A Relational-learning Perspective to Multi-label Chest X-ray Classification
This addresses the challenge of incorporating auxiliary information like label dependencies in medical imaging for radiologists, though it is incremental with a modest performance gain.
The paper tackled multi-label chest X-ray classification by reformulating it as a link prediction problem using a knowledge graph, which improved predictive performance to an AUC of 83.5%, a gain of about 1% over state-of-the-art discriminative methods.
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.