Font Size: a A A

Study On Human Brain Default Mode Network Based On Independent Component Analysis Using Functional Magnetic Resonance Imaging

Posted on:2008-06-05Degree:MasterType:Thesis
Country:ChinaCandidate:Z J LiuFull Text:PDF
GTID:2178360272977026Subject:Biomedical engineering
Abstract/Summary:PDF Full Text Request
Independent component analysis (ICA) has been widely used in functional magnetic resonance imaging (fMRI) data processing field in recent years. Usually there are two ways ICA applicated in fMRI data processing: spatial independent component analysis (sICA) and temporal independent component analysis (tICA). Due to the spatial dimension is far larger than temporal dimension, sICA plays a dominant role in fMRI data processing. If we need to analysis a group of fMRI data, Group ICA becomes an effective method, which can relief the burden of computation and obtain a statistical result.Human brain default mode network is a special functional network, which is responsible for monitoring the environment both inside and outside our body, keeping us conscious and so on. Epilepsy is a harmful nervous system disease. Sustaining epilepsy episode may change the default mode network of human brain. Therefore the study on epileptic default mode network using fMRI will be of benefit to the clinical diagnosis and treament of epilepsy.In this paper, the models of both sICA and Group ICA are deeply studied. Then a group of fMRI data obtained on a visual paradigm is processed using Group ICA methods and the performance is evaluated. Finally a method based on ICA is introduced to study the human brain default mode network. The result shows that this method is effective. The details are shown as follows:(1) The sICA model of fMRI data is in depth studied. Mixed data is created and then processed using sICA. The result is evaluated in the aspects of time accuracy and shows that sICA can effectively separate the mixed data into several spatial independent components.(2) Three Group ICA methods are studied and two of them are used to deal with a group of fMRI data obtained on a visual paradigm. The results are evaluated in the metric of time accuracy. At last, the application range of each method is proposed according to the respective mathematical model.(3) A method based on ICA is proposed to obtain the human brain default mode network. The default mode network of twelve TLE patients and twelve healthy controls is isolated using the proposed method. An intra-group analysis is performed to get a statistical map of each group and an inter-group analysis is performed to find the differences between two groups. The results show that the method is capable of detecting the default mode network of human brain and the TLE's default mode network represents higher activation than healthy controls'. The results will be of benefit to the clinical diagnosis and treament of epilepsy.
Keywords/Search Tags:Independent Component Analysis (ICA), functional Magnetic Resonance Imaging (fMRI), Group ICA, Default Mode Network, Temporal Lobe Epileptic (TLE)
PDF Full Text Request
Related items