Neural Fine-Gray: Monotonic neural networks for competing risks
This work addresses a critical issue in medical survival analysis by reducing bias in risk scores, though it is incremental as it builds on existing neural network methods with specific constraints.
The paper tackled the problem of biased survival estimation in time-to-event modeling when ignoring competing risks, by using constrained monotonic neural networks to model each competing survival distribution, achieving exact likelihood maximization with reduced computational cost and demonstrating effectiveness on one synthetic and three medical datasets.
Time-to-event modelling, known as survival analysis, differs from standard regression as it addresses censoring in patients who do not experience the event of interest. Despite competitive performances in tackling this problem, machine learning methods often ignore other competing risks that preclude the event of interest. This practice biases the survival estimation. Extensions to address this challenge often rely on parametric assumptions or numerical estimations leading to sub-optimal survival approximations. This paper leverages constrained monotonic neural networks to model each competing survival distribution. This modelling choice ensures the exact likelihood maximisation at a reduced computational cost by using automatic differentiation. The effectiveness of the solution is demonstrated on one synthetic and three medical datasets. Finally, we discuss the implications of considering competing risks when developing risk scores for medical practice.