Automatic Analysis of Heart Failure by Statistical and Nonlinear Measures of Heart Rate Variability

The work is devoted to the search for new methods of automatic analysis of the early stage of congestive heart failure (CHF). The possibility of detecting pathology using statistical and nonlinear heart rate parameters is being investigated. Methods of recognition of CHF against the background of normal rhythm and atrial fibrillation are proposed. The significance of the parameters depending on the problem being solved is experimentally evaluated. Threshold classification rules for the detection of CHF have been obtained.