Amazon Software Sifts Medical Records for Data to Improve Care, Cut Costs

What’s New?

Amazon has introduced a new cloud-based software service that reportedly will sift through patient medical records in seconds searching for actionable intelligence to provide hospitals and physicians with clinical decision support and coding while reducing costs. Amazon Comprehend Medical can identify health information such as medical conditions and medications and determines their relationship to each other (for example, medication dosage and strength). The software can be accessed through a simple application program interface with no machine-learning expertise required.

This is the latest move by the online retailer into the health care space after previously beginning to sell medical supplies through its business marketplace. And earlier this year, Amazon paid $1 billion to buy PillPack Inc., giving the online retailer the ability to ship prescription drugs.

What It Means

If the new service delivers on its promise, it could provide significant efficiency gains for health care organizations by providing the ability to harness clinical notes to drive predictive analytics and assist in personalized treatments. Currently, hospitals have to undergo costly, time-consuming steps to extract meaningful data from unstructured medical text such as medical notes, prescriptions, audio interview transcripts, and pathology and radiology reports.

As TechNewsWorld reports, health care providers and insurers currently have to write and maintain a set of customized medical coding classification rules for natural language software. A change to a single classification code name can have a ripple effect on dozens of hard-coded rules, all of which then must be updated, or data will be missed or classified incorrectly. Amazon Comprehend Medical lets users create models that reliably understand this medical information, vastly improving medical coding and the ability to mine this data more quickly to provide clinical support.

Amazon says that one day, the richness of the information may be able to help patients manage their own health, including medication management, proactively scheduling care visits, or empowering them to make better-informed decisions about their health and eligibility.

The Wall Street Journal reports that Amazon software developers trained the system using a process known as deep learning to recognize the many ways a doctor might record notes. During testing, the software performed as well or better than other published efforts, and can extract data on patients’ diseases, prescriptions, lab orders and procedures, a senior leader with Amazon’s health care and artificial intelligence efforts told the Journal.

Fred Hutchison Cancer Research Center in Seattle, which helped Amazon test and train its algorithm, is using the software to identify patients eligible for studies of experimental drugs. Amazon says the center used the software to analyze millions of clinical notes and cut the time needed to process each document from “hours to seconds.”

Still, patients may question how Amazon will ensure that their privacy is protected with machine-learning analysis. Amazon says patient data is encrypted and that no data processed will be stored or used for training its algorithms. The company also notes that Comprehend Medical is HIPAA-compliant.