Table 1:
Summary of 23 studies that reported on the performance of speech recognition in a clinical documentation task.
| Author | Date | Title | Study Duration (d) | N | n Reports DT | n Reports SR | Accuracy Rate DT (%) | Accuracy Rate SR (%) | Mean Words/ Report DT | Mean Words/ Report SR | Mean Errors/ Report DT | Mean Errors/ Report SR | Mean Dictation Time DT (s) | Mean Dictation Time SR (s) | Mean Editing Time DT (s) | Mean Editing Time SR (s) | Mean TAT DT (min) | Mean TAT SR (min) | n Error Classes | SR Technology Utilized | Article SR Disposition |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Al-Aynati and Chorneyko | 2003 | Comparison of voice-automated transcription and human transcription in generating pathology reports | 21 | 7 | 206 | 206 | 99.60 | 93.60 | 0.398 | 6.655 | 56.80 | 115.05 | IBM Via-Voice Pro 8 with Pathology Vocabulary | + | |||||||
| Basma et al. | 2011 | Error rates in breast imaging reports: comparison of automatic speech recognition and dictation transcription | 487 | 33 | 307 | 308 | 0.225 | 0.698 | 12 | Speech Magic (6.1 SP2, Nuance) | – | ||||||||||
| Bhan et al. | 2008 | Effect of voice recognition on radiologist reporting time | 14 | 5 | 224 | 244 | 14.3 | 83.10 | 94.20 | Powerscribe SR System 4.7 1b, Nuance | + | ||||||||||
| Chang et al. | 2011 | Nonclinical errors using voice recognition dictation software for radiology reports: a retrospective audit | 183 | 19 | 1010 | 0.477 | 6 | Powerscribe 3.5 | Exclude | ||||||||||||
| Chapman et al. | 2000 | Contribution of a speech recognition system to a computerized pneumonia guideline in the emergency department | 42 | Whole Dept. | 400 | 327 | 753 | 133 | Dragon Naturally Speaking | + | |||||||||||
| Chleborad et al. | 2013 | Evaluation of voice-based data entry to an electronic health record system for dentistry | 3 | 131.79 | Dent Voice | + | |||||||||||||||
| David et al. | 2014 | Error rates in physician dictation: quality assurance and medical record production | 1 | 1425 | 966 | 0.315 | 15 | – | |||||||||||||
| Hundt et al. | 1999 | Speech processing in radiology | 4 | 200 | 200 | 88.90 | 135.4 | 150.00 | 204.00 | 66.00 | 1410 | 486 | Philips SP 6000 | + | |||||||
| Ilgner et al. | 2006 | Free-text data entry by speech recognition software and its impact on clinical routine | 1 | 68 | 93.15 | 424.00 | 3720 | 1440 | IBM ViaVoice, German v10 | + | |||||||||||
| Issenman et al. | 2004 | Use of voice recognition software in an outpatient pediatric specialty practice | 30 | 1 | 72 | 99.97 | 90.80 | 259.0 | 225.0 | 0.070 | 19.600 | 240.00 | 180.00 | 17.00 | 336.00 | Dragon Naturally Speaking v6 | – | ||||
| Kanal et al. | 2001 | Initial evaluation of a continuous speech recognition program for radiology | 6 | 72 | 89.70 | 4 | IBM MedSpeak v1.1 | + | |||||||||||||
| Koivikko et al. | 2008 | Improvement of report workflow and productivity using speech recognition – a follow-up study | 274 | 30 | 6037 | 15558 | 1486 | 280 | Philips SpeechMagic | + |
| Author | Date | Title | Study Duration (d) | N | n Reports DT | n Reports SR | Accuracy Rate DT (%) | Accuracy Rate SR (%) | Mean Words/ Report DT | Mean Words/ Report SR | Mean Errors/ Report DT | Mean Errors/ Report SR | Mean Dictation Time DT (s) | Mean Dictation Time SR (s) | Mean Editing Time DT (s) | Mean Editing Time SR (s) | Mean TAT DT (min) | Mean TAT SR (min) | n Error Classes | SR Technology Utilized | Article SR Disposition |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Krishnaraj et al. | 2010 | Voice recognition software: effect on radiology report turnaround time at an academic medical center | 548 | 30 | 149049 | 156843 | 1680 | 762 | Talk Technology v3.0 | + | |||||||||||
| McGurk et al. | 2008 | The effect of voice recognition software on comparative error rates in radiology reports | 7 | Whole Dept. | 727 | 1160 | 0.021 | 0.048 | 3 | Talk Technology V2, DNS | – | ||||||||||
| Mohr et al. | 2003 | Speech recognition as a transcription aid: a randomized comparison with standard transcription | 28 | Whole Dept. | 1287 | 1745 | 276.00 | 246.00 | 1248.00 | 1176.00 | LTI Model-Building Software | – | |||||||||
| Pezzullo et al. | 2008 | Voice recognition dictation: Radiologist as transcriptionist | 91 | 7 | 100 | 100 | 99.90 | 96.00 | 225.6 | 181.7 | 0.120 | 4.400 | 131.00 | 251.00 | 10.00 | 129.00 | Talk Technology V2.1.28, AGFA | – | |||
| Quint et al. | 2008 | Frequency and spectrum of errors in final radiology reports generated with automatic speech recognition technology | 84 | 88 | 265 | 0.362 | 11 | – | |||||||||||||
| Ramaswamy et al. | 2000 | Continuous speech recognition in MR imaging reporting: Advantages, disadvantages, and impact | 274 | 44 | 4552 | 5072 | 92.70 | 95.0 | 60.0 | 5268 | 2616 | 3 | MedSpeak/Radiology v1.2 | + | |||||||
| Rana et al. | 2005 | Voice recognition for radiology reporting: is it good enough? | 30 | 4 | 220 | 220 | 0.660 | 0.890 | 23.37 | 31.70 | ASR Medispeak | + | |||||||||
| Rosenthal et al. | 1998 | Computers in radiology: computer-based speech recognition as a replacement for medical transcription | 14 | 9 | 686 | 656 | 3737 | 1459 | MedSpeak/Radiology v1.2 | + | |||||||||||
| Smith et al. | 1990 | Recognition accuracy with a voice-recognition system designed for anesthesia record keeping | 3 | 31 | 30 | 95.30 | AARK systems X2 | + | |||||||||||||
| Vorbeck et al. | 2000 | Report generation using digital speech recognition in radiology | 2 | 450 | 450 | 99.60 | 94.50 | 303.00 | 253.00 | 329 | 275 | Philips SP 6000 | + | ||||||||
| Zemmel and Neil | 1996 | Status of voicetype dictation for windows for the emergency physician | 7 | 14 | 91.00 | IBM VoiceType Dictation | – |