Back to the list
Congress: ECR25
Poster Number: C-13620
Type: Poster: EPOS Radiologist (scientific)
Authorblock: P. Bombinski1, P. Paczuski2, K. Paczuski2; 1Warsaw/PL, 2Legionowo/PL
Disclosures:
Przemyslaw Bombinski: Consultant: CMO at upmedic
Paweł Paczuski: CEO: upmedic
Krzysztof Paczuski: Founder: upmedic
Keywords: Artificial Intelligence, Computer applications, eHealth, CAD, PACS, Teleradiology, Computer Applications-General, Cost-effectiveness, Efficacy studies, Patterns of Care, Workforce
Purpose Efficient preparation of radiology reports is a cornerstone of modern diagnostic imaging workflows. Radiologists are under increasing pressure to manage high volumes of cases while ensuring accurate and detailed reporting [1,2] , which directly influences patient outcomes. However, traditional methods of report preparation, such as manual keyboard entry and voice dictation, come with significant limitations [3].Manual typing is time-consuming and prone to errors, especially under high workloads. It requires radiologists to focus on repetitive tasks rather than diagnostic reasoning, leading...
Read more Methods and materials This study utilized the upmedic platform, a tool specifically designed for creating structured radiological reports (Fig 1). During the preparatory phase, triggers - pre-defined phrases that describe abnormal findings and can be raised by voice command or mouse click - were automatically generated using large language models (LLMs). These models were trained on datasets containing previously completed radiology reports, received from each radiologist. A set of individual triggers was automatically generated for each radiologist. Subsequently, radiologists reviewed the triggers, with...
Read more Results The use of triggers significantly accelerated the report preparation process (Tab. 1). Specifically, trigger-based reporting resulted in a mean 322,6 words/min to be entered into the report, which is 16,5x more compared to typing (mean 19,6 words/min) and 1.86x more compared to voice dictation (mean 173,4 words/min). Additionally, trigger-based reporting resulted in a mean 2614,3 characters/min to be entered into the report, which is 17x more compared to typing (mean 153,7 characters/min) and 1.9x more compared to voice dictation (mean...
Read more Conclusion The findings of this study demonstrate that using pre-defined phrases (triggers) significantly accelerated the report preparation process compared to traditional methods such as manual keyboard entry and voice dictation. The significant reduction in spoken words and keystrokes further highlights the ergonomic advantages of this approach. By decreasing the physical and cognitive workload associated with repetitive tasks, triggers may alleviate some of the factors contributing to radiologist burnout. Additionally, these results suggest that adopting trigger-based methods can streamline workflows in busy...
Read more References Troupis CJ, Knight RAH, Lau KK. What is the appropriate measure of radiology workload: Study or image numbers? J Med Imaging Radiat Oncol. 2024 Aug;68(5):530-539. doi: 10.1111/1754-9485.13713. Epub 2024 Jun 5. PMID: 38837555.  Kasalak O, Alnahwi H, Toxopeus R, Pennings JP, Yakar D, Kwee TC, Work overload and diagnostic errors in radiology, European Journal of Radiology, Volume 167, 2023, 111032, ISSN 0720-048X, https://doi.org/10.1016/j.ejrad.2023.111032. (https://www.sciencedirect.com/science/article/pii/S0720048X23003467) Motyer RE, Liddy S, Torreggiani WC, Buckley O. Frequency and analysis of non-clinical errors made in radiology reports...
Read more
GALLERY