A Comparative Study of CT Perfusion Postprocessing Tools in Medium/Distal Vessel Occlusion Stroke

Area of Science:

• Neurology
• Radiology
• Medical Imaging

Background:

• Computed tomography perfusion (CTP) is validated for large vessel occlusion stroke.
• Its accuracy for medium or distal vessel occlusions (MDVO) is less established.
• Discrepancies in CTP postprocessing tools may affect MDVO stroke management.

Purpose of the Study:

• To evaluate discrepancies between two widely used CTP postprocessing tools.
• To assess the agreement in lesion volume and spatial distribution for MDVO stroke.
• To compare penumbra volumes calculated by different CTP software.

Main Methods:

• Retrospective single-center cohort study of acute isolated MDVO stroke patients.
• Comparison of CTP data processed by Syngo.via and RapidAI.
• Segmentation and analysis of perfusion lesions, infarct core, and penumbra volumes.
• Statistical comparison of volume and spatial distribution agreement.

Main Results:

• Agreement criteria met in less than 30% of patients.
• Disagreement on tissue-at-risk volume and distribution in up to 79.6%.
• Infarct core presence and volume disagreed in up to 86% of cases.
• Significant differences in penumbra volumes between tools (P < .001).

Conclusions:

• Commercially available CTP postprocessing tools yield discrepant results in MDVO stroke.
• These discrepancies highlight potential limitations in current CTP analysis for MDVO.
• Further standardization or validation of CTP tools for MDVO is warranted.