Neural Networks for Time-To-Event Analysis

Neural Networks for Time-To-Event Analysis

Type: Master Thesis

Student: Pierre Springer

Supervisor: Sebastian Pölsterl, Christian Wachinger

Status: Running

 

Abstract:

In clinical studies, often the main interest is to predict the time to an adverse event. However, the exact time of an event will remain unknown for a subset of individuals, simply because some remained event-free before the study ended. This is called right censoring. While many machine learning algorithms have been adopted to perform learn from censored data, approaches for deep neural networks have been not been studied extensively yet. In this thesis, the objective is to implement a wide variety of networks and loss functions for time-to-event analysis and perform a comparative study on synthetic and real-world datasets covering a wide range of diseases.