Background: Heart dynamics have been studied via different approaches such as heart rate variability; however, recent analyses have been incorporated from the fields of theoretical physics and mathematics to develop diagnostic methodologies of clinical utility. Objective: To show the diagnostic capacity of a methodology designed to discriminate normal dynamics from acute cardiovascular disease dynamics of patients in intensive care units. Methodology: The data of Holter registries were taken from 120 subjects, from which 30 were clinically diagnosed as normals and 90 exhibited different cardiac pathologies. The minimal and maximal heart rates, as well as the number of heartbeats for each hour, were obtained to generate chaotic attractors. These were measured through grids according to the Box-Counting method to determine if the dynamic is acutely pathological, normal, or in evolution. The sensitivity and specificity, as well as the Kappa coefficient, were calculated to determine the effectiveness of the mathematical method with respect to the clinical Gold Standard for diagnosis. Results: The values of fractal dimension did not allow the differentiation of normals from disease. However, values obtained with the Kp grid did enable differentiation of normality versus abnormality, with values of sensitivity and specificity of 100% and a Kappa coefficient equal to 1. Conclusions: A clinical application of a methodology based on dynamical systems, fractal geometry, and a chaotic mathematical law was conducted which allowed perfect differentiation of normality from acute cardiovascular disease in patients in an intensive care unit.
Full Text Finder