Comparison of Machine Learning Approaches Toward Assessing the Risk of Developing Cardiovascular Disease as a Long-Term Diabetes Complication

Abstract

The estimation of long-term diabetes complications risk is essential in the process of medical decision making. Guidelines for the management of Type 2 Diabetes Mellitus (T2DM) advocate calculating the Cardiovascular Disease (CVD) risk to initiate appropriate treatment. The objective of this study is to investigate the use of sophisticated machine learning techniques toward the development of personalized models able to predict the risk of fatal or nonfatal CVD incidence in T2DM patients. The important challenge of handling the unbalanced nature of the available dataset is addressed by applying novel ensemble strategies. Hybrid Wavelet Neural Networks (HWNNs) and Self-Organizing Maps (SOMs) constitute the primary models for building ensembles following a subsampling approach. Different methods for combining the decisions of the primary models are applied and comparatively assessed. Data from the 5-year follow up of 560 patients with T2DM are used for development and evaluation purposes. The highest discrimination performance (Area Under the Curve (AUC): 71.48%) is achieved by taking into account both the HWNN- and SOM- based primary models' outputs. The proposed method is superior to the Binomial Linear Regression (BLR) model justifying the need to apply more sophisticated techniques in order to produce reliable CVD risk scores.

A Comparative Analysis of Machine Learning Algorithms to Build a Predictive Model for Detecting Diabetes Complications

“…The figure below illustrates the five approaches for feature selection. Algorithms 2015 -2019 ANN [8], [19], [2], [20], [21],[11] K-means [22] Logistic [22], [9], [2], [20], [23], [24], [11] Decision Tree [1], [20], [3], [25], [24], [10], [26] SVM [27], [2], [1], [20], [3], [21],[24] KNN [1], [21], [24] Random Forest [1], [24] Naive Bayes [3], [23], [25], [24] Designed algorithms [28], [13], [5], [29], [9] Table 1: Shows the trends of used algorithms in previous literature.…”

An explainable XGBoost–based approach towards assessing the risk of cardiovascular disease in patients with Type 2 Diabetes Mellitus

“…Data collected from a 5-year follow up of 560 T2DM patients at the Hippokration General Hospital of Athens, including 41 patients (7.32%) with CVD incidents during their follow-up period, were used for development and evaluation purposes [15]. Out of the 41 patients, four patients experienced stroke and the rest experienced CHD.…”

“…In order to avoid over-fitting, an ensemble learning method, based on a sub-sampling approach, was applied towards the generation of appropriate training subsets from the original data. Multiple individual models were trained on these subsets and combined to produce the final CVD risk scores [15]. In particular, the initial training dataset was divided into positive (for CVD) and negative (for CVD) instances.…”

Roles of Anxiety and Depression in Predicting Cardiovascular Disease Among Patients With Type 2 Diabetes Mellitus: A Machine Learning Approach

“…In recent years, with ever-growing data from hospitals, there are great benefits of employing machine learning technology to provide insights, augment prevention, and reduce costs in healthcare settings. Multiple machine learning models have been applied for producing the risk of developing complications in diabetic patients (Zarkogianni et al, 2018 ). Nowak et al developed a cardiovascular risk prediction system using gradient-boosted machine learning and lasso regularized Cox regression and revealed that multiprotein arrays could be useful in identifying individuals with T2DM who were at the highest risk of CVD (Nowak et al, 2018 ).…”