ABSTRACT
With the development of artificial intelligence, pulse diagnosis has been standardized and objectified. However, there is a lack of research on the extraction and dimensionality reduction of hypertensive pulse features. We propose two effective features for distinguishing pulses of disease samples and a fusion dimensionality reduction method that combines linear and nonlinear dimensionality reduction. The results show that the proposed features and dimensionality reduction method make the classification accuracy of hypertension pulse feature reach 94.23%, and the training time of the classifier is reduced by 47 seconds, which improves the performance in terms of both accuracy and time.
- R. Velik, An objective review of the technological developments for radial pulse diagnosis in traditional Chinese medicine, Euro. J. Integr. Med. 7 (4) (2015) 321–331.Google ScholarCross Ref
- L. Xu, K. Wang, D. Zhang, Y. Li, Z. Wan, J. Wang, Objectifying researches on traditional Chinese pulse diagnosis, Inform. Med. Sloven. 8 (1) (2003) 56–63.Google Scholar
- Jl A, Qiang W B. Multi-modal bioelectrical signal fusion analysis based on different acquisition devices and scene Settings: overview, challenges, and novel orientation[J]. Information Fusion 79 (2022) 229-247.Google ScholarDigital Library
- Rosemarie, Velik. An objective review of the technological developments for radial pulse diagnosis in Traditional Chinese Medicine[J]. European Journal of Integrative Medicine, 2015, 7(4):321-331.Google ScholarCross Ref
- Liu Lei, Wu Qiufeng, Zhang Hongzhi, A review of pulse diagnosis objectification research[J]. Intelligent Computers and Applications, 2013(03):24-28.Google Scholar
- Zhang Z, Zhang Y, Song H, A Sensor-Based Wrist Pulse Signal Processing and Lung Cancer Recognition[J]. Journal of Biomedical Informatics, 2018:S153204641830011X.Google Scholar
- Zj A, Cg A, Jin Z A, Features fusion of multichannel wrist pulse signal based on KL-MGDCCA and decision level combination[J]. Biomedical Signal Processing and Control, 57. STFTGoogle Scholar
- WANG L, WANG K, XU L. Lempel-ziv decomposition based arrhythmic pulses recognition[C]//Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005.27th Annual International Conference of the. IEEE, 2006: 4606-4609.Google Scholar
- WANG K, XU L, LI Z, Approximate entropy based pulse variability analysis: Computer-based medical systems[C]//2003. Proceedings. 16th IEEE Symposium. IEEE, 2003: 236-241.Google Scholar
- Wang Ke, Xiao Pengfeng, Feng Xuezhi, Improved two-dimensional discrete Hilbert transform-based image edge detection method[J]. Journal of Surveying and Mapping, 2012, 41(3):8.Google Scholar
- Kuang Xue, Li Zhi, Wang Yongjun, An improved HHT-based method for pulse signal analysis[J]. Computer Engineering, 2019, 45(9):6.Google Scholar
- Guo C-L, Chen H-C, Wang Z-F. Research on human pulse feature recognition and classification based on HHT[J]. Electronic Measurement Technology, 2021, 44(19):6.Google Scholar
- Rosemarie, Velik. An objective review of the technological developments for radial pulse diagnosis in Traditional Chinese Medicine[J]. European Journal of Integrative Medicine, 2015, 7(4):321-331.Google ScholarCross Ref
- Bae J H, Jeon Y J, Lee S, A feasibility study on age-related factors of wrist pulse using principal component analysis[C]//2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, 2016.Google Scholar
Index Terms
- Hypertensive Pulse Diagnosis Method Based on Hilbert-Huang Transform and Feature Fusion Dimensionality Reduction
Recommendations
Wrist pulse diagnosis of stable coronary heart disease based on acoustics waveforms
Highlights- This paper presented for the first time the stratification of pulse into five layers. In addition, for the first time, acoustic sensors were used to collect ...
Abstract Background and objectiveAs a common pathological pulse, unsmooth pulse has important diagnostic value in traditional Chinese medicine (TCM). In modern pulse diagnosis, unsmooth pulse plays an important role in the diagnosis ...
Feature Extraction for Pulse Waveform in Traditional Chinese Medicine by Hemodynamic Analysis
BIBM '09: Proceedings of the 2009 IEEE International Conference on Bioinformatics and BiomedicinePulse diagnosis is one of important diagnosis methods in Traditional Chinese Medicine(TCM). Feature extraction of TCM pulse has received more and more attention in recent years. While most of previous methods for feature extraction of TCM pulse have no ...
Kernel-based nonlinear dimensionality reduction for electrocardiogram recognition
The human heart is a complex system that reveals many clues about its condition in its electrocardiogram (ECG) signal, and ECG supervising is the most important and efficient way of preventing heart attacks. ECG analysis and recognition are both ...
Comments