╝╚═╝ ╚═╝╚═╝ ╚═╝ ╚═╝╚═╝ ╚═══╝
Japan Accelerates Digital Health Acceptance
Japan's PMDA has not issued guidance equivalent to FDA's computational modeling credibility framework, evaluating in silico evidence on a case-by-case basis through consultation processes. However, the agency has aggressively modernized its digital health regulatory infrastructure.
The "Dash for SaMD2" program (launched January 2024) targets a six-month maximum review for all Software as a Medical Device submissions, with an expanded SaMD review team and subscription-based consultation services. PMDA's five-year plan (2024-2028) prioritizes digital health acceleration.
The IDATEN system ("Immediate Deployment and Acceleration of Technologies for the Nation") allows pre-approved improvement plans for AI-based SaMD, eliminating regulatory re-approval requirements for each adaptive AI update—a significant reduction in administrative burden for evolving devices.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🇯🇵 PMDA DIGITAL HEALTH INITIATIVES ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ ⚡ DASH FOR SaMD2 (January 2024) │ ║
║ │ ══════════════════════════════ │ ║
║ │ │ ║
║ │ ┌─────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ TARGET: 6-month maximum review for ALL SaMD │ │ ║
║ │ │ │ │ ║
║ │ │ ████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░ │ │ ║
║ │ │ ◄────── 6 MONTHS ──────► │ │ ║
║ │ │ │ │ ║
║ │ │ • Expanded SaMD review team │ │ ║
║ │ │ • Subscription-based consultation services │ │ ║
║ │ │ • Part of PMDA 5-year plan (2024-2028) │ │ ║
║ │ │ │ │ ║
║ │ └─────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 🤖 IDATEN SYSTEM │ ║
║ │ ═══════════════ │ ║
║ │ "Immediate Deployment and Acceleration of Technologies │ ║
║ │ for the Nation" │ ║
║ │ │ ║
║ │ ┌─────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ BEFORE IDATEN AFTER IDATEN │ │ ║
║ │ │ ═════════════ ════════════ │ │ ║
║ │ │ │ │ ║
║ │ │ AI Update #1 → Re-approval AI Update #1 ──┐ │ │ ║
║ │ │ AI Update #2 → Re-approval AI Update #2 ──┤ Pre- │ │ ║
║ │ │ AI Update #3 → Re-approval AI Update #3 ──┼─approved │ │ ║
║ │ │ AI Update #4 → Re-approval AI Update #4 ──┤ plan │ │ ║
║ │ │ AI Update #5 → Re-approval AI Update #5 ──┘ │ │ ║
║ │ │ │ │ ║
║ │ │ ❌ Repeated regulatory burden ✅ Single approval │ │ ║
║ │ │ │ │ ║
║ │ └─────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ 📅 IMDRF 28th Management Committee: Sapporo, Sept 15-19, 2025 ║
║ 📋 PMDA AI Action Plan: September 2025 ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
██╗███╗ ███╗██████╗ ██████╗ ███████╗
██║████╗ ████║██╔══██╗██╔══██╗██╔════╝
██║██╔████╔██║██║ ██║██████╔╝█████╗ 🌍 GLOBAL HARMONIZATION
██║██║╚██╔╝██║██║ ██║██╔══██╗██╔══╝
██║██║ ╚═╝ ██║██████╔╝██║ ██║██║
╚═╝╚═╝ ╚═╝╚═════╝ ╚═╝ ╚═╝╚═╝
IMDRF Finalizes Key Harmonization Documents
IMDRF N88 ("Good Machine Learning Practice for Medical Device Development"), finalized in January 2025, establishes ten guiding principles building on the 2021 FDA-Health Canada-MHRA joint framework.
IMDRF N81 ("Medical Device Software Considerations for Device and Risk Characterization"), also finalized January 2025, expands beyond traditional SaMD to include all medical device software including embedded software (SiMD), aligning terminology with EU MDR/IVDR "medical device software" (MDSW).
