Table 4.

Performance of the proposed deep learning framework with different dimensions of features. 3D ResNet was adopted as the feature extractor.

Number of dimensions	Sensitivity	Specificity	Precision	F1-score	AUCa
256	93.09	51.09	89.39	91.12	0.83
2048	94.57	54.35	90.17	92.26	0.86

^aAUC: area under the curve.