Google’s Derm Assist AI tool

Google is making a big move into healthcare with its announcement last week of a new AI skin care tool, Derm Assist, which is set to launch in Europe later this year. The new tool received a Class I medical device CE mark from European regulators, and is backed by landmark research that was published in Nature Medicine this time last year, showing how Google’s AI model was trained on more than 16,000 real-world dermatology cases and achieved accuracy on a par with US board-certified dermatologists. Another more recent study in JAMA Network indicated the AI tool also significantly improved diagnostic accuracy of non-derma specialists such as GPs.

In its blog announcement, Google said: “With our CE marked AI-powered dermatology assist tool, a web-based application that we aim to make available for early testing in the EU later this year, it’s easier to figure out what might be going on with your skin. Simply use your phone’s camera to take three images of the skin, hair or nail concern from different angles. You’ll then be asked questions about your skin type, how long you’ve had the issue and other symptoms that help the AI to narrow down the possibilities. The AI model analyses all of this information and draws from its knowledge of 288 conditions to give you a list of possible conditions that you can then research further. It’s not meant to be a replacement for diagnosis, but rather a good place to start.”

Google said that it sees billions of Google Searches every year related to skin, nail and hair issues, but that it remains difficult for people to describe what they see on their skin through words alone. According to the FT, Google cited studies showing that people only diagnose themselves correctly 13% of the time when it comes to skin conditions, presenting a huge opportunity for the Derm Assist tool. In the blog announcement, Google provided a visual walkthrough of how the tool will work on smartphones, starting with a prompt to submit a picture of your “skin concern” (examples include moles, birthmarks and rashes), with the visual information then analysed by the AI before being converted into a gallery of suggested conditions.

Google is keen to emphasis the inclusivity of its AI model, stating: “To make sure we’re building for everyone, our model accounts for factors like age, sex, race and skin types – from pale skin that does not tan to brown skin that rarely burns. We developed and fine-tuned our model with de-identified data encompassing around 65,000 images and case data of diagnosed skin conditions, millions of curated skin concern images and thousands of examples of healthy skin – all across different demographics.”

Review New Product Activity and the latest digital health developments across the industry with Innovation in CHC 2021, the latest report from CHC New Products Tracker. For more information, or to order your copy, please contact melissa.lee@NicholasHall.com

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s