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 📜 IMDRF N88: GOOD MACHINE LEARNING PRACTICE (January 2025) ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 🔟 GUIDING PRINCIPLES FOR AI/ML MEDICAL DEVICES │ ║
║ │ │ ║
║ │ ┌─────┬─────────────────────────────────────────────────────────┐ │ ║
║ │ │ 1 │ Multi-disciplinary expertise throughout lifecycle │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 2 │ Good software engineering and security practices │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 3 │ Representative clinical study participants & datasets │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 4 │ Independent training and test datasets │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 5 │ Reference datasets based on best available methods │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 6 │ Model design tailored to available data │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 7 │ Focus on human-AI team performance │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 8 │ Testing under realistic conditions │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 9 │ Clear, essential user information │ │ ║
║ │ ├─────┼─────────────────────────────────────────────────────────┤ │ ║
║ │ │ 10 │ Post-deployment monitoring and risk management │ │ ║
║ │ └─────┴─────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ │ 📌 Builds on: 2021 FDA-Health Canada-MHRA joint framework │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 📜 IMDRF N81: MEDICAL DEVICE SOFTWARE (January 2025) │ ║
║ │ │ ║
║ │ Expands scope beyond traditional SaMD: │ ║
║ │ │ ║
║ │ ┌───────────────┐ ┌───────────────┐ ┌───────────────┐ │ ║
║ │ │ SaMD │ + │ SiMD │ = │ MDSW │ │ ║
║ │ │ (standalone) │ │ (embedded) │ │ (all med sw) │ │ ║
║ │ └───────────────┘ └───────────────┘ └───────────────┘ │ ║
║ │ │ ║
║ │ Aligns with EU MDR/IVDR "medical device software" terminology │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ 📅 IMDRF Strategic Plan 2026-2030 (N91 FINAL:2025): December 2025 ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
█████╗ ███████╗███╗ ███╗███████╗ ██╗ ██╗ █████╗ ██╗ ██╗ ██╗ ██╗ ██████╗
██╔══██╗██╔════╝████╗ ████║██╔════╝ ██║ ██║██╔══██╗██║ ██║ ██║ ██║██╔═████╗
███████║███████╗██╔████╔██║█████╗ ██║ ██║███████║██║ ██║ ███████║██║██╔██║
██╔══██║╚════██║██║╚██╔╝██║██╔══╝ ╚██╗ ██╔╝╚════██║╚██╗ ██╔╝ ╚════██║████╔╝██║
██║ ██║███████║██║ ╚═╝ ██║███████╗ ╚████╔╝ ██║ ╚████╔╝ ██║╚██████╔╝
╚═╝ ╚═╝╚══════╝╚═╝ ╚═╝╚══════╝ ╚═══╝ ╚═╝ ╚═══╝ ╚═╝ ╚═════╝
ASME V&V 40-2018: The Technical Foundation
ASME V&V 40-2018 ("Assessing Credibility of Computational Modeling through Verification and Validation: Application to Medical Devices") remains the current standard with no formal revisions published. The standard holds FDA-recognized consensus standard status (FR Recognition Number 5-122, recognized January 14, 2019).
Significant supplementary development is underway through six active working groups. The VVUQ 40.1 technical report, led by Brandon Lurie (W.L. Gore & Associates), provides an end-to-end example applying V&V 40-2018 to a fictional tibial tray durability model.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 📐 ASME VERIFICATION & VALIDATION STANDARDS ECOSYSTEM ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ ┌───────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ ASME V&V 40-2018 │ │ ║
║ │ │ ══════════════════ │ │ ║
║ │ │ MEDICAL DEVICES │ │ ║
║ │ │ (Core Standard) │ │ ║
║ │ │ │ │ ║
║ │ │ FDA Recognition: 5-122 │ │ ║
