A Relational-learning Perspective to Multi-label Chest X-ray Classification

TL;DR

This paper introduces a knowledge graph-based framework for multi-label chest X-ray classification, enhancing predictive performance and enabling easy integration of auxiliary information like label dependencies.

Contribution

It proposes a novel knowledge graph reformulation of multi-label classification that improves accuracy and allows incorporation of domain knowledge in chest X-ray analysis.

Findings

Achieved 83.5% AUC on CheXpert dataset.

Outperformed state-of-the-art discriminative methods.

Demonstrated flexibility in adding auxiliary information.

Abstract

Multi-label classification of chest X-ray images is frequently performed using discriminative approaches, i.e. learning to map an image directly to its binary labels. Such approaches make it challenging to incorporate auxiliary information such as annotation uncertainty or a dependency among the labels. Building towards this, we propose a novel knowledge graph reformulation of multi-label classification, which not only readily increases predictive performance of an encoder but also serves as a general framework for introducing new domain knowledge. Specifically, we construct a multi-modal knowledge graph out of the chest X-ray images and its labels and pose multi-label classification as a link prediction problem. Incorporating auxiliary information can then simply be achieved by adding additional nodes and relations among them. When tested on a publicly-available radiograph dataset (CheXpert), our relational-reformulation using a naive knowledge graph outperforms the state-of-art by achieving an area-under-ROC curve of 83.5%, an improvement of "sim 1" over a purely discriminative approach.