A prediction model for childhood obesity risk based on maternal thyroid status and related parameters using machine learning: a mother–newborn–offspring study in a mild-to-moderate iodine deficiency area
Authors' conclusion
Machine learning approaches show promise in predicting childhood obesity risk using maternal and dietary factors during pregnancy. If validated, these findings could support interventions to reduce childhood obesity rates.
Study context
- Population
- Mother–newborn–offspring prospective cohort in a mild-to-moderate iodine deficiency region; maternal iodine status and thyroid function (incl. FT3) assessed in the second half of pregnancy87 parameters
- Participants
- 191 participants
- Comparison
- Six machine learning models predicting overweight at 2 years
- Outcome
- Overweight at 2 years (gender-adjusted weight percentile >85%)
- Follow-up
- 2 years
Methodological detailsmachine learning, prospective mother–newborn–offspring cohort, 5-fold cross-validation, 80% training / 20% test split, synthetic datasets
- Machine learning
- Prospective mother–newborn–offspring cohort
- 5-fold cross-validation
- 80% training / 20% test split
- Synthetic datasets
Performance
As reported in the abstract
Prediction of overweight at 2 years on unseen test data
| Test or model | Accuracy | Kappa | Outcome F1-score | Weighted F1 |
|---|---|---|---|---|
| Neural networks (Nnet) | 0.743 | 0.347 | 0.500 | 0.769 |
Neural networks (Nnet)
Accuracy0.743
Kappa0.347
Outcome F1-score0.500
Weighted F10.769
Further findings
Significant predictors
- Gravidity, maternal height, newborn head circumference, maternal consumption and dietary intake of iodine-rich foods (popsicle, selected fish, yogurt) and FT3
Original publication
A prediction model for childhood obesity risk based on maternal thyroid status and related parameters using machine learning: a mother–newborn–offspring study in a mild-to-moderate iodine deficiency area. Int J Obes (Lond). Published 26 December 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →