{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,12]],"date-time":"2026-09-12T19:44:01Z","timestamp":1789242241186,"version":"build-2803163510"},"reference-count":40,"publisher":"Public Library of Science (PLoS)","issue":"11","license":[{"start":{"date-parts":[[2020,11,9]],"date-time":"2020-11-09T00:00:00Z","timestamp":1604880000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61633006"],"award-info":[{"award-number":["61633006"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["81872247"],"award-info":[{"award-number":["81872247"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:sec id=\"sec001\">\n<jats:title>Background<\/jats:title>\n<jats:p>Pneumothorax can lead to a life-threatening emergency. The experienced radiologists can offer precise diagnosis according to the chest radiographs. The localization of the pneumothorax lesions will help to quickly diagnose, which will be benefit for the patients in the underdevelopment areas lack of the experienced radiologists. In recent years, with the development of large neural network architectures and medical imaging datasets, deep learning methods have become a methodology of choice for analyzing medical images. The objective of this study was to the construct convolutional neural networks to localize the pneumothorax lesions in chest radiographs.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec002\">\n<jats:title>Methods and findings<\/jats:title>\n<jats:p>We developed a convolutional neural network, called CheXLocNet, for the segmentation of pneumothorax lesions. The SIIM-ACR Pneumothorax Segmentation dataset was used to train and validate CheXLocNets. The training dataset contained 2079 radiographs with the annotated lesion areas. We trained six CheXLocNets with various hyperparameters. Another 300 annotated radiographs were used to select parameters of these CheXLocNets as the validation set. We determined the optimal parameters by the AP<jats:sub>50<\/jats:sub> (average precision at the intersection over union (IoU) equal to 0.50), a segmentation evaluation metric used by several well-known competitions. Then CheXLocNets were evaluated by a test set (1082 normal radiographs and 290 disease radiographs), based on the classification metrics: area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV); segmentation metrics: IoU and Dice score. For the classification, CheXLocNet with best sensitivity produced an AUC of 0.87, sensitivity of 0.78 (95% CI 0.73-0.83), and specificity of 0.78 (95% CI 0.76-0.81). CheXLocNet with best specificity produced an AUC of 0.79, sensitivity of 0.46 (95% CI 0.40-0.52), and specificity of 0.92 (95% CI 0.90-0.94). For the segmentation, CheXLocNet with best sensitivity produced an IoU of 0.69 and Dice score of 0.72. CheXLocNet with best specificity produced an IoU of 0.77 and Dice score of 0.79. We combined them to form an ensemble CheXLocNet. The ensemble CheXLocNet produced an IoU of 0.81 and Dice score of 0.82. Our CheXLocNet succeeded in automatically detecting pneumothorax lesions, without any human guidance.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec003\">\n<jats:title>Conclusions<\/jats:title>\n<jats:p>In this study, we proposed a deep learning network, called, CheXLocNet, for the automatic segmentation of chest radiographs to detect pneumothorax. Our CheXLocNets generated accurate classification results and high-quality segmentation masks for the pneumothorax at the same time. This technology has the potential to improve healthcare delivery and increase access to chest radiograph expertise for the detection of diseases. Furthermore, the segmentation results can offer comprehensive geometric information of lesions, which can benefit monitoring the sequential development of lesions with high accuracy. Thus, CheXLocNets can be further extended to be a reliable clinical decision support tool. Although we used transfer learning in training CheXLocNet, the parameters of CheXLocNet was still large for the radiograph dataset. Further work is necessary to prune CheXLocNet suitable for the radiograph dataset.<\/jats:p>\n<\/jats:sec>","DOI":"10.1371\/journal.pone.0242013","type":"journal-article","created":{"date-parts":[[2020,11,9]],"date-time":"2020-11-09T18:29:30Z","timestamp":1604946570000},"page":"e0242013","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":38,"title":["CheXLocNet: Automatic localization of pneumothorax in chest radiographs using deep convolutional neural networks"],"prefix":"10.1371","volume":"15","author":[{"given":"Hongyu","family":"Wang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hong","family":"Gu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0545-3044","authenticated-orcid":true,"given":"Pan","family":"Qin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jia","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2020,11,9]]},"reference":[{"issue":"2","key":"pone.0242013.ref001","doi-asserted-by":"crossref","first-page":"545","DOI":"10.1378\/chest.10-1302","article-title":"Interpretation of plain chest roentgenogram","volume":"141","author":"S Raoof","year":"2012","journal-title":"Chest"},{"key":"pone.0242013.ref002","doi-asserted-by":"crossref","unstructured":"Cire\u015fan DC, Giusti A, Gambardella LM, Schmidhuber J. Mitosis detection in breast cancer histology images with deep neural networks. In: International conference on medical image computing and computer-assisted intervention. Springer; 2013. p. 411\u2013418.","DOI":"10.1007\/978-3-642-40763-5_51"},{"key":"pone.0242013.ref003","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. In: International Conference on Medical image computing and computer-assisted intervention. Springer; 2015. p. 234\u2013241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"pone.0242013.ref004","doi-asserted-by":"crossref","first-page":"179","DOI":"10.1007\/978-3-319-46976-8_19","volume-title":"Deep Learning and Data Labeling for Medical Applications","author":"M Drozdzal","year":"2016"},{"issue":"2","key":"pone.0242013.ref005","doi-asserted-by":"crossref","first-page":"e0212094","DOI":"10.1371\/journal.pone.0212094","article-title":"Automatic diagnostics of tuberculosis using convolutional neural networks analysis of MODS digital images","volume":"14","author":"S Lopez-Garnier","year":"2019","journal-title":"PloS one"},{"issue":"6","key":"pone.0242013.ref006","doi-asserted-by":"crossref","first-page":"1856","DOI":"10.1109\/TMI.2019.2959609","article-title":"Unet++: Redesigning skip connections to exploit multiscale features in image segmentation","volume":"39","author":"Z Zhou","year":"2019","journal-title":"IEEE transactions on medical imaging"},{"key":"pone.0242013.ref007","article-title":"Deep Multi-Scale 3D Convolutional Neural Network (CNN) for MRI Gliomas Brain Tumor Classification","author":"H Mzoughi","year":"2020","journal-title":"Journal of Digital Imaging"},{"key":"pone.0242013.ref008","doi-asserted-by":"crossref","first-page":"101781","DOI":"10.1016\/j.artmed.2019.101781","article-title":"Multi-planar 3D breast segmentation in MRI via deep convolutional neural networks","volume":"103","author":"G Piantadosi","year":"2020","journal-title":"Artificial Intelligence in Medicine"},{"issue":"2","key":"pone.0242013.ref009","doi-asserted-by":"crossref","first-page":"574","DOI":"10.1148\/radiol.2017162326","article-title":"Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks","volume":"284","author":"P Lakhani","year":"2017","journal-title":"Radiology"},{"issue":"11","key":"pone.0242013.ref010","doi-asserted-by":"crossref","first-page":"e1002686","DOI":"10.1371\/journal.pmed.1002686","article-title":"Deep learning for chest radiograph diagnosis: A retrospective comparison of the CheXNeXt algorithm to practicing radiologists","volume":"15","author":"P Rajpurkar","year":"2018","journal-title":"PLoS medicine"},{"issue":"11","key":"pone.0242013.ref011","article-title":"Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: a cross-sectional study","volume":"15","author":"JR Zech","year":"2018","journal-title":"PLoS medicine"},{"issue":"11","key":"pone.0242013.ref012","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pmed.1002697","article-title":"Automated detection of