Predicting risk of obesity in overweight adults using interpretable machine learning algorithms

Abstract

ObjectiveTo screen for predictive obesity factors in overweight populations using an optimal and interpretable machine learning algorithm.MethodsThis cross-sectional study was conducted between June 2011 and January 2012. The participants were randomly selected using a simple random sampling technique. Seven commonly used machine learning methods were employed to construct obesity risk prediction models. A total of 5,236 Chinese participants from Ningde City, Fujian Province, Southeast China, participated in this study. The best model was selected through appropriate verification and validation and suitably explained. Subsequently, a minimal set of significant predictors was identified. The Shapley additive explanation force plot was used to illustrate the model at the individual level.ResultsMachine learning models for predicting obesity have demonstrated strong performance, with CatBoost emerging as the most effective in both model validity and net clinical benefit. Specifically, the CatBoost algorithm yielded the highest scores, registering 0.91 in the training set and an impressive 0.83 in the test set. This was further corroborated by the area under the curve (AUC) metrics, where CatBoost achieved 0.95 for the training set and 0.87 for the test set. In a rigorous five-fold cross-validation, the AUC for the CatBoost model ranged between 0.84 and 0.91, with an average AUC of ROC at 0.87 ± 0.022. Key predictors identified within these models included waist circumference, hip circumference, female gender, and systolic blood pressure.ConclusionCatBoost may be the best machine learning method for prediction. Combining Shapley’s additive explanation and machine learning methods can be effective in identifying disease risk factors for prevention and control.

Chronic Diseases: A Comprehensive Review of Applications for Prediction of Obesity

“…This real-time monitoring and adjustment can significantly improve adherence to diet and exercise programs, which is often a major hurdle in traditional obesity management strategies [60,61]. Additionally, ML helps in segmenting populations based on their risk and response to different treatments, which can lead to more targeted and effective public health interventions [7,62]. This capability is particularly important in managing obesity at the population level, where one-size-fits-all approaches have often failed [63].…”

“…This persistent challenge underscores the need for innovative approaches that transcend traditional methodologies. Machine Learning (ML), with its ability to harness large volumes of diverse data to uncover patterns and insights beyond human discernment, presents a promising frontier in the battle against obesity [6][7][8].…”

“…The integration of ML into obesity research and management is not without challenges [13,14]. Issues such as data privacy, algorithmic bias, and the digital divide pose significant barriers to the widespread adoption of these technologies [7,15]. Moreover, the effectiveness of ML-driven interventions must be scrutinized through rigorous, multidisciplinary research to ensure they deliver…”

Implementation of K-Nearest Neighbors, Naïve Bayes Classifier, Support Vector Machine and Decision Tree Algorithms for Obesity Risk Prediction

“…Based on research [8][9] [10], this research will classify obesity datasets to compare four algorithms, namely K-NN, Naïve Bayes Classifier, SVM and Decision Tree Algorithms. In research [1][11] [12], the K-NN algorithm, Naïve Bayes Classifier and Decision Tree Algorithms became superior algorithms in their research.…”