Developing an individualized survival prediction model for colon cancer

Abstract

In this work a 5-year survival prediction model was developed for colon cancer using machine learning methods. The model was based on the SEER dataset which, after preprocessing, consisted of 38,592 records of colon cancer patients. Survival prediction models for colon cancer are not widely and easily available. Results showed that the performance of the model using fewer features is close to that of the model using a larger set of features recommended by an expert physician, which indicates that the first may be a good compromise between usability and performance. The purpose of such a model is to be used in Ambient Assisted Living applications, providing decision support to health care professionals.