moderate and large pneumothorax on frontal chest X-rays using deep convolutional neural networks: A retrospective study","volume":"15","author":"AG Taylor","year":"2018","journal-title":"PLoS medicine"},{"issue":"5","key":"pone.0242013.ref013","doi-asserted-by":"crossref","first-page":"1197","DOI":"10.1109\/TMI.2018.2881415","article-title":"Synthesizing chest x-ray pathology for training deep convolutional neural networks","volume":"38","author":"H Salehinejad","year":"2018","journal-title":"IEEE transactions on medical imaging"},{"issue":"7","key":"pone.0242013.ref014","doi-asserted-by":"crossref","first-page":"e0236621","DOI":"10.1371\/journal.pone.0236621","article-title":"Deep transfer learning artificial intelligence accurately stages COVID-19 lung disease severity on portable chest radiographs","volume":"15","author":"J Zhu","year":"2020","journal-title":"Plos one"},{"key":"pone.0242013.ref015","article-title":"Deep learning covid-19 features on cxr using limited training data sets","author":"Y Oh","year":"2020","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"pone.0242013.ref016","first-page":"1","article-title":"Covid-19: automatic detection from x-ray images utilizing transfer learning with convolutional neural networks","author":"ID Apostolopoulos","year":"2020","journal-title":"Physical and Engineering Sciences in Medicine"},{"key":"pone.0242013.ref017","doi-asserted-by":"crossref","unstructured":"Wang X, Peng Y, Lu L, Lu Z, Bagheri M, Summers RM. Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2017. p. 2097\u20132106.","DOI":"10.1109\/CVPR.2017.369"},{"key":"pone.0242013.ref018","doi-asserted-by":"crossref","unstructured":"Papandreou G, Chen LC, Murphy KP, Yuille AL. Weakly-and semi-supervised learning of a deep convolutional network for semantic image segmentation. In: Proceedings of the IEEE international conference on computer vision; 2015. p. 1742\u20131750.","DOI":"10.1109\/ICCV.2015.203"},{"key":"pone.0242013.ref019","doi-asserted-by":"crossref","unstructured":"Vuola AO, Akram SU, Kannala J. Mask-RCNN and U-net ensembled for nuclei segmentation. In: 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019). IEEE; 2019. p. 208\u2013212.","DOI":"10.1109\/ISBI.2019.8759574"},{"key":"pone.0242013.ref020","doi-asserted-by":"crossref","unstructured":"Girshick R, Donahue J, Darrell T, Malik J. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2014. p. 580\u2013587.","DOI":"10.1109\/CVPR.2014.81"},{"key":"pone.0242013.ref021","unstructured":"Ren S, He K, Girshick R, Sun J. Faster r-cnn: Towards real-time object detection with region proposal networks. In: Advances in neural information processing systems; 2015. p. 91\u201399."},{"key":"pone.0242013.ref022","doi-asserted-by":"crossref","unstructured":"He K, Gkioxari G, Doll\u00e1r P, Girshick R. Mask r-cnn. In: Proceedings of the IEEE international conference on computer vision; 2017. p. 2961\u20132969.","DOI":"10.1109\/ICCV.2017.322"},{"issue":"7553","key":"pone.0242013.ref023","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"Y LeCun","year":"2015","journal-title":"nature"},{"key":"pone.0242013.ref024","doi-asserted-by":"crossref","unstructured":"Lin TY, Doll\u00e1r P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2017. p. 2117\u20132125.","DOI":"10.1109\/CVPR.2017.106"},{"key":"pone.0242013.ref025","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2016. p. 770\u2013778.","DOI":"10.1109\/CVPR.2016.90"},{"key":"pone.0242013.ref026","doi-asserted-by":"crossref","unstructured":"Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2015. p. 3431\u20133440.","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"pone.0242013.ref027","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1007\/978-1-4612-4380-9_35","volume-title":"Breakthroughs in statistics","author":"PJ Huber","year":"1992"},{"issue":"10","key":"pone.0242013.ref028","doi-asserted-by":"crossref","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"SJ Pan","year":"2009","journal-title":"IEEE Transactions on knowledge and data engineering"},{"key":"pone.0242013.ref029","doi-asserted-by":"crossref","unstructured":"Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, et al. Microsoft coco: Common objects in context. In: European conference on computer vision. Springer; 2014. p. 740\u2013755.