This paper presents modeling and simulation of a multi-input multi-output (MIMO) adaptive control system (ACS) for human arterial blood pressure (HABP) by multiple drug inputs -infusion speed (IS) control of inotropic...This paper presents modeling and simulation of a multi-input multi-output (MIMO) adaptive control system (ACS) for human arterial blood pressure (HABP) by multiple drug inputs -infusion speed (IS) control of inotropic agent (IA) and vasoactive agent (VA). The MIMO ACS is able to choose the most appropriate IS of IA and VA at good IS in order to maintain the aortic pressure (AOP) and central vain pressure (CAP) at desired levels, and at the same time to increase the cardiac output (CO).This ACS simulation consists of 3 parts: the system model (SM), the identifier (ID), and the explicit multivariable self-turning controller (MSC). The SM is a 2-input 2-output bilinear model with nonwhite system noise. The variable ID is capable of estimating the variable onset delay model (BNVD). The ID is capable of estimating the variable onset delay online by compressing the value in gaining the unbiased parameter estimation with an improved generalized least-squares (LS) algorithm. The MSC employs the minimum variance one-step-ahead control law. These three parts make a closed-loop control system successfully for heart disease patients during and after the operation.展开更多
Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom.This paper proposes a hybrid a...Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom.This paper proposes a hybrid approach based on the support vector machine (SVM) technique and genetic algorithm (GA) to select an optimum feature subset from the feature group extracted from the skin micro-images. An adaptive GA is introduced for maintaining the convergence rate. With the proposed method, the average cross validation accuracy is increased from 88.25% using all features to 96.92% using only selected features provided by a classifier for classification of 5 classes of skin symptoms. The experimental results are satisfactory.展开更多
This paper presents realistic data mining based on the data of B-type ultrasonic detection and diagnosis for cholrcystolithiasis (gallbladder stone in biliary tract) recorded by a district central hospital in Shanghai...This paper presents realistic data mining based on the data of B-type ultrasonic detection and diagnosis for cholrcystolithiasis (gallbladder stone in biliary tract) recorded by a district central hospital in Shanghai during the past several years. Computer simulation and modeling is described.展开更多
文摘This paper presents modeling and simulation of a multi-input multi-output (MIMO) adaptive control system (ACS) for human arterial blood pressure (HABP) by multiple drug inputs -infusion speed (IS) control of inotropic agent (IA) and vasoactive agent (VA). The MIMO ACS is able to choose the most appropriate IS of IA and VA at good IS in order to maintain the aortic pressure (AOP) and central vain pressure (CAP) at desired levels, and at the same time to increase the cardiac output (CO).This ACS simulation consists of 3 parts: the system model (SM), the identifier (ID), and the explicit multivariable self-turning controller (MSC). The SM is a 2-input 2-output bilinear model with nonwhite system noise. The variable ID is capable of estimating the variable onset delay model (BNVD). The ID is capable of estimating the variable onset delay online by compressing the value in gaining the unbiased parameter estimation with an improved generalized least-squares (LS) algorithm. The MSC employs the minimum variance one-step-ahead control law. These three parts make a closed-loop control system successfully for heart disease patients during and after the operation.
文摘Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom.This paper proposes a hybrid approach based on the support vector machine (SVM) technique and genetic algorithm (GA) to select an optimum feature subset from the feature group extracted from the skin micro-images. An adaptive GA is introduced for maintaining the convergence rate. With the proposed method, the average cross validation accuracy is increased from 88.25% using all features to 96.92% using only selected features provided by a classifier for classification of 5 classes of skin symptoms. The experimental results are satisfactory.
文摘This paper presents realistic data mining based on the data of B-type ultrasonic detection and diagnosis for cholrcystolithiasis (gallbladder stone in biliary tract) recorded by a district central hospital in Shanghai during the past several years. Computer simulation and modeling is described.
