Survival-oriented embeddings for improving accessibility to complex data structures

TL;DR

This paper introduces a hazard-regularized variational autoencoder that enhances interpretability in survival analysis, specifically applied to abdominal CT scans for liver tumor prognosis, addressing critical transparency issues in medical AI.

Contribution

It presents a novel deep learning model that improves interpretability in survival analysis, tailored for clinical applications involving complex medical imaging data.

Findings

Effective interpretation of deep models in survival analysis.

Application to liver tumor CT scans demonstrates clinical relevance.

Improved transparency in AI-driven prognosis tools.

Abstract

Deep learning excels in the analysis of unstructured data and recent advancements allow to extend these techniques to survival analysis. In the context of clinical radiology, this enables, e.g., to relate unstructured volumetric images to a risk score or a prognosis of life expectancy and support clinical decision making. Medical applications are, however, associated with high criticality and consequently, neither medical personnel nor patients do usually accept black box models as reason or basis for decisions. Apart from averseness to new technologies, this is due to missing interpretability, transparency and accountability of many machine learning methods. We propose a hazard-regularized variational autoencoder that supports straightforward interpretation of deep neural architectures in the context of survival analysis, a field highly relevant in healthcare. We apply the proposed approach to abdominal CT scans of patients with liver tumors and their corresponding survival times.