Among various technologies in the healthcare world, artificial intelligence has found a much-welcomed role in the tedious but detail-oriented world of medical transcription – freeing physicians to spend more time with patients than on their computers.
Jared Pelo, MD, chief clinical product officer, Nuance Healthcare, helps lead the ambient clinical intelligence team and is working to improve the usability and versatility of A.I. products designed to remove much of the time-consuming work of medical documentation through automation. He said a major technological rollout planned for 2022 is the debut of a new “express mode” for Nuance’s Dragon speech recognition software. According to Pelo, the product more accurately and automatically transcribes patient-doctor conversations for EHR use.
“It’s a doctor and a patient having a normal conversation. That conversation is recorded. The doctor is not thinking about their documentation; they’re just thinking about the patient,” Pelo said of a current Dragon experience during an episode of the MGMA Business Solutions podcast. “Then, that recording is put through our A.I. cloud. We use speech recognition, natural language processing and natural language generation to go ahead and generate a clinical note. That clinical note then goes through a review process and is put in the EHR.
“[Express mode] is where we’ve collected enough data for certain users in their A.I. draft notes – the notes that go to a reviewer are getting high enough quality that they will be going straight to the provider. The provider will review those notes and decide if they want to edit them themselves or if they want to send them to review. That’s our big goal. That’s what we’re marching towards is how quickly can we make A.I. that’s scalable available to the masses of clinicians who really need this help in their day-to-day lives.”
Hear more from Pelo in this episode of the MGMA Business Solutions podcast:
MGMA · Business Solutions: How AI Offsets Administrative Burdens, Creates Better Practice Efficiencies
Pelo admitted that this is a quantum leap from the world of in-person medical scribes, the grunt-work job he remembers from his days as a medical student and as a first-time healthcare manager.
“I kind of fell in love with this idea of someone else doing my documentation just early in my training,” he said. “Now, we’ll skip forward six years to when I’m medical director at this small emergency department. I just kept thinking, OK, there’s two things that I want. One is I want the best doctors I can possibly hire. And two is I want to make them as efficient as possible. I just was super-frustrated with the inability to be in this small town to hire medical scribes. The talent wasn’t there. I knew what it took to train scribes, retain scribes, pay scribes, all of that overhead. So, that’s really where I started thinking about ambient intelligence. What if we just started using technology to capture the right data to be able to move down this path? It was kind of a turning point in my career.”
Additional resources:
Pelo said that A.I. applications serve as the ultimate replacement for the often thankless work done by medical scribes, who, like a past version of himself, are often medical students with a short commitment to doing that important transcription work.
“A.I. continues to get smarter and smarter. You don’t have to retrain it. It’s scalable. It’s going to be with you forever,” Pelo said. “It doesn’t call in sick. You don’t have to give it an extra computer in the office. You don’t have to make sure that A.I. does HIPAA training every year. Great medical scribes have been a life raft for many physicians over the last 10 years, but I think we’re at the point where A.I. is getting good enough where we need to start investing in solutions that will be with us for our whole career, not for just the next six months.”