DeepScribe raises $5.2M to transcribe medical notes with AI

Mobile

Join Transform 2021 this July 12-16. Register for the AI event of the year.


DeepScribe, an AI-driven platform for medical record-taking, today announced it has raised $5.2 million in a seed round led by Bee Partners. The company says the funding will be put toward product research and development, as well as hiring efforts.

Physicians spend on average 16 minutes updating electronic health records (EHRs) for each patient visit, a 2020 study in the Annals of Internal Medicine found. Typically, chart review (33%), documentation (24%), and ordering (17%) account for most note-taking activity. But the compressed time frame tends to promote errors. According to research published in the Journal of the American Medical Association, just 18% of a typical medical note is manually entered by the doctor, leaving 46% copied and pasted from other parts of the medical record and 36% imported by the electronic medical software.

DeepScribe, launched in 2019 by Akilesh Bapu, Matthew Ko, and Kairui Zeng and based on technology developed at the University of California, Berkeley, is designed to transcribe patient notes even when dealing with accents, interruptions, multiple speakers, and more. The platform combines AI with rules-based natural language processing, as well as the output of three speech recognition engines — including its own — each with distinct advantages. For example, DeepScribe leverages IBM’s speech recognition technology to recognize medications and conditions and Google’s tech for conversational speech.

“Though difficult to pull off, this approach makes DeepScribe the only AI widely accepted by medical professionals that can draft medical notes from the natural patient conversation,” Bapu told VentureBeat via email. “On average, physicians only have to make one correction or less after 20 days of using the product. The rules-based natural language processing portion of DeepScribe’s approach also allows providers to custom-tailor their notes without having to retrain deep learning models, which can be unpredictable at times.”

DeepScribe’s platform captures both sides of doctor-patient interactions and distills them to their essence without the need for the doctor’s intervention. Even in the absence of wake words and instructions, DeepScribe can listen to an appointment in the background via an iOS or web-based app, take notes, and automatically summarize the discussion into an entry in a doctor’s EHR system. The transcribed summaries are customizable, allowing doctors to create tailored notes including long sentences or short, a conversational or official tone, and a summary of everything discussed or just what’s medically relevant.

DeepScribe interfaces with major EHR platforms, including AdvancedMD, AthenaHealth, Claimpower, Elation, DrChrono, and PracticeFusion. The startup also builds custom integrations that can transfer notes into discrete fields within EHR systems.

“Many doctors are turning to scribes — human and mechanical — to reduce the strain of EHR paperwork. The problem is, human scribes are too expensive for many practices, and AI scribes aren’t accurate enough to rely on. DeepScribe is the best of both worlds: quality equal to or better than a human scribe at a cost any physician can afford,” Bapu continued. “Unlike other AI scribes, which need wake words or explicit direction, DeepScribe is an ‘ambient AI’ that [simply] listens in.”

Medical transcription market

The closure of DeepScribe’s financing comes after Microsoft acquired Nuance Communications, an AI and speech technology company with a large health provider client base, for $19.7 billion. The global medical transcription market is anticipated to reach $4.89 billion by 2027, up from $1.32 billion in 2019, according to Fortune Business Insights. Beyond DeepScribe, Nvidia, Google, and Amazon are among those who’ve built health note transcription solutions.

The increasing investment in the segment is in part motivated by research showing that EHR responsibilities can have a negative impact on patient care. A Stanford Medicine-commissioned survey found that 7 out of 10 physicians believe EHRs “greatly” contribute to physician burnout. The same study suggests 6 out of 10 doctors think EHRs need a “complete” overall.

DeepScribe claims to have over 300 customers who collectively see 10,000 patients a week across 18 different specialties.

“The pandemic helped accelerate our growth as the dire need of frontline providers put a magnifying glass on the inefficiencies leading providers to see fewer patients,” Bapu said. “It also became unsanitary and a friction point to have in-person scribes, given the spread of COVID, along with the increased frequency of virtual visits. These two factors increased the demand for our product and exploded our growth.”

Industry Ventures and Stage II Capital participated in the seed round, along with existing investors Tsingyuan Ventures, 1984 Ventures, Wavemaker 360, Supernode Ventures, Skydeck, Plug and Play, and Sequoia Scout Fund. DeepScribe has 26 employees.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Products You May Like

Leave a Reply

Your email address will not be published. Required fields are marked *