Google Releases New Dermatology AI Recognition Program

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
Get help with your application

Use all the free resources available to you from SDN: articles, guides, expert advising, forums discussions, and school research.

litokhmaikid

Full Member
10+ Year Member
Advertisement - Members don't see this ad

New tool will allow people to upload 3 pictures online, answer questions about symptoms, and get a differential of top 3 diagnoses presented to them. Essentially a patient-centered VisualDx. "For each matching condition, the tool will show dermatologist-reviewed information and answers to commonly asked questions, along with similar matching images from the web."

How do you think this will affect dermatology patient and referral volume? On one hand, I could see it increasing patient volume by encouraging people to access care for skin disease that people previously ignored, but it could also reduce a lot of the benign derm visits (lipoma, congenital nevus, etc.)
 
Last edited:

New tool will allow people to upload 3 pictures online, answer questions about symptoms, and get a differential of top 3 diagnoses presented to them. Essentially a patient-centered VisualDx. "For each matching condition, the tool will show dermatologist-reviewed information and answers to commonly asked questions, along with similar matching images from the web."

How do you think this will affect dermatology patient and referral volume? On one hand, I could see it increasing patient volume by encouraging people to access care for skin disease that people previously ignored, but it could also reduce a lot of the benign derm visits (lipoma, congenital nevus, etc.)

Last year at this time, my practice closed down and just offered teledermatology services.

The minute our state allowed us to re-open, we cancelled our contract with the teledermatology company. People were unable and unwilling to use that technology.

I can tell you from experience the images that people upload will not be of sufficient quality to render a diagnosis. (Heck, we had a mandatory photography course during residency and I'd argue half of the residents who graduated still hadn't learned to take quality photos with the fancy clinic-supplied equipment)

This will be a novelty that maybe some very tech-savvy people will utilize as a trial. It will not impact clinical volume whatsoever.

You also have to remember Google is behind this project. If you've utilized Google technology over the last decade, you're also probably aware of how quickly they start and abandon projects.
 
Last edited: