Research On Smart Garments Oriented Medical Diagnosis Application Based On Multi-fusion Information

Wearable Electronics Intelligent Clothing is a kind of Interdisciplinary technology, which is a bold innovation on clothing design. It comes with the context of the continuous development of computer hardware technology, various highly integrated and ultra-miniature components, as well as the continual emergence of new computer theory and technology. On the other hand, medical diagnosis based on Multi-source information fusion becomes more and more important with the more and more attention paid to the health. The research work can be divided into two categories, one is a decision-making system of physiological characteristics based on rules, and the other is the Modern multi - information fusion method such as Bayesian Decision, Neural Network and so on. The paper achieved a system of Smart Garments Oriented Medical Diagnosis Application Based on Multi-fusion Information with the combination of two key technologies, the smart Garments and the medical diagnosis based on Multi-source information fusion.First of all, the paper optimizes the current algorithm. On the one hand, we study the multi-source information fusion algorithm, compare them and then select the one which is more reliability and whose recognition rate is higher. Then we focus on wavelet support vector machine (Wavelet Support Vector Machine) in a multi - physiological data fusion in medical diagnosis applications. So the Wavelet Support Vector Machine used on medical diagnosis based on Multi-source information fusion is the most important. On the other hand, we study on the Parallel Support Vector Machine Training Algorithm based on Sequential minimal optimization in order to improve the efficiency of the training. At the same time, we achieve the Parallel Sequential minimal optimization Algorithm Based on LabVIEW, combining the advantage in parallel computing of LabVIEW. Besides that, we confirm the Parallel Computing arithmetic makes the training rate higher with the increase of processors through the experiment.Secondly, it introduces the architecture of Smart Garments Oriented Medical Diagnosis .The system is divided into Signal Fetching, Signal Processing and Signal Response. It collects physiological information through the front-end sensors and sends them to the background by wireless devices, and then the background monitoring software process the information accordingly to get result sent to the users. The Monitoring software uses C # language based on Windows platform, what is also a part of independent research of the whole system.Finally, the paper gives two Medical Diagnosis applications on Smart Garment, Intensive Care and Myocardial Ischemic Diagnosis. It proves the feasibility of the method by training and testing the data of Authoritative database. The paper also realizes the user interface for these two applications. Furthermore, it analyses the performance of Wavelet Support Vector Machine (WSVM) and proves that the Convergence Performance and Classification accuracy of this method is better than Wavelet Neural Network and Gauss Kernel Support Vector Machine.Above all, the Smart Garments Oriented Medical Diagnosis Application integrates the technology of sensors, wearable computing, and digital signal processing, embedded systems and so on. Medical diagnostic function is embedded into the clothing in order to achieve the online medical diagnosis. This diagnosis method is very important for the long-term monitoring of chronic diseases and acute temporary-aided diagnosis.