【作者】 胡萌；

【导师】 唐孝威； 李光；

【作者基本信息】 浙江大学 ， 凝聚态物理， 2008， 博士

【摘要】 大脑作为控制中枢是人体最复杂的器官。目前对大脑功能的研究还处于起步阶段。作为研究大脑工作机制的新兴交叉学科,神经信息学将神经科学和信息科学结合起来,利用现代化信息工具来研究大脑的神经处理机制。脑电分析恰好是一种使用现代化信息工具来研究大脑电活动的手段。目前,脑电研究因为其高效率,低成本,无损害,成为了众多研究手段中不可缺少的一员。本文通过研究人在不同的状态下(模拟高空缺氧,针刺足三里穴,单次电流刺激)的脑电变化,研究大脑在负荷状态、受激状态下的反应。为了研究高空缺氧对大脑的影响,通过面罩给予试被不同氧气含量的混合气体来模拟不同的高空高度。分别记录正常和高空缺氧时的脑电。结合线性和非线性方法研究正常状态和缺氧状态下脑电的区别,同时利用神经行为测试来辅证脑电分析的结果。结果显示,通过脑电可以客观、定量地区分正常状态和缺氧状态。为了探讨针灸治疗的机理,按照临床疗程给被试足三里穴进行针灸。利用脑电记录针灸的过程并研究期间脑电的变化趋势。将脑电非线性分析结合大脑的皮层分区理论,尝试解释针灸治疗的机理。结果显示,针刺足三里会引起大脑相关脑区电活动的改变,为解释针灸的机理提供了又一证据。为提出一种更有效的神经功能测量方法,给被试单次电流刺激并记录脑电。采用特征提取和模式识别方法在单次刺激的脑电中识别事件相关电位。结果显示单次电流刺激的脑电可以识别出事件相关电位。本研究可以克服传统的神经感觉功能检测方法受被试主观性影响较大的缺点。同时,在识别事件相关电位时,放弃多次叠加的传统方法,避免多次测量引起的神经疲劳。本次研究有希望发展成一种客观的,快速的神经感觉功能测试方法。综上所述,通过脑电可以识别大脑在负荷状态、受激状态下的变化。脑电分析为神经信息学研究大脑神经处理机制提供了有力的手段。更多还原

【Abstract】 Brain is the most complex organ in a human body. Nowadays, the research on brain is still at a preliminary stage. As a rising interdisciplinary subject, neuroinformatics not only applies modern methods of informatics to neurology, but also studies the information processing mechanism of neurological systems. For neuroinformatic study, electroencephalography (EEG) analysis becomes one of the most important research methods due to its safety, low cost and high temporal resolution.In this paper, the EEG responses to different conditions (simulated high altitude hypoxia, acupuncture at Zusanli acupoint, electrical current stimulation) are studied.Via a pilot mask, a man breathing at different altitudes are simulated by adjusting the oxygen proportion. Normal and hypoxia EEGs are recorded, while the hypoxia is evaluated via neurobehavioral tests. Both linear and nonlinear algorithms are used to recognize hypoxia EEG from normal ones, The experimental results illustrate that the normal and hypoxia status can be quantificationally classified according to EEG.Acupunture at Zusanli acupoint is carried out according to the clinical therapy procedure. EEG during whole procedure is recorded. Based on the theory of functional regions on brain, the mechanism of acupuncture is analyzed using nonlinear algorithms. The results show that the acupuncture at Zusanli acupoint causes the activity changes on the relative functional regions of the brain.A subject receives electrical current stimuli while the EEG is recorded. The method to recognize the event relative potential (ERP) according to a single current stimulation trial is studied. Support vector machine (SVM) is used for EEG pattern recognition. In comparison with conventional methods, this algorithm can be used to test the sensory threshold objectively. Futhermore, due to only single stimulus needed for ERP recognition, the fatigue of nerve induced by repetitious stimuli is avoided. This method is promising to develop an objective and rapid method for quantitative sensory testing.Our experimental results show that EEG can be used to indicate the statuses of brain under different conditions and provide an efficient tool for neuroinformatics research.更多还原