Table 3.
Performance of the baseline 2D and the proposed 3D models.
Task and model |
Size (MB) | Data set | Accuracy, mean (SD) | Recall, mean (SD) | Specificity, mean (SD) | Precision, mean (SD) | F1-score, mean (SD) | AUROCa, mean (SD) | P value | ||||||||||
1 | |||||||||||||||||||
|
3D | 14.2 | EENTb | 0.878 (0.017) | 0.853 (0.032) | 0.913 (0.067) | 0.933 (0.045) | 0.89 (0.012) | 0.00959 (0.00011) | .003 | |||||||||
|
2D | 274 | EENT | 0.861 (0.019) | 0.845 (0.028) | 0.883 (0.052) | 0.909 (0.036) | 0.875 (0.016) | 0.00939 (0.00013) | N/Ac | |||||||||
|
3D | 14.2 | WUd | 0.843 (0.015) | 0.756 (0.047) | 0.934 (0.021) | 0.924 (0.018) | 0.83 (0.022) | 0.00933 (0.0001) | .05 | |||||||||
|
2D | 274 | WU | 0.821 (0.023) | 0.744 (0.078) | 0.901 (0.046) | 0.891 (0.039) | 0.808 (0.036) | 0.00918 (0.00012) | N/A | |||||||||
2 | |||||||||||||||||||
|
3D | 14.2 | EENT | 0.783 (0.04) | 0.808 (0.025) | 0.77 (0.054) | 0.652 (0.06) | 0.721 (0.042) | 0.00853 (0.0003) | <.001 | |||||||||
|
2D | 274 | EENT | 0.67 (0.037) | 0.716 (0.144) | 0.646 (0.119) | 0.523 (0.044) | 0.596 (0.036) | 0.00744 (0.00025) | N/A | |||||||||
|
3D | 14.2 | WU | 0.812 (0.033) | 0.614 (0.085) | 0.878 (0.031) | 0.626 (0.078) | 0.618 (0.069) | 0.00826 (0.00055) | <.001 | |||||||||
|
2D | 274 | WU | 0.676 (0.103) | 0.479 (0.224) | 0.741 (0.185) | 0.41 (0.086) | 0.411 (0.096) | 0.00714 (0.00049) | N/A |
aAUROC: area under the receiver operating characteristic curve.
bEENT: Eye, Ear, Nose, and Throat Hospital of Fudan University.
cN/A: not applicable.
dWU: Wuhan Union Hospital.