║ │ │ Status: UNCHANGED │ │ ║
║ │ │ │ │ ║
║ │ └─────────────┬─────────────┘ │ ║
║ │ │ │ ║
║ │ ┌─────────────────────────┼─────────────────────────┐ │ ║
║ │ │ │ │ │ ║
║ │ ▼ ▼ ▼ │ ║
║ │ ┌───────────┐ ┌───────────┐ ┌───────────┐ │ ║
║ │ │ VVUQ 40.1 │ │ VVUQ 20.1 │ │ VVUQ 60.1 │ │ ║
║ │ │ Technical │ │ Multivar. │ │ Software │ │ ║
║ │ │ Report │ │ Metrics │ │ Selection │ │ ║
║ │ │ (H1 2025) │ │ (2024) │ │ (2024) │ │ ║
║ │ └───────────┘ └───────────┘ └───────────┘ │ ║
║ │ │ ║
║ │ ┌─────────────────────────┬─────────────────────────┐ │ ║
║ │ ▼ ▼ ▼ │ ║
║ │ ┌───────────┐ ┌───────────┐ ┌───────────┐ │ ║
║ │ │ VVUQ 70 │ │ VVUQ 80 │ │ WG 3 │ │ ║
║ │ │ Machine │ │ Pharma │ │ Patient- │ │ ║
║ │ │ Learning │ │ Products │ │ Specific │ │ ║
║ │ │ (New SC) │ │ (New SC) │ │ Modeling │ │ ║
║ │ └───────────┘ └───────────┘ └───────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
█████╗ ██╗ ██╗██╗ ██████╗███████╗███╗ ██╗███╗ ██╗ █████╗
██╔══██╗██║ ██║██║██╔════╝██╔════╝████╗ ██║████╗ ██║██╔══██╗
███████║██║ ██║██║██║ █████╗ ██╔██╗ ██║██╔██╗ ██║███████║
██╔══██║╚██╗ ██╔╝██║██║ ██╔══╝ ██║╚██╗██║██║╚██╗██║██╔══██║
██║ ██║ ╚████╔╝ ██║╚██████╗███████╗██║ ╚████║██║ ╚████║██║ ██║
╚═╝ ╚═╝ ╚═══╝ ╚═╝ ╚═════╝╚══════╝╚═╝ ╚═══╝╚═╝ ╚═══╝╚═╝ ╚═╝
Avicenna Alliance Expands Influence
The Avicenna Alliance continues as Europe's primary advocacy organization for in silico medicine, achieving several milestones in 2025. Most significantly, TÜV SÜD joined as a partner—the first major Notified Body to formally affiliate with the Alliance, signaling growing NB receptivity to computational evidence.
The Alliance's flagship publication "Toward Good Simulation Practice" (published February 2024, Springer Open Access) has surpassed 50,000 downloads. The 144-expert collaboration, including 13 FDA modeling specialists, establishes best practices for computational evidence in regulatory submissions.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🏛️ AVICENNA ALLIANCE - EUROPE'S IN SILICO MEDICINE HUB ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 🆕 MILESTONE: TÜV SÜD PARTNERSHIP (2025) │ ║
║ │ ═══════════════════════════════════════ │ ║
║ │ │ ║
║ │ First major Notified Body to formally affiliate with the Alliance │ ║
║ │ │ ║
║ │ ┌─────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ TÜV SÜD AVICENNA ALLIANCE │ │ ║
║ │ │ ════════ ════════════════ │ │ ║
║ │ │ │ │ ║
║ │ │ • Leading Notified Body • In silico advocacy │ │ ║
║ │ │ • EU, US, UK, Asia • Policy development │ │ ║
║ │ │ • 150+ years experience • Regulatory engagement │ │ ║
║ │ │ • 30,000+ employees • 144+ expert network │ │ ║
║ │ │ ↓ ↓ │ │ ║
║ │ │ └─────────┬─────────┘ │ │ ║
║ │ │ │ │ │ ║
║ │ │ 🤝 STRATEGIC PARTNERSHIP 🤝 │ │ ║
║ │ │ │ │ │ ║
║ │ │ Growing NB receptivity to │ │ ║
║ │ │ computational evidence │ │ ║
║ │ │ │ │ ║
║ │ └─────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 📚 "TOWARD GOOD SIMULATION PRACTICE" (Feb 2024) │ ║
║ │ │ ║
║ │ ┌───────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ 📥 Downloads: 50,000+ │ │ ║
║ │ │ 👥 Authors: 144 experts │ │ ║
║ │ │ 🇺🇸 FDA: 13 modeling specialists │ │ ║
║ │ │ 📖 Publisher: Springer (Open Access) │ │ ║
║ │ │ │ │ ║
║ │ └───────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ 🔬 ACTIVE TASK FORCES: ║
║ • EMA Collaboration Task Force (bi-monthly meetings) ║
║ • Notified Bodies Task Force (Team-NB engagement) ║
║ • Global Harmonization Task Force (Marc Horner, Ansys) ║
║ ║
║ 💶 EU VIRTUAL HUMAN TWINS INITIATIVE: €100M+ Commission investment ║
