Learning to diagnose from scratch by exploiting dependencies among labels

TL;DR

This paper introduces a method using LSTMs to exploit label dependencies for multi-label chest X-ray diagnosis, achieving state-of-the-art results without pre-training, addressing data scarcity and clinical accuracy needs.

Contribution

It presents a novel approach leveraging inter-label dependencies with LSTMs for multi-label medical image classification, avoiding pre-trained models and improving diagnostic accuracy.

Findings

Achieved state-of-the-art performance on NIH chest X-ray dataset.

Demonstrated effectiveness of LSTMs in modeling label dependencies.

Proposed alternative evaluation metrics relevant to clinical practice.

Abstract

The field of medical diagnostics contains a wealth of challenges which closely resemble classical machine learning problems; practical constraints, however, complicate the translation of these endpoints naively into classical architectures. Many tasks in radiology, for example, are largely problems of multi-label classification wherein medical images are interpreted to indicate multiple present or suspected pathologies. Clinical settings drive the necessity for high accuracy simultaneously across a multitude of pathological outcomes and greatly limit the utility of tools which consider only a subset. This issue is exacerbated by a general scarcity of training data and maximizes the need to extract clinically relevant features from available samples -- ideally without the use of pre-trained models which may carry forward undesirable biases from tangentially related tasks. We present and evaluate a partial solution to these constraints in using LSTMs to leverage interdependencies among target labels in predicting 14 pathologic patterns from chest x-rays and establish state of the art results on the largest publicly available chest x-ray dataset from the NIH without pre-training. Furthermore, we propose and discuss alternative evaluation metrics and their relevance in clinical practice.