Magnetic resonance imaging (MRI) is a very versatile imaging modality which can be used to acquire several different types of images. Some examples include anatomical images, images showing local brain activation and images depicting different types of pathologies. Brain activation is detected by means of functional magnetic resonance imaging (fMRI). This thesis presents two methods for adaptive spatial filtering of fMRI data.
Contents
1 Introduction
1.1 Detection of brain activity
1.2 Separation of fat and water
1.3 How to read this thesis
1.4 Contributions
1.5 Publications
1.6 Notation
1.7 Abbreviations
2 Magnetic resonance
2.1 Properties of elementary particles
2.2 Magnetism and spin
2.3 Radio-frequency pulses
2.4 Relaxation
2.5 Magnetic resonance spectroscopy
3 Magnetic resonance imaging
3.1 Spatial encoding
3.2 Slice selection
3.3 Pulse sequences
3.4 Image contrast
3.5 Further reading
4 Functional magnetic resonance imaging
4.1 Neural activity and blood oxygenation
4.2 Paradigms
4.3 BOLD models
4.4 Detecting BOLD-like signals
4.5 Sensitivity and specificity
4.6 Exploratory analysis
4.7 Evaluation of analysis methods
4.7.1 Receiver operating characteristics curves
4.8 Pre-processing of data
4.8.1 Registration
4.8.2 Detrending
4.9 Visualization
5 Analysis based on canonical correlation analysis
5.1 Canonical correlation analysis
5.2 CCA in fMRI data analysis
5.2.1 Filter kernels
5.2.2 Restricted CCA
5.2.3 Analysis
5.3 Rotational invariance
5.3.1 Isotropic filtering
5.3.2 Anisotropic filtering
5.3.3 Experiments
6 Analysis based on bilateral filtering
6.1 Introduction
6.2 Bilateral filtering
6.3 Method
6.3.1 Measuring time sequence similarities
6.3.2 Anatomical similarities
6.3.3 Combining signal and anatomical similarities
6.4 Experiments
6.4.1 Two-dimensional data analysis
6.4.2 Three-dimensional data analysis
6.4.3 Data without sharp edges
6.5 Discussion
7 Robust correlation estimation
7.1 Theory
7.1.1 Correlation, GLM and CCA
7.1.2 Weighted correlation
7.2 Method
7.3 Application to analysis of functional MRI data
7.4 Application to detection of partially occluded objects
7.5 Discussion
8 Phase sensitive image reconstruction
8.1 Obesity, a big problem
8.2 Imaging of adipose tissue
8.2.1 T1 weighted imaging
8.2.2 Dixon imaging
8.3 Phase estimation and correction
8.3.1 Phase estimation using region growing
8.3.2 Phase estimation using the inverse gradient
9 Review of papers
9.1 Paper I: On rotational invariance in adaptive spatial filtering offMRI data
9.2 Paper II: Signal and anatomical constraints in adaptive filtering offMRI data
9.3 Paper III: Robust correlation analysis with an application to func-tional MRI
9.4 Paper IV: Phase sensitive reconstruction for water/fat separation in MR imaging using inverse gradient
10 Discussion
10.1 Adaptive spatial filtering of fMRI data
10.2 Robust correlation estimation
10.3 Phase sensitive image reconstruction
Author: Rydell, Joakim
Source: Linköping University
Download URL 2: Visit Now