Deep Learning-Based Diagnosis of Femoropopliteal Artery Steno-Occlusion Using Maximum Intensity Projection Images of CT Angiography.

In the temporal validation dataset, while RDNet recorded the highest AUC in the half-set approach, all other models demonstrated an increasing trend in AUC performance with greater numbers of utilized images For multi-region ensemble model, lower threshold consistently yielded higher predictive performance in both the internal test set and the temporal validation dataset ( Table 5 ).

also in 63,971 other papers