Radiology reporting is undergoing the same digital revolution as image acquisition. A recent pilot study took one of the first looks at the effect of structured reporting/speech recognition (SR) on radiologist report creation time, as compared with
Radiology reporting is undergoing the same digital revolution as image acquisition.
A recent pilot study took one of the first looks at the effect of structured reporting/speech recognition (SR) on radiologist report creation time, as compared with conventional dictation.
SR systems may have the potential to replace conventional dictation and transcription, but results of this study showed clerical times and viewing-only times were both longer for SR, by 115 seconds and 105 seconds per report, respectively.
"This is likely due to limitations of the image review station at the SR site, factors that will be corrected by integrated imaging-reporting workstations," said lead author Dr. Curtis Langlotz, president of eDict Systems.
Langlotz also expects reporting-viewing times, which were 76 seconds longer per report for SR, to decrease as users gain more experience with the system and as reporting macros are developed.
While the pilot study was limited to one radiologist still learning to use the SR system, other caution lights flash on the road to SR. The rise of PACS and speech recognition technologies has created new problems within some radiology departments. These technologies, designed to enhance communications capabilities, have seriously reduced direct clinician-radiologist communication.
"Fast and easy access to patient images and reports has had a detrimental effect on face-to-face consultations with clinicians, which were commonplace before PACS and now have almost completely disappeared," said Dr. David B. Hayt, director of radiology at Elmhurst Hospital in Queens, NY, in a paper in the September issue of the Journal of Digital Imaging (2001;14(3):149-57).
Radiologist-clinician exchanges, which occurred frequently before a final report was dictated, often resulted in better understanding of the clinical problem and a more meaningful final report, Hayt said. Although conferencing features facilitate communication within the PACS, they are not used by clinicians.
"The dilemma is that as information about patients is made more available to the hospital staff, less information is provided about patients to the radiologists," he said.
SR systems may benefit the hospital, its staff, and the patients served by reducing clinician time awaiting a diagnostic report and patients' waiting time in the clinic and emergency room. But they do not necessarily benefit the radiologists who use them. Speech recognition dictation systems slow down the individual productivity of the radiologists' dictation process by at least 25%, Hayt said.
"Radiologists are assuming the role of transcriptionists as well as diagnosticians. Mistakes occur that would not with the use of a traditional dictation system and professional transcriptionists," he said.
Comparative AI Study Shows Merits of RapidAI LVO Software in Stroke Detection
February 6th 2025The Rapid LVO AI software detected 33 percent more cases of large vessel occlusion (LVO) on computed tomography angiography (CTA) than Viz LVO AI software, according to a new comparative study presented at the International Stroke Conference (ISC).
Study: Mammography AI Leads to 29 Percent Increase in Breast Cancer Detection
February 5th 2025Use of the mammography AI software had a nearly equivalent false positive rate as unassisted radiologist interpretation and resulted in a 44 percent reduction in screen reading workload, according to findings from a randomized controlled trial involving over 105,000 women.