Bionic models for identification of biological systems

International Conference on Information Technologies in Business and Industry 2016

计算机科学;经济学;工业技术

Gerget, O.M.^1

Tomsk Polytechnic University, 30, Lenina ave., Tomsk

634050, Russia^1

关键词: Adverse outcomes;    Biomedical data;    Bionic model;    Clinical decision support systems;    Control actions;    Hyper-parameter optimizations;    Hyperparameters;    Pregnant woman;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/803/1/012046/pdf
DOI  :  10.1088/1742-6596/803/1/012046

来源: IOP

【 摘 要 】

This article proposes a clinical decision support system that processes biomedical data. For this purpose a bionic model has been designed based on neural networks, genetic algorithms and immune systems. The developed system has been tested on data from pregnant women. The paper focuses on the approach to enable selection of control actions that can minimize the risk of adverse outcome. The control actions (hyperparameters of a new type) are further used as an additional input signal. Its values are defined by a hyperparameter optimization method. A software developed with Python is briefly described.