Abstract
Federated learning (FL) plays a crucial role in COVID-19 detection by enabling collaborative analysis of medical data while preserving privacy. In this context, medical data is typically dispersed across various medical institutions. FL allows these institutions to collectively train a deep learning model without the need to centralize or transfer the data. But Federated Averaging (FedAvg), the default setting of FL, does not consider personalizing the deep learning model to each institution and cannot ensure model performance when data heterogeneity heavily exists among institutions. In this paper, we investigate Personalized Federated Averaging (PerFedAvg), a personalized variant of the FL framework that aims to train a model that can be tailored to each institution after a few iterations of local gradient descent on the data of that institution while safeguarding the privacy of that institution’s patients. We used data from a publicly available Kaggle dataset that contained chest X-ray images of COVID-19, pneumonia, and normal patients, and created two federated settings that would reflect the heterogeneity of data in terms of distribution, and quantity across medical institutions in the real world. We experimentally demonstrate that PerFedAvg outperforms the standard FedAvg technique in such settings with an increase in accuracy by 3.11%-9.30%. With the help of this research, nations and institutions might quickly construct effective artificial intelligence during pandemics, easing the load of centralized aggregation of copious volumes of sensitive data.
Similar content being viewed by others
Data availability
The datasets generated during and/or analyzed during the current study are available in the Kaggle repository https://www.kaggle.com/datasets/prashant268/chest-xray-covid19-pneumonia.
References
Yuki K, Fujiogi M, Koutsogiannaki S (2020) COVID-19 pathophysiology: A review. Clin Immunol 215:108427
Liu B et al. Experiments of federated learning for COVID-19 Chest X-Ray Images. arXiv, 5 July 2020. arXiv.org, http://arxiv.org/abs/2007.05592
Watson OJ, Barnsley G, Toor J, Hogan AB, Winskill P, Ghani AC (2022) Global impact of the first year of COVID-19 vaccination: a mathematical modelling study. Lancet Infect Dis 22(9):1293–1302. https://doi.org/10.1016/S1473-3099(22)00320-6
Roccetti M (2023) Excess mortality and COVID-19 deaths in Italy: A peak comparison study. Math Biosci Eng : MBE 20(4):7042–7055. https://doi.org/10.3934/mbe.2023304
World Health Organization. (n.d.). The true death toll of COVID-19: Estimating global excess mortality. World Health Organization. https://www.who.int/data/stories/the-true-death-toll-of-covid-19-estimating-global-excess-mortality
Mei X, Lee HC, Diao Ky et al (2020) Artificial intelligence–enabled rapid diagnosis of patients with COVID-19. Nat Med 26:1224–1228. https://doi.org/10.1038/s41591-020-0931-3
Charney AW, Simons NW, Mouskas K, Lepow L, Cheng E, Le Berichel J, Chang C, Marvin R, Del Valle DM, Calorossi S, Lansky A, Walker L, Patel M, Xie H, Yi N, Yu A, Kang G, Mendoza A, Liharska LE, Moya E, Merad M (2020) Sampling the host response to SARS-CoV-2 in hospitals under siege. Nat Med 26(8):1157–1158. https://doi.org/10.1038/s41591-020-1004-3
Sigel K, Swartz T, Golden E, Paranjpe I, Somani S, Richter F, De Freitas JK, Miotto R, Zhao S, Polak P, Mutetwa T, Factor S, Mehandru S, Mullen M, Cossarini F, Bottinger E, Fayad Z, Merad M, Gnjatic S, Aberg J, Glicksberg BS (2020) Coronavirus 2019 and people living with human immunodeficiency virus: Outcomes for hospitalized patients in New York City. Clin Infect Dis : Off Pub Infect Dis Soc Am 71(11):2933–2938. https://doi.org/10.1093/cid/ciaa880
Xu Z, Shi L, Wang Y, Zhang J, Huang L, Zhang C, Liu S, Zhao P, Liu H, Zhu L, Tai Y, Bai C, Gao T, Song J, Xia P, Dong J, Zhao J, Wang FS (2020) Pathological findings of COVID-19 associated with acute respiratory distress syndrome. Lancet Respir Med 8(4):420–422. https://doi.org/10.1016/S2213-2600(20)30076-X
Chung M, Bernheim A, Mei X, Zhang N, Huang M, Zeng X, Cui J, Xu W, Yang Y, Fayad ZA, Jacobi A, Li K, Li S, Shan H (2020) CT Imaging features of 2019 novel coronavirus (2019-nCoV). Radiology 295(1):202–207. https://doi.org/10.1148/radiol.2020200230
