Google launches an AI health tool for skin conditions

[ad_1]

Google is taking one of the most significant steps taken by a large technology company toward health, launching an AI-powered tool that will help consumers self-diagnose hundreds of skin diseases.

Derm Assist is the first of its kind to be launched in Europe this year before targeting nearly 2 billion people worldwide who suffer from skin diseases ranging from acne to melanoma.

Users upload images of their medical condition through the Derm Assist website and answer questions about their symptoms. An AI model analyzes the information and generates a list of possible matching conditions. The service will be free for all Internet users, whether Google users or not.

“The tool is not intended to provide a diagnosis. . . rather, we look forward to giving you access to authorized information so that you can make a more informed decision about your next step, ”Google said.

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

Google chose dermatology as its first goal for AI-based health, due to the huge number of people affected by skin conditions. Approximately 10 billion searches are done on Google each year related to skin, nail and hair problems, and studies have shown that people are only correctly diagnosed 13% of the time, the search giant said.

Google’s new artificial intelligence tool Derm Assist © Google

“Skin diseases as a category are a huge global burden: people turn to Google to investigate their skin concerns. Most cases can be cured, but half the world’s population faces a shortage. dermatologist review, “said Dr. Peggy Bui, Google Health’s chief product officer and internal medicine specialist at the University of California, San Francisco.

The Derm Assist system is based on a machine learning algorithm formed in more than 16,000 cases of real-world dermatology. In accordance with a study since last year, the tool is able to identify skin conditions with the same accuracy as dermatologists certified by the US Board

Some of the information provided to users is reviewed by human dermatologists. If a user mentions alarming symptoms, such as inability to breathe, additional alerts advise them to see a doctor immediately.

A study published at JAMA Network Open found that the AI ​​tool also significantly improved the diagnostic accuracy of non-specialists, such as GPs and professional nurses, who used it to help them diagnose skin conditions.

“Our observations suggest that AI has the potential to increase the capacity of[generalist doctors and nurses]. . . to diagnose and choose skin conditions more effectively, “wrote study author Yuan Liu and his team in the peer-reviewed paper.” Improving the diagnostic accuracy of non-derived cases… could have huge implications for healthcare systems. ”

Eric Topol, a professor of molecular medicine at the Scripps Research Institute and an expert in AI and health, said: “This should happen at some point, as it was the first major use case of AI. of deep learning in medicine with some validation. in 2017 “.

To prevent the loss of skin cancer cases through false negatives, the algorithm was designed to be prudent in decision making. “When we designed this, we said we wanted to optimize for high sensitivity, especially for alarming or appalling conditions,” Dr. Bui.

To address privacy issues about users’ health data, Google said it would not use uploaded images to target advertising and would only save images to further form the Derm Assist algorithm, if users gave them permission. explicit to do so.

“Users have control over their data with the option to save, delete or donate data for research,” said Dr. Bui. “We hope to encourage donation, as the algorithms are as good as the data with which it was formed. . . We will continue to improve the model by obtaining other data sets from other sources, in addition to the data provided.

[ad_2]

Source link

Leave a Reply

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