Comparison between mathematical methods to estimate blood glucose levels from ECG signals

• DOI: 10.1016/j.biosx.2024.100474
• Publication Date: 2024-03-26T07:34:49Z
• Authors: Oscar Ivan Corona...; Adriana del Carme...; José Antonio Guti...; Luis Alfredo Cast...; Eilen García Rodr...
• Keywords: Cubic regression; Quadratic regression; Diabetes; Blood glucose concentration; 02 engineering and technology; Linear regression; 0210 nano-technology; TP248.13-248.65; Electrocardiogram; Biotechnology

Diabetes mellitus, known as diabetes, is a chronic disease that affects the control of blood glucose concentration levels, it mostly adults (type 2 diabetes), but can also occur in children 1 or childhood well pregnant women (gestational diabetes). one diseases with highest prevalence and high mortality worldwide. has no cure, continuous monitoring to maintain levels normal ranges reduces possibility suffering from gastrointestinal problems, vision loss, limb amputations (such diabetic foot) damage vital organs such heart kidneys, among other associated complications. This article compares results estimation by using linear, quadratic cubic regression considering electrical characteristics generated cardiac conduction (HR, HRV, T-wave peak, QT interval) recorded on single-lead electrocardiogram (VII), used non-invasive model. The best estimate was obtained regression. validation performed Clarke grid having 77.78 % data A zone 22.22 B Pearson correlation value 0.94103