machine learning AI

AI for machine learning

The evolution of machine learning is still in its early stages, but in HLTH 2022 conference this week, the companies shared how they are working to fine-tune their approaches to AI. These efforts include everything from improving the quality of the patient data that underpins the algorithms, which have been criticized for not reflecting a sufficiently diverse patient population, to making it easier for healthcare organizations to validate their effectiveness. Health tech companies also highlighted the different approaches they are taking to how they work with vendors to pilot machine learning algorithms and bring them to market.

Dr. John Halamka, President of Mayo Clinic Platform, used his talk at HLTH to highlight an initiative to evaluate and reduce bias in patient data to improve the effectiveness of machine learning algorithms. Launched three years ago, the Mayo Clinic Platform has built an ecosystem to coordinate collaborations with health technology companies to enable healthcare innovation.

Halamka’s report on “algorithmic underservice” notes that when healthcare organizations use an algorithm, they often have no idea whether it’s working well or not. The aim of its AI validation platform, Mayo Clinic Platform _ Validationis to provide clinical validation of machine learning algorithms.

The Mayo Clinic Platform also partners with other healthcare organizations to set standards and reporting models as part of The AI ​​Health Coalition. In addition to Mayo, other founding members include UC Berkeley, Duke Health, Johns Hopkins University, MITRE, Stanford Medicine, and UC San Francisco. Industry members include Change Healthcare, Google, Microsoft and SAS.

Halamka announced that a webinar is planned for December 7 that will offer a preview of plans for a public-private partnership to create a national registry to evaluate the utility of various algorithms in health care.

“We think we need a national set of security standards for algorithms,” Halamka said. The registry will host the metadata for algorithms produced in healthcare.

AI market

Health technology companies are also developing marketplaces to improve how collaborative partners, such as providers, payers and research groups, select algorithms.

“Data may be the new oil, but data needs to be refined,” said Wavemaker Three-Sixty Health General Partner Jay Goss. One of the companies in his portfolio, Health gradient, partners with medical data providers around the world (typically hospitals and imaging centers) to prepare annotated medical images for research labs and AI corporations so they don’t have to make one-time deals with hospitals to get the data. Companies can search through segmented and labeled surveys or request a customized dataset, spending less time tracking data and more time developing new tools.

AI hubs

Ferrum Hello developed a program that allows health systems to evaluate machine learning algorithms without exposing their de-identified patient data to the cloud or otherwise forcing them to centralize that data. The company it is part of United Healthcare Accelerator 2022 Cohort, exhibited in the accelerator pavilion. Ferrum’s approach allows these tests to be performed on-site, behind a firewall, an approach that David Miller, Ferrum’s vice president of sales – West, said is designed to de-risk their business for hospitals and health systems. Its marketplace algorithms are FDA approved.

“We’re testing the algorithms using de-identified hospital patient data to show how they work for them,” West said. “We let our customers try it before they buy it.”

The company’s four AI centers include: oncology, orthopedics, cardiovascular and breast care.

Reducing physician burnout

DeepScribe exhibited as part of Plug and Play accelerator conference imprint. Its automated physician natural language processing software automatically summarizes a physician’s conversation with their patient and automatically populates those notes into EMR fields. Among the EMR companies it works with are athenaealth, dr chrono, AdvancedMD and Claimpower.

Earlier this year, DeepScribe shut down a $30 million Series A round to help the company grow. The business is designed to eliminate the need for a GP, saving clinicians money.

The approach to validating algorithms seems like a natural progression in machine learning, much like the rise of digital health applications, followed by the need to validate them to ensure adoption by healthcare organizations skeptical of overly high-tech technologies. It’s a natural progression that balances the interest in machine learning with the recognition that healthcare algorithms are not created equal.

photo: Hemera Technologies, Getty Images

Source link

Leave a Reply

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