FedMCC: Federated multi-center clustering algorithm to improve privacy healthcare.

In recent years, healthcare data from various sources such as clinical institutions, patients, and pharmaceutical industries have become increasingly abundant. However, due to the complex healthcare system and data privacy concerns, aggregating and utilizing these data in a centralized manner can be challenging. Federated learning (FL) has emerged as a promising solution for distributed training in edge computing scenarios, utilizing on-device user data while reducing server costs. In traditional FL, a central server trains a global model sampled client data randomly, and the server combines the collected model from different clients into one global model. However, for not independent and identically distributed (non-i.i.d.) datasets, randomly selecting users to train server is not an optimal choice and can lead to poor model training performance. To address this limitation, we propose the Federated Multi-Center Clustering algorithm (FedMCC) to enhance the robustness and accuracy for all clients. FedMCC leverages the Model-Agnostic Meta-Learning (MAML) algorithm, focusing on training a robust base model during the initial training phase and better capturing features from different users. Subsequently, clustering methods are used to ensure that features among users within each cluster are similar, approximating an i.i.d. training process in each round, resulting in more effective training of the global model. We validate the effectiveness and generalizability of FedMCC through extensive experiments on public healthcare datasets. The results demonstrate that FedMCC achieves improved performance and accuracy for all clients while maintaining data privacy and security, showcasing its potential for various healthcare applications.

Association of COL4A2 Gene Polymorphisms with Lacunar Stroke in Xinjiang Han Populations.

Recent studies have shown that variants in the COL4A2 genes are associated with sporadic cerebral small vessel disease. The aim of the study was to investigate the relationship between COL4A2 gene polymorphisms and lacunar stroke in Xinjiang Han populations. The improved multiple ligase detection reaction (iMLDR) method was used to analyze the genotypes of seven single-nucleotide polymorphisms (SNPs) in the COL4A2 gene (rs3803230, rs391859, rs4103, rs445348, rs76425569, rs7990383, rs9515185) in a case-control study of 406 lacunar stroke patients and 425 controls. The GG genotype of rs3803230 (adjusted OR = 1.303, 95% CI = 1.146-1.480, P < 0.001) and the GA/AA genotype of rs76425569 (adjusted OR = 1.744, 95% CI = 1.306-2.329, P < 0.001) showed significant increases in the risk of lacunar stroke. The G-A haplotype of rs3803230-rs76425569 carried a significant increase in the risk of lacunar stroke (OR = 1.616, 95% CI = 1.292-2.022, P < 0.001). Hypertension stratification analyses demonstrated that the GA/AA genotype of rs76425569 was significantly associated with lacunar stroke in the hypertensive group (adjusted OR = 1.316, 95% CI = 1.083-1.598, P = 0.006). In the non-hypertensive group, the GG genotype of rs3803230 (adjusted OR = 1.584, 95% CI = 1.257-1.997, P < 0.001) and GA/AA genotype of rs76425569 were significantly associated with lacunar stroke (adjusted OR = 1.312, 95% CI = 1.054-1.635, P = 0.015). The TT genotype of rs4103 was significantly associated with lacunar stroke in the non-hypertensive group (adjusted OR = 1.355, 95% CI = 1.152-1.594, P < 0.001). This study demonstrates that the COL4A2 gene could play a role in the pathogenesis of lacunar stroke in the Han population of China.