Research on feature mining algorithm and disease diagnosis of pulse signal based on piezoelectric sensor

• Published in: Informatics in medicine unlocked Vol. 26; p. 100717
• Main Authors: Lin, Fan, Zhang, Jincheng, Wang, Zhongmin, Zhang, Xiaokang, Yao, Ruiling, Li, Yan
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 2021; Elsevier
• Subjects: Pathological feature mining; Physiological signal feature; Pulse diagnosis; Pulse wave analysis; Pathological feature mining; Pulse diagnosis; Physiological signal feature; Pulse wave analysis
• ISSN: 2352-9148; 2352-9148
• DOI: 10.1016/j.imu.2021.100717

The human pulse contains various information reflecting the internal environment of the human body. However, the classical method of pulse diagnosis in traditional Chinese medicine (TCM) has the disadvantages of relying too much on the doctor's experience and the diagnosis result is too subjective. Based on the principle of TCM pulse diagnosis, the use of photoelectric sensors to collect the pulse signals of multiple healthy people and patients with chronic diseases, and organize the detailed pulse information into a data set and analyze it with algorithms, is a solution to overcome this problem through modern technology. However, this method is still difficult to understand the patient's physiological condition in detail, and it is also difficult to explain the internal connection between abnormal pulse conditions and their physiological conditions. In the experiment, after denoising, smoothing, and eliminating the baseline drift of the subjects' pulse data, we designed two algorithms to describe the difference between the two-dimensional images of the pulse data of normal people and patients with chronic diseases. The specific feature values obtained are converted into a multi-dimensional array and trained in a support vector machine (SVM) classifier. The classification accuracy is higher than the basic temporal features. Experimental results show that it is feasible to use specific feature mining algorithms for disease detection. Through analysis, this paper found the pathological characteristics reflected in the two-dimensional pulse image, discovered the internal connection between the pulse waveform characteristics of the human body and the disease, and tried to describe it through algorithms, trying to establish a method for detecting specific diseases using photoelectric signals. •When we record a large number of pulse images of patients and normal people and analyze the pulse image data sets of normal people and patients, we find that the pulse images of some patients can extract some characteristic data to distinguish the differences between them and normal people. We try to extract these data and use a computer algorithm to distinguish, to achieve the effect of identifying specific patients and diseases.Through this paper, we verify a pulse image analysis, combined with an algorithm had designed a method that can effectively determine normal people and patients with chronic diseases. When processing a large number of pulse images, we also design two unique sampling algorithms to extract the data we need from the data set of MATLAB software and complete the division. For how to extract the pulse data from the data set, this paper provides a solution.•According to the existing pulse diagnosis research, although pulse waveform analysis (PWA) and disease diagnosis based on machine learning and pattern recognition technology has extensively experimented on different technologies and instruments. However, the factors that affect the characteristic value of the pulse waveform, such as the specification and type of the sensor used, the measurement process, the subject's gender, age, biological clock, physiological state, and disease, are often not considered in the disease classification results of this model. And how these factors are reflected as the specific values in the pulse condition of the subjects, and what influence do they have on the internal characteristics of the pulse condition data, we often need to explore these factors when we complete the diagnosis of PWA disease. In the actual experiment process, due to the influence of many of the above factors, it is difficult to use the extracted features as the feature to determine that the subject has a specific disease. It is not enough to use the only SVM to segment the pulse data between normal people and patients. However, we are trying to establish a method to identify the characteristics of pathological data in human pulse images, to determine the subject's disease, and make the classification results more credible.