Mathematical-physical prediction of cardiac dynamics using the proportional entropy of dynamic systems

Juan Mauricio Pardo Oviedo, Signed Prieto, Javier Rodríguez-Velásquez, Dario Dominguez, Martha Melo, Fernán Mendoza, Catalina Correa, Yolanda Soracipa , Laura Pinilla, Nathalia Ramirez

Research output: Contribution to journalArticlepeer-review

Abstract

Heart dynamic characterized within the context of dynamical systems theory allows differentiating and
predicting normal cardiac states, different levels of disease, as well as evolution towards disease. The purpose of this study is to apply a previously developed methodology to 450 electrocardiographic registers to establish its effectiveness and comparing it with clinical conventional diagnosis. The methodology was applied to 50 normal Holters and 400 Holters with different pathologies. After masking the diagnostic conclusions, the minimum and maximum heart rate and the total number of beats each hour were used to construct an attractor for each Holter in a phase space, by means of which the probability, entropy and their proportions were evaluated in ordered pairs of heart rate. Measures were compared with the physical and mathematical parameters of normality and disease previously settled down. Diagnostic conclusions and dates from medical history of each Holter were unmasked, to calculate sensibility, specificity and coefficient Kappa respect to Gold-standard. This methodology allowed mathematically differentiating the normal, acute and chronic disease dynamics and the evolution among these states. Sensibility and specificity of 100% were obtained and Kappa coefficient was equal to 1, demonstrating its diagnostic utility.
Original languageEnglish (US)
Pages (from-to)370 - 381
Number of pages11
JournalThe Journal of Medicine and Medical Sciences
Volume4
Issue number9
StatePublished - 2013

Fingerprint

Dive into the research topics of 'Mathematical-physical prediction of cardiac dynamics using the proportional entropy of dynamic systems'. Together they form a unique fingerprint.

Cite this