Exploiting intelligent decision supports for model-driven biomedical system analysis

Both linear modeling and nonlinear modeling provide special means to simplify problem-solving in science and engineering, which also represents a critical stage for intelligent decision support systems. Model-driven decision support systems have numerous real world applications, such as the feature extraction, image compression, pattern recognition, medical diagnosis and telecommunication. In this article, an intelligent decision support system has been exploited for biomedical system analysis. The PCA (Principal Component Analysis) and ICA (Independent Component Analysis) based linear and nonlinear models have been implemented onto the sample characterization of the biomedical systems for decision making, together with applications of neural networks training. The result indicates the effectiveness of the proposed approaches, where the merit and drawback are discussed between linear and nonlinear modeling. Numerical simulations have also been made.