Early detection of type 2 diabetes mellitus using machine learning-based prediction models

Abstract

Most screening tests for T2DM in use today were developed using multivariate regression methods that are often further simplified to allow transformation into a scoring formula. The increasing volume of electronically collected data opened the opportunity to develop more complex, accurate prediction models that can be continuously updated using machine learning approaches. This study compares machine learning-based prediction models (i.e. Glmnet, RF, XGBoost, LightGBM) to commonly used regression models for prediction of undiagnosed T2DM. The performance in prediction of fasting plasma glucose level was measured using 100 bootstrap iterations in different subsets of data simulating new incoming data in 6-month batches. With 6 months of data available, simple regression model performed with the lowest average RMSE of 0.838, followed by RF (0.842), LightGBM (0.846), Glmnet (0.859) and XGBoost (0.881). When more data were added, Glmnet improved with the highest rate (+ 3.4%). The highest level of variable selection stability over time was observed with LightGBM models. Our results show no clinically relevant improvement when more sophisticated prediction models were used. Since higher stability of selected variables over time contributes to simpler interpretation of the models, interpretability and model calibration should also be considered in development of clinical prediction models. Type 2 diabetes mellitus (T2DM) is very common and is responsible for very considerable morbidity, mortality. Furthermore, it is a substantial financial drain both on individuals/families, health systems and societies. Of major concern is that the incidence and prevalence of T2DM are increasing rapidly-globally. In 2017, it was estimated that 425 million people had any type of diabetes (approx. 5.5% of worldwide population) of which 90% had T2DM and according to projection estimations the prevalence is going to increase substantially in the coming years; by 2,045, for example, a 48% increase of prevalence from the above numbers is expected or in absolute numbers an estimated 629 million people (approx. 6.6% of the worldwide population) are expected to be suffering from any type of diabetes 1. T2DM can lead to substantially increased risk of macrovascular and microvascular disease, especially in those with inadequate glycaemic control 2. Progression of T2DM from impaired fasting glucose is typically slow and more importantly, its symptoms may remain undetected for many years. Delays in diagnosis are an important contributory factor to poor control and risk of complications 3. Data mining is nowadays applied to various fields of science, including healthcare and medicine. Often applied are pattern recognition, disease prediction and classification using various data mining techniques 4. Due to the increased prevalence of T2DM, various techniques have been used to build predictive models and models for early disease diagnosis, such as logistic and Cox proportional hazard regression models 5-7 , Random Forest 8,9 , boosted ensembles 10,1...

Predicting bloodstream infection outcome using machine learning

“…We implemented the gradient-boosting predictor trained with the LightGBM [37] Python package. LightGBM has shown effectiveness on clinical and patient tabular data in particular, and was adopted by many recently published models [38][39][40][41][42][43]. Missing values were inherently handled by the LightGBM predictor [37,44,45].…”

Multi-Tier Ensemble Learning Model With Neighborhood Component Analysis to Predict Health Diseases

“…While the DT model obtained the best recall values: 90% and 100%, and RF obtained the best AUC value: 83% for the PID dataset, and NB obtained the highest AUC value: 84% for the Kurmitola dataset. In contrast, Kopitar et al [17] utilized several ML models: Glmnet, RF, XGBoost, and LightGBM, to predict undiagnosed type 2 diabetes mellitus [17]. The authors obtained a dataset from EHR in 10 healthcare centers in Slovenia in which 3,723 participant records were applicable in their development to predict type 2 diabetes mellitus [17].…”

“…In contrast, Kopitar et al [17] utilized several ML models: Glmnet, RF, XGBoost, and LightGBM, to predict undiagnosed type 2 diabetes mellitus [17]. The authors obtained a dataset from EHR in 10 healthcare centers in Slovenia in which 3,723 participant records were applicable in their development to predict type 2 diabetes mellitus [17]. Their utilized model gained the highest performance regarding the lowest average RMSE value of 83.8% to predicting diabetes [17].…”

“…The authors obtained a dataset from EHR in 10 healthcare centers in Slovenia in which 3,723 participant records were applicable in their development to predict type 2 diabetes mellitus [17]. Their utilized model gained the highest performance regarding the lowest average RMSE value of 83.8% to predicting diabetes [17]. In comparison, Haq et al [18] developed a diagnosis system after employing ML approaches to detect diabetes.…”

Machine Learning-based Web Application for Early Diagnosis of Diabetes

“…It shows strategic calculations were more important than precision. The restriction was that only Type 2 Diabetes was under consideration [19]. Younus et al proposed an algorithm that is based on random forest and attempted to detect the complicated areas of patients with type 2 diabetes.…”