Research of COVID-19 CT Image Segmentation Based on PRAU-Net
Aiming at the problems of small lesion area,large variation in shape structure and noise in CT images of COVID-19,we proposed a parallel residual attention U-Net medical image segmentation method based on encoder-decoder architecture.Firstly,the model extracted features by residual inception attention(RIA)in the encoder.RIA adopted a residual structure to combine deeper parallel convolution block and channel attention mechanisms to capture richer features.Secondly,features of different scales were fused by skip connection to obtain richer global information.Lastly,the global attention module was used in the decoder to enable the network focus on relevant features,which effectively reduced the influence of noise in CT images.To verify the effectiveness of the proposed method,we have conducted experiments on segmentation dataset nr.2,CC-CCII and COVID19_1110.Experimental results show that proposed method is more accurate than the classical methods.Compared with classical segmentation methods such as U-Net,the Dice co-efficient increases by 1.12%～14.84%and the sensitivity increases by 0.7%～24.63%.To further demonstrate the segmentation per-formance,the Segmentation dataset nr.2 is extended by using generative adversarial network,the PRU-NET method and several classical segmentation networks are used to verify the method.It is showed that expanding the small sample dataset can effectively improve the segmentation performance,the Dice coefficient of PRAU-Net method was increased from 0.836 4 to 0.858 3.
万方数据
维普
