{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2022,3,29]],"date-time":"2022-03-29T17:05:27Z","timestamp":1648573527416},"reference-count":40,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2021,11,9]],"date-time":"2021-11-09T00:00:00Z","timestamp":1636416000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Conventionally, convolutional neural networks (CNNs) have been used to identify and detect thorax diseases on chest x-ray images. To identify thorax diseases, CNNs typically learn two types of information: disease-specific features and generic anatomical features. CNNs focus on disease-specific features while ignoring the rest of the anatomical features during their operation. There is no evidence that generic anatomical features improve or worsen the performance of convolutional neural networks for thorax disease classification in the current research. As a consequence, the relevance of general anatomical features in boosting the performance of CNNs for thorax disease classification is investigated in this study. We employ a dual-stream CNN model to learn anatomical features before training the model for thorax disease classification. The dual-stream technique is used to compel the model to learn structural information because initial layers of CNNs often learn features of edges and boundaries. As a result, a dual-stream model with minimal layers learns structural and anatomical features as a priority. To make the technique more comprehensive, we first train the model to identify gender and age and then classify thorax diseases using the information acquired. Only when the model learns the anatomical features can it detect gender and age. We also use Non-negative Matrix Factorization\u00a0(NMF) and Contrast Limited Adaptive Histogram Equalization (CLAHE) to pre-process the training data, which suppresses disease-related information while amplifying general anatomical features, allowing the model to acquire anatomical features considerably faster. Finally, the model that was earlier trained for gender and age detection is retrained for thorax disease classification using original data. The proposed technique increases the performance of convolutional neural networks for thorax disease classification, as per experiments on the Chest X-ray14 dataset. We can also see the significant parts of the image that contribute more for gender, age, and a certain thorax disease by visualizing the features. The proposed study achieves two goals: first, it produces novel gender and age identification results on chest X-ray images that may be used in biometrics, forensics, and anthropology, and second, it highlights the importance of general anatomical features in thorax disease classification. In comparison to state-of-the-art results, the proposed work also produces competitive results.<\/jats:p>","DOI":"10.7717\/peerj-cs.738","type":"journal-article","created":{"date-parts":[[2021,11,9]],"date-time":"2021-11-09T10:53:47Z","timestamp":1636455227000},"page":"e738","source":"Crossref","is-referenced-by-count":0,"title":["Gender and age detection assist convolutional neural networks in classification of thorax diseases"],"prefix":"10.7717","volume":"7","author":[{"given":"Mumtaz","family":"Ali","sequence":"first","affiliation":[{"name":"Computer Science, Huazhong University of Science and Technology, Wuhan, Hubei, China"},{"name":"Computer Systems Engineering, Sukkur IBA University, Sukkur, Sukkur, Sindh, Pakistan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Riaz","family":"Ali","sequence":"additional","affiliation":[{"name":"Computer Sceince, Sukkur IBA University, Sukkur, Sukkur, Sindh, Pakistan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"4443","published-online":{"date-parts":[[2021,11,9]]},"reference":[{"key":"10.7717\/peerj-cs.738\/ref-1","doi-asserted-by":"crossref","DOI":"10.1155\/2018\/4168538","article-title":"Deep convolutional neural networks for chest diseases detection","volume":"2018","author":"Abiyev","year":"2018","journal-title":"Journal of Healthcare Engineering"},{"key":"10.7717\/peerj-cs.738\/ref-2","first-page":"1","article-title":"Chest X-ray abnormality detection based on squeezenet","author":"Akpinar","year":"2020"},{"key":"10.7717\/peerj-cs.738\/ref-3","article-title":"Padchest: a large chest X-ray image dataset with multi-label annotated reports","author":"Aurelia\u00a0Bustos","year":"2019"},{"issue":"1","key":"10.7717\/peerj-cs.738\/ref-4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13640-018-0395-2","article-title":"Machine learning hyperparameter selection for contrast limited adaptive histogram equalization","volume":"2019","author":"Campos","year":"2019","journal-title":"EURASIP Journal on Image and Video Processing"},{"key":"10.7717\/peerj-cs.738\/ref-5","doi-asserted-by":"publisher","first-page":"101554","DOI":"10.1016\/j.bspc.2019.04.031","article-title":"Dualchexnet: dual asymmetric feature learning for thoracic disease classification in chest X-rays","volume":"53","author":"Chen","year":"2019","journal-title":"Biomedical Signal Processing and Control"},{"key":"10.7717\/peerj-cs.738\/ref-6","first-page":"53","article-title":"Chest breadths to predict individuals\u2019 age\u2014a case based view","author":"Domingues","year":"2016"},{"key":"10.7717\/peerj-cs.738\/ref-7","doi-asserted-by":"publisher","first-page":"e303","DOI":"10.7717\/peerj-cs.303","article-title":"A multi-task pipeline with specialized streams for classification and segmentation of infection manifestations in covid-19 scans","volume":"6","author":"El-bana","year":"2020","journal-title":"PeerJ Computer Science"},{"key":"10.7717\/peerj-cs.738\/ref-8","doi-asserted-by":"crossref","DOI":"10.1007\/s40134-019-0333-9","article-title":"Deep learning for chest radiology: a review","volume":"7","author":"Feng","year":"2019","journal-title":"Current Radiology Reports"},{"key":"10.7717\/peerj-cs.738\/ref-9","article-title":"The why and how of nonnegative matrix factorization","author":"Gillis","year":"2014"},{"issue":"2020","key":"10.7717\/peerj-cs.738\/ref-10","doi-asserted-by":"publisher","first-page":"259","DOI":"10.1016\/j.patrec.2018.10.027","article-title":"Multi-label chest X-ray image classification via category-wise residual attention learning","volume":"130","author":"Guan","year":"2020","journal-title":"Pattern Recognition Letters"},{"key":"10.7717\/peerj-cs.738\/ref-11","article-title":"Multilabel chest X-ray image classification via category-wise residual attention learning","author":"Guan","year":"2018","journal-title":"Pattern Recognition Letters"},{"key":"10.7717\/peerj-cs.738\/ref-12","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1016\/j.patrec.2019.11.040","article-title":"Thorax disease classification with attention guided convolutional neural network","volume":"131","author":"Guan","year":"2020","journal-title":"Pattern Recognition Letters"},{"key":"10.7717\/peerj-cs.738\/ref-13","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.7717\/peerj-cs.738\/ref-14","first-page":"4700","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"issue":"1","key":"10.7717\/peerj-cs.738\/ref-15","doi-asserted-by":"publisher","first-page":"190","DOI":"10.3390\/electronics9010190","article-title":"Fusion high-resolution network for diagnosing chestX-ray images","volume":"9","author":"Huang","year":"2020","journal-title":"Electronics"},{"key":"10.7717\/peerj-cs.738\/ref-16","doi-asserted-by":"crossref","first-page":"64279","DOI":"10.1109\/ACCESS.2019.2916849","article-title":"A novel approach for multi-label chest X-ray classification of common thorax diseases","volume":"7","author":"Imane Allaouzi","year":"2019","journal-title":"IEEE ACCESS"},{"key":"10.7717\/peerj-cs.738\/ref-17","article-title":"Squeezenet: alexnet-level accuracy with 50x fewer parameters and<0.5 mb model size","author":"Iandola","year":"2016"},{"key":"10.7717\/peerj-cs.738\/ref-18","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v33i01.3301590","article-title":"Chexpert: a large chest radiograph dataset with uncertainty labels and expert comparison","author":"Irvin","year":"2019"},{"key":"10.7717\/peerj-cs.738\/ref-19","first-page":"4039","article-title":"Precision\u2013recall versus accuracy and the role of large data sets","author":"Juba","year":"2019"},{"issue":"6","key":"10.7717\/peerj-cs.738\/ref-20","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","article-title":"Imagenet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Communications of the ACM"},{"key":"10.7717\/peerj-cs.738\/ref-21","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1016\/j.compmedimag.2019.05.005","article-title":"Sdfn: segmentation-based deep fusion network for thoracic disease classification in chest X-ray images","volume":"75","author":"Liu","year":"2019","journal-title":"Computerized Medical Imaging and Graphics"},{"key":"10.7717\/peerj-cs.738\/ref-22","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1016\/j.neucom.2016.12.025","article-title":"Data augmentation for face recognition","volume":"230","author":"Lv","year":"2017","journal-title":"Neurocomputing"},{"key":"10.7717\/peerj-cs.738\/ref-23","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP.2019.8682952","article-title":"Multi-attention