This is a Data Science for Social Good project childrens' growth curves a few years into the future in order to determine which kids are at risk of obesity.

Our partner, NorthShore University HealthSystem, provided us with the height and weight measurements of over thousands children, records captured over the past 6 years by its hospitals’ EMR system.

We analyzed this data and found that adiposity rebound, when a child’s BMI trajectory dips and then rebounds, can be used as a predictor of obesity.

This wiki documents our work creating a predictive model that could eventually be the foundation of a clinical tool used by pediatricians to detect kids at risk of becoming overweight or obese, prompting early intervention.

• [Problem: The Obesity Epidemic and the Use of Growth Charts] (./wiki/problem)
• [Data: Longitudinal Height and Weight Measurements] (./wiki/data)
• [Exploratory Analysis: Divergence in Growth Patterns] (./wiki/Exploratory-Analysis)
• [Methodology: Calculating Adiposity Rebound] (./wiki/methodology)
• [Results: Predicting Obesity with Adiposity Rebound] (./wiki/results)
• [Moving Forward: Implementing Predictive Tools in Practice] (./wiki/future-work)