Synthetic data augmentation using GAN for improved liver lesion classification. Frid-Adar, M., Klang, E., Amitai, M., Goldberger, J., & Greenspan, H. In Proceedings - International Symposium on Biomedical Imaging, volume 2018-April, pages 289–293, April, 2018. IEEE. ISSN: 19458452 _eprint: 1801.02385![Synthetic data augmentation using GAN for improved liver lesion classification [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex In this paper, we present a data augmentation method that generates synthetic medical images using Generative Adversarial Networks (GANs). We propose a training scheme that first uses classical data augmentation to enlarge the training set and then further enlarges the data size and its diversity by applying GAN techniques for synthetic data augmentation. Our method is demonstrated on a limited dataset of computed tomography (CT) images of 182 liver lesions (53 cysts, 64 metastases and 65 hemangiomas). The classification performance using only classic data augmentation yielded 78.6% sensitivity and 88.4% specificity. By adding the synthetic data augmentation the results significantly increased to 85.7% sensitivity and 92.4% specificity.
@inproceedings{frid-adar_synthetic_2018,
title = {Synthetic data augmentation using {GAN} for improved liver lesion classification},
volume = {2018-April},
isbn = {978-1-5386-3636-7},
url = {https://doi.org/10.1109%2Fisbi.2018.8363576},
doi = {10.1109/ISBI.2018.8363576},
abstract = {In this paper, we present a data augmentation method that generates synthetic medical images using Generative Adversarial Networks (GANs). We propose a training scheme that first uses classical data augmentation to enlarge the training set and then further enlarges the data size and its diversity by applying GAN techniques for synthetic data augmentation. Our method is demonstrated on a limited dataset of computed tomography (CT) images of 182 liver lesions (53 cysts, 64 metastases and 65 hemangiomas). The classification performance using only classic data augmentation yielded 78.6\% sensitivity and 88.4\% specificity. By adding the synthetic data augmentation the results significantly increased to 85.7\% sensitivity and 92.4\% specificity.},
booktitle = {Proceedings - {International} {Symposium} on {Biomedical} {Imaging}},
publisher = {IEEE},
author = {Frid-Adar, Maayan and Klang, Eyal and Amitai, Michal and Goldberger, Jacob and Greenspan, Hayit},
month = apr,
year = {2018},
note = {ISSN: 19458452
\_eprint: 1801.02385},
keywords = {\#nosource, Data augmentation, Generative adversarial network, Image synthesis, Lesion classification, Liver lesions},
pages = {289--293},
}
Downloads: 0
{"_id":"c27fGZNJCNjaR47WG","bibbaseid":"fridadar-klang-amitai-goldberger-greenspan-syntheticdataaugmentationusingganforimprovedliverlesionclassification-2018","authorIDs":[],"author_short":["Frid-Adar, M.","Klang, E.","Amitai, M.","Goldberger, J.","Greenspan, H."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Synthetic data augmentation using GAN for improved liver lesion classification","volume":"2018-April","isbn":"978-1-5386-3636-7","url":"https://doi.org/10.1109%2Fisbi.2018.8363576","doi":"10.1109/ISBI.2018.8363576","abstract":"In this paper, we present a data augmentation method that generates synthetic medical images using Generative Adversarial Networks (GANs). We propose a training scheme that first uses classical data augmentation to enlarge the training set and then further enlarges the data size and its diversity by applying GAN techniques for synthetic data augmentation. Our method is demonstrated on a limited dataset of computed tomography (CT) images of 182 liver lesions (53 cysts, 64 metastases and 65 hemangiomas). The classification performance using only classic data augmentation yielded 78.6% sensitivity and 88.4% specificity. By adding the synthetic data augmentation the results significantly increased to 85.7% sensitivity and 92.4% specificity.","booktitle":"Proceedings - International Symposium on Biomedical Imaging","publisher":"IEEE","author":[{"propositions":[],"lastnames":["Frid-Adar"],"firstnames":["Maayan"],"suffixes":[]},{"propositions":[],"lastnames":["Klang"],"firstnames":["Eyal"],"suffixes":[]},{"propositions":[],"lastnames":["Amitai"],"firstnames":["Michal"],"suffixes":[]},{"propositions":[],"lastnames":["Goldberger"],"firstnames":["Jacob"],"suffixes":[]},{"propositions":[],"lastnames":["Greenspan"],"firstnames":["Hayit"],"suffixes":[]}],"month":"April","year":"2018","note":"ISSN: 19458452 _eprint: 1801.02385","keywords":"#nosource, Data augmentation, Generative adversarial network, Image synthesis, Lesion classification, Liver lesions","pages":"289–293","bibtex":"@inproceedings{frid-adar_synthetic_2018,\n\ttitle = {Synthetic data augmentation using {GAN} for improved liver lesion classification},\n\tvolume = {2018-April},\n\tisbn = {978-1-5386-3636-7},\n\turl = {https://doi.org/10.1109%2Fisbi.2018.8363576},\n\tdoi = {10.1109/ISBI.2018.8363576},\n\tabstract = {In this paper, we present a data augmentation method that generates synthetic medical images using Generative Adversarial Networks (GANs). We propose a training scheme that first uses classical data augmentation to enlarge the training set and then further enlarges the data size and its diversity by applying GAN techniques for synthetic data augmentation. Our method is demonstrated on a limited dataset of computed tomography (CT) images of 182 liver lesions (53 cysts, 64 metastases and 65 hemangiomas). The classification performance using only classic data augmentation yielded 78.6\\% sensitivity and 88.4\\% specificity. By adding the synthetic data augmentation the results significantly increased to 85.7\\% sensitivity and 92.4\\% specificity.},\n\tbooktitle = {Proceedings - {International} {Symposium} on {Biomedical} {Imaging}},\n\tpublisher = {IEEE},\n\tauthor = {Frid-Adar, Maayan and Klang, Eyal and Amitai, Michal and Goldberger, Jacob and Greenspan, Hayit},\n\tmonth = apr,\n\tyear = {2018},\n\tnote = {ISSN: 19458452\n\\_eprint: 1801.02385},\n\tkeywords = {\\#nosource, Data augmentation, Generative adversarial network, Image synthesis, Lesion classification, Liver lesions},\n\tpages = {289--293},\n}\n\n","author_short":["Frid-Adar, M.","Klang, E.","Amitai, M.","Goldberger, J.","Greenspan, H."],"key":"frid-adar_synthetic_2018","id":"frid-adar_synthetic_2018","bibbaseid":"fridadar-klang-amitai-goldberger-greenspan-syntheticdataaugmentationusingganforimprovedliverlesionclassification-2018","role":"author","urls":{"Paper":"https://doi.org/10.1109%2Fisbi.2018.8363576"},"keyword":["#nosource","Data augmentation","Generative adversarial network","Image synthesis","Lesion classification","Liver lesions"],"downloads":0},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/3522498/collections/DIPSR9D9/items?key=GIOIfx1olqrQ1vxbE1TrXAIi&format=bibtex&limit=100","creationDate":"2021-03-03T03:31:06.302Z","downloads":0,"keywords":["#nosource","data augmentation","generative adversarial network","image synthesis","lesion classification","liver lesions"],"search_terms":["synthetic","data","augmentation","using","gan","improved","liver","lesion","classification","frid-adar","klang","amitai","goldberger","greenspan"],"title":"Synthetic data augmentation using GAN for improved liver lesion classification","year":2018,"dataSources":["zeSfpfAFdsRF2mNyj"]}