Low-dose CT Perfusion with Sparse-view Filtered Back Projection in Acute Ischemic Stroke

Area of Science:

• Medical Imaging
• Radiology
• Neuroimaging

Background:

• Computed tomography perfusion (CTP) is valuable for quantifying ischemic stroke volume but involves significant radiation dose.
• Sparse-view CT (sCT) reduces dose by acquiring fewer projections, but artifacts can affect image quality.

Purpose of the Study:

• To evaluate the feasibility of dose reduction in CTP using sparse-view acquisition with filtered back projection (FBP).
• To assess the impact of sparse-view CTP on the accuracy of ischemic lesion volume measurements.

Main Methods:

• Simulated full-view (984 views) and sparse-view (492, 328, 246, 164 views) CTP datasets from 38 acute ischemic stroke patients.
• Generated cerebral blood flow (CBF) and time-to-maximum (Tmax) maps using deconvolution.
• Measured ischemic volumes based on CBF < 30% and Tmax > 6 s, comparing sparse-view results to full-view reference using statistical tests and Bland-Altman analysis.

Main Results:

• Significant differences in lesion volumes were observed for 164-view CBF and 246/164-view Tmax compared to full-view (p < 0.001).
• Mean volume differences ranged from -1.0 ± 2.8 ml to -12.5 ± 19.9 ml across different sparse-view settings and parameters.
• Despite some differences, the study explored the potential for significant dose reduction.

Conclusions:

• Sparse-view CTP with FBP shows potential for substantial radiation dose reduction, up to a factor of 3.
• Careful selection of sparse-view parameters is necessary to maintain acceptable accuracy in ischemic volume estimation.
• This approach could make CTP more widely applicable in acute ischemic stroke management.