Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture

Abstract Small intracranial aneurysms (SIAs) (< 5 mm) are increasingly detected due to advanced imaging, but predicting rupture risk remains challenging. Rupture, though rare, can cause devastating subarachnoid hemorrhage. This study analyzed 141 SIAs (101 unruptured, 40 ruptured) using semi-auto...

Full description

Saved in:
Bibliographic Details
Main Authors: Vanessa M. Swiatek, Samuel Voss, Florian Sprenger, Igor Fischer, Hafez Kader, Klaus-Peter Stein, Roland Schwab, Sylvia Saalfeld, Ali Rashidi, Daniel Behme, Philipp Berg, I. Erol Sandalcioglu, Belal Neyazi
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-08478-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849335108825251840
author Vanessa M. Swiatek
Samuel Voss
Florian Sprenger
Igor Fischer
Hafez Kader
Klaus-Peter Stein
Roland Schwab
Sylvia Saalfeld
Ali Rashidi
Daniel Behme
Philipp Berg
I. Erol Sandalcioglu
Belal Neyazi
author_facet Vanessa M. Swiatek
Samuel Voss
Florian Sprenger
Igor Fischer
Hafez Kader
Klaus-Peter Stein
Roland Schwab
Sylvia Saalfeld
Ali Rashidi
Daniel Behme
Philipp Berg
I. Erol Sandalcioglu
Belal Neyazi
author_sort Vanessa M. Swiatek
collection DOAJ
description Abstract Small intracranial aneurysms (SIAs) (< 5 mm) are increasingly detected due to advanced imaging, but predicting rupture risk remains challenging. Rupture, though rare, can cause devastating subarachnoid hemorrhage. This study analyzed 141 SIAs (101 unruptured, 40 ruptured) using semi-automatic morphological analysis and high-resolution, image-based blood flow simulations from 3D rotational angiography. Advanced morphological and hemodynamic parameters were extracted, with clustering applied to address multicollinearity. Univariate logistic regression identified cluster representatives, and forward selection highlighted the maximum height, Neck inflow rate, and Non-sphericity index as rupture predictors, though only the latter two were significant. Clinical variables like age, sex, and comorbidities were also assessed but failed to predict rupture risk. The full model showed overfitting, with a pseudo-R2 of 0.142 on the training set but only 0.032 on the test set. A simplified model using just Neck inflow rate and Non-sphericity index performed similarly poorly (pseudo-R2 of 0.034). Multiple machine learning classifiers were evaluated, with similar performance across models, supporting the model-independence of the results. Overall, neither morphological, hemodynamic, nor clinical variables reliably predicted rupture risk, highlighting the limitations of current methods and underscoring the need for prospective studies and multimodal approaches that integrate imaging biomarkers and compare small and large aneurysms for better risk stratification.
format Article
id doaj-art-da89c50bf20e47629d2b3f89ebfe9af4
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-da89c50bf20e47629d2b3f89ebfe9af42025-08-20T03:45:23ZengNature PortfolioScientific Reports2045-23222025-07-0115111210.1038/s41598-025-08478-1Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm ruptureVanessa M. Swiatek0Samuel Voss1Florian Sprenger2Igor Fischer3Hafez Kader4Klaus-Peter Stein5Roland Schwab6Sylvia Saalfeld7Ali Rashidi8Daniel Behme9Philipp Berg10I. Erol Sandalcioglu11Belal Neyazi12Department of Neurosurgery, Otto-von-Guericke UniversityDepartment of Fluid Dynamics and Technical Flows, Otto-von-Guericke UniversityDepartment of Neurosurgery, Otto-von-Guericke UniversityDepartment of Neurosurgery, University Hospital Düsseldorf and Heinrich-Heine-UniversityAutonomous Multisensor Systems Group, Institute for Intelligent Cooperating Systems, Otto-von-Guericke UniversityDepartment of Neurosurgery, Otto-von-Guericke UniversityDepartment of Neuroradiology, Otto-von-Guericke UniversityForschungscampus STIMULATEDepartment of Neurosurgery, Otto-von-Guericke UniversityDepartment of Neuroradiology, Otto-von-Guericke UniversityDepartment of Medical Engineering, Otto-von-Guericke UniversityDepartment of Neurosurgery, Otto-von-Guericke UniversityDepartment of Neurosurgery, Otto-von-Guericke UniversityAbstract Small intracranial aneurysms (SIAs) (< 5 mm) are increasingly detected due to advanced imaging, but predicting rupture risk remains challenging. Rupture, though rare, can cause devastating subarachnoid hemorrhage. This study analyzed 141 SIAs (101 unruptured, 40 ruptured) using semi-automatic morphological analysis and high-resolution, image-based blood flow simulations from 3D rotational angiography. Advanced morphological and hemodynamic parameters were extracted, with clustering applied to address multicollinearity. Univariate logistic regression identified cluster representatives, and forward selection highlighted the maximum height, Neck inflow rate, and Non-sphericity index as rupture predictors, though only the latter two were significant. Clinical variables like age, sex, and comorbidities were also assessed but failed to predict rupture risk. The full model showed overfitting, with a pseudo-R2 of 0.142 on the training set but only 0.032 on the test set. A simplified model using just Neck inflow rate and Non-sphericity index performed similarly poorly (pseudo-R2 of 0.034). Multiple machine learning classifiers were evaluated, with similar performance across models, supporting the model-independence of the results. Overall, neither morphological, hemodynamic, nor clinical variables reliably predicted rupture risk, highlighting the limitations of current methods and underscoring the need for prospective studies and multimodal approaches that integrate imaging biomarkers and compare small and large aneurysms for better risk stratification.https://doi.org/10.1038/s41598-025-08478-1Small intracranial aneurysmsRupture risk assessmentSemiautomatic neck curve reconstructionComputational fluid dynamicsPredictive modeling
spellingShingle Vanessa M. Swiatek
Samuel Voss
Florian Sprenger
Igor Fischer
Hafez Kader
Klaus-Peter Stein
Roland Schwab
Sylvia Saalfeld
Ali Rashidi
Daniel Behme
Philipp Berg
I. Erol Sandalcioglu
Belal Neyazi
Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture
Scientific Reports
Small intracranial aneurysms
Rupture risk assessment
Semiautomatic neck curve reconstruction
Computational fluid dynamics
Predictive modeling
title Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture
title_full Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture
title_fullStr Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture
title_full_unstemmed Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture
title_short Predictive modeling and machine learning show poor performance of clinical, morphological, and hemodynamic parameters for small intracranial aneurysm rupture
title_sort predictive modeling and machine learning show poor performance of clinical morphological and hemodynamic parameters for small intracranial aneurysm rupture
topic Small intracranial aneurysms
Rupture risk assessment
Semiautomatic neck curve reconstruction
Computational fluid dynamics
Predictive modeling
url https://doi.org/10.1038/s41598-025-08478-1
work_keys_str_mv AT vanessamswiatek predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT samuelvoss predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT floriansprenger predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT igorfischer predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT hafezkader predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT klauspeterstein predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT rolandschwab predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT sylviasaalfeld predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT alirashidi predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT danielbehme predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT philippberg predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT ierolsandalcioglu predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture
AT belalneyazi predictivemodelingandmachinelearningshowpoorperformanceofclinicalmorphologicalandhemodynamicparametersforsmallintracranialaneurysmrupture