Résumé
Objective: to show that RadR can reduce the time taken to produce reports compared with the use of conventional dictation in CT scans for suspected stroke, without any loss of information.Method: RadR and conventional dictation was used by 5 radiologists to produce 130 reports about 10 stroke cases with blinding of the diagnosis. Dictation times were retrieved from video recording and analyzed after pairing. At the same time, the volume of information was analyzed using a reference report. Finally, a survey was used to perform qualitative and comparative evaluation of the methods concluded this study.Results: RadR achieved a net time gain of 77 seconds (-14%) compared to conventional dictation (p=0.029; Student's t-test). The reports produced by RadR contained a greater volume of information (p<0.05; McNemar test). The RadR method had a higher quality level than the conventional dictation method, at 8.1/10 versus 5.3/10 (p=0.032; Student's t-test).Conclusion: when producing CT scan reports for suspected stroke, RadR saves time compared to conventional dictation and allows for greater volume of information. RadR could be seen as a tool for standardizing the quality of reports, while at the same time being an educational tool for less experienced radiologists.