本文關(guān)鍵詞： 模式定位 稀疏表示 多變量模式分析 功能連接　出處：《控制理論與應(yīng)用》2017年06期 　論文類型：期刊論文

【摘要】：在腦成像數(shù)據(jù)分析中,基于稀疏表示的模式定位算法在群組水平分析中具有非常優(yōu)秀的性能,但在單個(gè)數(shù)據(jù)集的情況下結(jié)果還不盡如人意.為此,文中在先前研究的基礎(chǔ)上提出了一種改進(jìn)算法,通過基于原始數(shù)據(jù)集生成多個(gè)派生數(shù)據(jù)集的方法,來改善算法在單個(gè)數(shù)據(jù)集分析中的不足.仿真結(jié)果表明改進(jìn)后算法在性能上有顯著的提高.文章隨后將該改進(jìn)算法應(yīng)用于帕金森病異常功能連接模式定位分析之中,得到廣泛分布于全腦的與該疾病相關(guān)的269個(gè)異常功能連接,由此對(duì)算法的有效性進(jìn)行了驗(yàn)證,并可能有助于加強(qiáng)對(duì)與該疾病相關(guān)的病理生理機(jī)制的了解.
[Abstract]:In the analysis of brain imaging data, the pattern location algorithm based on sparse representation has excellent performance in group level analysis, but the result is not satisfactory in the case of a single data set. Based on the previous research, an improved algorithm is proposed to generate multiple derived datasets based on the original data set. The simulation results show that the performance of the improved algorithm is significantly improved. Then, the improved algorithm is applied to the localization analysis of abnormal functional connection mode for Parkinson's disease. A total of 269 abnormal functional connections related to the disease were obtained, which proved the effectiveness of the algorithm and may contribute to better understanding of the pathophysiological mechanisms associated with the disease.
【作者單位】： 華南理工大學(xué)自動(dòng)化科學(xué)與工程學(xué)院;
【基金】：Supported by National Natural Science Foundation of China(91120305) Guangdong Natural Science Foundation(2014A030312005)
【分類號(hào)】：R742.5;TP391.41

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前2條

1 李軍輝;基于字典學(xué)習(xí)和稀疏表示的癲癇檢測(cè)[D];山東大學(xué);2016年

2 陳珊珊;基于稀疏表示和特征提取的癲癇腦電分類識(shí)別方法研究[D];濟(jì)南大學(xué);2016年

，

本文編號(hào)：1538808