Recent advances in speech recognition technology are making it easier for healthcare providers to reconcile the conflicting demands of free-text reporting and structured data.

M*Modal, a maker of healthcare-specific speech recognition software, will demonstrate the latest release of its Fluency for Imaging Reporting next week at the RSNA meeting (Booth 1317). The product features flags and alerts that help radiologists remember to dictate all the relevant information they need to comply with regulatory requirements and new coding demands, without making them fill in structured documents. Some of the new features include:

* Actionable findings: Feedback loops automatically detect 20 findings that need immediate attention from referring physicians, and allow radiologists to communicate them to referring physicians from within their normal workflow.

* Physician Quality Reporting System measures: Radiologists can be reminded to include all required PQRS information in their dictation so that they receive incentive payments and adjustments.

* ICD-10 reminders: Radiologists can be reminded to include information required for complete ICD-10 coding, such as the gender of the patient and the laterality of the images, if needed. The software also reminds them to supply complete organ information on abdominal ultrasounds, which is key to getting fully reimbursed.

The software can pull data from other systems, such as PACS, EHR, RIS, and the imaging modalities themselves. For example, it can automatically populate the radiology report with radiation dose information from CT scanners so that the provider is in compliance with new Joint Commission requirements on dose reporting.

The software processes and interprets the radiologist’s speech during the dictation, allowing for real-time alerts, says Russ Cardwell, M*Modal’s Vice President of Imaging Solutions. “It understands the full meaning of what’s being spoken, and makes a higher quality document.” For example, if the radiologist reports a pneumothorax (a collapsed lung), the software can tell whether it’s a current condition, possibly requiring immediate attention, or part of the patient’s history.

The software also reminds the radiologist to follow up with the referring physician on incidental findings--for example, an x-ray for chest pain that shows a small and probably unrelated lung nodule. “Our technology enables you to mark it, flag it for follow-up, and start closing the care gap,” Cardwell said.

The included alerts are optional; providers can choose which ones to turn on in order to prevent alert fatigue. Some disappear automatically, without having to be manually closed, when the radiologist dictates the appropriate information.

M*Modal has about 200 customers for its radiology reporting products, and 10 are live with the new alerting capabilities. Cardwell declined to give pricing information, but said the cost of the system is about 25 to 35 percent of the cost of transcription.

Cardwell expects that meeting attendees will challenge the accuracy of the product with a variety of accents, but it will be ready for anything. “Our cloud intelligence service takes snippets from the speech of all our users and makes its speech recognition engine more accurate,” he said. ”It’s heard the Pakistani from the bayou before.”

Register or login for access to this item and much more

All Health Data Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access