Google has launched an AI health tool for skin conditions


Google has taken one of the most important steps in a major healthcare tech company, launching an AI -enabled tool that can help consumers self -diagnose hundreds of skin conditions.

Derm assist is the first of its kind and will launch in Europe this year before targeting nearly 2bn people worldwide who suffer from skin conditions ranging from acne to melanoma.

Users will need to sign in to their Google accounts, upload images of their medical condition through the Derm assist website and answer questions about their symptoms.

An AI model analyzes the information and makes a list of possible pairing conditions. The service will be free to all internet users, Google or not.

“The tool is not designed to provide a diagnosis… However we hope it will give you access to authoritative information so you can make a more informed decision about your next step,” according to Google.

The launch follows three years of progress at Google, which has long viewed healthcare as a market ripe for disruption of advanced artificial intelligence. It comes as rivals Apple, Amazon and Microsoft are also pushing into the potentially lucrative space, building healthcare services for consumers, doctors and pharmaceutical companies.

Google has chosen dermatology as its primary target for AI -driven healthcare because of the large number of people affected by skin conditions. Nearly 10bn Google searches are conducted each year related to skin, nail and hair issues, and studies show that people correctly diagnose themselves 13 percent of the time, the search giant said.

The new AI tool in Derm Help © Google

“Skin diseases as a category are a huge burden on the world – people turn to Google to research their skin concerns. Most cases are resolved, but half of the world’s population is face a critical shortage of dermatologists, ”said Dr. Peggy Bui, product manager at Google Health and an internal medicine specialist at the University of California, San Francisco.

The Derm assist system is based on a machine learning algorithm trained in more than 16,000 real-world dermatology cases. Agreed to a study from last year, the tool is able to identify skin conditions as accurately as dermatologists have been certified by the U.S. board.

Some of the information provided by users has been reviewed by human dermatologists. If a user mentions any frightening symptoms, such as shortness of breath, additional alerts advise that they see a doctor immediately.

A study published at the JAMA Network Open found that the AI ​​tool has also been able to improve diagnostic accuracy in non-specialists such as GPs and trained nurses, using it to help them make diagnoses of skin conditions.

“Our observations suggest that AI has the potential to increase the ability to[generalist doctors and nurses]. . . to diagnose and test skin conditions more effectively, ”study author Yuan Liu and his team wrote in the peer-reviewed paper. “Improving the accuracy of diagnosing cases that have not been referred… Could have many implications for healthcare systems.”

Eric Topol, a professor of molecular medicine at the Scripps Research Institute, and expert in AI and healthcare, says: in 2017. ”

To avoid the disappearance of skin cancer cases through incorrect negatives, the algorithm is designed to be careful in making decisions. “When we designed it, we said we wanted it to be optimized for high sensitivity, particularly in alarming or frightening conditions,” Dr Bui said.

To address privacy concerns about users ’health data, Google said it will not use uploaded images to target advertising, and will only save images to further train the Derm assist algorithm, if users give them explicit permission to do so.

“Users have access to their data with the option to store, delete or donate for research,” Dr Bui said. “We hope to encourage donation, because the algorithms are just as good as the data they train… We will continue to improve the model by getting other sets of data from other sources, in addition to the donated data. “



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *