Abstract
Thoracic diseases are very serious health problems that plague a large numberof people. Chest X-ray is currently one of the most popular methods to diagnosethoracic diseases, playing an important role in the healthcare workflow.However, reading the chest X-ray images and giving an accurate diagnosis remainchallenging tasks for expert radiologists. With the success of deep learning incomputer vision, a growing number of deep neural network architectures wereapplied to chest X-ray image classification. However, most of the previous deepneural network classifiers were based on deterministic architectures which areusually very noise-sensitive and are likely to aggravate the overfitting issue.In this paper, to make a deep architecture more robust to noise and to reduceoverfitting, we propose using deep generative classifiers to automaticallydiagnose thorax diseases from the chest X-ray images. Unlike the traditionaldeterministic classifier, a deep generative classifier has a distributionmiddle layer in the deep neural network. A sampling layer then draws a randomsample from the distribution layer and input it to the following layer forclassification. The classifier is generative because the class label isgenerated from samples of a related distribution. Through training the modelwith a certain amount of randomness, the deep generative classifiers areexpected to be robust to noise and can reduce overfitting and then achieve goodperformances. We implemented our deep generative classifiers based on a numberof well-known deterministic neural network architectures, and tested our modelson the chest X-ray14 dataset. The results demonstrated the superiority of deepgenerative classifiers compared with the corresponding deep deterministicclassifiers.