In the US, the FDA classifies devices into three classes. Class I devices (low risk, e.g., bandages) require general controls. Class II devices (moderate risk, e.g., infusion pumps) typically go through the 510(k) predicate pathway, demonstrating substantial equivalence to an existing cleared device. Class III devices (high risk, e.g., implantable pacemakers) require a Premarket Approval (PMA) with clinical evidence. In the EU, the Medical Device Regulation (MDR 2017/745) replaced the older MDD in 2021, imposing stricter clinical evidence requirements and post-market surveillance obligations. China’s NMPA follows a similar three-class framework with increasing localization requirements for clinical data.

AI/ML-based medical devices — from radiology AI (detecting lung nodules, diabetic retinopathy) to ECG interpretation algorithms and sepsis prediction tools — are being regulated by the FDA under a software-as-a-medical-device (SaMD) framework. The FDA’s AI/ML action plan addresses the challenge of “continuously learning” algorithms that change post-approval. CE marking under EU MDR requires clinical validation of AI performance across intended use populations. Key players include Siemens Healthineers, Philips, GE HealthCare, and a large ecosystem of AI diagnostics startups (Aidoc, Viz.ai, Paige.ai). IP in AI medical devices is concentrated in algorithm architecture, training data curation, and clinical validation methodologies.

Point-of-care (POC) diagnostics bring clinical testing to the patient — at the bedside, in a pharmacy, or in a low-resource setting — eliminating the need to send samples to a central lab. COVID-19 accelerated POC adoption dramatically, validating platforms like Abbott BinaxNOW and demonstrating that rapid molecular testing could be deployed at global scale. The key technical challenges are sensitivity/specificity matching lab standards, sample preparation simplification, and result connectivity. Optical microfluidic sensors, nanopore sequencing, and droplet digital PCR are enabling the next generation of POC platforms with single-molecule detection sensitivity and multiplexed pathogen panels.

Continuous glucose monitors — led by Abbott FreeStyle Libre and Dexcom G-series — have transformed diabetes management by providing real-time glucose readings every few minutes without fingerstick calibration. Growth drivers include expanding insurance coverage, approval for Type 2 diabetes (a much larger population than Type 1), integration with insulin pumps to form closed-loop artificial pancreas systems, and emerging use in metabolic health monitoring for non-diabetic populations. Key IP areas include electrochemical sensor chemistry, biocompatible adhesive formulations (critical for 14-day wear), and CGM-to-insulin pump algorithm IP. Abbott and Dexcom dominate current filings, but Medtronic, Roche, and Asian manufacturers are intensifying competition.

Based on recent patent trends, the highest-activity segments are: AI-assisted diagnostics and medical imaging (driven by deep learning algorithm patents from Siemens, GE, and AI startups); minimally invasive surgical devices and robotics (Intuitive Surgical, Medtronic Hugo, CMR Surgical); continuous monitoring wearables (CGM, cardiac monitoring, neural interfaces); cell and gene therapy manufacturing equipment (Miltenyi, Lonza, Thermo Fisher); and advanced biomaterials for implants and tissue engineering (biopolymers, plasma coatings, nanocomposites). China has become a major filer in IVD, medical imaging hardware, and biomaterials. The US retains leadership in software-driven and high-complexity device categories.