network for thoracic disease classification and localization","author":"Ma","year":"2019"},{"key":"10.7717\/peerj-cs.738\/ref-24","article-title":"Large scale automated reading of frontal and lateral chest X-rays using dual convolutional neural networks","author":"Rubin","year":"2018"},{"issue":"3","key":"10.7717\/peerj-cs.738\/ref-25","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1002\/ima.22399","article-title":"Pre-trained convolutional neural networks as feature extractors for diagnosis of breast cancer using histopathology","volume":"30","author":"Saxena","year":"2020","journal-title":"International Journal of Imaging Systems and Technology"},{"key":"10.7717\/peerj-cs.738\/ref-26","first-page":"618","article-title":"Grad-cam: visual explanations from deep networks via gradient-based localization","author":"Selvaraju","year":"2017","journal-title":"IEEE International Conference on Computer Vision (ICCV), Venice"},{"key":"10.7717\/peerj-cs.738\/ref-27","first-page":"802","article-title":"Convolutional lstm network: a machine learning approach for precipitation nowcasting","author":"Shi","year":"2015"},{"key":"10.7717\/peerj-cs.738\/ref-28","first-page":"1","article-title":"Going deeper with convolutions","author":"Szegedy","year":"2015"},{"key":"10.7717\/peerj-cs.738\/ref-29","first-page":"69","article-title":"Dualanet: dual lesion attention network for thoracic disease classification in chest X-rays","author":"Teixeira","year":"2020"},{"key":"10.7717\/peerj-cs.738\/ref-30","first-page":"12","article-title":"Convolutional neural network","author":"V\u00e9stias","year":"2021"},{"issue":"3","key":"10.7717\/peerj-cs.738\/ref-31","doi-asserted-by":"crossref","first-page":"738","DOI":"10.1109\/TMI.2015.2493530","article-title":"Histopathological image classification using discriminative feature-oriented dictionary learning","volume":"35","author":"Vu","year":"2015","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"10.7717\/peerj-cs.738\/ref-32","first-page":"3462","article-title":"ChestX-ray8: hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases","author":"Wang","year":"2017"},{"key":"10.7717\/peerj-cs.738\/ref-33","doi-asserted-by":"crossref","first-page":"101846","DOI":"10.1016\/j.media.2020.101846","article-title":"Triple attention learning for classification of 14 thoracic diseases using chest radiography","volume":"67","author":"Wang","year":"2020","journal-title":"Medical Image Analysis"},{"key":"10.7717\/peerj-cs.738\/ref-34","article-title":"Detecting thoracic diseases via representation learning with adaptive sampling","author":"Wang","year":"2020","journal-title":"Neurocomputing"},{"key":"10.7717\/peerj-cs.738\/ref-35","article-title":"Margin sample mining loss: a deep learning based method for person re-identification","author":"Xiao","year":"2017"},{"key":"10.7717\/peerj-cs.738\/ref-36","first-page":"105790D","article-title":"Using deep learning for detecting gender in adult chest radiographs","volume":"10579","author":"Xue","year":"2018","journal-title":"SPIE Medical Imaging"},{"key":"10.7717\/peerj-cs.738\/ref-37","first-page":"54","article-title":"The why and how of nonnegative matrix factorization","author":"Xue","year":"2018"},{"key":"10.7717\/peerj-cs.738\/ref-38","article-title":"Weakly supervised deep learning for thoracic disease classification and localization on chest X-rays. [cs.CV]","author":"Yan","year":"2018"},{"key":"10.7717\/peerj-cs.738\/ref-39","article-title":"Learning to diagnose from scratch by exploiting dependencies among labels. [cs.CV]","author":"Yao","year":"2018"},{"key":"10.7717\/peerj-cs.738\/ref-40","doi-asserted-by":"crossref","first-page":"474","DOI":"10.1016\/B978-0-12-336156-1.50061-6","volume-title":"Graphics gems iv","author":"Zuiderveld","year":"1994"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-738.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-738.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-738.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-738.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,11,9]],"date-time":"2021-11-09T10:54:02Z","timestamp":1636455242000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-738"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,9]]},"references-count":40,"alternative-id":["10.7717\/peerj-cs.738"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.738","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,11,9]]},"article-number":"e738"}}