Voice recognition technology implementation in surgical pathology: advantages and limitations

Arch Pathol Lab Med. 2011 Nov;135(11):1476-81. doi: 10.5858/arpa.2010-0714-OA.


Context: Voice recognition technology (VRT) has been in use for medical transcription outside of laboratories for many years, and in recent years it has evolved to a level where it merits consideration by surgical pathologists.

Objective: To determine the feasibility and impact of making a transition from a transcriptionist-based service to VRT in surgical pathology.

Design: We have evaluated VRT in a phased manner for sign out of general and subspecialty surgical pathology cases after conducting a pilot study. We evaluated the effect on turnaround time, workflow, staffing, typographical error rates, and the overall ability of VRT to be adapted for use in surgical pathology.

Results: The stepwise implementation of VRT has resulted in real-time sign out of cases and improvement in average turnaround time from 4 to 3 days. The percentage of cases signed out in 1 day improved from 22% to 37%. Amendment rates for typographical errors have decreased. Use of templates and synoptic reports has been facilitated. The transcription staff has been reassigned to other duties and is successfully assisting in other areas. Resident involvement and exposure to complete case sign out has been achieved resulting in a positive impact on resident education.

Conclusions: Voice recognition technology allows for a seamless workflow in surgical pathology, with improvements in turnaround time and a positive impact on competency-based resident education. Individual practices may assess the value of VRT and decide to implement it, potentially with gains in many aspects of their practice.

MeSH terms

  • Feasibility Studies
  • Humans
  • Pathology, Surgical / instrumentation
  • Pathology, Surgical / methods*
  • Pilot Projects
  • Speech Recognition Software*