【作者】 赵静；

【导师】 李刚；

【作者基本信息】 天津大学 ， 精密仪器及机械， 2011， 博士

【摘要】 舌诊作为中医望诊的一部分,对于诊断疾病起着非常重要的作用。随着科学技术的发展,舌诊客观化受到越来越多的关注,然而近几十年来进展较小,主要原因是现有舌诊客观化方法以舌色和纹理为主要研究内容,然而同色异谱现象说明了三原色来表征舌体生理病理信息存在严重的不足,所以,要真正实现中医舌诊的客观化,本文认为:既要以中医基本特征——系统、辨证为基础,又要从现代科学的研究方法出发,挖掘舌体的更多信息。光谱法以其快速、简便、无创的优点在生物医学领域得到了广泛的应用,基于此,本研究将光谱和高光谱技术运用到舌诊疾病快速筛查中来,与单纯的提取舌色相比,光谱法能更全面的反映人舌体所携带的生理和病理信息,在中医舌诊客观化研究的方法论上向前迈进了一步。本文的主要研究内容包括:1.突破现有舌诊客观化研究简单、机械模仿中医感官诊断的惯性,在中医系统、辨证的基础上,首次将光谱法结合数据挖掘和黑箱理论运用到舌诊疾病的快速筛查中来,通过对采集来的样本进行数据建模和预测,验证了光谱法用于舌诊疾病快速筛查的可行性,同时也说明该方法有可能为临床提供一种简便、快速的疾病筛查手段。2.同时采集同一被测舌体信息的可见、近红外两个波段光谱信息来探究不同波段的选取对于建模效果的影响。对现有数据进行建模分析发现,与可见光波段相比,近红外波段用来建立疾病快速筛查模型可以得到更高的预测准确度。3.为了解决光谱采集受环境光干扰严重的问题,本研究首次将光调制技术用于提高光谱法舌诊测量的信噪比。系统采用调制盘对参考光信号强度进行调制,并进行了模拟实验,实验结果表明光谱法舌诊研究中运用光调制技术可以很好的抑制环境杂散光的干扰,为高信噪比舌诊光谱数据采集提供了一种新途径。4.为了获取更多舌体信息,将高光谱技术用到舌诊的研究中来,鉴于现有高光谱仪价格昂贵,本文提出了一种基于X-Y振镜扫描式舌诊高光谱采集系统。通过进行模拟实验,验证了该系统可以同时得到被测对象的高光谱图谱信息,而且对于被测对象表面信息能够较准确的进行测定,该X-Y振镜扫描式高光谱采集系统可以用于舌体高光谱信息的采集,为舌体信息获取提供了一种新途径。更多还原

【Abstract】 Tongue inspection as a part of traditional Chinese medicine plays an important role in diagnosis of disease. With the development of the technology, tongue characterization received wide attention, but has made little progress in recently few decades. The main reason is that these researches which were limited to tongue color measurement didn’t analysis tongue information as a whole. In order to solve the shortage of the above research methods, this paper argues that: the research that can make tongue characterization come true should not only from the basis features of the traditional Chinese medicine, but also make the techonolgy as a basic tool to develop more tongue information. A method based on spectroscopy and hyperspectral for tongue inspection on rapid screening disease was present in this research. Comparing with former methods, spectroscopy can reflect more physiological and pathological information of tongue, and a forward step can be made for tongue characterization on the metholology.The main contents of this paper are as follows:(1) Breakthrough the inertia that simple and mechanical imitation of the traditional Chinese medicine diagnosis in the current tongue characterization, and the visible-near infrared spectroscopy combined with black box theory was used to diagnosis disease of tongue inspection for the first time on the basis features of traditional Chinese medicine. A mathematic prediction model was built in this research. The experiment results showed that application of the spectroscopy for disease diagnosis is promising, and this method may provide a new way for clinical disease diagnosis.(2) In order to discover which band was more relevant to the results of model, two spectrometers were used in this research. After built the model and analysis the existing data, we can conclude that the near-infared band has the higher correlation with the disease classification than the visible band for the existing data.(3) In order to solve the problem that the collected spectrum was influenced seriously by the environment light, the optical modulation was used to improve the signal to noise ratio of spectroscopy-tongue inspection for the first time. Reference light was modulated by reticle in this system, and an experiment was made to test the ability of this method. The experiment result showed that the optical modulation can remove the influence of the environment light, and this method can provide a new way to improve the signal to noise ratio of spectroscopy for tongue inspection.(4) Hyperspectral technology was used for tongue inspection in this work to get more information of tongue. A hyperspectral acquisition system based on X-Y scanning galvanometer was also proposed due to the expensive of current equipment. An experiment was made to test the ability of this system. Simulation results showed that X-Y galvanomer scanning system can obtain the hyperspectral data and can correctly obtain surface information of the measured object, and the results also showed that it can be used to collect the hyperspectral data of tongue and provide a new way for tongue information collection.更多还原