Related Experiment Video
Updated: Sep 2, 2025
Comprehensive Endovascular and Open Surgical Management of Cerebral Arteriovenous Malformations
Published on: October 20, 2017
Machine learning for predicting hemorrhage in pediatric patients with brain arteriovenous malformation
Satvir Saggi1, Ethan A Winkler1, Simon G Ammanuel1
11Department of Neurological Surgery, University of California, San Francisco.
Insights
Machine learning models identified bAVM size and patient age as key predictors of ruptured brain arteriovenous malformations (bAVMs) in children. This approach revealed crucial factors missed by traditional regression, improving prediction of hemorrhagic presentation.
Area of Science:
- Neurology
- Medical Informatics
Background:
- Ruptured brain arteriovenous malformations (bAVMs) in children carry significant risks of morbidity and mortality.
- Predicting hemorrhagic presentation in pediatric bAVMs is challenging due to limited prior research.
Purpose of the Study:
- To identify predictors of hemorrhagic presentation in pediatric patients with bAVMs.
- To compare machine learning (ML) with traditional regression for identifying these predictors.
Main Methods:
- Retrospective cohort study of 186 pediatric patients over 19 years.
- Applied three ML algorithms (random forest, gradient boosted decision trees, AdaBoost).
- Utilized logistic regression for comparative analysis.
Main Results:
- ML models consistently identified bAVM size and patient age as primary predictors.
- ML identified bAVM location and concurrent arterial aneurysm as significant factors.
- Logistic regression highlighted left-sided bAVMs, small size, and concurrent aneurysms.
Conclusions:
- ML approaches can uncover predictors of bAVM hemorrhagic presentation missed by conventional regression.
- This study enhances understanding of pediatric bAVM rupture risk factors.
Objective:
Ruptured brain arteriovenous malformations (bAVMs) in a child are associated with substantial morbidity and mortality. Prior studies investigating predictors of hemorrhagic presentation of a bAVM during childhood are limited. Machine learning (ML), which has high predictive accuracy when applied to large data sets, can be a useful adjunct for predicting hemorrhagic presentation. The goal of this study was to use ML in conjunction with a traditional regression approach to identify predictors of hemorrhagic presentation in pediatric patients based on a retrospective cohort study design.
Methods:
Using data obtained from 186 pediatric patients over a 19-year study period, the authors implemented three ML algorithms (random forest models, gradient boosted decision trees, and AdaBoost) to identify features that were most important for predicting hemorrhagic presentation. Additionally, logistic regression analysis was used to ascertain significant predictors of hemorrhagic presentation as a comparison.
Results:
All three ML models were consistent in identifying bAVM size and patient age at presentation as the two most important factors for predicting hemorrhagic presentation. Age at presentation was not identified as a significant predictor of hemorrhagic presentation in multivariable logistic regression. Gradient boosted decision trees/AdaBoost and random forest models identified bAVM location and a concurrent arterial aneurysm as the third most important factors, respectively. Finally, logistic regression identified a left-sided bAVM, small bAVM size, and the presence of a concurrent arterial aneurysm as significant risk factors for hemorrhagic presentation.
Conclusions:
By using an ML approach, the authors found predictors of hemorrhagic presentation that were not identified using a conventional regression approach.
