Artificial intelligence at heart of new Amazon medical documentation program

This image published by Amazon is an illustrative example of the application experience that health care developers can provide users with the online giant's AWS HealthScribe.

Online giant Amazon hopes its new artificial intelligence (AI) program will make physician documentation as easy as ordering new clothes, gadgets, and housewares.

Amazon Web Services Inc. (AWS) announced the new AWS HealthScribe program eligible for use under the federal Health Insurance Portability and Accountability Act (HIPAA). It aims to empower health care software providers “to build clinical applications that use speech recognition and generative AI to save clinicians time by generating clinical documentation.”

The program will “create robust transcripts, extract key details (e.g., medical terms and medications), and create summaries from doctor-patient discussion that can then be entered into an electronic health record (EHR) system,” the company said in a news release.

“Our healthcare customers and partners tell us they want to spend more time creating innovative clinical care and research solutions for their patients while spending less time building, maintaining, and operating foundational health data capabilities,” Bratin Saha, AWS vice president of machine learning and artificial intelligence services, said in the news release. “That is why AWS has invested in building a portfolio of AI-powered, high-performance, and population-scale health applications so that clinicians can spend more time with the patients during the face-to-face or telehealth visits.

“Documentation is a particularly time-consuming effort for healthcare professionals, which is why we are excited to leverage the power of generative AI in AWS HealthScribe and reduce that burden,” Saha said.

Online giant Amazon said its AWS HealthScribe program is designed responsibly with each summary sentence generated by artificial intelligence also linked back to the consultation transcript.

Amazon executives Jason Mark, Sarthak Handa, and Tehsin Syed described the capabilities in an accompanying description published online.

“Imagine a busy day at the clinic, where clinicians spend their day juggling appointments as they try to provide quality care to every patient,” they said, depicting a scenario by now familiar to physicians, other clinicians, and medical students.

The industry requires meticulous documentation, and physicians “struggle between providing compassionate care and maintaining accurate records.”

“This burden takes a toll on both clinicians and patients. Patients receive less attention from their health care providers while the clinicians (face) a higher risk of burnout and decreased job satisfaction,” the Amazon executives said. Medical scribes can help, but they can be expensive to hire, train, and retain, and may face burnout themselves.

Enter AI, advancing with “significantly improved contextual understanding” and uniquely suited for work in health care. The Amazon leaders noted implementing AI can be challenging due to complexity, security, and trust.

AWS HealthScribe aims to provide:

• Rich consultation transcripts
• Speaker role identification
• Transcript segmentation, including periods of silence and small talk
• Summarized clinical notes
• Evidence mapping with reference to the original consultation transcript
• Structured medical terms for conditions, medications, and treatments

Health care software vendors 3M Health Information Systems, ScribeEMR, and Babylon already are using AWS HealthScribe in their clinical applications, according to AWS. The company’s summary included testimonial quotations from leaders of those businesses.