Format

Send to

Choose Destination
Neuroimage. 2015 Apr 15;110:11-21. doi: 10.1016/j.neuroimage.2015.01.028. Epub 2015 Jan 14.

A human post-mortem brain model for the standardization of multi-centre MRI studies.

Author information

1
Department of Neurology, University Medical Center of the Johannes Gutenberg University Mainz, Germany; Neuroimaging Center (NIC) of the Focus Program Translational Neuroscience (FTN), Johannes Gutenberg University, Mainz, Germany.
2
Department of Radiology and Nuclear Medicine, St. Josef Hospital, Ruhr-University Bochum, Bochum, Germany.
3
Institute for Neuropathology, Justus Liebig University, Giessen, Germany.
4
Institute of Microanatomy and Neurobiology, University Medical Center of the Johannes Gutenberg University Mainz, Germany.
5
Department of Medicine, Surgery and Neuroscience, University of Siena, Italy.
6
Department of Neurology, St. Josef Hospital, Ruhr-University of Bochum, Bochum, Germany.
7
Department of Radiology, Münster University Hospital, Münster, Germany.
8
Department of Neurology, Münster University Hospital, Münster, Germany.
9
Department of Neurology, University Medical Center of the Johannes Gutenberg University Mainz, Germany; Neuroimaging Center (NIC) of the Focus Program Translational Neuroscience (FTN), Johannes Gutenberg University, Mainz, Germany. Electronic address: frauke.zipp@unimedizin-mainz.de.
10
Brain Imaging Center (BIC), Goethe University, Frankfurt am Main, Germany.

Abstract

Multi-centre MRI studies of the brain are essential for enrolling large and diverse patient cohorts, as required for the investigation of heterogeneous neurological and psychiatric diseases. However, the multi-site comparison of standard MRI data sets that are weighted with respect to tissue parameters such as the relaxation times (T1, T2) and proton density (PD) may be problematic, as signal intensities and image contrasts depend on site-specific details such as the sequences used, imaging parameters, and sensitivity profiles of the radiofrequency (RF) coils. Water or gel phantoms are frequently used for long-term and/or inter-site quality assessment. However, these phantoms hardly mimic the structure, shape, size or tissue distribution of the human brain. The goals of this study were: (1) to validate the long-term stability of a human post-mortem brain phantom, performing quantitative mapping of T1, T2, and PD, and the magnetization transfer ratio (MTR) over a period of 18months; (2) to acquire and analyse data for this phantom and the brain of a healthy control (HC) in a multi-centre study for MRI protocol standardization in four centres, while conducting a voxel-wise as well as whole brain grey (GM) and white matter (WM) tissue volume comparison. MTR, T2, and the quotient of PD in WM and GM were stable in the post-mortem brain with no significant changes. T1 was found to decrease from 267/236ms (GM/WM) to 234/216ms between 5 and 17weeks post embedment, stabilizing during an 18-month period following the first scan at about 215/190ms. The volumetric measures, based on T1-weighted MP-RAGE images obtained at all participating centres, revealed inter- and intra-centre variations in the evaluated GM and WM volumes that displayed similar trends in both the post-mortem brain as well as the HC. At a confidence level of 95%, brain regions such as the brainstem, deep GM structures as well as boundaries between GM and WM tissues were found to be less reproducible than other brain regions in all participating centres. The results demonstrate that a post-mortem brain phantom may be used as a reliable tool for multi-centre MR studies.

KEYWORDS:

Multi-centre MRI studies; Post-mortem brain; Quantitative MRI; Standardization

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center