![]() ![]() When the system makes an error, it is important to correct the error by voice and not by the keyboard, because the program will not improve its accuracy unless it “hears” the corrections. ![]() Although geared to the North American accent, the system can quickly adapt to recognize other speech patterns. When first enrolling in a program, a pathologist reads preselected passages for about 5 to 10 minutes to allow the program to create a voice profile associated with the speaking style and accent of the user. VR programs, such as Dragon Naturally Speaking–Medical Solutions ( have extensive built-in dictionaries. Difficulties often result from resistance to innovation. It is important, however, to be patient during the initial start-up period. The technical difficulties inherent in prototypical VR programs have been overcome, and the fluidity with which continuous (versus discrete) speech is captured is impressive. The installation of a VR system does not guarantee a satisfactory outcome, and not because of the technology. Regarding turnaround, the advent of the Tissue Tek Continuous Rapid Processor and the Tissue Tek Auto Tek Automated Embedding System (Sakura Finetek USA, Inc., Torrance, Calif.) may eventually allow some leveling of the playing field. There is, of course, another negative aspect: the faster one is able to work, the faster one is expected to work. Once the pathologist has overcome the initial barriers to VR, the independence offered by this technology is remarkable. ![]() For uncomplicated cases (generally the bulk of one's workload), the turnaround time can be exceptional the final report, once triggered, can be autofaxed almost immediately. There is a certain satisfaction from being able to control the entire composition of the gross and microscopic report without resorting to intermediaries. ![]() This factor, however, diminishes if one is conscientious in “training” one's computer. Granted, they are already required to review and approve text, but the additional act of primary composition on the screen subtly increases the intensity of the work, especially when first learning VR. On the negative side is the fact that pathologists will absorb the ab initio focus on text and composition. It would be a mistake, however, to attempt a significant reduction or re-allocation of staff until there is full confidence in the completely installed and tested module. Because VR creates text from dictated words, the required level of secretarial support may decline, and because financial reality imposes a will of its own, eventual staff reduction becomes almost a quid pro quo during negotiations with administrators. Also, VR systems do not take vacations, become ill, or require pension benefits. On the positive side are speed and the lure of rapid turnaround. The technology allows dictation at up to 160 words per minute.Ī successful switch to VR has major implications (positive and negative) for one's patterns of practice and staffing. The system deploys Voiceover, a server-based version of the Dragon core engine (developed by Voicebrook, Lake Success, NY that integrates with applications not built for speech. The Department of Pathology at North Shore University Hospital (Manhasset, NY) has introduced voice recognition (VR) technology-the automatic conversion of speech to text ( Fig. Digital recording has replaced tape recording in many places and offers several advantages in managing workflow, such as the automatic assignment of priorities in typing or the monitoring of individual work products. However, cassette contamination and the poor resistance of these machines to chemical onslaught are well-known problems. NOCHOMOVITZ, in Modern Surgical Pathology (Second Edition), 2009 REPORT DICTATION AND VOICE RECOGNITIONįor simple report dictation, an ordinary tape recorder suffices when there are no alternatives. ![]()
0 Comments
Leave a Reply. |