Gradient-based Explanations for Deep Learning Survival Models
This work addresses the 'black box' problem in survival models for personalized medicine, though it is incremental as it extends existing gradient methods to a new domain.
The authors tackled the lack of interpretability in deep learning survival models by proposing gradient-based explanation methods, resulting in GradSHAP(t) outperforming existing methods like SurvSHAP(t) and SurvLIME in speed-accuracy trade-offs.
Deep learning survival models often outperform classical methods in time-to-event predictions, particularly in personalized medicine, but their "black box" nature hinders broader adoption. We propose a framework for gradient-based explanation methods tailored to survival neural networks, extending their use beyond regression and classification. We analyze the implications of their theoretical assumptions for time-dependent explanations in the survival setting and propose effective visualizations incorporating the temporal dimension. Experiments on synthetic data show that gradient-based methods capture the magnitude and direction of local and global feature effects, including time dependencies. We introduce GradSHAP(t), a gradient-based counterpart to SurvSHAP(t), which outperforms SurvSHAP(t) and SurvLIME in a computational speed vs. accuracy trade-off. Finally, we apply these methods to medical data with multi-modal inputs, revealing relevant tabular features and visual patterns, as well as their temporal dynamics.