Tech giant Amazon is reaching deeper into healthcare with a new speech recognition service for clinical documentation.
At its Re:Invent conference in Las Vegas Monday, Amazon Web Services announced the launch of Amazon Transcribe Medical to convert clinician and patient speech to text. The service is currently available in northern Virginia on the East Coast and Oregon on the West Coast.
AWS launched Amazon Transcribe in 2017 as an automatic speech recognition service to enable developers to add speech-to-text capability to their applications.
In 2018, AWS rolled out Amazon Comprehend Medical, a machine learning tool that pulls out medically relevant information such as patient diagnoses, symptoms, medical test details, treatments and dosages while simultaneously highlighting any protected health information. That tool allows developers to process unstructured medical text.
Transcribe Medical is a machine-learning-powered medical transcription service that utilizes automatic speech recognition applications to offer a more efficient and accurate medical note-taking experience, AWS said in a press release.
According to AWS, Transcribe Medical is HIPAA-eligible, meaning healthcare organizations can use these services in compliance with the Health Insurance Portability and Accountability Act (HIPAA). The service also offers an easy-to-use application programming interface that can integrate with voice-enabled applications and any device with a microphone.
Clinical speech recognition is one of many areas in which Amazon is competing with Microsoft and Google. Microsoft is collaborating with Nuance Communications to develop technology to "listen" to physician-patient conversations and automatically document in an electronic health record (EHR). Stanford University is working with Google on a digital scribe pilot project to use voice assistants during patient encounters.
As tech giants move further into healthcare, it raises some thorny issues around connections to patients' health data. Google is facing significant blowback, including scrutiny from regulators and lawmakers, following news that it is collecting personal health information on millions of patients in a partnership with Ascension.
AWS executives say Transcribe Medical helps solve the current challenge with getting accurate medical transcriptions. Many physicians use a recorder to dictate notes that are sent to a third party who manually transcribes the voice file, an expensive and time-consuming process that takes multiple days to complete, the company said.
"Others choose to use human scribes, which can be distracting and uncomfortable for patients and clinicians. Some organizations have tried to use existing medical transcription software, but complex medical language can be difficult to transcribe, leading to inefficiency and poor accuracy that can cause serious consequences," AWS said.
By utilizing voice-enabled technologies, Transcribe Medical automatically translates audio streams into medical speech and is designed to be deployed at scale to patients’ healthcare providers, allowing for affordable, consistent, secure and accurate note-taking for clinical staff and facilities, AWS officials said.
In a blog post, Julien Simon, a technical evangelist at AWS, said the service will help reduce the documentation burden for physicians and clinicians. A 2017 study by the University of Wisconsin and the American Medical Association found that primary care physicians in the U.S. spend six hours per day entering their medical reports in EHR systems.
"I don’t think that anyone would argue that doctors should go back to paper reports: working with digital data is so much more efficient. Still, could they be spared these long hours of administrative work? Surely, that time would be better spent engaging with patients, and getting a little extra rest after a busy day at the hospital?" Simon wrote.
The medical transcription service enables physicians to easily and quickly dictate their clinical notes and see their speech converted to accurate text in real time, without any human intervention, Simon wrote.
Clinicians can use natural speech and do not have to explicitly call out punctuation like “comma” or “full stop.” This text can then be automatically fed to downstream applications such as EHR systems, he said.
Amazon has been rapidly expanding its reach in the healthcare space, most notably with its acquisition of PillPack last year. The tech giant announced its employee healthcare program, called Amazon Care, in September and said the new service combines "the best of both virtual and in-person care" by offering virtual visits, in-person primary care visits at patients' homes or offices and prescription delivery.
The company's acquisition of startup Health Navigator builds on that service by providing preliminary and final diagnoses and treatments on a digital platform.