Towards modelling hazard factors in unstructured data spaces using gradient-based latent interpolation
This work addresses the need for more interpretable models in survival analysis for fields like digital health, predictive maintenance, and churn analysis, representing an incremental improvement over existing methods.
The paper tackles the problem of interpretability in deep learning-based survival analysis by proposing a multi-task VAE with survival objective and a novel method called HazardWalk, which models hazard factors in the original data space and is evaluated on simulated and CT imaging datasets.
The application of deep learning in survival analysis (SA) allows utilizing unstructured and high-dimensional data types uncommon in traditional survival methods. This allows to advance methods in fields such as digital health, predictive maintenance, and churn analysis, but often yields less interpretable and intuitively understandable models due to the black-box character of deep learning-based approaches. We close this gap by proposing 1) a multi-task variational autoencoder (VAE) with survival objective, yielding survival-oriented embeddings, and 2) a novel method HazardWalk that allows to model hazard factors in the original data space. HazardWalk transforms the latent distribution of our autoencoder into areas of maximized/minimized hazard and then uses the decoder to project changes to the original domain. Our procedure is evaluated on a simulated dataset as well as on a dataset of CT imaging data of patients with liver metastases.