Google Unveils AI Web App to Identify Common Skin Conditions
Google introduces a new web-based dermatology assist tool that uses smartphone cameras to help users identify skin, hair, and nail issues. The tech giant also reveals an AI system designed to screen for potential tuberculosis cases using chest X-rays.
Google unveils a new AI-powered dermatology assist tool that helps users identify common skin, hair, and nail conditions. The web-based application works by having users take three photos of an affected area with their smartphone camera and answer a series of questions about their symptoms and skin type. An AI model trained on 288 different conditions then analyzes this data to provide a list of possible matches.
The search giant emphasizes that this tool is not a diagnostic device and does not replace professional medical advice. Instead, it serves as an informational resource that provides dermatologist-reviewed answers, frequently asked questions, and similar images from search results. The system accounts for diverse factors like age, sex, race, and skin type to ensure the results are as accurate and inclusive as possible.
In addition to the dermatology tool, Google reveals a separate AI system designed to screen for potential tuberculosis patients using chest X-rays. This deep learning model is built on de-identified data from nine countries to account for a wide range of demographics, and it potentially reduces the cost per positive TB case detected by up to 80 percent. Both health-focused tools are part of upcoming pilot programs and research studies planned for later this year.