Related Experiment Video
Updated: Aug 11, 2026
Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion
Published on: January 15, 2016
Development and Validation of a Prognostic Model for Independent Walking in Children With Cerebral Palsy Based on
1Children's Rehabilitation Department, The Third Affiliated Hospital of Zhengzhou University, Zhengzhou, China.
Insights
Machine learning models accurately predict independent walking in children with cerebral palsy (CP). These tools offer personalized prognosis, aiding clinicians in assessing walking ability and timing for better patient care.
Area of Science:
- Pediatric Neurology
- Rehabilitation Medicine
- Biostatistics
Background:
- Independent walking is a crucial developmental milestone for children with cerebral palsy (CP).
- Predicting walking ability aids in early intervention and personalized care planning.
- Current prediction methods may lack precision and generalizability.
Purpose of the Study:
- To develop and validate machine learning (ML) models for predicting independent walking in children with CP.
- To identify key predictors of walking ability and its timing.
- To create a clinical tool for personalized prognosis.
Main Methods:
- Retrospective cohort study using data from a national CP registry (n=807).
- Follow-up assessments via telephone interviews.
- Development and validation of ML models including logistic regression, XGBoost, and multilayer perceptron.
Main Results:
- 69.5% of children achieved independent walking by age 6.
- Key predictors identified: neonatal asphyxia, GMFCS level, sitting age, CP type, MRI, GMFM-88, epilepsy, intellectual disability, preterm birth, low birth weight.
- ML models showed excellent predictive performance (AUC > 0.945); logistic regression was highest.
- A nomogram was developed for clinical application, with good internal validation but potential overestimation for ages 5-6.
Conclusions:
- ML models accurately predict independent walking in children with CP.
- The developed nomogram serves as an interpretable tool for clinicians.
- External validation is recommended to confirm generalizability.
Objective:
To develop and validate machine learning-based models for predicting independent walking ability in children with cerebral palsy (CP).
Design:
Retrospective cohort study.
Setting:
Data were collected from a national CP registry platform and follow-up assessments were conducted through telephone interviews.
Participants:
Children with CP (n=807) registered between January 2016 and December 2020, with follow-up data collected from October 2022 to March 2023.
Interventions:
Not applicable.
Main Outcome Measures:
The primary outcome was independently walking before the age of 6 years.
Results:
Among the 807 participants, 561 (69.5%) achieved independent walking. Univariate Cox regression identified several predictive factors, including neonatal asphyxia, bilirubin encephalopathy, Gross Motor Function Classification System level before age of 2 years, age of independent sitting, type of CP, magnetic resonance imaging classification, Gross Motor Function Measure-88 scores, epilepsy, intellectual disability, early preterm birth, and very low birth weight (P<.05). Machine learning models demonstrated excellent predictive performance, with logistic regression achieving the highest area under the curve (AUC=0.947), followed by XGBoost (AUC=0.946) and multilayer perceptron (AUC=0.945). Cox proportional hazard models identified key predictors for the timing of independent walking, with a nomogram constructed for clinical application. Internal validation confirmed model reliability, although calibration curves indicated potential overestimation for ages 5-6 years.
Conclusions:
Machine learning models accurately predict independent walking ability in children with CP, although calibration analyses indicated potential overestimation for children aged 5-6 years. The proposed nomogram provides clinicians with an interpretable tool for personalized prognosis. Although internal validation demonstrated excellent performance, future external validation in multicenter cohorts will be critical to confirm generalizability.
