Correlation-based perfusion mapping using time-resolved MR angiography: A feasibility study for patients with suspicions of steno-occlusive craniocervical arteries
- 117 Downloads
To explore the feasibility of using correlation-based time-delay (CTD) maps produced from time-resolved MR angiography (TRMRA) to diagnose perfusion abnormalities in patients suspected to have steno-occlusive lesions in the craniocervical arteries.
Materials and methods
Twenty-seven patients who were suspected to have steno-occlusive lesions in the craniocervical arteries underwent both TRMRA and brain single-photon emission computed tomography (SPECT). TRMRA was performed on the supra-aortic area after intravenous injection of a 0.03 mmol/kg gadolinium-based contrast agent. Time-to-peak (TTP) maps and CTD maps of the brain were automatically generated from TRMRA data, and their quality was assessed. Detection of perfusion abnormalities was compared between CTD maps and the time-series maximal intensity projection (MIP) images from TRMRA and TTP maps. Correlation coefficients between quantitative changes in SPECT and parametric maps for the abnormal perfusion areas were calculated.
The CTD maps were of significantly superior quality than TTP maps (p < 0.01). For perfusion abnormality detection, CTD maps (kappa 0.84, 95% confidence interval [CI] 0.67-1.00) showed better agreement with SPECT than TTP maps (0.66, 0.46-0.85). For perfusion deficit detection, CTD maps showed higher accuracy (85.2%, 95% CI 66.3-95.8) than MIP images (66.7%, 46-83.5), with marginal significance (p = 0.07). In abnormal perfusion areas, correlation coefficients between SPECT and CTD (r = 0.74, 95% CI 0.34-0.91) were higher than those between SPECT and TTP (r = 0.66, 0.20-0.88).
CTD maps generated from TRMRA were of high quality and offered good diagnostic performance for detecting perfusion abnormalities associated with steno-occlusive arterial lesions in the craniocervical area.
• Generation of perfusion parametric maps from time-resolved MR angiography is clinically useful.
• Correlation-based delay maps can be used to detect perfusion abnormalities associated with steno-occlusive craniocervical arteries.
• Estimation of correlation-based delay is robust for low signal-to-noise 4D MR data.
KeywordsMagnetic resonance angiography Perfusion Stenosis Computer-assisted image processing Comparative study
Alberta Stroke Program Early CT score
Correlation-based time delay
Digital subtraction cerebral angiography
Gadolinium-based contrast agent
High-resolution contrast-enhanced MR angiography
Internal carotid artery
Middle cerebral artery
Maximal intensity projection
Region of interest
Time-resolved, multiphasic MR angiography
Time to peak
Time-resolved angiography with stochastic trajectories
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2017R1D1A1B03033829).
Compliance with ethical standards
The scientific guarantor of this publication is Bum-soo Kim.
Conflict of interest
The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
Statistics and biometry
No complex statistical methods were necessary for this paper.
Written informed consent was waived by the Institutional Review Board.
Institutional Review Board approval was obtained.
• cross-sectional study
• performed at one institution
- 20.NamY, Jang J, Lee S, Kim B, AhnMI (2016) Automated extraction of arterial and venous function from time-resolved multiphasic MR angiography. Proc Int Soc Magn Reson Med Sci Meet Exhib Int Soc Magn Reson Med Sci Meet Exhib:1423Google Scholar