Method for analyzing functional MRI brain images

Patent metadata

Field	Value
Publication number	US-9633430-B2
Application number	US-201114374853-A
Country	US
Kind code	B2
Filing date	Dec 28, 2011
Priority date	Dec 28, 2011
Publication date	Apr 25, 2017
Grant date	Apr 25, 2017

Abstract

A method for analyzing fMRI brain data, comprising: collecting the fMRI data including spatial information and temporal information from subjects; preprocessing the fMRI data; computing independent components (ICs) and their corresponding time course for each individual subjects; constructing an initial functional connectivity pattern; constructing a classifier based on the functional connectivity pattern; and applying the classifier to functional connectivity patterns of individual subjects for statistical analysis or diagnosis. The method may be used in fMRI based studies of a brain function and brain disorder diagnosis.

First claim

What is claimed is:

1. A method for analyzing brain functional magnetic resonance imaging (fMRI) data, comprising steps of:
collecting the fMRI data including spatial information and temporal information from subjects;
preprocessing the fMRI data;
computing independent components (ICs) and their corresponding time course for each individual subjects;
constructing an initial functional connectivity pattern;
constructing a classifier based on the functional connectivity pattern; and
applying the classifier to functional connectivity patterns of individual subjects for statistical analysis or diagnosis,
wherein the functional connectivity pattern includes a spatial functional connectivity pattern and a corresponding temporal functional connectivity pattern,
wherein the corresponding temporal functional connectivity pattern includes temporal functional connectivity patterns for one frequency, or temporal functional connectivity patterns for different frequencies, and
wherein the temporal functional connectivity pattern for one frequency is a vector consisting of elements of correlation coefficients between the temporal functional connectivity patterns for one frequency, or between the temporal functional connectivity patterns for different frequencies.

2. The method of claim 1 , wherein the step of preprocessing comprises:
re-sampling time points so that the time points for collecting all of pixels of a function image corresponding to one time point are the same substantially;
removing translation and rotation between the functional images collected at different time points due to a head movement by a rigid transformation; and
applying a spatial normalization to a normal template.

3. The method of claim 1 , wherein the step of computing independent components and their corresponding time course for each individual subjects comprises:
concatenating the fMRI data from all subjects temporally to form a large 4D fMRI image;
performing an independent component analysis (ICA) the concatenated image to generate group independent components; and
mapping the group independent components to each individual subject by back-reconstruction to obtain subject specific independent components.

4. The method of claim 1 , wherein the time courses are band pass filtered to obtain temporal signals at different frequencies.

5. The method of claim 1 , wherein the step of constructing an initial functional connectivity pattern comprise determining portion of the independent components based on a prior knowledge.

6. The method of claim 1 , wherein the initial functional connectivity pattern is constructed as an empty set.

7. The method of claim 1 , further comprising measuring a similarity or distance between different spatial functional connectivity patterns using a Riemannian distance metric.

8. The method of claim 1 , wherein a similarity or a distance between two spatial functional connectivity patterns is described using a principal angle based Riemannian distance metric.

9. The method of claim 1 , wherein the similarity or distance between different temporal functional connectivity patterns is measured using Euclidean metrics.

Assignees

• Fan Yong
• Inst Automation Cas

Inventors

• Fan Yong

Classifications

• G06T2207/30016

  Brain · CPC title

• G01R33/4806

  Functional imaging of brain activation · CPC title

• A61B2576/026

  for the brain · CPC title

• G16H30/40

  for processing medical images, e.g. editing · CPC title

• G06T2207/20081

  Training; Learning · CPC title