Decision - Making Analysis Of Female Qi And Blood Pathology Based On Pulse Characteristic Parameters And Facial Temperature Distribution

The theory of TCM holds that qi and blood are the fundamental for life. What the body needs most is qi and blood, which, especially for women, have effects on every aspect of the body.Pulse signals contain important pathophysiologic information in the human body. The principle of the pulse diagnosis of TCM is also based on the facts that pulse signals may better characterize various important information in the human body, including zang-fu viscera, vascular and qi-blood information and various information in the pulse signal reflects the changes in waveforms. With the advent of technology, the objectification and digitization of pulse diagnosis of TCM is increasingly applied to all aspects of disease diagnosis. Therefore, people may correspond to different pathologies by extracting the change information of pulse waveforms, thus assisting physicians to diagnose diseases.Chloasma faciei is a typical disease in the pathology of qi and blood of TCM. In the field of medicine, the diagnosis of chloasma has its corresponding standard, including grades that represent the size and shades of color of chloasma and types that represent the site of chloasma. However, grading and typing of chloasma depends on physician’s subjective judgment and lacks an objective and unified standard, which may affect physician’s judgment of patient conditions and sequential treatment. Therefore, this work is to obtain the facial temperature distribution by taking human facial infrared thermographs, thereby making an assistant judgment of patient conditions of chloasma faciei.Patients with chloasma also usually have menoxenia, which show a relation between its etiology and pathogenesis. TCM holds that menoxenia is a direct factor affecting chloasma faciei. Menoxenia correspond to different TCM syndrome type. Starting with the extraction of time-domain characteristic points of pulse signals, this work summarized and sorted out the typical syndrome types of the pathology of qi and blood and the female-specific pathophysiologic characteristics, and investigated the corresponding relation between the characteristic parameters of pulse waveforms and the syndrome types and manifestations of the pathology of qi and blood, thus assisting physicians to make sound judgments of specific etiology of gynecologic diseases.The research contents of this article include the following aspects:(1) The generation mechanism of the pulse wave and the relation between characteristics of pulse signals and human pathophysiology. The pulse is generated by human blood circulation system, and its waveforms are affected by various physical conditions and closely related to the human physiological information. Research on the corresponding relation between the characteristics of pulse waves and diseases is of importance in clinical diagnosis and treatment.(2) Pulse signal acquisition. In this work, HK-2000C pulse sensor was selected to acquire pulse waves of the human body, and all pulse data were actually collected from the Gynecology Clinic, Yunnan Provincial Hospital of Traditional Chinese Medicine.(3) Pulse signal preprocessing. For actually collected pulse wave data, pulse wave signals included various noises and invalid components. This work proposed a method based on the AR model spectral estimation to remove random noises from pulse signals, and used modified LIP algorithm for cycle partition of the whole pulse signal, cubic spline interpolation method for removing baseline drift, and non-matching continuation algorithm for cutting invalid components of signals backwards in sequence and extracting average waveforms of signals and normalize the signal in a single cycle.(4) Extraction of characteristic information of pulse waves. This work selectively extracted the time-domain characteristic information of the pulse signal, specifically, using the eigenvalue method to extract six time-domain characteristic points of the pulse signal. Based on the summarization of other eigenvalue extraction algorithms, a modified differential algorithm was proposed. After analyzing the special significance of six characteristic points in waveforms, this algorithm made the characteristic point extraction more accurate and effective by combining AR model denoising and wavelet single-branch reconstruction algorithms with first-order and three-order differentials based on the disadvantage that the differential method is poorly robust for noises.(5) Judgment of distribution of chloasma faciei. This work proposed two methods to extract the human facial temperature distribution from infrared thermographs taken by an infrared imaging system. The temperature data extracted were saved in Excel in the order of original image array. An interface was designed to achieve the determination of the position of the temperature data within a temperature value or range to help us analyze and investigate the distribution of chloasma.(6) Analysis of the corresponding relation between the characteristic information of pulse signals and the pathological information. First, based on the patient information, statistics and classification were made of the characteristic information of pulse signals and the pathological data of two syndromes related to the pathology of qi and blood (chloasma and menoxenia), then characteristics of pulse signals, TCM syndrome types of related syndromes, and the correlation and corresponding relation of specific pathological characteristics were analyzed and summarized, and finally statistical tests were done and final research conclusions were made.