Advances in AI combined with other technologies, like smartphone cameras, can unlock new ways for people to be informed and take control of their health. Google is set to pilot an AI-powered dermatology assist tool to help people understand what's going on with issues related to their body's largest organ and identify dermatological issues by simply using a camera phone.
Each year there were almost ten billion Google Searches related to skin, nail, and hair issues. Two billion people globally suffer from dermatologic issues, with many taking the first step to diagnosing them using the Google Search bar, but it can be difficult to describe a skin condition with words alone. Three simple images capturing the skin, hair, or nail condition from different angles, and a series of questions, help narrow down the possibilities. The AI model analyzes this information and draws from its knowledge of 288 conditions to provide a list of possible matching conditions that can be researched further.
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. The tool is not intended to be a diagnosis or replace medical advice, but it can help bridge the gap globally in the shortage of specialists by providing access to authoritative information to inform decision making.
The Google dermatological tool is the culmination of over three years of machine-learning research, product development, and the publication of several peer-reviewed papers that validate the AI model.
A landmark study, featured in Nature Medicine, debuted Google's deep-learning approach to assessing skin diseases and showed that its AI system can achieve accuracy that is on par with US board-certified dermatologists. A recent paper in JAMA Network Open demonstrated how non-specialist doctors can use AI-based tools to improve their ability to interpret skin conditions.
The tool takes into account factors like age, sex, race, and skin type, from pale skin that does not tan to brown skin that rarely burns. Google developed and fine-tuned its model with de-identified data encompassing approximately 65,000 images and case data of diagnosed skin conditions, millions of curated skin-concern images, and thousands of examples of healthy skin across different demographics.
Recently, the AI model that powers the tool successfully passed clinical validation, and the tool has been CE marked as a Class I medical device in the EU. Derm Assist will launch in Europe this year.
2 Article(s) Remaining