Screening of early gastric cancer using Pre-Activation Squeeze-and-Excitation ResNet

Abstract

To propose a quick and accurate method for screening early gastric cancer based on Pre-Activation Squeeze- and-Exception ResNet (PASE-ResNet) gastroscopy images in limited labeled data sets. We developed an algorithm based on Pre-Activation Squeeze- and-Exception ResNet for early gastric cancer screening. To focus on the taskrelated image region and enhance the feature expression ability of model, we combined the Squeeze-and-Exception (SE) module with the residual module in PreAct-ResNet to adjust the weight of the feature channel. The strategy of local screening + global sliding window was adopted to improve the performance of early cancer screening. After data expansion, 18 400 set subgraphs were obtained, and the gastroscopy images were examined using the PASE-ResNet model by sliding window. The results of experiments showed that the proposed algorithm had good performance for screening early gastric cancer with an accuracy of 98.03%, a sensitivity of 98.96% and a specificity of 96.52%. The PASE-ResNet can achieve a high accuracy for screening early gastric cancer.