Skip to main content

Table 1 Performance of reader 3 (technician) and the artificial intelligence in correctly deciding on the necessity of contrast injection in the validation set (n = 100) with the consensus of two experienced radiologists (R1 and R2) as reference standard.

From: Improving workflow in prostate MRI: AI-based decision-making on biparametric or multiparametric MRI

  DCE necessary DCE not necessary Agreement AUC (95% CI) Sensitivity Specificity
Consensus (R1/R2) 36/100 (36%) 64/100 (64%) Ref Ref Ref Ref
Technician (R3) 70/100 (70%) 30/100 (30%) 0.55 0.765 (0.669–0844) 63.9% 89.1%
Artificial Intelligence (AI) 56/100 (56%) 44/100 (44%) 0.54 0.881 (0.801–0.937) 94.4% 68.8%
  1. Agreement: kappa with Consensus as reference standard; AUC: Area-under-the-curve; Sensitivity and Specificity of the artificial intelligence based on ROC analysis with a maximized Youden index and high sensitivity to avoid re-examinations