Real-time model-based cerebral perfusion calculation for ischemic stroke

Hao Sun; Bao Li; Jincheng Liu; Xiaolu Xi; Liyuan Zhang; Yanping Zhang; Guangfei Li; Huamei Guo; Kenan Gu; Tongna Wang; Chuanqi Wen; Youjun Liu

doi:10.1016/j.cmpb.2023.107916

Real-time model-based cerebral perfusion calculation for ischemic stroke

Comput Methods Programs Biomed. 2024 Jan:243:107916. doi: 10.1016/j.cmpb.2023.107916. Epub 2023 Nov 11.

Authors

Hao Sun¹, Bao Li¹, Jincheng Liu¹, Xiaolu Xi², Liyuan Zhang¹, Yanping Zhang¹, Guangfei Li¹, Huamei Guo¹, Kenan Gu¹, Tongna Wang¹, Chuanqi Wen¹, Youjun Liu³

Affiliations

¹ Department of Biomedical Engineering, Faculty of Environment and Life, Beijing University of Technology, No. 100 Pingleyuan, Chaoyang, Beijing 100124, China.
² Wuhan United Imaging Healthcare Surgical Technology Co., Ltd. Hubei, China.
³ Department of Biomedical Engineering, Faculty of Environment and Life, Beijing University of Technology, No. 100 Pingleyuan, Chaoyang, Beijing 100124, China. Electronic address: lyjlma@bjut.edu.cn.

PMID: 37976610
DOI: 10.1016/j.cmpb.2023.107916

Abstract

Background and objectives: Clinical diagnosis of ischemic stroke commonly relies on examining cerebral perfusion changes by using computed tomography perfusion (CTP) techniques. However, the radiation dose in CTP is quite higher in comparison to computed tomography angiography (CTA), with associated costs and time.

Methods: Hence, this study established a lumped-parameter model (LPM) of brain tissue microcirculation (BTM) based on CTA, aiming to achieve real-time calculation of cerebral perfusion. After validation of calculated flow results with clinical data, the BTM-LPM model was used to examine the changes in cerebral perfusion following ischemic stroke, in which the effects of nine anatomical structures of the Circle of Willis (CoW) together with various distribution patterns of stenosis in the feeding arteries were considered.

Results: When compared the calculated flow results from BTM-LPM with the clinically measured data of literature, the mean squared error (MSE) value for the feeding arteries was 3.9 % and its total value for microcirculatory flow in each region was 0.1 %. Notably, the calculation time was 35.6 s. In the case of the CoW missing the left and right posterior communicating artery, a 60 % stenosis of the basilar artery is likely to cause ischemic damage to some temporal and occipital lobes of the right and left hemispheres. While in the case of the CoW missing the anterior communicating artery and the left posterior communicating artery, ischemic damage to the entire frontal lobe and parts of the temporal and parietal lobes of the left hemisphere was found when 80 % stenosis occurred in the left internal carotid artery.

Conclusions: The BTM-LPM proposed in this study could accurately calculate cerebral perfusion in real time and demonstrated the importance of CoW anatomy in different ischemic injuries to cerebral tissue. The calculated cerebral perfusion would be a reference value for early diagnosis and preoperative planning of different ischemic injuries to cerebral tissue, thereby the BTM-LPM holds great promising for replacing CTP examination.

Keywords: Cerebral perfusion; Ischemic stroke; Lumped-parameter model.

MeSH terms

Carotid Stenosis*
Cerebral Angiography / methods
Cerebrovascular Circulation
Constriction, Pathologic
Humans
Ischemic Stroke*
Microcirculation
Perfusion
Stroke* / diagnostic imaging