2022

Rautiainen, Ilkka; Äyrämö, Sami

Overweight and obesity are an increasing phenomenon worldwide. Reliable and accurate prediction of future overweight or obesity early in the childhood could enable effective interventions by experts. While a lot of research has been done using explanatory modeling methods, capability of machine learning, and predictive modeling, in particular, remain mainly unexplored. In predictive modeling, the models are validated with previously unseen examples, giving a more accurate estimate of their performance and generalization ability in real-life scenarios. Our objective was to find and review existing overweight or obesity research from the perspectives of childhood data and predictive modeling. Thirteen research articles and three review articles were identified as relevant for this review. In general, prediction models with high performance either have a short time span to predict and/or are based on late childhood data. Logistic regression is currently the most often used method in forming the prediction models, although recently more complex models have also been applied. In addition to child’s own weight and height information, maternal weight status and body mass index were often used as predictors in the models. More recent research has started to focus on a wider variety of other predictors as well.