Esophageal squamous cell carcinoma (ESCC) is one of the highest incidence and mortality cancers in the world. An effective survival prediction model can improve the quality of patients’ survival. Therefore, a parameter-optimized deep belief network based on the improved Archimedes optimization algorithm is proposed in this paper for the survival prediction of patients with ESCC. Firstly, a combination of features significantly associated with the survival of patients is found by the minimum redundancy and maximum relevancy (MRMR) algorithm. Secondly, a DBN network is introduced to make predictions for survival of patients. Aiming at the problem that the deep belief network model is affected by parameters in the construction process, this paper uses the Archimedes optimization algorithm to optimize the learning rate α and batch size β of DBN. In order to overcome the problem that AOA is prone to fall into local optimum and low search accuracy, an improved Archimedes optimization algorithm (IAOA) is proposed. On this basis, a survival prediction model for patients with ESCC is constructed. Finally, accuracy comparison tests are carried out on IAOA-DBN, AOA-DBN, SSA-DBN, PSO-DBN, BES-DBN, IAOA-SVM, and IAOA-BPNN models. The results show that the IAOA-DBN model can effectively predict the five-year survival rate of patients and provide a reference for the clinical judgment of patients with ESCC.

Esophageal squamous cell carcinoma (ESCC) is one of the highest incidence and mortality cancers in the world. An effective survival prediction model can improve the quality of patients’ survival. In this study, ten indicators related to the survival of patients with ESCC are founded using genetic algorithm feature selection. The prognostic index (PI) for ESCC is established using the binary logistic regression. PI is divided into four stages, and each stage can reasonably reflect the survival status of different patients. By plotting the ROC curve, the critical threshold of patients’ age could be found, and patients are divided into the high-age groups and the low-age groups. PI and ten survival-related indicators are used as independent variables, based on the bald eagle search (BES) and least-squares support vector machine (LSSVM), and a survival prediction model for patients with ESCC is established. The results show that five-year survival rates of patients are well predicted by the bald eagle search-least-squares support vector machine (BES-LSSVM). BES-LSSVM has higher prediction accuracy than the existing particle swarm optimization-least-squares support vector machine (PSO-LSSVM), grasshopper optimization algorithm-least-squares support vector machine (GOA-LSSVM), differential evolution-least-squares support vector machine (DE-LSSVM), sparrow search algorithm-least-squares support vector machine (SSA-LSSVM), bald eagle search-back propagation neural network (BES-BPNN), and bald eagle search-extreme learning machine (BES-ELM).