Heart Rate Variability And The Lorenz Diagram

The heart is an important organ which gives dynamic power in the human body. Heart and cardiovascular diseases do serious harm to human health, and are also the main causes which lead to human deaths and disabilities. Therefore, the related study of those diseases has been of great importance to scientific researchers. ECG has been used for one hundred years, and is still the most common and effective way to check heart diseases. This paper is a study of heart rate variability which is extracted from the ECG signals. Heart rate variability (HRV) analysis is a non-invasive detection method for evaluation of the autonomic nervous regulation of cardiac function, and it has an important diagnostic value to analyze the related diseases.In this paper, we choose the data records as samples for analysis from the MIT-BIH ECG database of normal subjects and patients with congestive heart failure, atrial fibrillation or sudden cardiac death. Based on the nonlinear properties of ECG signals and the research and summary of the HRV analysis, we lay special stress on the use of Lorenz plots analysis of heart rate variability. The previous studies mainly focused on the differentiation of the shape of 2-D images, which provided only a limited amount of information, with poor differentiation effects. Therefore, we tried to extend Lorenz plots from 2-D to 3-D space, even to 4-D space. Based on the distribution properties of 3-D and 3-D differentiation Lorenz plots in normal groups, we established a set of quantitative parameters respectively calculated by computer, according to the sphere diameter and the cone with the diameter, angle, length, volume etc. We proposed the standards of the quantitative analysis of 3-D Lorenz plots, based on the calculated quantitative results. We analyzed and compared the different groups with the standards. On this basis, by increasing the one-dimensional axes we can further come to the 4-D Lorenz plots analysis. Based on the dynamic change characteristics of each Lorenz plots, especially on the characteristics of the Lorenz plots in the normal group, which moves with relative concentration in a certain direction and rapidly, we have done a comparative analysis of each group. The findings suggest that these high-dimensional Lorenz plots can effectively distinguish between normal groups and the other cardiovascular groups. To facilitate HRV analysis, we have adopted MATLAB GUI to program HRV analysis software and achieved graphical visualization of Lorenz plots. It can be observed directly from the Multi-angle of 3-D Lorenz plots and the 4-D Lorenz plots dynamic visual images. The application software can not only analyze the 3-D and 3-D differentiation Lorenz plots quantification parameters, but also calculate and analyze parameters in time-domain and complexity. It is of a certain practical value in scientific research and clinical application.