Dissertation/Thesis Abstract

Automatic Real-Time Targeting of Single-Voxel Magnetic Resonance Spectroscopy
by Storrs, Judd, Ph.D., University of Cincinnati, 2010, 118; 3432351
Abstract (Summary)

Magnetic resonance spectroscopy (MRS) is a non-invasive and non-destructive in vivo technique available on magnetic resonance imaging (MRI) scanners that is used to measure biochemical profiles from localized regions, or volumes-of-interest (VOIs), inside the body. A confounding factor for interpretation and analysis of MRS is spatial inconsistency in selection of VOIs for data collection, which may obscure biochemical alterations and reduce the statistical power of a study. Because VOI selection is performed manually by the MRI operator, consistency both between sessions and among subjects requires careful protocol design and experienced staff. Inter-subject anatomic variation, imprecise experimental protocols, and inter-operator variation contribute to VOI positioning error.

In this work, automatic targeting of VOIs using a standard anatomic atlas was hypothesized to improve spatial consistency for VOIs, both among subjects and between sessions. Subject anatomy is aligned to a template during acquisition of routine highresolution 3D anatomic imaging. Alignment is computed parallel to acquisition and completes prior to the end of the scan allowing immediate use of the template coordinate system for the next scan. Once aligned, preselected VOIs are transferred from the template for acquisition. Two real-time alignment techniques are compared. The first performs affine alignment of the subject to the ICBM452 template, and the second rigidly aligns subject anatomy between baseline and followup sessions.

The technique was developed using simulations based on archived data from 79 subjects randomly segregated into training (40 subjects for development) and testing groups (39 subjects for evaluation). The accuracy of real-time spatial normalization was evaluated as disagreement with SPM5-derived nonlinear normalization. Median disagreement within the brain was 1.9 mm (largest: 9.1 mm). For comparison, optimal affine alignment was computed directly from nonlinear SPM5 results and had a median disagreement of 1.7 mm (largest: 7.7 mm). Median inter-session (test-retest) disagreement was 0.4 mm (largest: 1.9 mm) for the ICBM452-based technique and 0.2 mm (largest: 0.7 mm) for the rigid-body technique. Comparable results on training and testing groups indicate good generalization of both techniques beyond the training group.

Automatic and manual prescription of MRS was compared for VOIs in the anterior cingulate gyrus (ACG) and left and right inferior frontal gyri (L-IFG and R-IFG). Automatic ICBM452-based selection of nonoblique VOIs improved inter-subject overlap by +19.6%, +29.6%, and +22.4% (ACG, L-IFG, and R-IFG), and inter-session overlap by +16.2%, +15.7%, and +11.3% compared to manual VOI selections. Automatic ICBM452- based prescription of oblique VOIs provided further improvements for inter-subject overlap (+1.0%, +1.4%, +1.1%) and inter-session overlap (+3.0%, +3.2%, +2.7%). Rigid-body coregistration further improved inter-session overlap compared to ICBM452-based normalization both for nonoblique (+3.0%, +3.2%, +2.7%) and oblique (+0.6%, +0.3%, +0.6%) VOIs. Quantitative comparisons of VOI tissue content demonstrated improved anatomic consistency for automatic prescriptions compared to manually selected VOIs.

The availability of rapid, atlas-consistent inter-subject alignment is expected to simplify experimental protocols while simultaneously improving study-wide consistency. These improvements are expected to increase statistical power for group comparisons, facilitate atlas-based research (including combined fMRI and MRS studies), and support the development of biomarkers.

Keywords: automatic prescription; brain mapping; magnetic resonance imaging (MRI); magnetic resonance spectroscopy (MRS); reproducibility; image registration.

Indexing (document details)
Advisor: Lee, Jing-Huei
Commitee: Ball, William S., Chu, Wen-Jang, Eliassen, James
School: University of Cincinnati
Department: Biomedical Engineering
School Location: United States -- Ohio
Source: DAI-B 72/02, Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Neurosciences, Biomedical engineering, Medical imaging
Keywords: Automatic prescription, Brain mapping, Image registration, Magnetic resonance imaging, Magnetic resonance spectroscopy, Reproducibility
Publication Number: 3432351
ISBN: 9781124360560
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest