Computerized speech recognition software has traditionally been used within the bounds of clinical radiology. But a new report in Current Problems in Diagnostic Radiology by Boston researchers suggests the software may also be useful in the development of academic papers in radiology.
“Speech recognition software in radiology offers immediate, real-time visualization of the transcribed words, rapid turnaround times, increased flexibility in the use of standardized templates, and reduced cost of labor for human transcription services,” wrote lead author Raul N. Uppot, MD, of Harvard Medical School and Massachusetts General Hospital in Boston, and colleagues.
Approximately 20 percent of radiologists institutions use speech recognition software to prepare documents other than clinical reports. Uppot and colleagues shared some of the benefits of utilizing transcription software to dictate academic materials.
Speech recognition software allows for faster composition of manuscripts.
Speech recognition can transcribe words at an average speed of 160 words per minute, compared to the average typist who can type only 38 to 40 words per minute. This improve efficiency may allow a user to “think out loud,” the authors wrote. The feature also helps composers express their ideas.
Faster composition is also aided by knowledge of the information before dictation. Additionally, an outline of the abstract, introduction, body, materials and methods, results, discussion and conclusion will aid in the development of the manuscript.
Text may be viewed in real-time and dictation allows for improved flow of the manuscript.
Manuscripts that are composed by dictation typically follow thought processes, which improve flow. Furthermore, visualization of the manuscript in real-time may be edited manually. A unique advantage of utilizing speech recognition software is its ability to recognize the speech patterns of individual radiologists.
Uppot and colleagues moted limitations of speech recognition software, including limits in usage time, incorporation of unintended words, and the potential for corrupted files.
Still, the authors concluded, speech recognition in academic radiology has potential for those who may want to produce scientific manuscripts using familiar technologies linked to daily clinical duties.
As a senior news writer for TriMed, Subrata covers cardiology, clinical innovation and healthcare business. She has a master’s degree in communication management and 12 years of experience in journalism and public relations.