Is that red spot on your arm just a pimple, or something more serious? Google wants to answer that for you with the help of artificial intelligence.
The search giant’s new dermatology feature, unveiled this week, uses machine learning to identify 288 different skin ailments ranging from acne to melanoma based on user-submitted photographs.
“In a matter of seconds, you will have a list of possible matching dermatological conditions,” said Dr. Karen DeSalvo, the company’s chief health officer and a former assistant secretary of health under the Obama administration, adding that the tool will be “inclusive of a range of skin types and tones.”
Google receives nearly 10 billion searches related to skin, nail and hair issues each year, the company said. But users may have difficulty describing their conditions through search terms alone. The company’s photo-based tool is designed to cover 90 percent of the most frequently searched dermatology-related questions.
Epidermis-minded Europeans will be able to access the feature within months because the EU has already cleared it as a “low-risk” medical device, Google said. The technology has not yet been approved for use in the US.
According to a study published in the medical journal Nature Medicine last year, Google’s tool is just as accurate at identifying skin conditions as a group of dermatologists — and outperforms non-specialists like primary care doctors and nurse practitioners.
However, Google says you shouldn’t say goodbye to the dermatologist’s office yet.
“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,” the company cautioned in a blog post. “Rather we hope it gives you access to authoritative information so you can make a more informed decision about your next step.”