Modeling the relationship between maternal health and infant behavioral characteristics based on machine learning

Area of Science:

• Maternal and Child Health
• Computational Biology
• Developmental Psychology

Background:

• Maternal health is crucial for infant development.
• Understanding the link between maternal well-being and infant outcomes is essential.
• Existing research often lacks comprehensive modeling of these complex relationships.

Purpose of the Study:

• To develop a mathematical model quantifying the impact of maternal health indicators on infant behavioral characteristics and sleep quality.
• To identify key maternal factors influencing infant development.
• To provide a data-driven foundation for targeted interventions.

Main Methods:

• Spearman's correlation coefficient for analyzing maternal indicators and infant behaviors.
• Machine learning models (Random Forest and Multilayer Perceptron) to link maternal health to infant behavior.
• Fuzzy C-means clustering for infant sleep quality classification and Random Forest regression for prediction.

Main Results:

• Identified key maternal health indicators influencing infant behavioral characteristics.
• Developed and validated predictive models for infant behavior based on maternal health.
• Successfully classified infant sleep quality and predicted it using maternal indicators.

Conclusions:

• Maternal health, encompassing physical and psychological aspects, is a significant determinant of infant behavioral characteristics and sleep quality.
• The developed models offer a robust framework for understanding and predicting infant developmental outcomes.
• This research provides valuable insights for developing effective maternal and infant health interventions.