Today, 85 percent of radiologists use a speech recognition system to record their reports while 15 percent continue to dictate a report and send it to a transcriptionist, according to new research from Reaction Data. In 2016, 83 percent used speech recognition, 8 percent dictated for transcription, and 9 percent used other means, the study reported.

WHY IT MATTERS

The Reaction Data report, which polled more than radiology executives and clinicians across the U.S., is significant in large part because it identifies the winners among vendors looking to capture the speech of U.S. radiologists.

Nuance dominates the radiology speech recognition market with an 81 percent share, followed by M*Modal at 13 percent, Dolbey at 5 percent and IBM at 1 percent, the study found.

The study also found that of the 15 percent of provider organizations that have not yet adopted speech recognition, 41 percent said they have no plans to ever adopt the technology. And on another note, executives and clinicians seem to be happy with their speech recognition technology, as only 2 percent said they are considering a new vendor.

When it comes to systems integration, more than half (53 percent) of provider organizations are integrating speech recognition into their PACS systems, the study found. Those numbers drop significantly as radiology information systems (28 percent) and electronic health records (19 percent) integration are a distant second and third.

This is not necessarily surprising since a PACS system is the primary system a radiologist uses by an order of magnitude over EHR and even RIS, the study added.

WHAT IS THE TREND

The dominance of Nuance among radiology speech recognition vendors is evident around the industry. Take, for example, Mercy Technology Services. The IT division of the sprawling St. Louis-based health system has developed a new cloud-based imaging platform and is commercializing it for other hospitals to deploy, built in part using Nuance technology.

The PACS system comprises best-of-breed enterprise viewer, vendor neutral archive, workflow orchestrator, speech recognition and reporting, according to Mercy, bundled as a secure software-as-a-service model aimed at small and midsize hospitals.

Mercy consolidated its own imaging platform with help from MTS, distilling nine legacy PACS systems into a single hosted technology. The system combines server-side image processing from Visage, a workflow orchestrator from Medicalis, and, notably, speech recognition from Nuance.

ON THE RECORD

Overwhelmingly, radiologists are using speech recognition, and this trend is unlikely to change anytime soon, the Reaction Data study said.

"And, despite the fact that no system is perfect, overall, radiologists are incredibly pleased with the performance of their chosen solution," the study added. "With 98 percent of organizations indicating they wouldn't switch vendors, it's also clear that all of the vendors are doing a great job, from a high level at least. Simply put, speech recognition has become a key part of a radiologist's workflow."

Twitter: @SiwickiHealthIT
Email the writer: bill.siwicki@himssmedia.com