TY - GEN
T1 - A Coarse Feature Reuse Deep Neural Network for CXR Lesion Detection
AU - Yang, Xinquan
AU - Li, Xuechen
AU - Shen, Linlin
AU - Cao, Min
AU - Zhou, Changen
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Lung disease screening using Chest x-ray (CXR) radiographs can obviously decrease the incidence of lung cancer. Using computer-aided diagnosis system to assist doctors in lung disease screening can greatly improve the diagnosis efficiency. In this paper, a coarse feature reuse deep neural network for CXR lesion detection is proposed. Firstly, we design a coarse feature reuse (CFR) block that can reuse low-level semantic features and extract high-level semantic information, which is used to replace the max-pooling layer in the shallow part of the network to achieve better feature extraction. A novel backbone network - RRCNet, which combines RepVGG block and Resblock, is proposed. The RepVggblock is used for better feature extraction at shallow layers and the Resblock is used for better feature fusion at deep layers. Extensive experiments on VinDr-CXR dataset demonstrate that our RRCNet-based detection network outperformes other classic detectors on both mAP (17.67%) and inference speed (0.1426s).
AB - Lung disease screening using Chest x-ray (CXR) radiographs can obviously decrease the incidence of lung cancer. Using computer-aided diagnosis system to assist doctors in lung disease screening can greatly improve the diagnosis efficiency. In this paper, a coarse feature reuse deep neural network for CXR lesion detection is proposed. Firstly, we design a coarse feature reuse (CFR) block that can reuse low-level semantic features and extract high-level semantic information, which is used to replace the max-pooling layer in the shallow part of the network to achieve better feature extraction. A novel backbone network - RRCNet, which combines RepVGG block and Resblock, is proposed. The RepVggblock is used for better feature extraction at shallow layers and the Resblock is used for better feature fusion at deep layers. Extensive experiments on VinDr-CXR dataset demonstrate that our RRCNet-based detection network outperformes other classic detectors on both mAP (17.67%) and inference speed (0.1426s).
KW - chest x-ray radiograph
KW - computer-aided detection
KW - medical image
KW - neural network
KW - pulmonary lesion detection
UR - http://www.scopus.com/inward/record.url?scp=85128828152&partnerID=8YFLogxK
U2 - 10.1109/ITME53901.2021.00070
DO - 10.1109/ITME53901.2021.00070
M3 - Conference contribution
AN - SCOPUS:85128828152
T3 - Proceedings - 11th International Conference on Information Technology in Medicine and Education, ITME 2021
SP - 307
EP - 313
BT - Proceedings - 11th International Conference on Information Technology in Medicine and Education, ITME 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th International Conference on Information Technology in Medicine and Education, ITME 2021
Y2 - 19 November 2021 through 21 November 2021
ER -
