Endre Grøvik1, Kyrre Eeg Emblem1, Ingrid Digernes1, Line Brennhaug Nilsen1, Cornelius Eichner2, Kourosh Jafari 2, Thomas Witzel2, Behroze Vachha2, Elizabeth Gerstner2, Jayashree Kalpathy-Cramer2, Kawin Setsompop 2, and Steven Stufflebeam 2
1Department for Diagnostic Physics, Oslo University Hospital, Oslo, Norway, 2Athinoula A. Martinos Center for Biomedical Imaging, Harvard Medical School, Boston, MA, United States
Synopsis
In recent years, Vessel Architectural Imaging (VAI) has emerged as a
promising tool in tumor diagnosis to reveal unique MRI-based information on vessel
architecture, hemodynamic efficacy and metabolic activity. Healthy control data
may further advance our knowledge on the VAI method and its underlying
mechanisms, as well as serve as study controls. Here we propose a set of healthy-tissue
template maps of all VAI derived parameters which may act as a toolbox to
identify anomalies of various vascular brain diseases and ultimately help
improve diagnostic and outcome assessment in clinical settings.
INTRODUCTION
Vessel Architectural Imaging (VAI) has
recently been introduced as a new paradigm in tumor diagnosis (1). The VAI
technique exploits a temporal shift in the vessel-size sensitive MRI signals to
revel new information on vessel calibers, architecture, hemodynamic efficacy
and metabolic activity (2). Here, we propose a set of templates for VAI
parameters derived from MRIs of volunteers, thus serving as a potential control
to identify anomalies of vascular brain diseases.
MATERIALS AND METHODS
In this IRB-approved study, ten healthy
volunteers underwent contrast-enhanced MRI (Skyra 3T, Siemens) including
gradient-echo spin-echo dynamic susceptibility contrast (DSC) imaging with simultaneous-multiple-slice
(SMS) acceleration (3). Parametric maps of mean vessel density, vessel size
index, oxygenation status (∆SO2), and voxel-wise dominance of arterioles and venules was estimated by
VAI as previously described (1) and used as proof-of-concept hereafter. The resulting parametrical maps were
spatially normalized in MNI-space using the
SPM software package. For all normalized maps, template maps were generated by
estimating the mean of all 10 healthy subjects. In addition, one glioblastoma patient
was included as a case example for comparison. RESULTS
This proof-of-concept study show the initial
framework of healthy-tissue template maps of vascular function in the brain. A
total of four template maps were generated based on the healthy subjects, as
shown in Figure 1. This figure also shows the comparison to a glioblastoma patient,
in which substantial differences in vascular function are shown, thus
demonstrating the potential clinical value. CONCLUSION
We propose a set of template maps of healthy
vascular function of the brain that may serve as a toolbox for early identification
of anomalies of vascular brain disease, with the ultimate goal of improving the
treatment decision making.Acknowledgements
No acknowledgement found.References
1. Emblem KE et al. Nat
Med. 2013;19(9):1178-1183 (doi: 10.1038/nm.3289)
2. Digernes I et al.
J Cereb Blood Flow Metab. 2017;37(6):2237-2248 (doi: 10.1177/0271678X17694187)
3. Eichner C et al.
Magn Reson Med. 2014;72(3):770-778 (doi: 10.1002/mrm.24960)