文献类型：期刊文献

英文题名：An enhanced privacy-preserving federated few-shot learning framework for respiratory disease diagnosis

作者：Wang, Ming[1];Duan, Zhaoyang[1];Xue, Dong[1];Liu, Fangzhou[2];Zhang, Zhongheng[3,4]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]Harbin Inst Technol, Sch Astronaut, Res Inst Intelligent Control & Syst, Harbin 150001, Peoples R China;[3]Zhejiang Univ, Sir Run Run Shaw Hosp, Dept Emergency Med,Sch Med, Prov Key Lab Precise Diag & Treatment Abdominal In, Hangzhou 310016, Peoples R China;[4]Shaoxing Univ, Sch Med, Shaoxing 312000, Peoples R China

年份：2026

卷号：190

外文期刊名：APPLIED SOFT COMPUTING

收录：;EI(收录号：20260219895073);WOS:【SCI-EXPANDED(收录号:WOS:001664399500001)】；

基金：This work was supported in part by the National Natural Science Foundation of China (Grant No. 62173147, 62373123, 82472243, and 82272180) , the China National Key Research and Development Program (No. 2023YFC3603104) , the Huadong Medicine Joint Funds of the Zhejiang Provincial Natural Science Foundation of China (No. LHDMD24H150001) , the China National Key Research and Development Program (No. 2022YFC2504500) , and the General Health Science and Technology Program of Zhejiang Province (No. 2024KY1099) .

语种：英文

外文关键词：Federated learning; Few-shot learning; Differential privacy; Respiratory disease diagnosis

摘要：The labor-intensive nature of medical data annotation presents a significant challenge for respiratory disease di agnosis, resulting in a scarcity of high-quality labeled datasets in resource-constrained settings. Moreover, patient privacy concerns complicate the direct sharing of local medical data across institutions, and existing centralized data-driven approaches, which rely on the amount of available data, often compromise data privacy. This study proposes a federated few-shot learning framework with privacy-preserving mechanisms to address the issues of limited labeled data and privacy protection in diagnosing respiratory diseases. In particular, a meta-stochastic gradient descent algorithm is proposed to mitigate the overfitting problem that arises from insufficient data when employing traditional gradient descent methods for neural network training. Furthermore, to ensure data privacy against gradient leakage, differential privacy noise from a standard Gaussian distribution is integrated into the gradients during the training of private models with local data, thereby preventing the reconstruction of med ical images. Given the impracticality of centralizing respiratory disease data dispersed across various medical institutions, a weighted average algorithm, called the Federated Averaging Algorithm (FedAvg), is employed to aggregate local diagnostic models from different clients, enhancing the adaptability of a model across diverse scenarios. Experimental results show that the proposed method yields compelling results with the implementa tion of differential privacy, while effectively diagnosing respiratory diseases using data from different structures, categories, and distributions.