Unsupervised Detection of Lung Nodules in Chest Radiography Using Generative Adversarial Networks
This addresses the challenge of detecting lung nodules in chest radiographs for medical imaging, but it is incremental as it modifies an existing GAN-based approach.
The paper tackled the problem of commonly missed lung nodules in chest radiographs by proposing P-AnoGAN, an unsupervised anomaly detection method, which achieved ROC-AUC scores of 91.17% and 87.89% in external validation and testing.
Lung nodules are commonly missed in chest radiographs. We propose and evaluate P-AnoGAN, an unsupervised anomaly detection approach for lung nodules in radiographs. P-AnoGAN modifies the fast anomaly detection generative adversarial network (f-AnoGAN) by utilizing a progressive GAN and a convolutional encoder-decoder-encoder pipeline. Model training uses only unlabelled healthy lung patches extracted from the Indiana University Chest X-Ray Collection. External validation and testing are performed using healthy and unhealthy patches extracted from the ChestX-ray14 and Japanese Society for Radiological Technology datasets, respectively. Our model robustly identifies patches containing lung nodules in external validation and test data with ROC-AUC of 91.17% and 87.89%, respectively. These results show unsupervised methods may be useful in challenging tasks such as lung nodule detection in radiographs.