This thesis focuses on the development and evaluation of survival analysis models within the framework of machine learning to predict disease progression. The primary objective is to explore and propose innovative computational approaches that can effectively estimate the time-to-event outcomes in medical conditions, particularly those with progressive nature. By leveraging survival models, this research aims to improve the accuracy and interpretability of predictions related to patient prognosis, treatment response, and risk stratification. The study will compare traditional statistical survival methods with modern machine learning-based approaches, assessing their effectiveness across different datasets and disease domains.
Relevant bibliography: