The value of artificial intelligence-based automatic segmentation in the identification of ruptured aneurysms in patients with multiple intracranial aneurysms and subarachnoid hemorrhage
DENG Xiqing
YIN Can
DUAN Shengwu
OUYANG Fan
Abstract：Objective To investigate the identification method of ruptured aneurysms in patients with multiple intracranial aneurysms and subarachnoid hemorrhage(SAH)based on artificial intelligence(AI)automatic segmentation technology,provid-ing more accurate and objective evidence for clinical decision-making.Methods A total of 88 patients with multiple intracra-nial aneurysms and subarachnoid hemorrhage(SAH)were enrolled.Based on rupture status,the aneurysms were categorized into a ruptured group(n=88)and an unruptured group(n=115).All aneurysms were analyzed using United Imaging's AI-based auto-segmentation software to obtain geometric parameters.Parameters such as aneurysm size,aspect ratio(AR),size ratio(SR),morphology,proximity to SAH,proximity to hematoma,and proximity to vasospasm were compared between the two groups to assess their role in identifying the ruptured aneurysm.Logistic regression analysis was used to screen and identify the in-dependent factors for determining the ruptured responsible aneurysms.Results A total of 203 aneurysms were included,of which 88 were ruptured.Univariate analysis showed statistically significant differences in aneurysm size,SR,irregular morphol-ogy,proximity to SAH,and proximity to hematoma between ruptured and unruptured groups(all P<0.01).Multivariate logistic regression analysis revealed that proximity to SAH(OR=226.84,95%CI:30.42-1691.49),irregular morphology(OR=8.93,95%CI:2.52-31.56),SR(OR=9.74,95%CI:3.36-28.19),and aneurysm size(OR=1.40,95%CI:1.12-1.76)were indepen-dent predictors for identifying ruptured aneurysms.Conclusion AI automatic segmentation technology provides a reference ba-sis for determining the ruptured responsible aneurysms in patients with multiple intracranial aneurysms and SAH by improving the accuracy and consistency of aneurysm geometric parameter measurement.
Keywords：Multiple intracranial aneurysmsSubarachnoid hemorrhageRuptured aneurysmTomographyX-ray com-putedArtificial intelligence
Publication Date：2025-11-30
Online Publishing Date：2026-01-09(First online date of this platform, not the publication date of the document)
Pages：5( 19-22,28 )