Heart Rate Variability Analysis Method Based on KNN Classification

  • Jingchuan ZhaoEmail author
  • Hongwei Zhuang
  • Ling Lu
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1146)


Complex fluctuations of heart rate variability (HRV) reflect the autonomic regulation of the heart. At present,the characteristics of different HRV signals are not obvious and the classification accuracy is not high which restricts their practical use. Therefore, this paper proposes a signal pattern recognition algorithm based on KNN classification for HRV signal analysis and recognition. HRV characteristic matrix is formed by HRV characteristic values in frequency domain which is extracted by AR spectrum analysis, time domain characteristic values and nonlinear characteristic values. Then the matrix is processed by the K-nearest neighbor classification algorithm. In order to verify the accuracy of the algorithm, four types of ECG data in MIT-BIH database are used to train and detect the algorithm in this paper. The proposed algorithm’s accuracy of HRV classification and recognition can be up to 93.5%, which is higher than other classification recognition algorithms of the same type. These results provide a new method for extraction and classification the eigenvalues of heart rate variability.


Heart rate variability AR spectrum analysis K-neighbor classification algorithm 



This research was supported by innovation team Science foundation of Engineering University of PAP (KYTD201905).


  1. 1.
    Hon, E.H., Lee, S.T.: Electronic evaluation of the fetal heart rate. Patterns preceding fetal death, further observations. Am. J. Obset. Gynecol. X 87, 814–826 (1965)Google Scholar
  2. 2.
    Liu, X., Miao, D., et al.: Study on heart rate variability of normal young men under different stress situations. Chin. Behav. Med. Sci. 11(6), 679–680 (2002). (in Chinese)MathSciNetGoogle Scholar
  3. 3.
    Shao, S., Wang, T., Song, C., et al.: A new measure of heart rate variability. J. Phys. 68(17), 1–12 (2019). (In Chinese)Google Scholar
  4. 4.
    Pan, Z.: Random Signals and Systems, pp. 23–56. National Defense Industry Press, Beijing (2013). (in Chinese)Google Scholar
  5. 5.
    Li, F.: Study on ECG signal processing based on wavelet transform. Master’s thesis of Nanjing University of Aeronautics and Astronautics (2009). (in Chinese)Google Scholar
  6. 6.
    Rao, J.: Study and Application of Heart Rate Variability Analysis Method. South China University of Technology, Guangzhou (2016). (in Chinese)Google Scholar
  7. 7.
    Chai, X., Wang, B., et al.: Study on the method to determine the optimal order in the analysis of heart rate variability autoregressive model. J. Biomed. Eng. 32(5), 958–964 (2015). (in Chinese)Google Scholar
  8. 8.
    Wang, Y.: Study on heart failure diagnosis model based on heart rate variability. Master’s thesis of Shandong University (2018). (in Chinese)Google Scholar
  9. 9.
    Yu, S.: Morphological recognition of ECG waveform. Master’s thesis of Zhejiang Normal University (2014). (in Chinese)Google Scholar
  10. 10.
    Khorrami, H., Moavenian, M.: A comparative study of DWT, CWT and DCT transformations in ECG arrhythmias classification. Expert Syst. Appl. 37(8), 5751–5757 (2010)CrossRefGoogle Scholar
  11. 11.
    Lisukor, M.: A qualitative comparison of artificial neural networks and support vector machines in ECG arrhythmias classification. Expert Syst. Appl. 37(4), 3088–3093 (2010)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Engineering University of PAPXi’anChina

Personalised recommendations