Keywords:
CT, Neuroradiology brain, CT-Quantitative, Computer Applications-Detection, diagnosis, Ischaemia / Infarction, Retrospective, Not applicable, Performed at one institution
Authors:
S. Ichikawa, H. YAMAMOTO; Kurashiki/JP
DOI:
10.26044/ecr2020/C-07900
Purpose
Computed tomography (CT) perfusion is a useful technique that enables the differentiation of salvageable ischemic penumbra from infarct core and can help identify patients most likely to benefit from acute revascularization therapies. The high radiation dose received from continuously scanning one location of the head during the first pass of iodinated contrast material has been a focus of interest to many articles [1, 2]. Reduction in tube current and tube voltage and the use of iterative reconstruction can lead to a remarkable reduction in the radiation dose while maintaining acceptable image quality [3].
A newly introduced Bayesian estimation algorithm can minimize the effects of oscillation, tracer delay, and low signal-to-noise ratio during estimation of the residue function of brain tissue, and yields more quantitatively accurate CT perfusion maps than singular value decomposition algorithms [4, 5]. Although there is a possibility that robustness to image noise of this algorithm can lead to a further reduction in the radiation dose, the effect of radiation dose reduction on quantitative interpretation of CT perfusion using the Bayesian estimation algorithm has not yet been investigated.
We hypothesized that low-dose CT perfusion with the Bayesian estimation algorithm yields similar results to what original CT perfusion does for assessment of acute ischemic stroke. In this study, we aimed to verify quantitative perfusion values obtained using the Bayesian estimation algorithm for simulated low-dose CT perfusion in patients with acute ischemic stroke.