中国医学装备2025，Vol.22Issue(9)：8-11,4.DOI:10.3969/j.issn.1672-8270.2025.09.002
基于深度学习图像分割对原发性高血压患者的脉络丛体积变化的研究
Study on changes of choroid plexus volume in patients with essential hypertension using deep learning-based image segmentation
摘要
Abstract
Objective:To investigate changes of choroid plexus volume in patients with essential hypertension using deep learning-based segmentation technique.Methods:According to the inclusion criteria,19 patients with essential hypertension who visited The 900th Hospital of People's Liberation Army Joint Service Support Force from December 2022 to May 2024 were divided into hypertension group.Additionally,19 healthy individuals,matched with the hypertension group in terms of gender,age,and years of education,were recruited as the healthy control group during the same period.All participants underwent data acquisition of three-dimensional T1(T1-3D)brain structural images using a Siemens 3.0T Trio magnetic resonance(MR)scanner.A pre-trained deep learning model,which based on the 3D U-Net architecture and generated by fully convolutional neural network,was employed for choroid plexus segmentation,and the total brain volume and lateral ventricle volume were calculated using the United Imaging Artificial Intelligence(AI)platform.Statistical analysis of the choroid plexus volume and other data for both groups were performed using SPSS27.0.Results:Compared with the healthy control group(1 866 mm3),the hypertensive group(2 047 mm3)showed an abnormal increase in choroid plexus volume(x2=3.900,P<0.05).Moreover,the choroid plexus volume was positively correlated with lateral ventricle volume and systolic blood pressure(r=0.849,P<0.05),and was negatively correlated with digit span(forward)in the hypertensive group(r=-0.374,P<0.05).Conclusion:The choroid plexus volume in patients with essential hypertension abnormally increased,which might be a factor promoting to changes in brain structure,it can affect further the cognitive function of patients with hypertension.关键词
原发性高血压/深度学习/图像分割/脉络丛/胶质淋巴系统Key words
Essential hypertension/Deep learning/Image segmentation/Choroid plexus/Glial lymphatic system分类
医药卫生引用本文复制引用
杨世泺,陈楚洁,虞浩,王晓阳,付丽媛..基于深度学习图像分割对原发性高血压患者的脉络丛体积变化的研究[J].中国医学装备,2025,22(9):8-11,4.基金项目
福建省科技计划项目(2021I0037) Fujian Provincial Science and Technology Plan Project(2021I0037) （2021I0037）
