A deep learning model can predict the risk for obesity for young children using only routinely collected electronic health record (EHR) data, according to a study published in the December issue of Obesity Pillars . Mehak Gupta, Ph.D.
, from Southern Methodist University in Dallas, and colleagues developed predictive models of childhood obesity using advanced machine learning methods applied to a large unaugmented EHR dataset (36,191 children aged 0 to 10 years). The researchers found that results were mostly better or comparable to existing studies, with the model achieving an area under the receiver operating characteristic curve above 0.8 in all cases (with most cases around 0.
9) for predicting obesity within the next three years for children 2 to 7 years of age. The model's robustness was validated, with top predictors matching important risk factors of obesity. "Our model can predict the risk of obesity for young children at multiple time points using only routinely collected EHR data, greatly facilitating its integration into clinical care ," the authors write.
"Our model can be used as an objective screening tool to provide clinicians with insights into a patient's risk for developing obesity so that early lifestyle counseling can be provided to prevent future obesity in young children." More information: Mehak Gupta et al, Reliable prediction of childhood obesity using only routinely collected EHRs may be possible, Obesity Pillars (2024). DOI: 10.
1016/j.obpill.2024.
100.
