A unified non-linear approach based on recurrence quantification analysis and approximate entropy: application to the classification of heart rate variability of age-stratified subjects

• Published in: Medical & biological engineering & computing Vol. 57; no. 3; pp. 741 - 755
• Main Authors: Singh, Vikramjit, Gupta, Amit, Sohal, J. S., Singh, Amritpal
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2019; Springer Nature B.V
• Subjects: Age; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Classification; Computer Applications; Data processing; Entropy; Geriatrics; Heart rate; Human Physiology; Imaging; Mathematical analysis; Multilayer perceptrons; Neural networks; Older people; Original Article; Radiology; Standard deviation; Statistical analysis; Support vector machines; Autonomic nervous system (ANS); Approximate entropy; Non-linear methods; Recurrence quantification analysis; Support vector machine; Heart rate variability; Information theory
• ISSN: 0140-0118; 1741-0444; 1741-0444
• DOI: 10.1007/s11517-018-1914-0

This paper presents a unified approach based on the recurrence quantification analysis (RQA) and approximate entropy (ApEn) for the classification of heart rate variability (HRV). In this paper, the optimum tolerance threshold ( r opt ) corresponding to ApEn max has been used for RQA calculation. The experimental data length ( N ) of RR interval series (RR i ) is optimized by taking r opt as key parameter. r opt is found to be lying within the recommended range of 0.1 to 0.25 times the standard deviation of the RR i , when N  ≥ 300. Consequently, RQA is applied to the age stratified RR i and indices such as percentage recurrence (%REC), percentage laminarity (%LAM), and percentage determinism (%DET) are calculated along with ApEn max , r opt min , r opt max , and an index namely the radius differential ( R D ). Certain standard HRV statistical indices such as mean RR, standard deviation of RR (or NN) interval (SDNN), and the square root of the mean squared differences of successive RR intervals (RMSSD) (Eur Hear J 17:354–381, 1996) are also found for comparison. It is observed that (i) R D can discriminate between the elderly and young subjects with a value of 0.1151 ± 0.0236 (mean ± SD) and 0.0533 ± 0.0133 (mean ± SD) respectively for the elderly and young subjects and is found to be statistically significant with p  < 0.05. (ii) Similar significant discrimination was obtained using r opt min with a value of 0.1827 ± 0.0382 (mean ± SD) and 0.2248 ± 0.0320 (mean ± SD) (iii) other significant indices were found to be %REC, %DET, %LAM, SDNN, and RMSSD; however, ApEn max was found to be insignificant with p  > 0.05. The above features of RR i time series were tested for classification using support vector machine (SVM) and multilayer perceptron neural network (MLPNN). Higher classification accuracy was achieved using SVM with a maximum value of 99.71%. Graphical abstract