.png){kind=icon}
.png){kind=icon}
Return
Pulmonary nodules recognition based on parallel cross-convolution
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-022-12908-x.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Lung cancer is the highest incidence rate and mortality rate in human beings. Pulmonary nodules are the early manifestation of lung cancer. The accurate identification of pulmonary nodules is of great significance for the treatment and response of early lung cancer, and can effectively improve the life expectancy of lung cancer patients. In order to improve the recognition rate of lung cancer, this paper proposes a lung cancer recognition method based on parallel cross convolution neural network, which aims to solve the problems of gray scale classification, image edge blur, local body effect, artifact and noise in low-dose CT images. Firstly, the low-dose CT images are preprocessed. Then, the parallel cross convolution neural network model is used to classify pulmonary nodules. The model uses two data streams with different steps to extract image features, and is optimized by improved excitation function and batch regularization method. The experimental results show that the accuracy of pccnn proposed in this paper for pulmonary nodules or lung cancer is 97.78%. This shows that the combination of lung parenchyma cutting and parallel cross convolution neural network model can extract image features more deeply and improve the accuracy of lung cancer recognition.
Keywords:
Lung cancer recognition
Parallel cross-convolution neural network
Segmentation of lung parenchyma
Low dose CT image
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W
Organization
No organization information available