Predicting body mass index in early childhood using data from the first 1000 days

Area of Science:

• Pediatrics
• Public Health
• Machine Learning

Background:

• The first 1000 days of life are critical for obesity prevention.
• Existing childhood obesity prediction models often lack prenatal and early infancy risk factors.

Purpose of the Study:

• To utilize machine learning to identify early life risk factors for childhood obesity.
• To develop accurate predictive models for body mass index (BMI) in early childhood.

Main Methods:

• LASSO regression identified 13 relevant features beyond standard measurements.
• Support vector regression with fivefold cross-validation was used to build BMI prediction models.
• Models were trained on 80% of patient data and validated on the remaining 20%.

Main Results:

• The models achieved high accuracy in predicting children's BMI at 30-36, 36-42, and 42-48 months.
• Mean average errors ranged from 0.96 to 1.00, with low standard deviations.
• Identified key risk factors from the first 1000 days influencing childhood BMI.

Conclusions:

• Machine learning models can effectively predict childhood BMI using early life data.
• These predictive tools can support clinical and public health initiatives for early obesity prevention.