基于高阶最小生成树的脑网络分析及对阿兹海默氏症患者的分类

发布时间：2018-06-07 03:47

本文选题：脑网络 + 最小生成树　；参考：《计算机应用》2017年11期

【摘要】：利用静息态功能磁共振成像技术来研究大脑的功能连接网络是当前脑疾病研究的重要方法之一。这种方法能准确地检测包括阿兹海默氏症在内的多种脑疾病。然而,传统的网络只是研究两个脑区之间相关程度,而且缺乏对大脑区域之间更深层次的交互信息和功能连接之间关联程度的研究。为了解决这些问题,提出了一种构建高阶最小生成树功能连接网络的方法,该方法不仅保证了功能连接网络的生理学意义,而且研究了网络中更复杂的交互信息,提高了分类的准确率。分类结果显示,基于高阶最小生成树功能连接网络的静息态功能磁共振成像分类方法大幅提高了阿兹海默氏症检测的准确率。
[Abstract]:Resting functional magnetic resonance imaging (fMRI) is one of the most important methods in the study of brain diseases. This method can accurately detect various brain diseases, including Alzheimer's disease. However, the traditional network only studies the degree of correlation between two brain regions, and there is a lack of research on the degree of correlation between deeper levels of interactive information and functional connections between brain regions. In order to solve these problems, this paper proposes a method to construct the high order minimum spanning tree functional connection network. This method not only guarantees the physiological significance of the functional connection network, but also studies the more complex interactive information in the network. The accuracy of classification is improved. The classification results show that the classification of resting functional magnetic resonance imaging (fMRI) based on high-order minimum spanning tree functional connection network greatly improves the accuracy of Alzheimer's disease detection.
【作者单位】：太原理工大学计算机科学与技术学院;
【基金】：国家自然科学基金资助项目(61373101,61472270,61402318,61672374) 山西省科技厅应用基础研究项目青年面上项目(201601D021073) 山西省教育厅高等学校科技创新研究项目(2016139)~~
【分类号】：R749.16;TP181

【相似文献】