Research Article
Interpretable Optimization Training Strategy-Based DCNN and Its Application on CT Image Recognition
Table 3
Comparison of pulmonary nodules’ recognition based on the LIDC-ID-RI datasets.
| Method | Test results | SEN (%) | SPE (%) | ACC (%) | FPF (%) |
| 1 | Method of literature [3] | 85.00 | 40.10 | 68.97 | 31.03 | 2 | Method of literature [4] | 92.37 | 73.24 | 85.20 | 14.80 | 3 | Method of literature [6] | 91.93 | 93.63 | 92.51 | 7.49 | 4 | Method of literature [5] | 96.24 | 92.11 | 94.40 | 5.60 | 5 | The proposed method | Scheme 1 | 56.80 | 47.20 | 52.00 | 48.00 | 6 | Scheme 2 | 83.20 | 80.80 | 82.00 | 18.00 | 7 | Scheme 3 | 62.40 | 61.60 | 62.00 | 38.00 | 8 | Scheme 4 | 96.40 | 95.60 | 96.00 | 4.00 |
|
|