║ 📅 VHT Platform Kickoff: June 26, 2025 ║
║ 🗺️ EDITH Project: €5M, 19 partners, VPH Institute coordination ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
███╗ ███╗██████╗ ██╗ ██████╗ ███╗ ███╗ ██████╗ ██████╗ ███████╗██╗███╗ ███╗
████╗ ████║██╔══██╗██║██╔════╝ ████╗ ████║██╔═══██╗██╔══██╗██╔════╝██║████╗ ████║
██╔████╔██║██║ ██║██║██║ ██╔████╔██║██║ ██║██║ ██║███████╗██║██╔████╔██║
██║╚██╔╝██║██║ ██║██║██║ ██║╚██╔╝██║██║ ██║██║ ██║╚════██║██║██║╚██╔╝██║
██║ ╚═╝ ██║██████╔╝██║╚██████╗ ██║ ╚═╝ ██║╚██████╔╝██████╔╝███████║██║██║ ╚═╝ ██║
╚═╝ ╚═╝╚═════╝ ╚═╝ ╚═════╝ ╚═╝ ╚═╝ ╚═════╝ ╚═════╝ ╚══════╝╚═╝╚═╝ ╚═╝
MDIC ModSim2025: 600+ Professionals Convene
The Medical Device Innovation Consortium's ModSim2025 symposium (November 4-6, 2025, Hyattsville, MD) convened over 600 MedTech professionals under the theme "Credible Models in the Age of AI." Dr. Michelle Tarver, FDA CDRH Director, delivered the plenary keynote emphasizing that progress depends on "uniting data, design, and delivery to accelerate patient access while safeguarding trust."
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🎯 MDIC ModSim2025 SYMPOSIUM ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ 📅 November 4-6, 2025 │ ║
║ │ 📍 College Park, Hyattsville, MD │ ║
║ │ 👥 600+ MedTech professionals │ ║
║ │ 🎨 Theme: "Credible Models in the Age of AI" │ ║
║ │ │ ║
║ │ ┌───────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ 🎤 KEYNOTE: Dr. Michelle Tarver (FDA CDRH Director) │ │ ║
║ │ │ │ │ ║
║ │ │ "Progress depends on uniting data, design, and │ │ ║
║ │ │ delivery to accelerate patient access while │ │ ║
║ │ │ safeguarding trust." │ │ ║
║ │ │ │ │ ║
║ │ └───────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ │ 📋 KEY FOCUS AREAS: │ ║
║ │ │ ║
║ │ ┌────────────────────┐ ┌────────────────────┐ │ ║
║ │ │ 🔮 Digital Twins │ │ ✅ VVUQ Methods │ │ ║
║ │ │ & AI-Based │ │ for Regulatory │ │ ║
║ │ │ Simulations │ │ Applications │ │ ║
║ │ └────────────────────┘ └────────────────────┘ │ ║
║ │ │ ║
║ │ ┌────────────────────┐ ┌────────────────────┐ │ ║
║ │ │ 📊 Model │ │ 🦴 Orthopedic │ │ ║
║ │ │ Credibility │ │ Digital Twins │ │ ║
║ │ │ Assessment │ │ to System-Level │ │ ║
║ │ └────────────────────┘ └────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ 🏆 MDIC ACHIEVEMENTS: ║
║ • Virtual Patient Model: Successful mock FDA submission ║
║ • bayesDP R package: Available on CRAN for Bayesian trial design ║
║ • ENRICHMENT Project (2019-2024): Playbook published, Frontiers paper ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
██████╗ ██████╗ ███╗ ██╗ ██████╗██╗ ██╗ ██╗███████╗██╗ ██████╗ ███╗ ██╗
██╔════╝██╔═══██╗████╗ ██║██╔════╝██║ ██║ ██║██╔════╝██║██╔═══██╗████╗ ██║
██║ ██║ ██║██╔██╗ ██║██║ ██║ ██║ ██║███████╗██║██║ ██║██╔██╗ ██║
██║ ██║ ██║██║╚██╗██║██║ ██║ ██║ ██║╚════██║██║██║ ██║██║╚██╗██║
╚██████╗╚██████╔╝██║ ╚████║╚██████╗███████╗╚██████╔╝███████║██║╚██████╔╝██║ ╚████║
╚═════╝ ╚═════╝ ╚═╝ ╚═══╝ ╚═════╝╚══════╝ ╚═════╝ ╚══════╝╚═╝ ╚═════╝ ╚═╝ ╚═══╝
Strategic Outlook: January 2026
The regulatory landscape for in silico medical device development as of January 2026 reflects maturing frameworks rather than revolutionary change. The FDA's 2023 guidance remains authoritative, with the May 2025 numaScrew MDDT qualification demonstrating practical implementation. Europe's December 2025 MDR simplification proposal represents a significant conceptual shift, explicitly legitimizing computational modeling for the first time at the EU level—though implementation awaits legislative adoption.
