Radiologists usually observe anatomical regions of chest X-ray images as well as the overall image before making a decision . Most existing deeplearning models only look at the entire X-rays for classification, failing to utilize important anatomical information . In this paper, we propose a novel multi-label chest x-ray classification model that accurately classifies the image finding and also localizes the findings to their correct anatomical regions . We also utilize a method to efficiently create an adjacency matrix for theanatomical regions using the correlation of the label across the differentregions. Detailed experiments and analysis of our results show the effectiveness of our method when compared to the current state-of-the-art multi-labour chest Xray image classification methods .

Author(s) : Nkechinyere N. Agu, Joy T. Wu, Hanqing Chao, Ismini Lourentzou, Arjun Sharma, Mehdi Moradi, Pingkun Yan, James Hendler

Links : PDF - Abstract

Code :

Keywords : classification - chest - label - anatomical - regions -

Leave a Reply

Your email address will not be published. Required fields are marked *