Table 4 The diagnostic performance of radiologists at different stages in TNVis-assisted diagnosis experiments
From: A deep learning based ultrasound diagnostic tool driven by 3D visualization of thyroid nodules
SR-1 | SR-2 | SR-3 | JR-1 | JR-2 | JR-3 | |
|---|---|---|---|---|---|---|
AUC | 0.73 [0.68, 0.77] | 0.76 [0.72, 0.80] | 0.82 [0.78, 0.85] | 0.59 [0.55, 0.64] | 0.68 [0.64, 0.73] | 0.77 [0.73, 0.81] |
Accuracy [%] | 73 [70,78] | 77 [73,81] | 83 [79,86] | 59 [55,64] | 68 [64,72] | 77 [74,81] |
Sensitivity [%] | 81 [76,86] | 84 [80,89] | 89 [86,93] | 60 [54,66] | 66 [60,71] | 79 [74,84] |
Specificity [%] | 64 [58,71] | 68 [61,74] | 74 [68,80] | 58 [51,65] | 71 [65,78] | 75 [69,81] |
PPV [%] | 74 [69,79] | 77 [72,82] | 82 [77,86] | 65 [58,71] | 74 [69,80] | 80 [75,85] |
NPV [%] | 73 [66,79] | 77 [71,83] | 85 [79,90] | 53 [47,60] | 62 [56,68] | 74 [68,80] |
F1 | 0.77 | 0.80 | 0.85 | 0.62 | 0.70 | 0.80 |
K [Kappa] | 0.46 | 0.53 | 0.64 | 0.18 | 0.36 | 0.54 |
