Skip to main content

Table 2 The performance indicators of three 3D-CNNs

From: Application of dual-stream 3D convolutional neural network based on 18F-FDG PET/CT in distinguishing benign and invasive adenocarcinoma in ground-glass lung nodules

Model Accuracy Sensitivity Specificity PPV NPV
Training set      
CT 3D-CNN 0.84 ± 0.03 0.90 ± 0.07 0.62 ± 0.16 0.90 ± 0.04 0.63 ± 0.15
PET 3D-CNN 0.92 ± 0.02 0.97 ± 0.03 0.76 ± 0.15 0.94 ± 0.04 0.88 ± 0.09
PET/CT 3D-CNN 0.93 ± 0.01 0.98 ± 0.01 0.76 ± 0.06 0.94 ± 0.02 0.90 ± 0.09
Testing set      
CT 3D-CNN 0.67 0.70 0.50 0.86 0.27
PET 3D-CNN 0.76 0.85 0.33 0.85 0.33
PET/CT 3D-CNN 0.85 0.96 0.33 0.87 0.67
  1. The results in the table are expressed as mean ± SD
  2. 3D-CNN, three-dimensional convolutional neural network; PPV, positive predictive value; NPV, negative predictive value