Finding the optimal deconvolution algorithm for MR perfusion in carotid stenosis: Correlations with angiographic
Chung-Jung Lin1, Sheng-Che Hung1, Feng-Chi Chang1
1Department of radiology, Taipei Veterans general hospital, 201, Shipai Road, Section 2, 11217 Taipei, Taiwan; School of medicine, National Yang Ming university, Taipei, Taiwan.
Journal of Neuroradiology = Journal De Neuroradiologie
|April 4, 2016
Summary
Different deconvolution algorithms impact Magnetic Resonance Perfusion (MRP) and Cerebral Circulation Time (CCT) correlations. Standard singular value decomposition showed the strongest correlation between CCT and Tmax, MTT, and CBF in carotid stenosis patients.
Area of Science:
- Neurology
- Radiology
- Medical Imaging
Background:
- Cerebral blood flow (CBF) parameters are crucial for assessing cerebrovascular diseases.
- Magnetic Resonance Perfusion (MRP) and angiography cerebral circulation time (CCT) are key imaging modalities.
- Deconvolution algorithms are essential for accurate MRP data analysis.
Purpose of the Study:
- To evaluate the influence of various deconvolution algorithms on the correlation between MRP-derived parameters (CBF, MTT, CBV, TTP, Tmax) and CCT.
- To compare the performance of standard singular value decomposition, block-circulant, and delay-corrected algorithms.
Main Methods:
- Retrospective analysis of 30 patients with unilateral carotid stenosis undergoing pre-stenting MRP and angiography.
- MRP acquired on a 1.5-T MR scanner; angiography using a bi-plane flat-detector system with a 12mL contrast bolus.
- Deconvolution methods included standard singular value decomposition, block-circulant, and two delay-corrected algorithms.
- CCT calculated as the time difference between cavernous ICA and parietal vein peak times.
Main Results:
- Standard singular value decomposition yielded the highest correlation between CCT and Tmax (r=0.65), followed by MTT (r=0.60) and CBF (r=-0.57).
- A significant correlation was observed between CCT and TTP (r=0.33).
- No significant correlation was found between CCT and CBV across all tested algorithms.
Conclusions:
- MRP with singular value decomposition effectively quantifies cerebral blood flow deficits in steno-occlusive disease within the angio-room.
- This integrated approach offers immediate hemodynamic assessment, potentially enhancing patient safety by avoiding additional radiation and iodine contrast.
- The findings support the use of SVD-based MRP for real-time hemodynamic evaluation in interventional suites.
