Table 1.
The comparison of performances of the models while experimented over the dataset of CT images.
Base | Metric | Quasi- ProtoPNet |
Ps-ProtoPNet (Singh & Yow, 2021b) |
Gen-ProtoPNet (Singh & Yow, 2021a) |
NP-ProtoPNet (Singh & Yow, 2021c) |
ProtoPNet (Chen et al., 2018) |
Base only |
---|---|---|---|---|---|---|---|
VGG-16 | Accuracy | 99.05 | 98.83 | 95.85 | 98.23 | 90.84 | 99.03 |
Precision | 0.98 | 0.96 | 0.93 | 0.93 | 0.89 | 0.98 | |
Recall | 0.99 | 0.98 | 0.95 | 0.95 | 0.91 | 0.99 | |
F1-score | 0.98 | 0.97 | 0.94 | 0.94 | 0.90 | 0.98 | |
VGG-19 | Accuracy | 99.15 | 98.53 | 98.17 | 98.23 | 96.54 | 98.71 |
Precision | 0.98 | 0.97 | 0.95 | 0.91 | 0.93 | 0.98 | |
Recall | 0.99 | 0.99 | 0.99 | 0.96 | 0.95 | 0.99 | |
F1-score | 0.98 | 0.98 | 0.97 | 0.93 | 0.94 | 0.98 | |
ResNet-34 | Accuracy | 99.29 ± 0.04 |
98.97 ± 0.05 |
98.40 ± 0.12 |
98.45 ± 0.07 |
97.05 ± 0.06 |
99.24 ± 0.10 |
Precision | 0.99 | 0.97 | 0.96 | 0.96 | 0.95 | 0.99 | |
Recall | 0.99 | 0.99 | 0.99 | 0.99 | 0.96 | 0.99 | |
F1-score | 0.99 | 0.98 | 0.97 | 0.97 | 0.96 | 0.99 | |
ResNet-152 | Accuracy | 99.26 ± 0.05 |
98.85 ± 0.04 |
95.90 ± 0.09 |
98.48 ± 0.06 |
88.20 ± 0.08 |
99.40 ± 0.05 |
Precision | 0.98 | 0.97 | 0.93 | 0.99 | 0.87 | 0.99 | |
Recall | 0.99 | 0.98 | 0.93 | 0.99 | 0.87 | 0.99 | |
F1-score | 0.98 | 0.97 | 0.93 | 0.99 | 0.87 | 0.99 | |
DenseNet-121 | Accuracy |
99.44 ± 0.04 |
99.24 ± 0.05 |
98.97 ± 0.02 |
98.83 ± 0.10 |
98.81 ± 0.07 |
99.32 ± 0.03 |
Precision | 0.99 | 0.98 | 0.98 | 0.99 | 0.98 | 0.99 | |
Recall | 0.99 | 0.99 | 0.99 | 0.98 | 0.98 | 0.99 | |
F1-score | 0.99 | 0.98 | 0.98 | 0.98 | 0.98 | 0.99 | |
DenseNet-161 | Accuracy | 99.37 ± 0.02 |
99.02 ± 0.03 |
98.87 ± 0.02 |
98.88 ± 0.03 |
98.76 ± 0.07 |
99.41 ± 0.07 |
Precision | 0.98 | 0.96 | 0.98 | 0.97 | 0.97 | 0.99 | |
Recall | 0.99 | 0.99 | 0.99 | 0.99 | 0.99 | 0.99 | |
F1-score | 0.99 | 0.97 | 0.98 | 0.97 | 0.98 | 0.99 |