Google Previews AI-Powered applications to diagnose skin conditions
2 min readYesterday on Google I / O, the company previewed the AI-Powered application that was intended to help diagnose dermatological problems to help people understand the challenges with their skin, hair, and nails. Doctor Peggy Bui, MD, Product Manager for Google Health, said that the application uses the same technique to help doctors detect diabetes eye diseases or lung cancer in CT scanning. This application allows users to get a more accurate diagnosis for dermatological problems such as rashes on their skin using the camera on their smartphones.
Google says almost 10 billion searches related to skin problems, nails, and hair every year. Globally, around 2 billion people face problems with skin, nails, or hair, and for many people, the search diagnosis starts on Google using keywords. AI-Powered Dermatology Assist Tool is a web-based application that aims Google as a pilot later this year to help diagnose problems with the skin.
This tool requires users to take three skin pictures, hair, or worries of nails from different angles. It submits a series of questions about skin types, how long they have problems, and other symptoms to help up to potential diagnosis. AI application analyzes information and uses it to compare with its Knowledge base 288 conditions to produce a list of possible compatibility that can be searched for more information.
For each suitable condition, this tool provides users of information and answers reviewed by dermatologists to general questions along with matching images from the web. Google is clear that this tool is not intended to provide a diagnosis or substitute for medical advice from qualified doctors.
Google notes that many skin conditions will need clinical reviews, direct checks, or additional testing such as biopsy. Search giants do expect this tool to give users access to authoritative information, helping them make decisions based on information about the next step in treating their conditions. This model contributes several factors, including age, gender, race, and skin type. It is also set with unidentified data which includes around 65,000 images and cases of cases of skin conditions diagnosed from all demographics. Those who are interested in the tool can register here to be notified when available.