IEEE 3652.1-2020 pdf free download

IEEE 3652.1-2020 pdf free download.IEEE Guide for Architectural Framework and Application of Federated Machine Learning.
or other channels, along with the environmental, socioeconomic, and behavior data. However, health-related data, cspccially patients’ data is highly sensitive and distributed in nature, thus collection and sharing of such data may bring critical legal and ethical privacy concerns. For example, if insurers learn a patient’s health data and find out he/she has severe or high medical cost diseases, they may refuse to provide insurance service. FML can overcome those obstacles by providing a federated machine learning model across organizations while keeping sensitive health data within the local environment. FML applications in the healthcare field may have different scenarios, including business-to-government (B2G). business-to-business (B2B), business-to- customer (B2C), or mixed models.
For common FML scenarios in healthcare there is a need for the collaborative building of FML models among different hospitals. companies, research institutions. etc. Direct-moving data between hospitals may raise concerns about security, privacy, and availability of medical data. FML can address these concerns, and the horizontal FML model should achieve better performance than models trained with single institutional data. As an example, applying horizontal FML in genetic studies allows for the comprehensive analysis of genes. and helps to discover the hidden patterns between genotype and phenotype: it also benefits diagnostic and treatment development of diseases, such as cancer. Currently. samples collected from a single institution are insuflicient to cover all the mutations in breast cancer type I and breast cancer type 2 (BRCA 1/2) genes. while FML provides a feasible and secured way of training an FML model predicting the risk of breast and ovarian cancer.
In contrast, a vertical FML model exploits the vertically-partitioned health data from different institutions, where the two data sets share the same patient set but differ in feature space (i.e., pathological data, multiomics data). This is usually the case in collaboration between hospitals and other medical companies with different data types. A typical example is to build a medication guidance system between hospitals and DNA! RNA sequencing companies when the hospitals have patients’ clinical data, and sequencing companies have patients’ genetic data. By building a vertical FML model, hospitals and companies can combine their features without revealing patients’ privacy. Researchers or clinicians should be able to build better diagnostic or predictive models with FML than with single institutional data. Vertical FML can also include both B2B and B2C as a mixed scenario.
For all scenarios mentioned above, government departments can be added as other parties in the FML for acting supervisors or management roles. IEEE 3652.1-2020 pdf free download.