Optimizing CTP in Posterior Circulation Infarction (POCI): A Comprehensive Analysis of CTP Postprocessing Algorithms...

Area of Science:

• Neuroimaging
• Stroke Imaging
• Cerebrovascular Imaging

Background:

• CT Perfusion (CTP) software uses various mathematical techniques for clinical maps.
• These techniques lack validation for Posterior Circulation Infarction (POCI).
• Algorithm differences in anterior circulation stroke studies affect map accuracy.

Purpose of the Study:

• To examine the influence of CTP processing algorithms on diagnostic accuracy.
• To determine optimal tissue parameters and thresholds for acute POCI.

Main Methods:

• Analyzed data from 58 POCI patients in the INSPIRE registry.
• Generated CTP parametric maps using five algorithms: SVD, SVDd, Partial-deconvolution, Stroke-stenosis, and Maximum Slope.
• Utilized ROC analysis and linear regression for voxel-based and volume-based analyses.

Main Results:

• Partial-deconvolution with Mean Transit Time (MTT) was optimal for penumbra (AUC=0.73) and core (AUC=0.70).
• Optimal MTT thresholds were >165% for core and >180% for penumbra.
• Algorithm choice influenced accuracy and optimal parameters; MTT was the most consistent parameter.

Conclusions:

• CTP is diagnostically useful for POCI.
• Significant variations exist in optimal parameters and thresholds across different algorithms.
• Clinicians must be aware of the specific CTP algorithm used and exercise caution when comparing vendor outputs.