Kenneth Crane BS, Presenter: Nothing to Disclose

Barton F. Branstetter MD, Abstract Co-Author: Nothing to Disclose

Juergen Fritsch PHD, Abstract Co-Author: Nothing to Disclose

Paul Joseph Chang MD, Abstract Co-Author: Nothing to Disclose

The transition from manual transcription services to speech recognition (SR) technology can temporarily decrease efficiency across the radiology department, as the SR system is individualized to its users. The purpose of this study was to determine whether a combination of manual transcription and SR could mitigate these transitional costs and improve radiologists’ acceptance of new SR systems.

We developed a software system that allows SR technology to be combined with traditional manual transcription services in the production of radiology reports. During the initial period of SR use, reports are first analyzed by the SR software, and then sent to a manual transcriptionist who proofreads the report for accuracy. This model, in which the transcriptionist is acting as a proxy for the radiologist in the proofreading of reports, allows radiologists to begin using the SR system without undergoing formal training sessions. When the error rate of the SR system becomes acceptably low, the manual transcriptionist can be excluded from the system. We measured the amount of time required for a transcriptionist to correct a SR report, and compared it to the time required to manually transcribe a report. We assessed the acceptance of the new system with regard to radiologists and transcriptionists, and compared radiologist productivity (number of cases and turnaround time) before and after the introduction of SR.

The overall acceptance of the combined system was excellent for both radiologists and transcriptionists. Transcriptionists were 3 times faster when proofreading than when transcribing reports. The combined system improved overall radiologist efficiency because the proofread reports had fewer recognition errors. As expected for any SR system, overall report turnaround time dropped dramatically.

SR can be introduced to radiology departments with an immediate positive impact on the efficiency of radiologists and transcriptionists. The combination of SR and manual transcription provides a smooth transition and improves acceptance of SR.

K.C.,B.F.B.,P.J.C.: Laboratory Support from Stentor, Inc., DARPA, NLM, NIH, US Air Force

Crane, K, Branstetter, B, Fritsch, J, Chang, P, Introducing Speech Recognition Technology without Disrupting Radiologist Efficiency: How to Have Your Cake and Eat It, Too.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4417876.html