Assessment of Heart Rate Complexity Recovery from Maximal Exercise Using Recurrence Quantification Analysis

2016 ◽  
pp. 157-168
Author(s):  
Rosangela Akemi Hoshi ◽  
Carlos Marcelo Pastre ◽  
Luiz Carlos Marques Vanderlei ◽  
Moacir Fernandes Godoy
Keyword(s):  
Heart Rate ◽  
Maximal Exercise ◽  
Recurrence Quantification Analysis ◽  
Recurrence Quantification ◽  
Quantification Analysis ◽  
Heart Rate Complexity

AUTOMATED DIAGNOSIS OF CARDIAC HEALTH USING RECURRENCE QUANTIFICATION ANALYSIS

2012 ◽  
Vol 12 (04) ◽  
pp. 1240014 ◽  
Author(s):  
M. MUTHU RAMA KRISHNAN ◽  
S. VINITHA SREE ◽  
DHANJOO N. GHISTA ◽  
EDDIE Y. K. NG ◽  
SWAPNA ◽  
...  
Keyword(s):  
Heart Rate ◽  
Probabilistic Neural Network ◽  
Gaussian Mixture ◽  
Recurrence Quantification Analysis ◽  
Ecg Signal ◽  
Operating Characteristics ◽  
Fuzzy Classifier ◽  
Cardiac Health ◽  
Recurrence Quantification ◽  
Quantification Analysis

The sum total of millions of cardiac cell depolarization potentials can be represented using an electrocardiogram (ECG). By inspecting the P-QRS-T wave in the ECG of a patient, the cardiac health can be diagnosed. Since the amplitude and duration of the ECG signal are too small, subtle changes in the ECG signal are very difficult to be deciphered. In this work, the heart rate variability (HRV) signal has been used as the base signal to observe the functioning of the heart. The HRV signal is non-linear and non-stationary. Recurrence quantification analysis (RQA) has been used to extract the important features from the heart rate signals. These features were fed to the fuzzy, Gaussian mixture model (GMM), and probabilistic neural network (PNN) classifiers for automated classification of cardiac bio-electrical contractile disorders. Receiver operating characteristics (ROC) was used to test the performance of the classifiers. In our work, the Fuzzy classifier performed better than the other classifiers and demonstrated an average classification accuracy, sensitivity, specificity, and positive predictive value of more than 83%. The developed system is suitable to evaluate large datasets.

Sign in / Sign up

Export Citation Format

Share Document