Large Scale Automated Reading of Frontal and Lateral Chest X-Rays using Dual Convolutional Neural Networks

TL;DR

This paper introduces a large-scale dataset and a novel dual CNN architecture for automated detection of thorax diseases in chest X-rays, demonstrating improved accuracy over separate models.

Contribution

It presents the largest chest X-ray dataset to date and a new DualNet architecture that processes frontal and lateral views simultaneously for better diagnosis.

Findings

DualNet outperforms separate classifiers on the dataset

Largest chest X-ray dataset available for deep learning

Improved accuracy in disease recognition

Abstract

The MIMIC-CXR dataset is (to date) the largest released chest x-ray dataset consisting of 473,064 chest x-rays and 206,574 radiology reports collected from 63,478 patients. We present the results of training and evaluating a collection of deep convolutional neural networks on this dataset to recognize multiple common thorax diseases. To the best of our knowledge, this is the first work that trains CNNs for this task on such a large collection of chest x-ray images, which is over four times the size of the largest previously released chest x-ray corpus (ChestX-Ray14). We describe and evaluate individual CNN models trained on frontal and lateral CXR view types. In addition, we present a novel DualNet architecture that emulates routine clinical practice by simultaneously processing both frontal and lateral CXR images obtained from a radiological exam. Our DualNet architecture shows improved performance in recognizing findings in CXR images when compared to applying separate baseline frontal and lateral classifiers.