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 🎯 THREE STRATEGIC CONSIDERATIONS FOR MANUFACTURERS ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ ① GLOBAL REGULATORY ASYMMETRY PERSISTS │ ║
║ │ ══════════════════════════════════════ │ ║
║ │ │ ║
║ │ ┌─────────────────────────────────────────────────────────────┐ │ ║
║ │ │ │ │ ║
║ │ │ 🇺🇸 FDA ████████████████████████████████████████ MATURE │ │ ║
║ │ │ │ │ ║
║ │ │ 🇪🇺 EU ██████████████████████░░░░░░░░░░░░░░░░░░ PENDING │ │ ║
║ │ │ │ │ ║
║ │ │ 🇯🇵 JP ████████████████░░░░░░░░░░░░░░░░░░░░░░░░ EVOLVING │ │ ║
║ │ │ │ │ ║
║ │ └─────────────────────────────────────────────────────────────┘ │ ║
║ │ │ ║
║ │ FDA acceptance substantially ahead of EU and Japanese frameworks │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ ② PHYSICS-BASED vs AI/ML: SEPARATE PATHWAYS │ ║
║ │ ═══════════════════════════════════════════ │ ║
║ │ │ ║
║ │ ┌─────────────────────────┐ ┌─────────────────────────┐ │ ║
║ │ │ PHYSICS-BASED │ │ AI/ML │ │ ║
║ │ │ ══════════════ │ │ ═════ │ │ ║
║ │ │ │ │ │ │ ║
║ │ │ 📜 FDA Nov 2023 │ │ 📜 FDA Jan 2025 │ │ ║
║ │ │ Credibility │ │ AI-Enabled │ │ ║
║ │ │ Framework │ │ Device Software │ │ ║
║ │ │ │ │ │ │ ║
║ │ │ 📐 ASME V&V 40 │ │ 📐 VVUQ 70 (new) │ │ ║
║ │ │ │ │ │ │ ║
║ │ └────────────┬────────────┘ └────────────┬────────────┘ │ ║
║ │ │ │ │ ║
║ │ └──────────────────────────────┘ │ ║
║ │ │ │ ║
║ │ FDA maintains explicit │ ║
║ │ distinction between pathways │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ ③ STANDARDIZATION PROVIDES COMMON LANGUAGE │ ║
║ │ ═══════════════════════════════════════════ │ ║
║ │ │ ║
║ │ ASME V&V 40 IMDRF N88/N81 │ ║
║ │ │ │ │ ║
║ │ └─────────┬───────────┘ │ ║
║ │ │ │ ║
║ │ ▼ │ ║
║ │ ┌────────────────────────────────┐ │ ║
║ │ │ COMMON TECHNICAL LANGUAGE │ │ ║
║ │ │ Even as national acceptance │ │ ║
║ │ │ varies │ │ ║
║ │ └────────────────────────────────┘ │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
What to Watch
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 👁️ CRITICAL DEVELOPMENTS TO MONITOR ║
║ ║
║ ┌─────────────────────────────────────────────────────────────────────────┐ ║
║ │ │ ║
║ │ ┌────┐ │ ║
║ │ │ 🇪🇺 │ EU proposal legislative progression │ ║
║ │ └────┘ → Parliament and Council review │ ║
║ │ → Expected adoption: Q2 2027 │ ║
║ │ → Feedback period: Open until March 4, 2026 │ ║
║ │ │ ║
║ │ ┌────┐ │ ║
║ │ │ 🇺🇸 │ FDA MDDT expansion │ ║
║ │ └────┘ → Additional computational tool qualifications │ ║
║ │ → Broader acceptance of virtual testing │ ║
║ │ │ ║
║ │ ┌────┐ │ ║
║ │ │ 📊 │ Bayesian virtual patient models │ ║
║ │ └────┘ → Clinical trial sample size reduction │ ║
║ │ → bayesDP R package adoption │ ║
║ │ │ ║
║ │ ┌────┐ │ ║
║ │ │ 🤖 │ Generative AI integration │ ║
