Prediction of metabolic syndrome based on sleep and work-related risk factors using an artificial neural network

Abstract

Background Metabolic syndrome (MetS) is a major public health concern due to its high prevalence and association with heart disease and diabetes. Artificial neural networks (ANN) are emerging as a reliable means of modelling relationships towards understanding complex illness situations such as MetS. Using ANN, this research sought to clarify predictors of metabolic syndrome (MetS) in a working age population. Methods Four hundred sixty-eight employees of an oil refinery in Iran consented to providing anthropometric and biochemical measurements, and survey data pertaining to lifestyle, work-related stressors and sleep variables. National Cholesterol Education Programme Adult Treatment Panel ІІI criteria was used for determining MetS status. The Management Standards Indicator Tool and STOP-BANG questionnaire were used to measure work-related stress and obstructive sleep apnoea respectively. With 17 input variables, multilayer perceptron was used to develop ANNs in 16 rounds of learning. ANNs were compared to logistic regression models using the mean squared error criterion for validation. Results Sex, age, exercise habit, smoking, high risk of obstructive sleep apnoea, and work-related stressors, particularly Role, all significantly affected the odds of MetS, but shiftworking did not. Prediction accuracy for an ANN using two hidden layers and all available input variables was 89%, compared to 72% for the logistic regression model. Sensitivity was 82.5% for ANN compared to 67.5% for the logistic regression, while specificities were 92.2 and 74% respectively. Conclusions Our analyses indicate that ANN models which include psychosocial stressors and sleep variables as well as biomedical and clinical variables perform well in predicting MetS. The findings can be helpful in designing preventative strategies to reduce the cost of healthcare associated with MetS in the workplace.

Prediction of metabolic and pre-metabolic syndromes using machine learning models with anthropometric, lifestyle, and biochemical factors from a middle-aged population in Korea

“…Similar performance results were reported in previous studies [ 35 , 41 ]. Moreover, in a study that measured lifestyle factors such as smoking status, physical activity, sleep time, shift work, and work-related stress in an Iranian working population to predict MetS using an artificial neural network, the results showed high predictive power with 89% accuracy, 82.5% sensitivity, and 92.2% specificity, significantly better than the traditional logistic regression analysis prediction model [ 16 ]. Furthermore, the study stressed the importance of lifestyle factors, such as work-related stress and sleep apnea, in addition to clinical blood markers, for achieving accurate prediction of the MetS status.…”

“…A supervised machine learning model was used for MetS prediction. The algorithms used to develop the model were decision tree, Gaussian Naïve Bayes (NB), K-nearest neighbor (KNN) [ 34 ], XGBoost, random forest (RF), logistic regression [ 15 , 18 , 35 ], support vector machine (SVM), multi-layer perceptron (MLP) [ 16 ], and 1-dimensional convolutional neural network (1D-CNN) [ 36 ]. Min-max normalization was applied to the data used in the analysis [ 37 ].…”

“…Moreover, a recent study attempted to use machine learning based on various clinical features to predict MetS [ 15 , 16 ]. A MetS prediction model for a working population using an artificial neural network was highly efficient, with an accuracy of 89%, higher than logistic regression analysis, the traditional prediction modeling technique [ 16 ]. Moreover, in a study that used the Naïve Bayes model to predict MetS based on clinical and genetic data of a normal-weight population, the area under the receiver operating characteristic curve (AUC) increased by 4% when genetic feature composed of single nucleotide polymorphism was added to the baseline clinical feature such as gender and age, indicating the importance of feature selection [ 15 ].…”

Machine learning models for predicting non-alcoholic fatty liver disease in the general United States population: NHANES database

“…They are also the established risk factors of NAFLD. The cut-off of 109.35 cm seems to be slightly higher than the general cut off value for metabolic syndrome (men, 102 cm and women, 80 cm)[ 25 ]. It is used to calculate the visceral adiposity index, which provides a good predictive capability[ 26 ].…”

Peer-to-Peer Tourism: Tourists’ Profile Estimation through Artificial Neural Networks

“…ANN presents a better degree of adjustment than other extrapolate methodologies such as hedonic pricing method, as stated in [62,63]. Although in these last studies, ANN is implemented in price determination, ANNs of MLP type had been also used in other kinds of estimations based on surveys [64][65][66].…”