AI Diagnostics, PCCPs & FDA Turnover
This week on Trying to Keep Up, Allison Komiyama and Michael Nilo are joined by former FDA reviewer and Canon Medical Informatics regulatory leader Jay Vaishnav, PhD, for a deep dive into AI-enabled medical devices, digital pathology, FDA review trends, and the future of software regulation.
The group unpacks several major FDA authorizations and breakthrough designations, including the DeepView AI® System for burn wound assessment and Cardiosense’s PCWP Analysis Software — a breakthrough AI-enabled heart failure monitoring tool with an approved Predetermined Change Control Plan (PCCP).
Jay shares perspective from both FDA and industry on why AI diagnostics are finally gaining traction, what companies still misunderstand about validation burden, and why PCCP adoption remains surprisingly low despite FDA’s push toward lifecycle-based AI oversight.
The episode also covers:
New De Novo authorizations for AI-enabled diagnostics
AI in digital pathology and cardiovascular monitoring
PCCPs and why companies struggle to implement them
Human factors guidance updates and FDA inconsistency frustrations
MDUFA V performance metrics and sponsor response delays
ISO 10993-1:2025 recognition updates
FDA staffing, hiring, and TAP program recruiting
Leadership turnover at FDA and what it actually means operationally
Why radiology and cardiology continue leading AI adoption
Plus: giraffe jokes, zebra jokes, crab-cheetah hybrids, office nostalgia, and a surprisingly emotional conversation about FDA reviewers packing up their offices during layoffs.
Links