║ │ └────┘ → Synthetic patient population generation │ ║
║ │ → Digital twin acceleration │ ║
║ │ │ ║
║ └─────────────────────────────────────────────────────────────────────────┘ ║
║ ║
║ 💡 THE STRATEGIC QUESTION HAS SHIFTED: ║
║ ║
║ ❌ "Whether to adopt computational methods" ║
║ │ ║
║ ▼ ║
║ ✅ "How aggressively to integrate them across the ║
║ development lifecycle—from early concept exploration ║
║ through regulatory submission to post-market surveillance" ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
╔═══════════════════════════════════════════════════════════════════════════════╗
║ ║
║ 📅 TIMELINE ║
║ ║
║ 2021 ───────────────────────────────────────────────────────────────── ║
║ │ ║
║ ├── May: IMAnalytics MRIxViP MDDT qualified ║
║ └── Nov: Virtual MRI Safety (MED Institute) MDDT qualified ║
║ ║
║ 2023 ───────────────────────────────────────────────────────────────── ║
║ │ ║
║ ├── Mar: MRI Temperature Rise MDDT qualified ║
║ └── Nov: 🏆 FDA CM&S Credibility Framework FINALIZED ║
║ ║
║ 2024 ───────────────────────────────────────────────────────────────── ║
║ │ ║
║ ├── Jan: Japan Dash for SaMD2 launched ║
║ ├── Feb: "Good Simulation Practice" published ║
║ └── ENRICHMENT Project completed ║
║ ║
║ 2025 ───────────────────────────────────────────────────────────────── ║
║ │ ║
║ ├── Jan: 🇺🇸 FDA AI-Enabled Device Draft Guidance ║
║ │ 🌍 IMDRF N88 & N81 finalized ║
║ ├── May: 🏆 numaScrew MDDT qualified (first mechanical test) ║
║ ├── Jun: EU VHT Platform kickoff ║
║ ├── Aug: FDA PCCP Final Guidance ║
║ ├── Sep: IMDRF 28th meeting (Sapporo) ║
║ ├── Nov: ModSim2025 (600+ attendees) ║
║ └── Dec: 🇪🇺 EU MDR/IVDR simplification proposal published ║
║ ║
║ 2026 ───────────────────────────────────────────────────────────────── ║
║ │ ║
║ ├── Jan: ◄─── YOU ARE HERE ║
║ └── Mar: EU feedback period closes (March 4) ║
║ ║
║ 2027 ───────────────────────────────────────────────────────────────── ║
║ │ ║
║ └── Q2: Expected EU proposal adoption ║
║ ║
╚═══════════════════════════════════════════════════════════════════════════════╝
╔═══════════════════════════════════════╗
║ 📋 DISCLAIMER 📋 ║
╚═══════════════════════════════════════╝
I am not a lawyer or financial adviser. Nothing in this article constitutes investment advice, legal advice, or a recommendation to buy or sell any securities. The information presented is for educational and informational purposes only. Readers should conduct their own research and consult with qualified professionals before making any investment or business decisions.
██████╗ ██████╗ ███████╗
██╔══██╗██╔═████╗██╔════╝
██████╔╝██║██╔██║███████╗
██╔═══╝ ████╔╝██║╚════██║
██║ ╚██████╔╝███████║
╚═╝ ╚═════╝ ╚══════╝.org
January 2026
Member discussion