Survival prediction and staging on MM EHR

Abstract

AI methods for survival analysis and risk staging has been a prevalent topic of research, and many models for both tasks have been introduced into the healthcare industry. Many deep learning methods for calculating hazards, predicting overall survival times and survival probability prediction have been developed, as well as many rule based risk staging schemes. The aim of this research work is to create an architecture that can conduct both tasks at once, and form a connection between survival curve of a patient and their risk group. We propose a novel methodology for the same, and report statistically significant results.