文献类型：会议论文

英文题名：Detecting Change Points in fMRI Data via Bayesian Inference and Genetic Algorithm Model

作者：Xiao, Xiuchun[1,3];Liu, Bing[2];Zhang, Jing[2];Xiao, Xueli[3];Pan, Yi[3,4]

机构：[1]Guangdong Ocean Univ, Coll Elect & Informat Engn, Zhanjiang 524025, Peoples R China;[2]Georgia State Univ, Dept Math & Stat, Atlanta, GA 30303 USA;[3]Georgia State Univ, Dept Comp Sci, Atlanta, GA 30303 USA;[4]Georgia State Univ, Dept Biol, Atlanta, GA 30303 USA

会议论文集：13th International Symposium on Bioinformatics Research and Applications (ISBRA)

会议日期：MAY 29-JUN 02, 2017

会议地点：Honolulu, HI

语种：英文

外文关键词：fMRI; Change point detection; Genetic algorithm; Bayesian inference

外文摘要：Dynamic functional connectivity detection in fMRI has been recently proved to be powerful for exploring brain conditions, and a variety of methods have been proposed. This paper mainly investigates the field of change point detection based on Bayesian inference and genetic algorithm. We define different indicator vectors as different individuals, which represent some possible change point distributions, and use Bayesian posterior probability to evaluate their fitness. Accordingly, we also present an improved genetic algorithm, which is applied to evolve the individuals toward the best one. Then, the most possible change points distribution could be resolved. The method has been applied to several synthesized data and simulation results reveal that the proposed method can detect change points in fMRI datasets with higher precision and lower time consumption.