Robust performance of AI in recognition of skin tumors in real life pathology environment: 98-99% specificity sensitivity, coping with all possible artifacts, differentiating all hard to diagnose types, H&E only, no other test or patient data https://t.co/nG1bbGeHW6
Screening in skin should have been implemented long ago. One way or another right things do happen. Dermatopathologists will get even more work🤷♀️
https://t.co/tPOf6nQldi
Thank you👏@FDADeviceInfo for the right move:
“Image Acquisition And/Or Optimization Guided By AI” and “Computer-Assisted Diagnostic Software For Lesions Suspicious For Cancer” will be permanently exempt from 510k requirement for premarket notification.
https://t.co/w1J2AmTZyC
Thank you👏@FDADeviceInfo for the right move:
“Image Acquisition And/Or Optimization Guided By AI” and “Computer-Assisted Diagnostic Software For Lesions Suspicious For Cancer” will be permanently exempt from 510k requirement for premarket notification.
https://t.co/w1J2AmTZyC
“Hundreds of billions of dollars to sharpen America’s innovative edge in markets for global leadership, including Artificial Intelligence and Biotechnology, is up for grabs”.
Ok, noted.
#AI#ArtificialIntelligence#Biotechnology
@PathPro@dpatweet Should be a solid reason and push for policies to keep up with new reality, no?:)
Moreover, all required data will be in place #dpachat
@apmg_glassy@dpatweet Definitely! And also image sharpness, software user friendliness and viewer speed (fast slide manipulation) + all relevant clinical information visible + decision support tools that save time for examination
@apmg_glassy@dpatweet Well, remote sign-out can be a dream job for many #pathologists. Flexible schedules, living where you’ve always wanted to live, peace of mind, postponed retirement, other projects - teaching, writing, consulting - you choose #DPAchat
@tampapath As well as adoption of workforce efficiency tools like automated triage (optimal assignment of cases) and quality control that will accelerate computational pathology too #dpachat
@DrBuiPathology@dpatweet Some US #dermatopathology labs were ready to remote sign-out before covid, they switched to it during lockdown. We work with them now to integrate an algorithm that recognizes diagnoses and triages cases btw pathologists of different expertise levels #dpachat
@PathPro@dpatweet Exactly, so they ensure business continuity.
You can make it even more efficient if you use diagnostic level AI to triage cases between pathologists of different level of experience and sub-specialty. Minimum consultations. Outsourcing. #dpachat
After many years of development the market for #AI in #DigitalPathology has finally emerged. Therefore we introduce our #DeepLearning algorithm for #SkinCancer diagnosis that's already achieved the accuracy of a specialized #dermatopathologist 👉 https://t.co/dZFickhtQU