the chief information officer says he works with
Google Cloud to create a synthetic data platform that will allow health insurance companies to better detect fraud and offer personal care to their members.
Anil Bhat said the plan is to use algorithms and statistical models to create approximately 1.5 to 2 petabytes of synthetic data, including artificially created datasets from medical history, health claims and other key medical data created in partnership with Google Cloud.
The ultimate goal, he said, is to test and train AI algorithms on large amounts of data, while reducing the privacy concerns associated with personal medical information. “More and more дадзе synthetic data will overtake and become the way people do AI in the future,” Mr Bhat said.
The anthem he used
Amazon Web Services, as a cloud service provider since 2017, used Google Cloud last year to analyze data and artificial intelligence capabilities as part of an ongoing effort to become more customer-focused and focus on members ’entire travel healthcare, said Mr. n Bhat. This is an ongoing effort that involves Anthem working with synthetic data.
This week the shareholders of Anthem are vote on the proposed rebranding companies in Elevance Health as part of the same effort.
Synthetic data applies to both real data that has been stripped of personal information and completely anonymous and artificial data that has been created from deep generative models, said Rita Giotti, vice president of global research on artificial intelligence and automation at research firm International Data. Corp. The anthem said it uses the second type.
The idea of synthetic data in enterprises has existed for decades, but it has recently begun to gain momentum as companies begin to make more use of AI itself and are in greater need of faster access to better datasets on which AI algorithms can be taught, Ms. Giotti said.
Anthem said the synthetic data will be used to test and train AI algorithms that identify things like fraudulent statements or deviations in human medical records, and these AI algorithms will be able to work on real members ’data.
Anthem is already using artificial intelligence algorithms to look for fraud and abuse in insurance claims, but a new synthetic data platform will scale it up. Personalizing member care and launching AI algorithms that determine when they may need medical intervention is a longer-term goal, Mr Bhat said.
In addition to mitigating privacy concerns, Ms. Giotti said another benefit of synthetic data is that they can reduce biases that exist in real datasets. However, she added, you can also get data sets that are worse than real.
“Changing the data will be very, very important,” Mr Bhatt said, adding that he believes that changing the synthetic data will ultimately be better than the company’s actual datasets.
“Synthetic data models, in our view, ultimately fuel the promise of being able to deliver big data,” said Chris Sokoloski, head of health and life science at Google Cloud. “We believe that this is actually what will push this industry forward.”
Write Isabel Busquette on Isabelle.Bousquette@wsj.com
Copyright © 2022 Dow Jones & Company, Inc. All rights reserved. 87990cbe856818d5eddac44c7b1cdeb8