Federated learning based futuristic biomedical big-data analysis and standardization.
Federated learning based futuristic biomedical big-data analysis and standardization.
复制标题
DOI：
10.1371/journal.pone.0291631
复制
发表时间：
2023
期刊：
影响因子：
3.7
通讯作者：
Zhao, Zhongming
中科院分区：
文献类型：
--
作者：
Fathima, Afifa Salsabil;Basha, Syed Muzamil;Ahmed, Syed Thouheed;Mathivanan, Sandeep Kumar;Rajendran, Sukumar;Mallik, Saurav;Zhao, Zhongming
医疗数据处理和分析在为未来的生物医学应用提供可靠的决策支持方面发挥着重要的影响。鉴于医疗数据的敏感性，为以应用程序为中心的处理量身定制的专门技术和框架势在必行。本文提出了一个概念化的分析和均匀的数据集，通过实施联邦学习（FL）。医疗大数据领域源于不同的起源，需要描述数据来源和属性范式，以促进特征提取和依赖性评估。管理数据收集框架的架构与远程数据传输有着错综复杂的联系，从而产生了有效的定制监督。该操作方法分为四个层次：数据源层、数据采集层、数据分类层和数据优化层。这一奋进的核心是多目标最优数据集（MooM），其特征是通过联邦学习模型的管道进行属性驱动的特征制图和聚类分类。特征同步和参数提取的编排在多层神经网络中进行，最终通过数据集标准化和标记提供了一种坚定的补救措施。实证研究结果反映了所提出的技术的有效性，拥有令人印象深刻的97.34%的准确率，在解开和聚类的远程医疗数据，促进了联邦模型的范围内的业务服务器。
Medical data processing and analytics exert significant influence in furnishing dependable decision support for prospective biomedical applications. Given the sensitive nature of medical data, specialized techniques and frameworks tailored for application-centric processing are imperative. This article presents a conceptualization for the analysis and uniformitarian of datasets through the implementation of Federated Learning (FL). The realm of medical big data stems from diverse origins, necessitating the delineation of data provenance and attribute paradigms to facilitate feature extraction and dependency assessment. The architecture governing the data collection framework is intricately linked to remote data transmission, thereby engendering efficient customization oversight. The operational methodology unfolds across four strata: the data origin layer, data acquisition layer, data classification layer, and data optimization layer. Central to this endeavor are multi-objective optimal datasets (MooM), characterized by attribute-driven feature cartography and cluster categorization through the conduit of federated learning models. The orchestration of feature synchronization and parameter extraction transpires across multiple tiers of neural networking, culminating in the provisioning of a steadfast remedy through dataset standardization and labeling. The empirical findings reflect the efficacy of the proposed technique, boasting an impressive 97.34% accuracy rate in the disentanglement and clustering of telemedicine data, facilitated by the operational servers within the ambit of the federated model.
登录
查看更多内容
影响因子：
9.8
作者：
Hughes, Laura D.;Tsueng, Ginger;DiGiovanna, Jack;Horvath, Thomas D.;Rasmussen, Luke V.;Savidge, Tor C.;Stoeger, Thomas;Turkarslan, Serdar;Wu, Qinglong;Wu, Chunlei;Su, Andrew I.;Pache, Lars
通讯作者：
Pache, Lars
影响因子：
2.2
作者：
Ahmed, Shakeel;Ramesh, N. V. K.;Reddy, B. Naresh Kumar
通讯作者：
Reddy, B. Naresh Kumar
影响因子：
10.6
作者：
Ahmed, Syed Thouheed;Kumar, V. Vinoth;Kim, Jungyoon
通讯作者：
Kim, Jungyoon
影响因子：
5.3
作者：
Ahmed, Syed Thouheed;Sandhya, M.;Sankar, Sharmila
通讯作者：
Sankar, Sharmila
影响因子：
15.2
作者：
Rieke, Nicola;Hancox, Jonny;Cardoso, M. Jorge
通讯作者：
Cardoso, M. Jorge