Amazon's AI tool for mining EHRs isn't new tech, academics say

Amazon's new EHR-mining tool marks another step in the e-commerce giant's move into healthcare. However, some industry players are skeptical of whether the new software is truly groundbreaking, according to Politico.

The software, dubbed Amazon Comprehend Medical, uses natural language processing and machine learning — two types of artificial intelligence — to review EHRs and unstructured clinical notes to highlight data that may help physicians improve patient care. Amazon said the insights will be able to help hospitals with clinical decision support, revenue cycle management and population health.

However, this type of natural language processing service isn't new. In fact, one academic informaticist who spoke with Politico said: "Is it news that a giant company with relatively unlimited resources can match the results of academics with paltry grants?"

In addition to academics, other companies in the health technology space have released similar products applying AI to patient data. UnitedHealth Group's OptumLabs, for example, offers a machine learning platform that works with a database containing information on 60 million patients. Allscripts and Cerner have also made advancements in this area.

The Veterans Health Administration, in particular, has used natural language processing to process and extract information from thousands of physicians' notes for the past decade.

Amazon worked with the Seattle-based Fred Hutchinson Cancer Research Center to train its EHR-mining software to extract relevant information from patient records. Emily SIlgard, a data scientist at the research center, told Politico that the new tool's benefit comes from its convenience. "That's a big barrier, especially in academic institutions that are notoriously low-resource," she said.

More articles on artificial intelligence:
CMS to launch AI challenge
NIH taps Nvidia to co-develop AI tools for brain, liver cancer clinical trials
NYU School of Medicine releases 10K MRI scans as part of Facebook collaboration

© Copyright ASC COMMUNICATIONS 2018. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months