Amazon confirmed on Tuesday that it’s launched a project to mine data from electronic medical records, as the company pushes deeper into the health-care market.
Coinciding with the Amazon Web Services re:Invent conference in Las Vegas this week, the company introduced Amazon Comprehend Medical, which “allows developers to process unstructured medical text and identify information such as patient diagnosis, treatments, dosages, symptoms and signs, and more,” according to a blog post.
The announcement follows a CNBC report in June that Amazon had spent several years working on an effort, internally dubbed Hera, to make better use of digitized medical data, capturing information that a physician may miss and helping remove inaccuracies. At that time, Amazon was targeting insurance companies as potential customers, pitching them on a product that could fill in gaps in data when doctors neglect to fully document a patient’s medical visit.
The company’s senior leader in health care and artificial intelligence, Taha Kass-Hout, told the Wall Street Journal on Tuesday that internal tests showed that the software performed as good or better than other published efforts to extract data on patients’ medical conditions, lab orders and procedures.
The company also confirmed that it’s teaming up with Fred Hutchinson Cancer Research Center in Seattle to apply machine learning to its data sets in a bid to prevent and cure cancers. Amazon said it’s working with the health center to evaluate “millions of clinical notes to extract and index medical conditions,” and it pointed to other partners, including pharmaceutical giant Roche.
It looks unlikely at this point that Amazon will compete directly with medical records companies like AllScripts and Cerner, as there are plenty of money-making opportunities to work with those vendors and provide services like population health and clinical trial support.
Rather, Amazon is most directly taking on UnitedHealth Group’s Optum, which is already in the space, as well as technology rivals Apple and Alphabet. Google is working with local health systems, notably the University of California, San Francisco, to get valuable information from medical records and better predict when a patient is likely to get sick. Apple is partnering with hospitals and clinics to provide patients with access to their medical information on an iPhone.
Amazon said the reason it got into this space is to help speed up the process of making sense of health data, which isn’t usually stored in ways that computers can understand and analyze.
“The majority of health and patient data is stored today as unstructured medical text, such as medical notes, prescriptions, audio interview transcripts, and pathology and radiology reports,” Amazon said in the blog post. “Identifying this information today is a manual and time consuming process, which either requires data entry by high skilled medical experts, or teams of developers writing custom code and rules to try and extract the information automatically.”