Regulatory strategy for AI/ML-enabled devices, training and validation best-practices, pre-sub questions, and more
FeaturedArticleAI/ML
FDA selected Limbic’s agentic therapy device for the TEMPO pilot. Its public evidence offers a rare look at layered AI architecture, prospective clinical testing, post-market mo...
J. David Giese
FeaturedArticleAI/ML
FDA's Aug 18, 2026 discussion paper proposes grading generative AI medical devices like physicians: benchmarks, then clinical confirmation, then live monitoring. What it says, w...
Yujan Shrestha
FeaturedArticleAI/ML
Most diagnostic AI teams clear FDA, then stall on reimbursement. FDA's TEMPO pilot and CMS's ACCESS model take on evidence and payment together. Here's how the two work, and wha...
J. David Giese
FeaturedArticleAI/ML
Can a frontier AI model replace your FDA regulatory consultant? We built a private 1,200-question benchmark, an opinionated distillation of 15 years of consulting judgment, and ...
Yujan Shrestha
FeaturedArticleAI/ML
FDA authorized 86 AI/ML devices in Q2 2026, but the sharper signal came after first clearance: 12 Special 510(k)s, 12 PCCPs, two new De Novo categories, and route choices that c...
Yujan Shrestha
FeaturedArticleAI/ML
FDA's draft MDUFA VI commitment letter (FY2028–2032) restructures how sponsors interact with the agency. Seven changes that will reshape submission planning for software and AI ...
Yujan Shrestha
FeaturedArticleAI/ML
UpDoc is now the clearest public example of an FDA-cleared LLM-enabled medical device. Here's what FDA actually cleared, why the clinical decision still sits behind a determinis...
Yujan Shrestha
FeaturedArticleAI/ML
For AI SaMD founders deciding how much to raise and how long the first FDA 510(k) will take. A 37-company benchmark: $16.0M market-comparable median, $11.0M capital-efficient, 2...
Yujan Shrestha
FeaturedArticleAI/ML
Deep dive into four FDA De Novo and 510(k) submissions for AI-powered whole slide image analysis in pathology: Paige Prostate (DEN200080), ArteraAI Prostate (DEN240068), Galen S...
Maddie Tran & Yujan Shrestha
FeaturedArticleAI/ML
For founders, CEOs, and CFOs at AI/ML medical device companies trying to decide how much to spend on a 510(k). A meta-analysis of 35 post-IPO AI/ML SaMD clearances shows $51.2M ...
Yujan Shrestha
FeaturedArticleAI/ML
FDA cleared 24 AI/ML SaMD applications in March 2026 — one every 31 hours. A breakdown of the patterns that matter for sponsors: radiology dominance, PCCP under-adoption, the lo...
Yujan Shrestha
FeaturedArticleAI/ML
FDA's glossary defines foundation models as AI models trained on large, typically unlabeled datasets that are adaptable across a wide range of contexts via fine-tuning, built on...
Yujan Shrestha
FeaturedArticleAI/ML
Fifty-one FDA-authorized pathology AI devices through April 2026, only seven of which are whole-slide imaging algorithms. This snapshot separates the hematology-analyzer volume ...
Yujan Shrestha
FeaturedArticleAI/ML
A deep dive into the 97 cleared cardiovascular AI/ML SaMD devices, the four archetypes that organize them, and what they reveal about the future of FDA-cleared software.
Yujan Shrestha
FeaturedArticleAI/ML
Why digital health companies keep dying before and after FDA authorization, the evidence-reimbursement trap strangling the sector, and a post-market-first framework for breaking...
Yujan Shrestha
FeaturedArticleAI/ML
AI-assisted medical device development is FDA-compliant when done right. The obstacle isn’t regulation. It’s the messy codebase you’ve been ignoring.
Yujan Shrestha
FeaturedArticleAI/ML
Comprehensive analysis of 2025 cardiology AI/ML clearances, market trends, capital events, and regulatory insights.
Yujan Shrestha
FeaturedArticleAI/ML
44 clearances. 18 in 2025 alone. The dental AI market has reached its tipping point. This report dissects the Pearl/Overjet duopoly controlling 34% of the market, the 172-day FD...
Yujan Shrestha
FeaturedArticleAI/ML
A data-driven look at 2025’s AI/ML 510(k) clearances: volume, review times, SaMD share, leading specialties, top manufacturers, and early PCCP adoption.
Yujan Shrestha
FeaturedArticleAI/ML
Deep dive into FDA 510(k) K252366, a2z‑Unified‑Triage, a radiology triage and notification device. Explains why the triage indication is likely the best way to get foundation mo...
Yujan Shrestha
FeaturedArticleAI/ML
Breakthrough Device Designation (BDD) is a free, fast FDA program that signals serious unmet need and potential meaningful improvement. For AI/ML SaMD teams, it de-risks pivotal...
Yujan Shrestha & George Hattub
FeaturedArticleAI/ML
Setting acceptance criteria incorrectly (too high or too low) can delay your FDA submission by weeks or trigger rejection. This guide analyzed 784 AI/ML medical device clearance...
Yujan Shrestha
FeaturedArticleAI/ML
This article shows how to achieve an AI/ML SaMD 510(k) submission in ~3 months by running a compact, well-powered MRMC study while continuously accruing standalone evidence thro...
Yujan Shrestha
FeaturedArticleAI/ML
Here we analyze over 200 FDA 510(k) and De Novo summaries to tease out best practices for AI/ML study designs and adjudication strategies specifically.
Yujan Shrestha
FeaturedArticleAI/ML
A practical guide for engineers new to medical imaging AI, highlighting common pitfalls—like coordinate-system errors and orientation mix-ups—and providing simple checks to avoi...
Yujan Shrestha
FeaturedArticleAI/ML
Ideas on how to get foundation models and generative AI FDA cleared.
Yujan Shrestha
FeaturedArticleAI/ML
Recent advancements in radiology have led to the development of foundation models integrating vision and language capabilities, transforming medical imaging analysis
Yujan Shrestha
ArticleAI/ML
We traced every AI SaMD De Novo's predicate descendants and found prospective enrollment falling toward zero by the third generation. Where you sit on your category's generation...
Yujan Shrestha
ArticleAI/ML
For founders and CFOs sizing the team a first AI SaMD submission needs. A census of 890 FDA clearance letters for AI software devices, a 24-company cohort with headcounts and re...
Yujan Shrestha
ArticleAI/ML
The myth: your 510(k) predicate must match your body region, modality, and product code. We screened all 1,457 AI/ML-flagged FDA records across six axes of predicate distance an...
Yujan Shrestha
ArticleAI/ML
If your protocol says 250 patients, the first reviewer question is: 250 what? Sensitivity is powered by condition-positive cases and specificity by negatives, so plan each quota...
Yujan Shrestha & Nishant Dahal
ArticleAI/ML
There is no single sample size for passing AUC 0.90. The pass rule, expected AUC, confidence interval method, and class mix set the number together. A worked 0.93 example lands ...
Yujan Shrestha & Nishant Dahal
ArticleAI/ML
FDA has selected Cadence as the second participant in TEMPO. Its clinician-supervised HypertensionOS software generates protocol-bound medication-initiation and titration propos...
J. David Giese
ArticleAI/ML
With predetermined change control plans (PCCPs), FDA has given manufacturers a great new regulatory tool. In this article, we discuss PCCP best practices and questions based on ...
J. David Giese
ArticleAI/ML
FDA publicly rejected Nihon Kohden's letter-to-file decision, so we pulled every warning letter FDA has published — 5,712 of them, 2015–2026 — classified them with AI, and mappe...
Yujan Shrestha
ArticleAI/ML
CMS and CDC are asking whether AI-assisted postanalytic interpretation, digital pathology, cloud analytics, and data-only medical-device software facilities should fall within t...
Yujan Shrestha
ArticleAI/ML
Learn how to write an FDA-ready AI model description covering inputs, outputs, architecture, training, selection, thresholds, pretrained components, limitations, and the submitt...
Yujan Shrestha
ArticleAI/ML
27 AI/ML medical devices authorized in April 2026. Radiology's share of clearances dropped from 79% to 56%, QIH remains the regulatory plumbing for image-processing AI, and PCCP...
Yujan Shrestha
ArticleAI/ML
TytoCare's De Novo cracked open ENT, a specialty with just five prior AI/ML clearances, and dropped a 510(k) on-ramp for everyone who follows. Inside the 330-day review: a new S...
Yujan Shrestha
ArticleAI/ML
From solo clinicians prototyping algorithms to enterprise teams shipping production software, tools like Cursor, Claude Code, ChatGPT Codex, GitHub Copilot, Lovable, Replit, Ant...
Yujan Shrestha
ArticleAI/ML
The U.S. Senate HELP Committee confirmed what we've been publishing for three years. An analysis of the regulatory signal, what it means for AI device companies, and why the com...
Yujan Shrestha
ArticleAI/ML
A step-by-step guide to using Monte Carlo simulation for estimating the sample size needed to pass Bland-Altman Limits of Agreement acceptance criteria with bootstrap confidence...
Yujan Shrestha & Nishant Dahal
GuidanceAI/ML
This FDA guidance clarifies which clinical decision support (CDS) software functions are excluded from the definition of a medical device.
Valeriia Zakrasniana
ArticleAI/ML
Yujans commentary on the new Non-invasive blood pressure measuring device FDA guidance document in response to WHOOP and others.
Yujan Shrestha
ArticleAI/ML
A deep dive into K232613 (CT Cardiomegaly)
Yujan Shrestha
ArticleAI/ML
6 FDA clearances in 10 days. The AI pace is accelerating. Coverage includes neuro, dental, diabetes, and ADHD. Key wins: Nobel Biocare’s implant SDK and Lumosity’s digital thera...
Yujan Shrestha
ArticleAI/ML
Fifteen AI/ML software 510(k) clearances landed in nine days in mid-December 2025. Here’s what got cleared, where it fits clinically, and what it signals for regulated AI momentum.
