A Comparison of Feature Selection and Forecasting Machine Learning Algorithms for Predicting Glycaemia in Type 1 Diabetes Mellitus

Abstract

Type 1 diabetes mellitus (DM1) is a metabolic disease derived from falls in pancreatic insulin production resulting in chronic hyperglycemia. DM1 subjects usually have to undertake a number of assessments of blood glucose levels every day, employing capillary glucometers for the monitoring of blood glucose dynamics. In recent years, advances in technology have allowed for the creation of revolutionary biosensors and continuous glucose monitoring (CGM) techniques. This has enabled the monitoring of a subject’s blood glucose level in real time. On the other hand, few attempts have been made to apply machine learning techniques to predicting glycaemia levels, but dealing with a database containing such a high level of variables is problematic. In this sense, to the best of the authors’ knowledge, the issues of proper feature selection (FS)—the stage before applying predictive algorithms—have not been subject to in-depth discussion and comparison in past research when it comes to forecasting glycaemia. Therefore, in order to assess how a proper FS stage could improve the accuracy of the glycaemia forecasted, this work has developed six FS techniques alongside four predictive algorithms, applying them to a full dataset of biomedical features related to glycaemia. These were harvested through a wide-ranging passive monitoring process involving 25 patients with DM1 in practical real-life scenarios. From the obtained results, we affirm that Random Forest (RF) as both predictive algorithm and FS strategy offers the best average performance (Root Median Square Error, RMSE = 18.54 mg/dL) throughout the 12 considered predictive horizons (up to 60 min in steps of 5 min), showing Support Vector Machines (SVM) to have the best accuracy as a forecasting algorithm when considering, in turn, the average of the six FS techniques applied (RMSE = 20.58 mg/dL).

Identifying Prenatal and Postnatal Determinants of Infant Growth: A Structural Equation Modelling Based Cohort Analysis

“…Some machine learning methods have emerged to identify the causal features from the large dataset [ 41 , 42 ]. These could be a new way forward to analyse the current dataset for future work.…”

Constrained IoT-Based Machine Learning for Accurate Glycemia Forecasting in Type 1 Diabetes Patients

“…In 2021, another study [ 27 ] compared different feature selection and forecasting machine learning algorithms for predicting glycemia in type 1 diabetes mellitus. The authors used a dataset of glucose readings collected from patients with type 1 diabetes and compared the performance of different feature selection techniques, including PCA and Lasso, with different machine learning algorithms, including random forest, SVM, and LSTM.…”

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

“…In order to evaluate the effect of the step of feature selection (FS) in improving the accuracy of the predicted glycemia in T1DM subjects, six FS algorithms including LR, RF, Multi-Layer Perceptron (MLP), Instance-Based K-nearest neighbor (IBk), Relief Attribute (Rlf), and PCA beside four predictive algorithms (RF, LR, SVM, and GP) were applied to a biomedical features dataset [ 145 ]. The outcomes showed that RF as both FS technique and predictive algorithm causes the best RMSE (18.54 mg/dL) throughout the 12 considered predictive horizons (up to 60 min in stages of 5 min).…”