Neural-network detection of hypoglycemic episodes in children with type 1 diabetes using physiological parameters

Abstract

The most common and highly feared adverse effect of intensive insulin therapy in patients with diabetes is the increased risk of hypoglycemia. Symptoms of hypoglycemia arise from the activation of the autonomous central nervous systems and from reduced cerebral glucose consumption. HypoMon is a non-invasive monitor that measures some physiological parameters continuously to provide detection of hypoglycemic episodes in Type 1 diabetes mellitus patients (T1DM). Based on heart rate, corrected QT interval of the ECG signal and skin impedance, a neural network detection algorithm has been developed to recognize the presence of hypoglycemic episodes. From a clinical study of 21 children with T1DM, associated with hypoglycemic episodes, their heart rates increased (1.16 +/- 0.16 vs. 1.03 +/- 0.11, P<0.0001), their corrected QT intervals increased (1.09 +/- 0.09 vs. 1.02 +/- 0.07, P<0.0001) and their skin impedances reduced significantly (0.66 +/- 0.19 vs. 0.82 +/- 0.21, P<0.0001). The overall data were obtained and grouped into a training set, a validation set and a test set, each with 7 patients randomly selected. Using a feedforward multi-layer neural network with 9 hidden nodes, and an algorithm developed from the training set and the validation set, a sensitivity of 0.9516 and specificity of 0.4142 were achieved for the test set. A more advanced neural network algorithm will be developed to improve the specificity of test sets in the near future.