Truthing microscope slides to validate artificial intelligence algorithms
Hello everyone, Let me apologize for the accidental announcement earlier. It was very old.
I’m contacting you now to let you know about a presentation, '' “A reader study on a 14-head microscope,” '' to be given at Pathology Informatics at 8:15am on Wednesday 5/23/2018. This work is feeding an FDA project titled, '' “High-throughput truthing of microscope slides to validate artificial intelligence algorithms analyzing digital scans of pathology slides: leveraging data collected in international "grand challenges”. '' Since writing the proposal, we have settled on a specific deliverable that many groups are excited about: collaborating with challenge organizers and other organizations to produce images + annotations to be qualified by FDA as medical device development tools (MDDTs). As MDDTs, the images + annotations can then be used by algorithm developers to test the performance of AI algorithms in an FDA submission with very little description and validation, easing the the regulatory burden for industry and the FDA.
- Presentation abstract and slides at https://nciphub.org/groups/eedapstudies/wiki/Presentation:AReaderStudyona14headMicroscope
- Project description: https://nciphub.org/groups/eedapstudies/wiki/HighThroughputTruthing
- FDA medical device development tools (MDDTs): https://www.fda.gov/MedicalDevices/ScienceandResearch/MedicalDeviceDevelopmentToolsMDDT