{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T14:03:35Z","timestamp":1791813815338,"version":"4.3.4"},"reference-count":32,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2024,7,10]],"date-time":"2024-07-10T00:00:00Z","timestamp":1720569600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,7,10]],"date-time":"2024-07-10T00:00:00Z","timestamp":1720569600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Cytomegalovirus retinitis (CMVR) is a significant cause of vision loss. Regular screening is crucial but challenging in resource-limited settings. A convolutional neural network is a state-of-the-art deep learning technique to generate automatic diagnoses from retinal images. However, there are limited numbers of CMVR images to train the model properly. Transfer learning (TL) is a strategy to train a model with a scarce dataset. This study explores the efficacy of TL with different pre-trained weights for automated CMVR classification using retinal images. We utilised a dataset of 955 retinal images (524 CMVR and 431 normal) from Siriraj Hospital, Mahidol University, collected between 2005 and 2015. Images were processed using Kowa VX-10i or VX-20 fundus cameras and augmented for training. We employed DenseNet121 as a backbone model, comparing the performance of TL with weights pre-trained on ImageNet, APTOS2019, and CheXNet datasets. The models were evaluated based on accuracy, loss, and other performance metrics, with the depth of fine-tuning varied across different pre-trained weights. The study found that TL significantly enhances model performance in CMVR classification. The best results were achieved with weights sequentially transferred from ImageNet to APTOS2019 dataset before application to our CMVR dataset. This approach yielded the highest mean accuracy (0.99) and lowest mean loss (0.04), outperforming other methods. The class activation heatmaps provided insights into the model's decision-making process. The model with APTOS2019 pre-trained weights offered the best explanation and highlighted the pathologic lesions resembling human interpretation. Our findings demonstrate the potential of sequential TL in improving the accuracy and efficiency of CMVR diagnosis, particularly in settings with limited data availability. They highlight the importance of domain-specific pre-training in medical image classification. This approach streamlines the diagnostic process and paves the way for broader applications in automated medical image analysis, offering a scalable solution for early disease detection.<\/jats:p>","DOI":"10.1038\/s41598-024-67121-7","type":"journal-article","created":{"date-parts":[[2024,7,10]],"date-time":"2024-07-10T07:02:41Z","timestamp":1720594961000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Selection of pre-trained weights for transfer learning in automated cytomegalovirus retinitis classification"],"prefix":"10.1038","volume":"14","author":[{"given":"Pitipol","family":"Choopong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Worapan","family":"Kusakunniran","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2024,7,10]]},"reference":[{"key":"67121_CR1","doi-asserted-by":"publisher","first-page":"1351","DOI":"10.1093\/cid\/cit494","volume":"57","author":"N Ford","year":"2013","unstructured":"Ford, N. et al. Burden of HIV-related cytomegalovirus retinitis in resource-limited settings: A systematic review. Clin. Infect. Dis. 57, 1351\u20131361 (2013).","journal-title":"Clin. Infect. Dis."},{"key":"67121_CR2","doi-asserted-by":"publisher","first-page":"14","DOI":"10.3390\/microorganisms8010055","volume":"8","author":"M Munro","year":"2019","unstructured":"Munro, M., Yadavalli, T., Fonteh, C., Arfeen, S. & Lobo-Chan, A. M. Cytomegalovirus retinitis in HIV and non-HIV individuals. Microorganisms. 8, 14 (2019).","journal-title":"Microorganisms."},{"key":"67121_CR3","doi-asserted-by":"publisher","first-page":"957","DOI":"10.1001\/archinte.158.9.957","volume":"158","author":"RJ Whitley","year":"1998","unstructured":"Whitley, R. J. et al. Guidelines for the treatment of cytomegalovirus diseases in patients with AIDS in the era of potent antiretroviral therapy: Recommendations of an international panel, International AIDS Society-USA. Arch. Intern. Med. 158, 957\u2013969 (1998).","journal-title":"Arch. Intern. Med."},{"key":"67121_CR4","doi-asserted-by":"crossref","unstructured":"CMV retinitis in China and SE Asia: The way forward [editorial]. BMC Infect. Dis. 11, 327 (2011).","DOI":"10.1186\/1471-2334-11-327"},{"key":"67121_CR5","doi-asserted-by":"publisher","first-page":"903","DOI":"10.2471\/BLT.14.142372","volume":"92","author":"D Heiden","year":"2014","unstructured":"Heiden, D. et al. Training clinicians treating HIV to diagnose cytomegalovirus retinitis. Bull. World Health Organ. 