Development and validation of a machine learning model to predict prognostic outcomes in infantile epileptic spasms
Caoxue Zuo1, Boen Xue2, Xiaofeng Mu3
1Department of Neurology, Wuhan Children's Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Insights
A new machine learning model accurately predicts poor seizure outcomes in infants with infantile epileptic spasms syndrome (IESS). This tool uses routine clinical data to aid treatment decisions and improve patient follow-up.
Area of Science:
- Pediatric Neurology
- Machine Learning in Medicine
- Epilepsy Research
Background:
- Infantile epileptic spasms syndrome (IESS) is a severe epilepsy syndrome in infants.
- Predicting seizure outcomes in IESS is crucial for timely intervention and management.
- Current prediction methods may lack accuracy and interpretability.
Purpose of the Study:
- To develop and validate a machine learning (ML) model for predicting seizure outcomes in infants with IESS.
- To identify independent prognostic factors for poor seizure outcomes in IESS.
- To compare the performance of different ML algorithms for IESS outcome prediction.
Main Methods:
- Retrospective study of pediatric patients with IESS.
- Random split into training (70%) and validation (30%) sets.
- Cox regression for prognostic factors, six ML algorithms for model development (XGBoost selected).
- Performance evaluation using AUROC, calibration curves, and decision curve analysis (DCA).
Main Results:
- 56% of patients experienced poor seizure outcomes.
- Tuberous sclerosis complex and malformations of cortical development on MRI were independent risk factors.
- The XGBoost ML model achieved an AUROC of 0.921, demonstrating excellent discrimination, calibration, and clinical utility.
Conclusions:
- An interpretable ML model effectively predicts poor seizure outcomes in IESS patients.
- The model utilizes routine clinical data, facilitating clinical decision-making.
- This tool can assist in optimizing follow-up planning for infants with IESS.
Objective:
To develop and validate a machine learning (ML) model for predicting seizure outcomes in infants with infantile epileptic spasms syndrome (IESS).
Methods:
This retrospective study enrolled pediatric patients diagnosed with infantile epileptic spasms syndrome (IESS) from Wuhan Children's Hospital. The cohort was randomly split into training and validation sets at a 7:3 ratio. Independent prognostic factors were identified using Cox regression analysis. Six machine learning algorithms were then applied to develop predictive models. Model performance was evaluated in terms of discrimination (e.g., AUROC), calibration (calibration curves), and clinical utility (decision curve analysis, DCA). The optimal model (XGBoost) was interpreted via decision tree visualization and SHAP analysis.
Results:
Poor seizure outcome was observed in 56% of the cohort. MRI findings of tuberous sclerosis complex or malformations of cortical development were independent risk factors. Among the models, XGBoost demonstrated the best overall performance, achieving an AUROC of 0.921 in the validation set, along with robust calibration and clinical utility.
Conclusion:
The developed ML model reliably and interpretably predicts poor seizure outcomes in IESS patients using routine clinical data, potentially aiding in clinical decision-making and follow-up planning.
Related Concept Videos
01:24Epilepsy and Seizures: Overview
Various factors can trigger epilepsy, including genetic factors, brain damage, metabolic causes, and unknown etiology. Diagnosis of epilepsy involves electroencephalography (EEG), which...
01:13Seizures: Classification
Seizures are typically classified into two main categories: focal and generalized seizures.
Focal Seizures
Focal seizures originate from specific regions of the brain. These seizures are further sub-classified into two types: