Research On Non-invasive Detection Methods Of Early Diabetic Autonomic Neuropathy

Various complications are the main damage of Diabetic Mellitus (DM), and among them Diabetic Autonomic Neuropathy (DAN) is recognized as the most common clinical pictures of nervous system disorders. Occurring about 60%~90%, which is the vital cause that leads to disable and death can be often seen in such characteristics as being taken bad potentially, deteriorating gradually, be seldom alleviating automatically, and even conning early than clinical signs in some cases. So the research on early DAN detecting methods has important clinical significances and social values.The diagnosis and treatment of DAN are influenced greatly by the complexity and variability of DAN patients' clinical signs, inconsistent detecting methods and diagnosing standards. While the research on non-invasive detection methods of diabetic neuropathy is so weak at present, especially the diagnosis and measurement to those early involvement of DAN. Furthermore, the quantitative standards of early DAN have not been established so far.Considering the demand for non-invasive early DAN detection and the neuropathy quantitative standards, we studies the epidemic situations and diagnosis relevant to DAN based on various research methods from domestic and overseas materials. Besides, this dissertation interprets systematically on the physiological basis of involvement, clinical signs and non-invasive detection methods of early DAN, especially puts the emphasis on Heart Rate Variability (HRV) and Galvanic Skin Response (GSR) Analysis as non-invasive methods of early DAN detection. The main research work is as follows:1. Studied wholly on the HRV analysis. Applying advancedtechnology and means on biomedical signal to RRI data analysis, we combined unevenly sampled power spectrum with conventional evenly sampled frequency domain analysis, and discussed some nonlinear HRV analysis methods based on chaos theory.2. Studied on the early detection of DAN by using GSR, a complementary method.3. Accomplished manufacturing the prototype instrument to detect and analyse early DAN based on virtual instrumental Workbench - LabVIEW and MATLAB.4. Experiments were performed upon some normal controls and hospitalized DM patients with prototype instrument. HRV analysis experiment compared 74 pairs of controls and DM patients, and GSR analysis experiment compared 75 pairs. Besides, some data analysis was carried out on the basis of the statistic results.The examination proved and confirmed that the selected non-invasive methods and quantitative indices can fulfill the requirement of diagnosis to early involvement of autonomic nerve system as well as further studies. The main innovation is as follows:1. Implemented quantitative indices of early DAN diagnosis and treatment through considering its physiological basis of involvement and applying some effective biomedical signal processing methods to HRV analysis combined with GSR analysis.2. Using HRV analysis as primary and GSR as a complementary method of early DAN detection, the work of designing the prototype instrument based on Virtual Workbench bas been finished. The prototype instrument has such virtues as non-invasive, secure, convenient and easy to be a pocket size device.