Evaluation of short-term predictors of glucose concentration in type 1 diabetes combining feature ranking with regression models

• Published in: Medical & biological engineering & computing Vol. 53; no. 12; pp. 1305 - 1318
• Main Authors: Georga, Eleni I., Protopappas, Vasilios C., Polyzos, Demosthenes, Fotiadis, Dimitrios I.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2015; Springer Nature B.V
• Subjects: Adult; Algorithms; Analysis; Artificial intelligence; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Blood Glucose - analysis; Blood Glucose - drug effects; Carbohydrates; Computer Applications; Construction; Datasets; Diabetes; Diabetes Mellitus, Type 1 - blood; Disease management; Exercise; Feature extraction; Female; Food; Glucose; Human Physiology; Humans; Hypoglycemic Agents - pharmacology; Hypoglycemic Agents - therapeutic use; Imaging; Insulin; Insulin - pharmacology; Insulin - therapeutic use; Lifestyles; Machine Learning; Male; Mathematical models; Meals; Medical; Middle Aged; Models, Statistical; Neural networks; Patients; Radiology; Regression Analysis; Special Issue - Original Article; Studies; Feature selection; Glucose level prediction; Type 1 diabetes; Machine learning
• ISSN: 0140-0118; 1741-0444
• DOI: 10.1007/s11517-015-1263-1

Glucose concentration in type 1 diabetes is a function of biological and environmental factors which present high inter-patient variability. The objective of this study is to evaluate a number of features, which are extracted from medical and lifestyle self-monitoring data, with respect to their ability to predict the short-term subcutaneous (s.c.) glucose concentration of an individual. Random forests (RF) and RReliefF algorithms are first employed to rank the candidate feature set. Then, a forward selection procedure follows to build a glucose predictive model, where features are sequentially added to it in decreasing order of importance. Predictions are performed using support vector regression or Gaussian processes. The proposed method is validated on a dataset of 15 type diabetics in real-life conditions. The s.c. glucose profile along with time of the day and plasma insulin concentration are systematically highly ranked, while the effect of food intake and physical activity varies considerably among patients. Moreover, the average prediction error converges in less than d /2 iterations ( d is the number of features). Our results suggest that RF and RReliefF can find the most informative features and can be successfully used to customize the input of glucose models.