论文部分内容阅读
在脑成像数据分析中,基于稀疏表示的模式定位算法在群组水平分析中具有非常优秀的性能,但在单个数据集的情况下结果还不尽如人意.为此,文中在先前研究的基础上提出了一种改进算法,通过基于原始数据集生成多个派生数据集的方法,来改善算法在单个数据集分析中的不足.仿真结果表明改进后算法在性能上有显著的提高.文章随后将该改进算法应用于帕金森病异常功能连接模式定位分析之中,得到广泛分布于全脑的与该疾病相关的269个异常功能连接,由此对算法的有效性进行了验证,并可能有助于加强对与该疾病相关的病理生理机制的了解.
In brain imaging data analysis, the pattern localization algorithm based on sparse representation has very good performance in group level analysis, but the result is not satisfactory in the case of a single data set.Therefore, based on previous studies This paper proposes an improved algorithm to improve the deficiencies of the algorithm in a single data set by generating multiple derived data sets based on the original data set.The simulation results show that the improved algorithm has a significant performance improvement.Then the article then The improved algorithm was applied to the positioning analysis of abnormal functional connectivity mode of Parkinson’s disease, and 269 abnormal functional connections related to this disease were found in the whole brain, thus the validity of the algorithm was verified and there may be Help to strengthen understanding of the pathophysiology of the disease.