Related Experiment Video
Updated: Jul 11, 2025
Setup of Consumer Wearable Devices for Exposure and Health Monitoring in Population Studies
Published on: February 3, 2023
Machine Learning Identifies Smartwatch-Based Physiological Biomarker for Predicting Disruptive Behavior in Children:
Magdalena Romanowicz1,2, Kyle S Croarkin3, Rana Elmaghraby4
1Department of Psychiatry and Psychology, Mayo Clinic, Rochester, Minnesota, USA.
Insights
Smartwatches can feasibly monitor children with disruptive behaviors. Machine learning accurately predicts behavior states using heart rate, sleep, and activity data, aiding interventions.
Area of Science:
- Pediatric Psychology
- Behavioral Science
- Digital Health
Background:
- Parents increasingly use smartwatches for child monitoring.
- Disruptive behaviors in children require effective monitoring and intervention.
- Existing methods for monitoring disruptive behaviors can be limited.
Purpose of the Study:
- To assess the feasibility and accuracy of smartwatch monitoring for predicting disruptive behaviors in hospitalized children.
- To identify physiological and activity-based biomarkers for behavior prediction.
- To explore the potential of machine learning in analyzing smartwatch data for behavioral insights.
Main Methods:
- Pilot study involving 10 children (aged 7-10) hospitalized for disruptive behaviors.
- Continuous behavioral phenotyping using smartwatch data (heart rate, sleep, motor activity).
- Supervised machine learning models trained to predict behavior states (calm, playful, disruptive), focusing on severe outbursts.
Main Results:
- Achieved 90% adherence for per-protocol smartwatch use.
- Machine learning models identified conditional dependencies between physiological/activity data and behavior states.
- Achieved 80.89% accuracy in predicting a child's behavior state via cross-validation.
Conclusions:
- Continuous smartwatch monitoring is feasible for children with severe disruptive behaviors.
- Machine learning can identify predictive biomarkers for impending disruptive behaviors.
- Future research can leverage smartwatch data for enhanced behavioral interventions and clinical trials.
Abstract:
Parents frequently purchase and inquire about smartwatch devices to monitor child behaviors and functioning. This pilot study examined the feasibility and accuracy of using smartwatch monitoring for the prediction of disruptive behaviors. The study enrolled children (N = 10) aged 7-10 years hospitalized for the treatment of disruptive behaviors. The study team completed continuous behavioral phenotyping during study participation. The machine learning protocol examined severe behavioral outbursts (operationalized as episodes that preceded physical restraint) for preparing the training data. Supervised machine learning methods were trained with cross-validation to predict three behavior states-calm, playful, and disruptive. The participants had a 90% adherence rate for per protocol smartwatch use. Decision trees derived conditional dependencies of heart rate, sleep, and motor activity to predict behavior. A cross-validation demonstrated 80.89% accuracy of predicting the child's behavior state using these conditional dependencies. This study demonstrated the feasibility of 7-day continuous smartwatch monitoring for children with severe disruptive behaviors. A machine learning approach characterized predictive biomarkers of impending disruptive behaviors. Future validation studies will examine smartwatch physiological biomarkers to enhance behavioral interventions, increase parental engagement in treatment, and demonstrate target engagement in clinical trials of pharmacological agents for young children.
