Revolutionizing Radiology Reporting With Artificial Intelligence-Based Voice Recognition: A Pilot Study on Lumbar Spine Magnetic Resonance Imaging

Abstract

Purpose This study evaluated whether the interpretation speed of routine lumbar spine magnetic resonance imaging (MRI) increases when using an artificial intelligence (AI)-based voice recognition system compared with the conventional keyboard typing method.

Materials and Methods We retrospectively reviewed 527 routine lumbar spine MRI images performed between November 2022 and February 2023. Two radiologists interpreted 292 and 235 images using conventional keyboard typing and dictation with an AI-based voice recognition system, respectively. Interpretation time, report character count, and turnaround time for the two methods were compared.

Results Interpretation time was significantly reduced by 21.7% using dictation with the AI-based voice recognition method compared with that using the conventional keyboard typing method (p < 0.05). However, no statistically significant differences were observed in the reported character count or turnaround time (p > 0.05).

Conclusion AI-based voice recognition system for interpreting lumbar spine MRI significantly reduced interpretation time compared with the conventional keyboard typing method, suggesting enhanced efficiency for radiologists.