An Open Source Matlab Tool For Heart Rate Turbulence And Its Application To Chronic Heart Failure Assessment

In this paper we describe a software package for the development of a Heart Rate Turbulence (HRT) evaluation tool, that provides also an instrument for HRT analysis. In fact, as HRT measurement represents an indirect assessment of baroreflex, it is of great importance in all those clinical situations that benefit from baroreflex evaluation, specially in terms of risk prediction, monitoring and stratification. Based on a review of the state of the art of software related to the HRT analysis, we motivate the development of this open source software platform which could be an interesting tool both for studying HRT or for performing clinical experiments for research purposes. We present the results of the application of our package to two ECG databases from Physionet of patients suffering from Chronic Heart Failure. We performed a discriminant analysis using the HRT parameters in order to distinguish between severe CHF from mild CHF. We adopted as discriminant function the Mahalanobis distances with stratified covariance estimates. The computed HRT parameters enables severe CHF patients to be distinguished from mild CHF ones with a sensitivity and a specificity rate of 100.0% and 72.7%, respectively. As proved by the results of this research, HRT measurements appear to be an effective instrument for Congestive Heart Failure severity assessment. In particular we found that the combined use of two of the HRT parameters computed through our Matlab tool, increases the discrimination performances, especially in terms of sensitivity and accuracy. We firmly believe that the results of our study and of previously published papers about HRT encourage further investigations concerning HRT parameters in many other clinical situations and, in this scenario, the developed Matlab tool could show its potentiality for future researches.