Yujan Shrestha
ArticleAI/ML
Analysis of K252086 (DTX Studio Assist), a strategic 510(k) clearance for a 'headless' SDK. This submission minimizes regulatory burden by offloading UI requirements, employs a ...
Yujan Shrestha
ArticleAI/ML
A client called out the “consultant thing”: hiding behind “ideally” instead of recommending a real plan. This piece explains why ideal talk wastes time, how to anchor on constra...
Yujan Shrestha
ArticleAI/ML
This article explains Bland–Altman analysis for evaluating agreement between two continuous measurement methods, especially in FDA-facing validation. It shows how bias and 95% l...
Yujan Shrestha
ArticleAI/ML
Most first-time teams burn over a year overbuilding documentation FDA never asks for. Here is a section-by-section breakdown of a cleared AI/ML SaMD 510(k) (Body Check CT Cardio...
Yujan Shrestha
ArticleAI/ML
Command Line as a Medical Device (CLaMD) represents a streamlined approach to FDA clearance for medical software. By utilizing command-line interfaces instead of graphical user ...
Yujan Shrestha
VideoAI/ML
Kris Huang led a discussion on key image processing techniques in medical imaging, focusing on CT and MRI scans. Topics included histogram equalization, morphological operations...
Kris Huang
VideoAI/ML
The document details the process and significance of Clinical Performance Assessments (CPAs) for medical devices, focusing on their role in regulatory approval. It highlights th...
Ethan Ulrich
VideoAI/ML
Discussion covered key takeaways from the RSNA conference, balancing safety with speed, and setting risk thresholds. Topics included statistical power, U.S. data needs, AI/ML sa...
Yujan Shrestha & 2 others
VideoAI/ML
This is a recording of the talk Yujan gave at the RSNA AI Theater
Yujan Shrestha
VideoAI/ML
The discussion explores generative AI (GenAI) in medical devices, focusing on FDA's considerations for regulation, risks, and the lifecycle management of such technologies. Key ...
Yujan Shrestha & J. David Giese
VideoAI/ML
Dr. Elena Sizikova's presentation highlights the role of synthetic data in addressing data scarcity and privacy challenges in medical imaging AI. Synthetic data offers benefits ...
Yujan Shrestha
VideoAI/ML
Yujan and David discussed the topic of FDA's performance in regulating AI medical devices, focusing on the ideal regulatory situation, the impact of AI on risk analysis, and the...
Yujan Shrestha & J. David Giese
VideoAI/ML
A 10x talk exploring the basics of prompt engineering, followed by an open-ended discussion.
VideoAI/ML
The talk covers device changes and their regulatory implications using FDA guidance. It focuses on two key documents for deciding when a 510(k) submission is needed for hardware...
Mary Vater
ArticleAI/ML
The article outlines a method for documenting AI/ML algorithms for FDA pre-submissions. It emphasizes clear algorithm descriptions to avoid delays. Key steps include visual runt...
Yujan Shrestha
ArticleAI/ML
Tips and FAQs on how to get a foundation model FDA 510(k) or De Novo cleared.
Yujan Shrestha
ArticleAI/ML
What is the difference between user needs and requirements?
Yujan Shrestha
ArticleAI/ML
Learn how to avoid invalidating your test dataset while still allowing yourself some flexibility to re-use the test data.
J. David Giese
VideoAI/ML
This video discusses the potential avenues for getting generative AI applications cleared through FDA.
Yujan Shrestha
ArticleAI/ML
This article outlines the process of developing an AI/ML algorithm from scratch and getting it FDA cleared. It covers the four phases of the process (Explore, Develop, Validate,...
Yujan Shrestha
ArticleAI/ML
This article provides a questionnaire for those looking to get their AI algorithm FDA cleared. It covers topics such as validation studies, non-ML and ML software items, and acc...
Yujan Shrestha
ArticleAI/ML
This article outlines the challenges of bringing radiology AI/ML to market, from clinical workflow to IT installation. It provides insights and lessons learned for those interes...
J. David Giese
VideoAI/ML
This webinar covers the four phases of the regulatory process: Explore, Develop, Validate, and Document. It also discusses the costs, time, and data requirements involved in the...
Yujan Shrestha
GuidanceAI/ML
This FDA guidance applies to AI/ML-enabled radiology software that assists in the detection of disease. Many of the principles and approaches, however, apply beyond this limited...
GuidanceAI/ML
This guidance is describes FDA’s thoughts on evaluating the performance of CADe radiological devices. Designing standalone and clinical performance testing studies for AI/ML-ena...
GuidanceAI/ML
This document outlines the technical performance assessment required for premarket submissions of quantitative imaging devices to the FDA. It provides guidance on the evaluation...
GuidanceAI/ML
Introduces standardized terms and definitions to foster a consistent understanding across the industry for ML-enabled devices.
ArticleAI/ML
This article outlines Innolitics' process for developing machine learning algorithms for FDA-cleared medical devices. It covers documentation requirements, common pitfalls (over...
Yujan Shrestha &
There are no resources matching these filter criteria.
Get To Market Faster
Our Medtech tips will help you get safe and effective Medtech software on the market faster. We cover regulatory process, AI/ML, software, cybersecurity, interoperability and more.
