Deep Learning Framework for Detection of Hypoglycemic Episodes in Children with Type 1 Diabetes

Publisher:
IEEE
Publication Type:
Conference Proceeding
Citation:
Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society 2016 (EMBC), 2016, pp. 3503 - 3506 (4)
Issue Date:
2016-08-16
Full metadata record
Files in This Item:
Filename Description Size
C6E82F6C-CE40-4440-9BF2-159E4CD37549 am.pdfAccepted Manuscript version189.41 kB
Adobe PDF
Most Type 1 diabetes mellitus (T1DM) patients have hypoglycemia problem. Low blood glucose, also known as hypoglycemia, can be a dangerous and can result in unconsciousness, seizures and even death. In recent studies, heart rate (HR) and correct QT interval (QTc) of the electrocardiogram (ECG) signal are found as the most common physiological parameters to be effected from hypoglycemic reaction. In this paper, a state-of-the-art intelligent technology namely deep belief network (DBN) is developed as an intelligent diagnostics system to recognize the onset of hypoglycemia. The proposed DBN provides a superior classification performance with feature transformation on either processed or un-processed data. To illustrate the effectiveness of the proposed hypoglycemia detection system, 15 children with Type 1 diabetes were volunteered overnight. Comparing with several existing methodologies, the experimental results showed that the proposed DBN outperformed and achieved better classification performance.
Please use this identifier to cite or link to this item: