Abstract: The morbidity and mortality of liver cancer in China are not optimistic. Early diagnosis and treatment have become the urgent means to change this situation. Therefore, a fusion of attention mechanism and dilated convolution 3D ELD_MobileNetV2 hepatic nodules classification model was proposed to classify abdominal dynamic enhanced magnetic resonance images. Firstly, the two-dimensional network structure was extended to three dimensions to avoid the loss of spatial information during feature extraction of magnetic resonance imaging (MRI); Secondly, based on the local cross-channel interaction strategy, an efficient channel attention mechanism combining local features and global features was embedded in the bottleneck structure of MobileNetV2 network to enhance the key feature extraction capability. Then, 3D dilated structure was introduced into depthwise convolution to improve the receptive field of the convolution kernel; Meanwhile, Leaky ReLU6 activation function was used to replace the original activation function to improve the robustness of the model. The model was tested and validated in 120 patients (30 cases each of hepatitis, cirrhotic nodules, dysplastic nodules, and hepatocellular carcinoma), and the experimental results show that the accuracy of the proposed model was improved by 8.34% compared with the original MobileNetV2.Compared with other networks such as AlexNet,VggNet16,ResNet50, ConvNeXt, 3D ELD_MobileNetV2 shows the best performance, with the accuracy of 79.17%, F1-Score of 0.688, 0.750, 0.848, 0.872, and micro-average AUC of 0.954, and macro-average AUC of 0.948. The model proposed in this study can better classify liver nodules in different periods, and is expected to provide help for the early diagnosis of liver cancer.

Key words: 3D MobileNetV2, efficient channel attention, dilated convolution, hepatic nodules classification, DCE-MRI