ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases

TL;DR

Addresses the need for large-scale, realistically labeled chest X-ray data to power data-hungry deep learning CAD systems. Introduces ChestX-ray8, a 108,948-image database with eight NLP-derived disease labels and a subset of bounding boxes for localization, and a unified weakly-supervised DCNN framework to detect and localize pathologies. Demonstrates strong multi-label classification performance, especially with ResNet-50 using a weighted loss and LSE pooling, and provides localization results across IoU thresholds to illustrate grounding feasibility and challenges. By detailing data construction, labeling quality, and benchmarking, the work aims to catalyze public data sharing and further research, with plans to expand disease coverage and integrate clinical data.

Abstract

The chest X-ray is one of the most commonly accessible radiological examinations for screening and diagnosis of many lung diseases. A tremendous number of X-ray imaging studies accompanied by radiological reports are accumulated and stored in many modern hospitals' Picture Archiving and Communication Systems (PACS). On the other side, it is still an open question how this type of hospital-size knowledge database containing invaluable imaging informatics (i.e., loosely labeled) can be used to facilitate the data-hungry deep learning paradigms in building truly large-scale high precision computer-aided diagnosis (CAD) systems. In this paper, we present a new chest X-ray database, namely "ChestX-ray8", which comprises 108,948 frontal-view X-ray images of 32,717 unique patients with the text-mined eight disease image labels (where each image can have multi-labels), from the associated radiological reports using natural language processing. Importantly, we demonstrate that these commonly occurring thoracic diseases can be detected and even spatially-located via a unified weakly-supervised multi-label image classification and disease localization framework, which is validated using our proposed dataset. Although the initial quantitative results are promising as reported, deep convolutional neural network based "reading chest X-rays" (i.e., recognizing and locating the common disease patterns trained with only image-level labels) remains a strenuous task for fully-automated high precision CAD systems. Data download link: https://nihcc.app.box.com/v/ChestXray-NIHCC

Figures (8)

• Figure 1: Eight common thoracic diseases observed in chest X-rays that validate a challenging task of fully-automated diagnosis.
• Figure 2: The circular diagram shows the proportions of images with multi-labels in each of 8 pathology classes and the labels' co-occurrence statistics.
• Figure 3: The dependency graph of text: "clear of focal airspace disease, pneumothorax, or pleural effusion".
• Figure 4: The overall flow-chart of our unified DCNN framework and disease localization process.
• Figure 5: A comparison of multi-label classification performance with different model initializations.