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"pone.0242013.ref030","unstructured":"Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprint arXiv:14126980. 2014."},{"issue":"1","key":"pone.0242013.ref031","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1007\/s11263-014-0733-5","article-title":"The pascal visual object classes challenge: A retrospective","volume":"111","author":"M Everingham","year":"2015","journal-title":"International journal of computer vision"},{"issue":"8","key":"pone.0242013.ref032","doi-asserted-by":"crossref","first-page":"857","DOI":"10.1002\/(SICI)1097-0258(19980430)17:8<857::AID-SIM777>3.0.CO;2-E","article-title":"Two-sided confidence intervals for the single proportion: comparison of seven methods","volume":"17","author":"RG Newcombe","year":"1998","journal-title":"Statistics in medicine"},{"key":"pone.0242013.ref033","unstructured":"Anuar A. SIIM\u2013ACR Pneumothorax Segmentation; 2019. https:\/\/github.com\/sneddy\/pneumothorax-segmentation."},{"issue":"8","key":"pone.0242013.ref034","doi-asserted-by":"crossref","first-page":"556","DOI":"10.1016\/j.jacr.2011.01.011","article-title":"White paper report of the 2010 RAD-AID Conference on International Radiology for Developing Countries: identifying sustainable strategies for imaging services in the developing world","volume":"8","author":"RD Welling","year":"2011","journal-title":"Journal of the American College of Radiology"},{"issue":"3","key":"pone.0242013.ref035","doi-asserted-by":"crossref","first-page":"573","DOI":"10.1148\/radiol.2019191225","article-title":"Deep Learning for Chest Radiograph Diagnosis in the Emergency Department","volume":"293","author":"EJ Hwang","year":"2019","journal-title":"Radiology"},{"issue":"1","key":"pone.0242013.ref036","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1200\/JCO.2009.23.2272","article-title":"Long-term results of the international adjuvant lung cancer trial evaluating adjuvant Cisplatin-based chemotherapy in resected lung cancer","volume":"28","author":"R Arriagada","year":"2010","journal-title":"J Clin Oncol"},{"issue":"8","key":"pone.0242013.ref037","doi-asserted-by":"crossref","first-page":"890","DOI":"10.1097\/00004424-199008000-00004","article-title":"Computer-displayed eye position as a visual aid to pulmonary nodule interpretation","volume":"25","author":"HL Kundel","year":"1990","journal-title":"Investigative radiology"},{"issue":"8","key":"pone.0242013.ref038","doi-asserted-by":"crossref","first-page":"1865","DOI":"10.1109\/TMI.2018.2806086","article-title":"Fully Convolutional Architectures for Multiclass Segmentation in Chest Radiographs","volume":"37","author":"AA Novikov","year":"2018","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"pone.0242013.ref039","article-title":"Image-to-Images Translation for Multi-Task Organ Segmentation and Bone Suppression in Chest X-Ray Radiography","author":"M Eslami","year":"2020","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"pone.0242013.ref040","doi-asserted-by":"crossref","unstructured":"Guendel S, Grbic S, Georgescu B, Liu S, Maier A, Comaniciu D. Learning to recognize abnormalities in chest x-rays with location-aware dense networks. In: Iberoamerican Congress on Pattern Recognition. Springer; 2018. p. 757\u2013765.","DOI":"10.1007\/978-3-030-13469-3_88"}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0242013","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2020,11,9]],"date-time":"2020-11-09T18:30:06Z","timestamp":1604946606000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0242013"}},"subtitle":[],"editor":[{"given":"Haoran","family":"Xie","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2020,11,9]]},"references-count":40,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2020,11,9]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0242013","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,11,9]]}}}