Abstract
Although much has been researched about the benefits of continuous speech recognition (CSR) in various fields such as radiology and electronic medical records, little evidence is available on the efficacy of CSR in cardiovascular medicine. In the present study, we compared conventional typing with a commercially available CSR system, AmiVoice(R) (Advanced Media, Inc., Tokyo, Japan) in making ambulatory electrocardiographic (AECG) reports. Report turnaround time by CSR contains additional time for correction of recognition errors. A total of 278 AECG reports generated between December 2009 and March 2010 were assessed. One hundred thirty-nine (50.0%) were dictated using CSR system and the same number of reports were generated by typing. Reporting time by CSR was statistically shorter than that by typing (median : 764 sec vs. 883 sec, p<0.001, Mann-Whitney U test) , although there was no significant difference in patient characteristics, total character counts or appreciable findings in reports between the two methods. Furthermore, this tendency still existed regardless of the number of reported findings associated with significant arrhythmia and myocardial ischemia. In conclusion, CSR dictation for AECG reporting, even considering the additional time required for corrections, may be a favorable alternative to conventional typing with the promise of time reduction in AECG report production.
