Objective analysis of human pulse based on traditional Chinese medicine pulse matrix

Pulse diagnosis, an important part of Chinese traditional medicine (TCM), has developed over thousands of years and is still valued and applied worldwide. The process of pulse diagnosis necessitates extensive training and practice by physicians, necessitating the objectification of pulse diagnosis. However, the existing studies on the definition and description of pulse waveforms are incomplete, and the standards for objectification vary, which to some extent affects the further development of pulse diagnosis in TCM. In this research, a concept of pulse matrix is introduced and the definition and description of 13 pulse signals are accomplished through a literature review combined with the characterization methods in time and frequency domains, and a new method of pulse recognition based on Spearman's rank correlation coefficient is introduced. For a single pulse signal, the recognition of the pulse was completed. For continuous pulse signals, a random pulse signal generator was constructed, and the recognition of continuous pulse. Through the recognition experiments of single pulse, continuous pulse and real pulse, the effectiveness and rationality of the method are proved. It provides a promotion for the modern development of TCM pulse diagnosis. With the continuous progress of technology and the deepening of interdisciplinary cooperation, it is believed that TCM will usher in a prosperous development in the near future. • Based on the concept of pulse diagnosis in traditional Chinese medicine, a concept of pulse matrix is proposed. Based on the concepts of traditional Chinese medicine and combining the descriptive ideas of mathematics and science, we have made an attempt to mathematize and objectify pulse diagnosis in Chinese medicine. TCM pulse diagnosis can be viewed in principle as the following process: the practitioner constructs a pulse matrix through experience and study. When taking the patient's pulse, the pulse signals obtained are assigned values to the pulse matrix coefficients based on experience, and the type of pulse is then derived. • Based on the literature and various sources, the standard pulse was objectified in conjunction with time-frequency domain analysis of the pulse waveform. Combining various kinds of data, and the concept of “position, number, shape and trend” in Chinese medicine, as well as signal analysis in the time-frequency domain, we have completed the definition and description of 13 kinds of standard pulse images. • A new pulse recognition scheme based on Spearman's correlation coefficient is proposed. The existing deep learning-based pulse condition research has higher requirements on the dataset, which limits the development of pulse recognition research in TCM to a certain extent. Therefore, in this paper, we propose a pulse recognition scheme based on Spearman's correlation coefficient. On this basis, for a single pulse signal, the recognition of the pulse was completed. For continuous pulse signals, a random pulse signal generator was constructed, and the recognition of continuous pulse. Finally, we calculated the similarity of the experimentally acquired real pulse signals and achieved good recognition results.