Tao ZHOU, Yuncan LIU, Senbao HOU, Xinyu YE, Huiling LU. REC-ResNet： Feature enhancement model for COVID-19 aided diagnosis[J]. Optics and Precision Engineering, 2023, 31(14): 2093
- Optics and Precision Engineering
- Vol. 31, Issue 14, 2093 (2023)
Fig. 1. Overall structure of REC-ResNet model
Fig. 2. Internal structure diagram of Stage2
Fig. 3. Residual adaptive feature fusion module
Fig. 4. Efficient feature enhanced Transformer
Fig. 5. Cross-level attention enhanced module
Fig. 6. Spatial attention
Fig. 7. Channel attention
Fig. 8. Chest X-Ray dataset sample
Fig. 9. Comparison of various evaluation index values of different CNN classification models
Fig. 10. Confusion matrix of classification results of different CNN models
Fig. 11. ROC curve of different CNN models
Fig. 12. Comparison of evaluation index values of ResNet50 classification model combining different attention mechanisms
Fig. 13. Confusion matrix of classification results of ResNet50 model combining different attention mechanisms
Fig. 14. ROC curve of ResNet50 model combining different attention mechanisms
Fig. 15. Comparison of evaluation index values of ablation experiment
Fig. 16. Confusion matrix of ablation experiment classification results
Fig. 17. ROC curve of all models in ablation experiment
Fig. 18. Three types of chest X-Ray images and corresponding heat maps
|
Table 1. Design of ablation experiments
|
Table 2. Comparison of classification performance of different CNN models
|
Table 3. Comparison of classification performance of ResNet50 model combining different attention mechanisms
|
Table 4. Results of ablation experiment（%）
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20