Skip to main content
. 2021 Jul 13;57(7):708. doi: 10.3390/medicina57070708

Table 3.

Diagnostic performance of models based on resident’s (level I) evaluation.

Level I Examiner Sensitivity
% (95% CI)
Specificity
% (95% CI)
Accuracy
% (95% CI)
Positive Predictive Value % (95% CI) Negative Predictive Value % (95% CI)
Subjective evaluation 94.1 (71.3–99.9) 75.8 (57.7–88.9) 82 (68.6–91.4) 66.7 (52–78.7) 96.2 (78.7–99.4)
GI-RADS 100 (80.5–100) 60.6 (42.1–77.1) 74 (59.7–85.4) 56.7 (46.1–66.6) 100
SR (n = 41) 100 (78.2–100) 63.6 (45.1–79.6) 76 (61.8–87) 58.6 (47.4–69) 100
SR + algorithm 100 (80.5–100) 97 (84.2–99.9) 98 (89.4–100) 94.4 (71.1–99.2) 100
ADNEX (cut off 10%) 100 (80.5–100) 75.8 (57.7–88.9) 84 (70.9–92.8) 68 (53.8–79.5) 100