Deep survival analysis with longitudinal X-rays for COVID-19

TL;DR

This paper introduces a deep learning method for time-to-event analysis that incorporates longitudinal X-ray images and non-imaging data, significantly improving COVID-19 patient outcome predictions.

Contribution

The paper presents a novel deep learning framework that integrates multiple time-dependent imaging studies into survival analysis, outperforming classical models on COVID-19 data.

Findings

Image sequences improve prediction accuracy.

Deep models reduce concordance error from 30-40% to 20%.

Models are robust against scanner artifacts.

Abstract

Time-to-event analysis is an important statistical tool for allocating clinical resources such as ICU beds. However, classical techniques like the Cox model cannot directly incorporate images due to their high dimensionality. We propose a deep learning approach that naturally incorporates multiple, time-dependent imaging studies as well as non-imaging data into time-to-event analysis. Our techniques are benchmarked on a clinical dataset of 1,894 COVID-19 patients, and show that image sequences significantly improve predictions. For example, classical time-to-event methods produce a concordance error of around 30-40% for predicting hospital admission, while our error is 25% without images and 20% with multiple X-rays included. Ablation studies suggest that our models are not learning spurious features such as scanner artifacts. While our focus and evaluation is on COVID-19, the methods we develop are broadly applicable.