Table 2 Performance of AI models in assessing tumor-infiltrating NTR

From: Noninvasive evaluation and clinical value prediction of tumor-infiltrating neutrophil-to-T-cell ratio in pancreatic ductal adenocarcinoma

Model	AUC (95% CI)	ACC (95% CI)	Sensitivity	Specificity	F1-score
Cuboid-model
Training	0.88 (0.83–0.93)	0.82 (0.76−0.88)	0.90	0.74	0.83
Internal validation	0.85 (0.77−0.93)	0.82 (0.74−0.90)	0.84	0.80	0.82
External validation	0.81 (0.71−0.91)	0.77 (0.67−0.87)	0.82	0.71	0.80
ROIex-3mm-model
Training	0.89 (0.84−0.94)	0.83 (0.77−0.89)	0.90	0.76	0.84
Internal validation	0.84 (0.76−0.92)	0.84 (0.77−0.91)	0.88	0.80	0.85
External validation	0.83 (0.70−0.96)	0.77 (0.64−0.90)	0.82	0.71	0.80
ROIex-5mm-model
Training	0.84 (0.78−0.90)	0.77 (0.71−0.83)	0.84	0.70	0.79
Internal validation	0.78 (0.68−0.88)	0.76 (0.67−0.85)	0.76	0.76	0.76
External validation	0.76 (0.65−0.87)	0.72 (0.61−0.82)	0.77	0.65	0.76
ROI-model
Training	0.90 (0.85−0.95)	0.86 (0.81−0.91)	0.92	0.80	0.87
Internal validation	0.88 (0.81−0.95)	0.84 (0.76−0.92)	0.84	0.84	0.84
External validation	0.85 (0.76−0.94)	0.79 (0.70−0.89)	0.82	0.76	0.82
Multimodal Cuboid-model
Training	0.91 (0.86−0.95)	0.87 (0.82−0.92)	0.92	0.82	0.88
Internal validation	0.89 (0.82−0.96)	0.86 (0.79−0.93)	0.88	0.84	0.86
External validation	0.86 (0.77−0.95)	0.82 (0.73−0.91)	0.86	0.76	0.84
Multimodal ROIex-model
Training	0.89 (0.84−0.94)	0.87 (0.82−0.92)	0.90	0.84	0.87
Internal validation	0.87 (0.79−0.95)	0.84 (0.76−0.92)	0.88	0.80	0.85
External validation	0.84 (0.75−0.93)	0.79 (0.70−0.89)	0.86	0.71	0.83
Multimodal ROI-model (PORCELAIN)
Training	0.94 (0.89–0.97)	0.89 (0.84−0.94)	0.94	0.84	0.90
Internal validation	0.92 (0.86−0.98)	0.88 (0.81−0.95)	0.88	0.88	0.88
External validation	0.90 (0.84−0.94)	0.85 (0.76−0.93)	0.86	0.82	0.86

ACC accuracy, AUC area under curve, ROI region of interest.

Back to article page

Quick links

Search

Quick links