Post-processing noise elimination method for performing ICA analysis of plural f MRI data
A data and complex technology, applied in the field of ICA analysis of complex fMRI data, can solve problems such as brain information loss and data discarding, and achieve the effect of solving data discarding and brain information loss
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment 1
[0016] Specific embodiment 1: Assume that the multiple fMRI data of a single subject collected under the existing exercise stimulation, the number of scans of the whole brain of the subject is 165. ICA is performed on the complex fMRI data to obtain task-related spatially activated brain area components containing a large amount of high-amplitude noise. The specific steps for post-processing and denoising are as follows: figure 1 shown.
[0017] In the first step, the number of independent components of the complex fMRI data of a single subject was estimated according to the information theory criterion, and the result was 30. The PCA (principle component analysis) method was used to reduce the complex fMRI data from 165 dimensions to 30 dimensions.
[0018] The second step is to use the complex EBM (entropy bound minimization) algorithm to perform ICA on the complex fMRI data after PCA dimensionality reduction, and obtain 30 complex ICA estimation components, from which the t...
specific Embodiment 2
[0024]Specific Example 2: For the complex fMRI data of 16 subjects collected under the existing exercise stimulation, the number of scans of the whole brain of each subject is 165. ICA was performed on the complex fMRI data of 16 subjects respectively, and the specific steps to obtain the group average task-related spatial activation brain area components of the final denoising were as follows: figure 2 shown.
[0025] In the first step, PCA was used to reduce the complex fMRI data of 16 subjects from 165 dimensions to 30 dimensions.
[0026] The second step is to use the complex EBM algorithm to perform ICA on the complex fMRI data of 16 subjects after PCA dimensionality reduction, and select the task-related spatial activation brain area components of each subject These components contain a large number of high-amplitude noise voxels.
[0027] In the third step, the phase correction method based on the time course component is selected for the Perform phase correction ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


