Automated segmentation and diagnosis of pneumothorax on chest X-rays with fully convolutional multi-scale ScSE-DenseNet: a retrospective study

• Published in: BMC medical informatics and decision making Vol. 20; no. S14; pp. 317 - 12
• Main Authors: Wang, Qingfeng, Liu, Qiyu, Luo, Guoting, Liu, Zhiqin, Huang, Jun, Zhou, Yuwei, Zhou, Ying, Xu, Weiyun, Cheng, Jie-Zhi
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 15.12.2020; BioMed Central; BMC
• Subjects: Algorithms; Annotations; Chest; Chest X-ray; Deep learning; Diagnosis; Diagnostic systems; Emergency medical services; Entropy (Information theory); fully convolutional DenseNet; Humans; Image Processing, Computer-Assisted; Image segmentation; Machine learning; Medical diagnosis; Medical imaging equipment; Modules; Pixels; Pneumothorax; Pneumothorax - diagnostic imaging; Pneumothorax segmentation and diagnosis; Radiographs; Radiography; Retrospective Studies; Spatial and channel squeezes and excitation; Spatial weighted cross-entropy loss; X-Rays; China; Pneumothorax segmentation and diagnosis; Spatial weighted cross-entropy loss; Spatial and channel squeezes and excitation; Chest X-ray; fully convolutional DenseNet
• DOI: 10.1186/s12911-020-01325-5
• ISSN: 1472-6947

Pneumothorax (PTX) may cause a life-threatening medical emergency with cardio-respiratory collapse that requires immediate intervention and rapid treatment. The screening and diagnosis of pneumothorax usually rely on chest radiographs. However, the pneumothoraces in chest X-rays may be very subtle with highly variable in shape and overlapped with the ribs or clavicles, which are often difficult to identify. Our objective was to create a large chest X-ray dataset for pneumothorax with pixel-level annotation and to train an automatic segmentation and diagnosis framework to assist radiologists to identify pneumothorax accurately and timely. In this study, an end-to-end deep learning framework is proposed for the segmentation and diagnosis of pneumothorax on chest X-rays, which incorporates a fully convolutional DenseNet (FC-DenseNet) with multi-scale module and spatial and channel squeezes and excitation (scSE) modules. To further improve the precision of boundary segmentation, we propose a spatial weighted cross-entropy loss function to penalize the target, background and contour pixels with different weights. This retrospective study are conducted on a total of eligible 11,051 front-view chest X-ray images (5566 cases of PTX and 5485 cases of Non-PTX). The experimental results show that the proposed algorithm outperforms the five state-of-the-art segmentation algorithms in terms of mean pixel-wise accuracy (MPA) with [Formula: see text] and dice similarity coefficient (DSC) with [Formula: see text], and achieves competitive performance on diagnostic accuracy with 93.45% and [Formula: see text]-score with 92.97%. This framework provides substantial improvements for the automatic segmentation and diagnosis of pneumothorax and is expected to become a clinical application tool to help radiologists to identify pneumothorax on chest X-rays.