Speech recognition's potential in mHealth continues to grow

Last fall, I noted how reliable speech recognition would be key to turning smartphones and tablets from basic record viewers into true tools for physician workflow. Without good keyboard options, most mobile devices simply aren't set up for physicians to make changes to patient records, enter notes, dictate orders and the like.

But a few moves are happening in the speech recognition market, giving hope to physicians who love their iPads, but have to keep finding a desktop or laptop to do real work.

Nuance, for instance, recently added some interesting functionality to its core speech recognition system, according to a story in USA Today. The company is pushing a new product that it can embed into your electronic health record system, allowing doctors to dictate directly into a patient's record.

What's more interesting: The system has the beginnings of artificial intelligence, which Nuance has been working on with University of Pittsburgh Medical Center. It can check medical information, identify inconsistencies in the record and even edit the text, Nuance officials tell USA Today. Researcher Rasu Shrestha explains it this way: If a doctor doesn't give the level of acuity for a patient's cardiac event, for example, the system will ask questions such as whether the problem was diastolic or systolic.

"A physician may know what to say, but not always how to say it," Shrestha says. "The idea is to get actionable data while you're interacting with the patient."

M-Modal, one of Nuance's main rivals also has a product similar to Nuance's that offers physicians the ability to record audio notes, dictation, etc., onto their mobile devices. "They don't want to point and click," M-Modal co-founder Juergen Fritsch tells USA Today.

Industry consultant Reda Chouffani adds that the system also can allow physicians to prescribe meds through voice recognition. "Jaws were dropping at a [recent industry] conference," he says.

But the new functionality, while exciting, doesn't mean that speech recognition is completely ready for prime time. A report last fall found that breast-imaging reports transcribed with speech recognition--rather than a live transcriptionist--were eight times as likely to have major errors.

There clearly remains a ways for the technology to go. Still, it's going to be exciting to see how fast it gets there. - Sara

Suggested Articles

The newly launched Center for Connected Health will be largest telehealth hub in the Philadelphia region, according to Penn Medicine.

The FDA commissioner wants to use additional funding under Trump's budget to advance digital health initiatives and integrate real-world data.

The FDA's approval of an app that uses AI to notify specialists of a potential stroke offers new possibilities for triage software that uses CDS.