Skip to main content

Table 5 Predictive efficacy of clinical, DLCT, and combined model for MSI status of gastric cancer

From: Dual-layer spectral-detector CT for predicting microsatellite instability status and prognosis in locally advanced gastric cancer

Models

AUC (95% CI)

Sensitivity (%)

Specificity (%)

Accuracy (%)

Training set

 Clinical

0.737 (0.660~0.813)

72.55 (37/51)

70.68 (94/133)

71.20 (131/184)

 DLCT

0.856 (0.792~0.920)

72.55 (37/51)

86.47 (115/133)

82.61 (152/184)

 Combined

0.880 (0.820~0.939)

80.39 (41/51)

81.20 (108/133)

80.98 (149/184)

Validation set

 Clinical

0.721 (0.596~0.846)

78.95 (15/19)

59.02 (36/61)

63.75 (51/80)

 DLCT

0.837 (0.725~0.949)

73.68 (14/19)

88.52 (54/61)

85.00 (68/80)

 Combined

0.879 (0.791~0.968)

78.95 (15/19)

75.40 (46/61)

76.25 (61/80)

  1. AUC Area under the curve, CI Confidence interval, DLCT Dual-layer spectral detector CT, MSI Microsatellite instability