Prognostic staging of esophageal cancer based on prognosis index and cuckoo search algorithm-support vector machine

•Find the most suitable machine learning modeling method: Lasso algorithm - Cuckoo search algorithm-support vector machine(CS-SVM).•Discover the most suitable statistical modeling method: Lasso algorithm - ROC curve - Multiple logistic regression.•Find the strong survival - related blood indicators: Neutrophil count(NEUT) and Prothrombin time(PT).•The established TNM-NPT prognostic staging system is superior to the TNM stages.•TNM-NPT prognostic staging system provides new ideas for clinical diagnosis and treatment of esophageal cancer.