92, 903\u2013908 (2014).","journal-title":"Bull. World Health Organ."},{"key":"67121_CR6","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1186\/1758-2652-14-41","volume":"14","author":"N Tun","year":"2011","unstructured":"Tun, N. et al. CMV retinitis screening and treatment in a resource-poor setting: Three-year experience from a primary care HIV\/AIDS programme in Myanmar. J. Int. AIDS Soc. 14, 41 (2011).","journal-title":"J. Int. AIDS Soc."},{"key":"67121_CR7","doi-asserted-by":"publisher","first-page":"1053","DOI":"10.1016\/j.ajo.2011.05.030","volume":"152","author":"S Ausayakhun","year":"2011","unstructured":"Ausayakhun, S. et al. Accuracy and reliability of telemedicine for diagnosis of cytomegalovirus retinitis. Am. J. Ophthalmol. 152, 1053-1058.e1 (2011).","journal-title":"Am. J. Ophthalmol."},{"key":"67121_CR8","doi-asserted-by":"publisher","first-page":"627","DOI":"10.1089\/tmj.2012.0233","volume":"19","author":"JM Shah","year":"2013","unstructured":"Shah, J. M. et al. Telemedicine screening for cytomegalovirus retinitis using digital fundus photography. Telemed. J. E Health 19, 627\u2013631 (2013).","journal-title":"Telemed. J. E Health"},{"key":"67121_CR9","doi-asserted-by":"publisher","first-page":"1052","DOI":"10.1001\/jamaophthalmol.2014.1108","volume":"132","author":"M Yen","year":"2014","unstructured":"Yen, M. et al. Telemedicine diagnosis of cytomegalovirus retinitis by nonophthalmologists. JAMA Ophthalmol. 132, 1052\u20131058 (2014).","journal-title":"JAMA Ophthalmol."},{"key":"67121_CR10","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1001\/jamaophthalmol.2014.4766","volume":"133","author":"C Jirawison","year":"2015","unstructured":"Jirawison, C. et al. Telemedicine screening for cytomegalovirus retinitis at the point of care for human immunodeficiency virus infection. JAMA Ophthalmol. 133, 198\u2013205 (2015).","journal-title":"JAMA Ophthalmol."},{"key":"67121_CR11","unstructured":"Kingkosol, P., Pooprasert, P., Choopong, P., Hunchangsith, B., Laksanaphuk, V. & Tantibundhit, C. Automated cytomegalovirus retinitis screening in fundus images. In 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). 20201996 (2002)."},{"key":"67121_CR12","doi-asserted-by":"publisher","first-page":"226","DOI":"10.1007\/s10916-018-1088-1","volume":"42","author":"SM Anwar","year":"2018","unstructured":"Anwar, S. M. et al. Medical image analysis using convolutional neural networks: A review. J. Med. Syst. 42, 226 (2018).","journal-title":"J. Med. Syst."},{"key":"67121_CR13","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1136\/bjophthalmol-2018-313173","volume":"103","author":"DSW Ting","year":"2019","unstructured":"Ting, D. S. W. et al. Artificial intelligence and deep learning in ophthalmology. Br. J. Ophthalmol. 103, 167\u2013175 (2019).","journal-title":"Br. J. Ophthalmol."},{"key":"67121_CR14","doi-asserted-by":"publisher","DOI":"10.1016\/j.preteyeres.2019.04.003","volume":"72","author":"DSW Ting","year":"2019","unstructured":"Ting, D. S. W. et al. Deep learning in ophthalmology: The technical and clinical considerations. Prog. Retin. Eye Res. 72, 100759 (2019).","journal-title":"Prog. Retin. Eye Res."},{"key":"67121_CR15","doi-asserted-by":"crossref","unstructured":"Srisuriyajan, P., Cheewaruangroj, N., Polpinit, P. & Laovirojjanakul, W. Cytomegalovirus retinitis screening using machine learning technology. Retina (2022) (publish ahead of print).","DOI":"10.1097\/IAE.0000000000003506"},{"key":"67121_CR16","doi-asserted-by":"crossref","unstructured":"Du, K.-F., Dong, L., Zhang, K. et al. Development and validation of a deep learning-based system for detection of AIDS-related cytomegalovirus retinitis in ultra-wide-field fundus images. SSRN Electron. J. (2021).","DOI":"10.2139\/ssrn.3854577"},{"key":"67121_CR17","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","volume":"22","author":"SJ Pan","year":"2010","unstructured":"Pan, S. J. & Yang, Q. A survey on transfer learning. IEEE Trans. Knowl. Data Eng. 22, 1345\u20131359 (2010).","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"67121_CR18","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2020.104115","volume":"128","author":"MA Morid","year":"2021","unstructured":"Morid, M. A., Borjali, A. & Del Fiol, G. A scoping review of transfer learning research on medical image analysis using ImageNet. Comput. Biol. Med. 128, 104115 (2021).","journal-title":"Comput. Biol. Med."