Google previews AI dermatology assistant tool

Google has released a preview of an artificial intelligence tool it has developed to help identify dermatologic issues.

The app is said to help spot skin, hair and nail conditions using the camera on your phone. Following uploading three images, the tool asks the user a series of questions, such as how long you have had the issue and for any other symptoms. The AI model then provides a list of possible conditions.

It’s not designed to replace clinical process, however, Google said the app can recognise up to 288 conditions to provide a list of possible matching conditions.

The dermatology assistant tool is to launch as a pilot later this year and has been CE marked as a Class I medical device in the EU.

Google said: “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 provide a diagnosis nor be a substitute for medical advice as many conditions require clinician review, in-person examination, or additional testing like a biopsy. Rather we hope it gives you access to authoritative information so you can make a more informed decision about your next step.”

The tool has been in development for three years, and peer-reviewed by academics and researchers. The most recent paper was published in JAMA Network Open and demonstrated how non-specialist doctors can use AI-based tools to improve their ability to interpret skin conditions.

Last week the company also announced its tool to support early detection of pulmonary tuberculosis via chest radiography, following an international study across 10 countries.

The deep learning system (DLS) was trained to detect active pulmonary tuberculosis (TB), with the study highlighting the tool sensitivity of 88% and its specificity 79%. The researchers noted that using the DLS as a prioritisation tool for confirmatory testing reduced the cost per positive case detected by 40-80% compared to using confirmatory testing alone.

The AI system produces a number between 0 and 1 that indicates the risk of TB. Google noted that for the system to be useful in a real-world setting, there needs to be agreement about what risk level indicates that patients should be recommended for additional testing.

To develop the tool further, research studies are now planned with Apollo Hospitals in India and the Centre for Infectious Disease Research in Zambia (CIDRZ) to take place later this year.