Skip to main content
. 2022 Jun 17;2022:7133972. doi: 10.1155/2022/7133972

Table 4.

Performance of the radiomic model and integrated model for predicting large-volume LNM in the PTMC and CPTC subgroups.

AUROC (95% CI) AUPR (95% CI) ACC (95% CI) SEN (95% CI) SPE (95% CI) PPV (95% CI) NPV (95% CI)
PTMC subgroup (n = 114)
 Integrated model 0.901 (0.790–1) 0.298 (0.261–0.335) 96 (90–99) 60 (15–95) 97 (92–99) 50 (12–88) 98 (93–100)
 Radiomic model 0.875 (0.794–0.957) 0.145 (0.131–0.159) 87 (79–93) 67 (22–96) 88 (80–93) 24 (7–50) 98 (93–100)
 Clinical model 0.752 (0.494–1) 0.016 (0.010–0.022) 80 (71–87) 60 (15–95) 81 (72–88) 12 (3–32) 98 (92–100)
CPTC subgroup (n = 192)
 Integrated model 0.897 (0.825–0.970) 0.539 (0.496–0.582) 92 (87–95) 68 (46–85) 95 (91–98) 68 (46–85) 95 (91–98)
 Radiomic model 0.877 (0.817–0.936) 0.463 (0.432–0.494) 83 (77–88) 72 (51–88) 84 (78–90) 41 (26–57) 95 (90–98)
 Clinical model 0.704 (0.593–0.814) 0.292 (0.245–0.339) 77 (70–83) 44 (24–65) 82 (75–88) 27 (14–43) 91 (85–95)

Significantly different (Delong test) p < 0.05 from clinical model. AUPR, area under the precision-recall curve; AUROC, area under the receiver operator characteristic curve; ACC, accuracy; CI, confidence interval; CPTC, conventional papillary thyroid cancer; NPV, negative predictive value; PPV, positive predictive value; PTMC, papillary thyroid microcarcinoma; SEN, sensitivity; SPE, specificity.