},{"key":"67121_CR19","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1186\/s12880-022-00793-7","volume":"22","author":"HE Kim","year":"2022","unstructured":"Kim, H. E. et al. Transfer learning for medical image classification: A literature review. BMC Med. Imaging 22, 69 (2022).","journal-title":"BMC Med. Imaging"},{"issue":"12968","key":"67121_CR20","first-page":"3","volume":"2021","author":"MR Hosseinzadeh Taher","year":"2021","unstructured":"Hosseinzadeh Taher, M. R., Haghighi, F., Feng, R., Gotway, M. B. & Liang, J. A systematic benchmarking analysis of transfer learning for medical image analysis. Domain Adapt. Represent. Transf. Afford Healthc. AI Resour. Divers. Glob. Health 2021(12968), 3\u201313 (2021).","journal-title":"Domain Adapt. Represent. Transf. Afford Healthc. AI Resour. Divers. Glob. Health"},{"key":"67121_CR21","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvcir.2021.103145","volume":"78","author":"Y Wen","year":"2021","unstructured":"Wen, Y., Chen, L., Deng, Y. & Zhou, C. Rethinking pre-training on medical imaging. J. Vis. Commun. Image Represent. 78, 103145 (2021).","journal-title":"J. Vis. Commun. Image Represent."},{"key":"67121_CR22","doi-asserted-by":"crossref","unstructured":"Huang, G., Liu, Z., Van Der Maaten, L. & Weinberger, K.Q. Densely connected convolutional networks. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 4700\u20134708 (2017).","DOI":"10.1109\/CVPR.2017.243"},{"key":"67121_CR23","doi-asserted-by":"crossref","unstructured":"Guillaumin, M. & Ferrari, V. Large-scale knowledge transfer for object localization in ImageNet. In 2012 IEEE Conference on Computer Vision and Pattern Recognition (2012).","DOI":"10.1109\/CVPR.2012.6248055"},{"key":"67121_CR24","unstructured":"Applications K. DenseNet. Accessed 16 Jan 2024 (2024)."},{"key":"67121_CR25","unstructured":"Society APT-O. APTOS 2019 Blindness Detection Kaggle. Accessed 9 May 2023 (2023)."},{"key":"67121_CR26","unstructured":"Rohith, R. APTOS 2019: Keras Baseline Kaggle. Accessed 16 Jan 2024 (2024)."},{"key":"67121_CR27","unstructured":"Rajpurkar, P., Irvin, J., Zhu, K. et al. CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning. cs.CV (2017)."},{"key":"67121_CR28","doi-asserted-by":"crossref","unstructured":"Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M. & Summers, R.M. 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. 2097\u20132106 (2017).","DOI":"10.1109\/CVPR.2017.369"},{"key":"67121_CR29","unstructured":"Chou, B. GitHub\u2014Brucechou1983\/CheXNet-Keras: This Project is a Tool to Build CheXNet-Like Models, Written in Keras. Accessed16 Jan 2024 (2024)."},{"key":"67121_CR30","doi-asserted-by":"crossref","unstructured":"Zhou, B., Khosla, A., Lapedriza, A., Oliva, A. & Torralba, A. Learning deep features for discriminative localization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2921\u20132929 (2016).","DOI":"10.1109\/CVPR.2016.319"},{"key":"67121_CR31","doi-asserted-by":"publisher","first-page":"2857","DOI":"10.1109\/TMI.2021.3060634","volume":"40","author":"F Haghighi","year":"2021","unstructured":"Haghighi, F., Taher, M. R. H., Zhou, Z., Gotway, M. B. & Liang, J. Transferable visual words: Exploiting the semantics of anatomical patterns for self-supervised learning. IEEE Trans. Med. Imaging 40, 2857\u20132868 (2021).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"67121_CR32","doi-asserted-by":"crossref","unstructured":"Haghighi, F., Hosseinzadeh Taher, M.R., Zhou, Z., Gotway, M.B. & Liang, J. Learning semantics-enriched representation via self-discovery, self-classification, and self-restoration. In: Medical Image Computing and Computer Assisted Intervention\u2014MICCAI 2020: 23rd International Conference, Lima, Peru, October 4\u20138, 2020, Proceedings, Part I 23. 137\u2013147 (2020).","DOI":"10.1007\/978-3-030-59710-8_14"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-024-67121-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-024-67121-7","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-024-67121-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,7,10]],"date-time":"2024-07-10T07:05:43Z","timestamp":1720595143000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-024-67121-7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7,10]]},"references-count":32,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,12]]}},"alternative-id":["67121"],"URL":"https:\/\/doi.org\/10.1038\/s41598-024-67121-7","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,7,10]]},"assertion":[{"value":"27 January 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 July 2024","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"10 July 2024","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"15899"}}