Click here to

Session: New Technologies and Image Reconstruction in CT and CBCT [Return to Session]

Estimating Ischemic Core Volume with Four Dynamic Images in CT Perfusion

K Chung*, D Lee, S Pandey, J Mandzia, A Khaw, T Lee, University Of Western Ontario, London, ON, CA


TU-F-TRACK 3-1 (Tuesday, 7/27/2021) 4:30 PM - 5:30 PM [Eastern Time (GMT-4)]

Purpose: Simplifying CT perfusion (CTP) protocols may encourage its use at less-experienced centers to guide patient selection for thrombectomy in acute ischemic stroke. In this simulation study, we demonstrate that the cerebral blood flow (CBF) map can be calculated with four images equivalent to standard of care non-contrast CT (NCCT) and multiphase CT angiography (mCTA) to measure the ischemic core volume required for patient selection.

Methods: Twenty-eight consecutive patients with acute ischemic stroke and CTP at admission were included in this study. The standard CTP protocol consisted of 22 dynamic images acquired over 60 s. NCCT and mCTA were simulated by manually selecting the four dynamic CTP images that corresponded to the pre-contrast baseline (NCCT), peak arterial enhancement, and two delayed images at 8 and 16 s after the peak (mCTA). Early or late timing of the peak and of the delayed images was also simulated by shifting the selection of contrast-enhanced images up to three time points (<9 s) before or after the true peak. CBF maps were generated by a model-based deconvolution algorithm from which ischemic core was identified by CBF<30% of the contralateral hemisphere. Bland-Altman analysis and the intraclass correlation coefficient (ICC) assessed agreement and reliability, respectively, of measured core volumes compared to those from a standard CTP study.

Results: With correct timing of the arterial peak, the mean difference (MD; limits of agreement) and ICC were 3.7 (–4.9 to 12.4) ml and 0.94 (0.72 to 0.98), respectively. MD and ICC were similarly good when the peak was timed early by up to 6 s or late by 3 s (MD<5.9 ml; ICC>0.86) but poor with earlier (≈9 s) or later timings (>3 s).

Conclusion: Core volumes measured with only four images equivalent to NCCT and mCTA were consistent to those from a standard CTP study.

Funding Support, Disclosures, and Conflict of Interest: K.J. Chung is supported by an NSERC Canada Graduate Scholarship - Master's award. T.Y. Lee licenses CT Perfusion software to GE Healthcare.



    Perfusion Imaging, CT, Stroke


    IM- CT: Perfusion imaging, CTA, coronary artery calcium scoring

    Contact Email