Relationship between Circle of Willis Variations and Cerebral or Cervical Arteries Stenosis Investigated by Computer Tomography Angiography and Multitask Convolutional Neural Network.

Jin Hou,Bin Liu,Ming Yong Gao,Ai Zhen Pan,Qing Yuan He,Jia Bin Lu,Wei Wang,Xiao Dong Zhang,Ya Bin Jin,Qiu Dian Wang,Xiaoxia Yin

doi:10.1155/2021/6024352

Abstract

Circle of Willis (CoW) is the most critical collateral pathway that supports the redistribution of blood supply in the brain. The variation of CoW is closely correlated with cerebral hemodynamic and cerebral vessel-related diseases. But what is responsible for CoW variation remains unclear. Moreover, the visual evaluation for CoW variation is highly time-consuming. In the present study, based on the computer tomography angiography (CTA) dataset from 255 patients, the correlation between the CoW variations with age, gender, and cerebral or cervical artery stenosis was investigated. A multitask convolutional neural network (CNN) was used to segment cerebral arteries automatically. The results showed the prevalence of variation of the anterior communicating artery (Aco) was higher in the normal senior group than in the normal young group and in females than in males. The changes in the prevalence of variations of individual segments were not demonstrated in the population with stenosis of the afferent and efferent arteries, so the critical factors for variation are related to genetic or physiological factors rather than pathological lesions. Using the multitask CNN model, complete cerebral and cervical arteries could be segmented and reconstructed in 120 seconds, and an average Dice coefficient of 78.2% was achieved. The segmentation accuracy for precommunicating part of anterior cerebral artery and posterior cerebral artery, the posterior communicating arteries, and Aco in CoW was 100%, 99.2%, 94%, and 69%, respectively. Artificial intelligence (AI) can be considered as an adjunct tool for detecting the CoW, particularly related to reducing workload and improving the accuracy of the visual evaluation. The study will serve as a basis for the following research to determine an individual's risk of stroke with the aid of AI.

Highlights

Circle of Willis (CoW) is the most important collateral pathway to allow blood communication between the contralateral cerebral hemisphere and carotid-basilar artery, depending on the integrity of the anterior and posterior parts of the CoW
Model Evaluation. e Dice coefficient (DC) was employed as the metric to evaluate the segmentation accuracy of the model on the validation group. e 61 datasets of the testing group were segmented by the model, and the segmentation accuracy was evaluated in the labelled images
Using the proposed model and head-neck computer tomography angiography (CTA) data, the segmentation accuracy for P1 was determined to be perfect at 100%, and A1 and posterior communicating arteries (Pco) were 99.2% and 94%, respectively, while for Aco, the accuracy significantly decreased to 69% (Figure 4)

Summary

Introduction

Circle of Willis (CoW) is the most important collateral pathway to allow blood communication between the contralateral cerebral hemisphere and carotid-basilar artery, depending on the integrity of the anterior and posterior parts of the CoW. In the study, using CTA data, the correlation between the variation in the CoW and factors, such as age, gender, and the stenosis of afferent or efferent arteries, were evaluated. In recent years, computeraided diagnosis has drawn much attention with the development of computer technologies, such as big data [6, 7] and deep learning [8]. It brings excellent progress in disease classification [9], anomaly detection [10, 11], and medical image segmentation [12, 13]. Automatic segmentation of cerebral vessels is required to perform automatic analysis of CoW variation. We proposed a multitask CNN-based-cerebral artery segmentation method and validated its performance in our clinical datasets. e study will serve as a basis for the following research to determine an individual’s risk of stroke with the aid of artificial intelligence (AI)

Methods

Results

Conclusion