Amazon’s newest service uses machine learning to extract medical data from patient records

Amazon has launched a new service that uses machine learning to extract key data from patient records and can potentially help healthcare providers and researchers save money, make treatment decisions and manage clinical trials. The company announced the service, called Amazon Comprehend Medical, on Tuesday, shortly after The Wall Street Journal reported on it. [Update: An AWS spokesperson clarified that Comprehend Medical is currently available only to AWS customers].

The cloud software combines text analysis and machine learning to read patient records that often consist of prescriptions, notes, audio interviews and test reports. Once those records are digitized and uploaded to Comprehend Medical, it picks out and organizes information about diagnoses, treatments, medication dosage and symptoms.

Amazon’s other recent forays into healthcare include paying almost $1 billion to acquire online prescription service PillPack and a new joint venture with Berkshire Hathaway and JP Morgan Chase to improve employee healthcare. It joins other large tech companies that are increasingly focused on healthcare. For example, earlier this year Apple launched a feature that lets customers view their hospital medical records on their iPhones, while Google recently hired former Geisinger CEO David Feinberg to unify and lead the healthcare initiatives across its businesses, including search, Google Brain, Google Fit and Nest.

In its announcement, Amazon said “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.” The company claimed that Comprehend Medical can accurately identify “medical conditions, anatomic terms, details of medical tests, treatments, and procedures.” In turn, patients can use the service to help manage different aspects of their treatment, including scheduling healthcare visits and prescription medicines or determining insurance eligibility.

Of course, the uploading of medical records to the cloud for machine-learning analysis might beg questions from patients about how Comprehend Medical will ensure their privacy. Amazon says patient data is encrypted and can only be unlocked by customers who have a key, and that no data processed will be stored or used for training its algorithms. Comprehend Medical complies with the Health Insurance Portability and Accountability Act (HIPAA).

Comprehend Medical is already being previewed by Roche Diagnostics, the Switzerland-headquartered pharmaceutical and diagnostics equipment company, and Fred Hutchison Cancer Research Center in Seattle, which is using it to identify patients for clinical trials. By using the software to analyze “millions of clinical notes,” Amazon says the center was able to reduce the time it needed to process each document “from hours, to seconds.”

In a statement, Matthew Trunnell, the CIO of Fred Hutchinson Cancer Research Center, which studies cancer and conducts clinical trials and volunteer studies on new treatments, said “For cancer patients and the researchers dedicated to curing them, time is the limiting resource. The process of developing clinical trials and connecting them with the right patients requires research teams to sift through and label mountains of unstructured medical record data. Amazon Comprehend Medical will reduce this time burden from hours per record to seconds. This is a vital step toward getting researchers rapid access to the information they need when they need it so they can find actionable insights to advance lifesaving therapies for patients.”