Kernel-Based Learning of Chest X-ray Images for Predicting ICU Escalation among COVID-19 Patients

TL;DR

This paper introduces GLIMARK, a generalized multiple kernel learning method tailored for predicting ICU escalation in COVID-19 patients using chest X-ray images, effectively handling diverse data types and capturing complex patterns.

Contribution

The paper extends multiple kernel learning to exponential family outcomes and applies it to COVID-19 X-ray data for ICU escalation prediction.

Findings

GLIMARK effectively models complex data patterns.

It accurately predicts ICU escalation from X-ray images.

Clinically meaningful features are extracted.

Abstract

Kernel methods have been extensively utilized in machine learning for classification and prediction tasks due to their ability to capture complex non-linear data patterns. However, single kernel approaches are inherently limited, as they rely on a single type of kernel function (e.g., Gaussian kernel), which may be insufficient to fully represent the heterogeneity or multifaceted nature of real-world data. Multiple kernel learning (MKL) addresses these limitations by constructing composite kernels from simpler ones and integrating information from heterogeneous sources. Despite these advances, traditional MKL methods are primarily designed for continuous outcomes. We extend MKL to accommodate the outcome variable belonging to the exponential family, representing a broader variety of data types, and refer to our proposed method as generalized linear models with integrated multiple additive regression with kernels (GLIMARK). Empirically, we demonstrate that GLIMARK can effectively recover or approximate the true data-generating mechanism. We have applied it to a COVID-19 chest X-ray dataset, predicting binary outcomes of ICU escalation and extracting clinically meaningful features, underscoring the practical utility of this approach in real-world scenarios.