Not applicable, Retrospective, Image verification, Computer Applications-Detection, diagnosis, MR, Neuroradiology brain, Artificial Intelligence and Machine Learning, Multicentre study
D. Sima1, G. Wilms1, T. Vande Vyvere2, W. Van Hecke3, D. Smeets1; 1Leuven/BE, 2Leuven, Be/BE, 3icometrix, Belgium/BE
With respect to timing, the median time for conventional radiological reporting was 7min17s (interquartile range 4min40s), while the median computer-aided reporting time was 4min37s (interquartile range 3min35s), with the latter significantly faster (paired t-test and Wilcoxon signed-rank test p<0.01). This indicates that the radiologist could complete 13 computer-aided reports per hour versus 8 conventional reports per hour, which is an increase of 60% (figure 5).
With respect to radiological findings, the computer-aided reports indicated 7 stable patients (normal atrophy, no lesion activity), 7 patients with slight disease activity (slightly abnormal atrophy rate and/or enlarging lesions), and 11 active patients (5 with new lesions, 10 with abnormal atrophy rate for their age). Conventional radiological reports indicated 19 stable patients (no lesion activity, no apparent atrophy) and 6 active patients (new lesion formation or lesion enlargement). See figure 6. All stable patients identified by the computer-aided technique were also deemed stable by conventional radiological reading. All active patients identified by conventional reading were also identified as active or slightly active in the software-based reporting scenario. However, the automatic brain MRI measurements indicated that several other patients might be (slightly) active, even if these were part of the stable group according to conventional radiological reading.