Wu Z, McGoogan JM (2020) Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention. JAMA 323(13):1239–1242. https://doi.org/10.1001/jama.2020.2648
Clerkin KJ, Fried JA, Raikhelkar J, Sayer G, Griffin JM, Masoumi A, Jain SS, Burkhoff D, Kumaraiah D, Rabbani L, Schwartz A, Uriel N (2020) COVID-19 and Cardiovascular Disease. Circulation 141(20):1648–1655. https://doi.org/10.1161/CIRCULATIONAHA.120.046941
Feki I, Ammar S, Kessentini Y, Muhammad K (2021) Federated learning for COVID-19 screening from Chest X-ray images. Appl Soft Comput 106:107330. https://doi.org/10.1016/j.asoc.2021.107330
Zhang W et al. (2021) Dynamic fusion-based federated learning for COVID-19 detection. IEEE Intern Things J
Zhang L, Shen B, Barnawi A et al (2021) FedDPGAN: Federated differentially private generative adversarial networks framework for the detection of COVID-19 pneumonia. Inf Syst Front 23:1403–1415. https://doi.org/10.1007/s10796-021-10144-6
McMahan HB, Moore E, Ramage D, Hampson S, Arcas BA (2016) Communication-efficient learning of deep networks from decentralized data. Int Conf Artif Intell Stat
Nguyen DC, Pham Q-V, Pathirana PN, Ding M, Seneviratne A, Lin Z, Dobre O, Hwang W-J (2022) Federated learning for smart healthcare: A Survey. ACM Comput Surv, 55 (3)1-37. https://doi.org/10.1145/3501296
Fallah A, Mokhtari A, Ozdaglar A (2020) Personalized federated learning with theoretical guarantees: a model-agnostic meta-learning approach. In: Proceedings of the 34th International Conference on Neural Information Processing Systems (NIPS'20). Curran Associates Inc., Red Hook, NY, USA, Article 300, 3557–3568
Chen F, Luo M, Dong Z, Li Z, He X (2018) Federated meta-learning with fast convergence and efficient communication. arXiv: Learning
McMahan HB, Ramage D, Talwar K, Zhang L (2017) Learning differentially private recurrent language models. IntConf Learn Rep
Li T, Sahu A, Zaheer M, Sanjabi M, Talwalkar A, Smith V (2020) Federated optimization in heterogeneous networks. In: Proceedings of Machine Learning and Systems pp. 429–450
Li T, Sanjabi M, Beirami A, Virginia Smith S (2020) Fair resource allocation in federated learning. in 8th international conference on learning representations, ICLR 2020, Addis Ababa, Ethiopia, April 26–30, 2020. OpenReview.net
Dinh CH, Tran NH, Nguyen TD (2020) Personalized federated learning with moreau envelopes. In: Proceedings of the 34th International Conference on Neural Information Processing Systems (NIPS'20). Curran Associates Inc., Red Hook, NY, USA, Article 1796, 21394–21405
Patel P (2020) Chest X-ray (covid-19 & pneumonia). Kaggle. Retrieved April 3, 2023, from https://www.kaggle.com/datasets/prashant268/chest-xray-covid19-pneumonia
ieee8023 (n.d.) IEEE8023/covid-chestxray-dataset: We are building an open database of COVID-19 cases with chest X-ray or CT images. GitHub. https://github.com/ieee8023/covid-chestxray-dataset
Mooney P (2018) Chest X-ray images (pneumonia). Kaggle. Retrieved April 3, 2023, from https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia
Agchung - overview. GitHub. (n.d.). Retrieved April 3, 2023, from https://github.com/agchung
Finn C, Abbeel P, Levine S. (2017) Model-agnostic meta-learning for fast adaptation of deep networks. In Proceedings of the 34th International Conference on Machine Learning - Volume 70 (ICML’17). JMLR.org, 1126–1135
Coronavirus death toll. Worldometer. (n.d.). Retrieved July 9, 2023, from https://www.worldometers.info/coronavirus/coronavirus-death-toll/
Acknowledgements
The authors, thanks to all the anonymous reviewers of Multimedia Tools and Applications Journal for their constructive remarks and fruitful suggestions to improve the manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
We do not have any conflict of interest related to the manuscript.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Sriram, D.S., Ranjan, A., Ghuge, V. et al. Personalized federated learning for the detection of COVID-19. Multimed Tools Appl 83, 29067–29084 (2024). https://doi.org/10.1007/s11042-023-16810-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16810-y