A Prognosis Method for Esophageal Squamous Cell Carcinoma Based on CT Image and Three-Dimensional Convolutional Neural Networks

Kaipeng Fan, Jifeng Guo, Bo Yang, Lin Wang, Lizhi Peng, Baosheng Li, Jian Zhu, Ajith Abraham

Abstract: Esophageal squamous cell carcinoma (ESCC) is one species cancer with high morbidity and mortality. Most cancer treatments need radiotherapy and chemotherapy. Before treatments, patients are often hesitant, thus delaying the time of treatment. If the effect of the treatment can be directly known according to the patient’s current condition, the patient can make decisions sensibly and enable to get timely treatment. In this paper, a prognosis method for esophageal squamous cell carcinoma is proposed. Three-dimensional convolutional neural networks (CNNs) is used to predict the effect of radiotherapy from computer tomography (CT) images of patients. The experiments confirm the predicted gross tumor volume are consistent with the actual values.