Non-invasive prediction of blood glucose trends during hypoglycemia

• Published in: Analytica chimica acta Vol. 1052; pp. 37 - 48
• Main Authors: Tronstad, Christian, Elvebakk, Ole, Staal, Odd Martin, Kalvøy, Håvard, Høgetveit, Jan Olav, Jenssen, Trond Geir, Birkeland, Kåre Inge, Martinsen, Ørjan G.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 04.04.2019; Elsevier BV
• Subjects: Adolescent; Adult; Blood; Blood glucose; Blood Glucose - metabolism; Clamps; Diabetes; Diabetes mellitus; Diabetes Mellitus, Type 1 - blood; Diabetes Mellitus, Type 1 - complications; Electric Impedance; Female; Glucose; Humans; Hypoglycemia; Hypoglycemia - blood; Hypoglycemia - complications; Information systems; Least-Squares Analysis; Machine learning; Male; Middle Aged; Models, Statistical; Multisensor; Neural network; Neural networks; Non invasive; Optical measurement; Predictions; Regression analysis; Regression models; Reproducibility of Results; Skin; Skin Temperature; Spectrum Analysis; Temperature measurement; Trends; Young Adult; Multisensor; Glucose; Neural network; Hypoglycemia; Machine learning; Non invasive
• ISSN: 0003-2670; 1873-4324; 1873-4324
• DOI: 10.1016/j.aca.2018.12.009

Over the last four decades, there has been a pursuit for a non-invasive solution for glucose measurement, but there is not yet any viable product released. Of the many sensor modalities tried, the combination of electrical and optical measurement is among the most promising for continuous measurements. Although non-invasive prediction of exact glucose levels may seem futile, prediction of their trends may be useful for certain applications. Hypoglycemia is the most serious of the acute complications in type-1 diabetes highlighting the need for a reliable alarm, but little is known about the performance of this technology in predicting hypoglycemic glucose levels and associated trends. We aimed to assess such performance on the way to develop a multisensor system for detection of hypoglycemia, based on near-infrared (NIR), bioimpedance and skin temperature measurements taken during hypoglycemic and euglycemic glucose clamps in 20 subjects with type-1 diabetes. Performance of blood glucose prediction was assessed by global partial least squares and neural network regression models using repeated double cross-validation. Best trend prediction was obtained by including all measurements in a neural network model. Prediction of glucose level was inaccurate for threshold-based detection of hypoglycemia, but the trend predictions may provide useful information in a multisensor system. Comparing NIR and bioimpedance measurements, NIR seems to be the main predictor of blood glucose while bioimpedance may act as correction for individual confounding properties. [Display omitted] •First study to assess NIR + bioimpedance for prediction of hypoglycemic glucose trends.•Prediction of trends could be useful in hypoglycemia detection.•Glucose level prediction was not sufficiently accurate for clinical use.•Neural network performed better than partial least squares as prediction model.•Bioimpedance seems to correct for individual variations in NIR-glucose relation.