TieNet: Text-Image Embedding Network for Common Thorax Disease Classification and Reporting in Chest X-rays

TL;DR

TieNet is a novel neural network that combines image and text data to improve classification and reporting of thorax diseases in chest X-rays, mimicking radiologist reasoning with high accuracy.

Contribution

The paper introduces TieNet, a multi-level attention CNN-RNN model that leverages radiological reports as prior knowledge for enhanced disease classification and automated report generation.

Findings

Achieves over 0.9 AUC in disease classification.

Improves classification AUC by 6% over baseline.

Successfully generates preliminary reports from X-ray images.

Abstract

Chest X-rays are one of the most common radiological examinations in daily clinical routines. Reporting thorax diseases using chest X-rays is often an entry-level task for radiologist trainees. Yet, reading a chest X-ray image remains a challenging job for learning-oriented machine intelligence, due to (1) shortage of large-scale machine-learnable medical image datasets, and (2) lack of techniques that can mimic the high-level reasoning of human radiologists that requires years of knowledge accumulation and professional training. In this paper, we show the clinical free-text radiological reports can be utilized as a priori knowledge for tackling these two key problems. We propose a novel Text-Image Embedding network (TieNet) for extracting the distinctive image and text representations. Multi-level attention models are integrated into an end-to-end trainable CNN-RNN architecture for highlighting the meaningful text words and image regions. We first apply TieNet to classify the chest X-rays by using both image features and text embeddings extracted from associated reports. The proposed auto-annotation framework achieves high accuracy (over 0.9 on average in AUCs) in assigning disease labels for our hand-label evaluation dataset. Furthermore, we transform the TieNet into a chest X-ray reporting system. It simulates the reporting process and can output disease classification and a preliminary report together. The classification results are significantly improved (6% increase on average in AUCs) compared to the state-of-the-art baseline on an unseen and hand-labeled dataset (OpenI).