Title
An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks.
Abstract
•This work investigates a computational method for automatic lung segmentation.•This method uses a database of chest X-Ray from the Montgomery County’s Tuberculosis Control Program.•The proposed method addresses the problem of dense abnormalities in chest X-Ray images to reconstruct the segmentation.•The method uses two deep convolutional neural networks to perform lung segmentation in chest X-Ray images.•The method achieved 97.54% of sensitivity, 96.79% of specificity, 96.79% accuracy, and 94% of dice index on lung segmentation in chest X-Ray images.
Year
DOI
Venue
2019
10.1016/j.cmpb.2019.06.005
Computer Methods and Programs in Biomedicine
Keywords
Field
DocType
Lung segmentation,Lung reconstruction,Chest x-ray,Convolutional neural networks
Computer vision,Computer science,Sørensen–Dice coefficient,Segmentation,Convolutional neural network,Artificial intelligence,Jaccard index,Lung segmentation,Lung field,Deep neural networks
Journal
Volume
ISSN
Citations 
177
0169-2607
8
PageRank 
References 
Authors
1.04
0
6