Classification and prediction of hypoglycemia in diabetic patients using machine learning techniques

Swaroopa Shastri; P. Sandhya

doi:10.53730/ijhs.v6nS4.10741

Authors

Swaroopa Shastri
swaroopas04@gmail.com
Assistant Professor, Department of CSE, Visvesvaraya Technological University, Center for Postgraduate Studies, Kalaburagi, Karnataka, India
P. Sandhya Associate Professor, Department of CSE, Visvesvaraya Technological University, Center for Postgraduate Studies, Mysuru, Karnataka, India

Keywords:

hypoglycaemia, microvascular, macrovascular consequences, artificial neural networks

Abstract

Type 1 diabetics can lower their risk of microvascular and macrovascular problems by carefully regulating their blood glucose levels. The downside is that these measures are incredibly challenging because of the wide diversity across individuals, as well as other factors that affect glycemic management. Keeping glucose levels under control is difficult due to the possibility of severe hypoglycemia in patients receiving intensive insulin therapy. In people with diabetes, hypoglycemia is a common complication, which has a negative impact on overall health and well-being. Improving patient safety by anticipating unfavourable glycemic events has become a practical approach to enhanced patient safety using machine learning decision assistance. This paper suggests the use of three machine learning techniques to solve the problem of diabetic safety: (1) Random Forest for continuous glucose predictions, (2) support vector machines for postprandial period predictions, and (3)artificial neural networks for overnight hypoglycemic predictions. It has the two different categorization and prediction capabilities already established. A major system feature is the overall reduction in bouts of hypoglycemia, which results in an increase in patient safety and provides better confidence in treatment decisions.

Downloads

Download data is not yet available.

References

Agiostratidou G, Anhalt H, Ball D, et al. Standardizing clinically meaningful outcome measures beyond hba1c for type 1 diabetes: a consensus report of the American Association of Clinical Endocrinologists, the American Association of Diabetes Educators, the American Diabetes Association, the Endocrine Society, JDRF International, The Leona M. and Harry B. Helmsley Charitable Trust, the Pediatric Endocrine Society, and the T1D Exchange. Diabetes Care 2017; 40(12): 1622–1630.

American Diabetes Association Workgroup on Hypoglycemia. Defining and reporting hypoglycemia in diabetes. Diabetes Care 2005; 28(5): 1245–1249.

Biagi L, Bertachi A, Gimenez M, et al. Individual categorisation of glucose profiles using compositional data analysis. Stat Methods Med Res 2019; 28(12): 3550–3567.

Biagi L, Bertachi A, Martín-Fernández JA, et al. Compositional data analysis of type 1 diabetes data. In: Proceedings of the 3rd international workshop on knowledge discovery in healthcare data (KDH) (ed K Bach, R Bunescu, O Farri, et al.), Stockholm, 13–14 July 2018, pp. 8–12. CEUR.

Breton MD, Patek SD, Lv D, et al. Continuous glucose monitoring and insulin informed advisory system with automated titration and dosing of insulin reduces glucose variability in type 1 diabetes mellitus. Diabetes Technol Ther 2018; 20(8): 531–540.

Buckingham BA, Bailey TS, Christiansen M, et al. Evaluation of a predictive low-glucose management system in-clinic. Diabetes Technol Ther 2017; 19(5): 288–292.

Contreras I and Vehi J. Artificial intelligence for diabetes management and decision support: literature review. J Med Internet Res 2018; 20(5): e10775.

Contreras I, Bertachi A, Biagi L, et al. Using grammatical evolution to generate short-term blood glucose prediction models. In: Proceedings of the 3rd international workshop on knowledge discovery in healthcare data (KDH) (ed K Bach, R Bunescu, O Farri, et al.), Stockholm, 13–14 July 2018, pp. 90–94. CEUR.

Contreras I, Oviedo S, Vettoretti M, et al. Personalized blood glucose prediction: a hybrid approach using grammatical evolution and physiological models. PLoS ONE 2017; 12(11): e0187754

Cryer P. Hypoglycaemia: the limiting factor in the glycaemic management of type I and type II diabetes. Diabetologia 2002; 45(7): 937–948.

Dataset Link : https://archive.ics.uci.edu/ml/datasets/diabetes

Dubosson F, Mordvanyuk N, López B, et al. Negative results for the prediction of postprandial hypoglycemias from insulin intakes and carbohydrates: analysis and comparison with simulated data. In: Proceedings of the 2nd international workshop on artificial intelligence for diabetes (AID) (ed P Herrero, B Lopez and C Martin), Vienna, 24 June 2017, pp. 25–29. Vienna: AIME.

Eljil KS, Qadah G and Pasquier M. Predicting hypoglycemia in diabetic patients using data mining techniques. In: 2013 9th international conference on innovations in information technology (IIT), Abu Dhabi, United Arab Emirates, 17–19 March 2013, pp. 130–135. New York: IEEE.

Garg S, Brazg RL, Bailey TS, et al. Reduction in duration of hypoglycemia by automatic suspension of insulin delivery: the in-clinic ASPIRE study. Diabetes Technol Ther 2012; 14(3): 205–209.

Garg SK, Weinzimer SA, Tamborlane WV, et al. Glucose outcomes with the in-home use of a hybrid closed-loop insulin delivery system in adolescents and adults with type 1 diabetes. Diabetes Technol Ther 2017; 19(3): 155–163.

Georga EI, Protopappas VC, Ardigo D, et al. A glucose model based on support vector regression for the prediction of hypoglycemic events under free-living conditions. Diabetes Technol Ther 2013; 15(8): 634–643.

Hidalgo JI, Colmenar JM, Kronberger G, et al. Data based prediction of blood glucose concentrations using evolutionary methods. J Med Syst 2017; 41: 142.

Hovorka R, Canonico V, Chassin LJ, et al. Nonlinear model predictive control of glucose concentration in subjects with type 1 diabetes. Physiol Meas 2004; 25(4): 905–920.

Kovatchev B and Cobelli C. Glucose variability: timing, risk analysis, and relationship to hypoglycemia in diabetes. Diabetes Care 2016; 39: 502–510.

Kovatchev BP. Metrics for glycaemic control – from HbA1c to continuous glucose monitoring. Nat Rev Endocrinol 2017; 13(7): 425–436.

Mahardika, I. M. R., Suyasa, I. G. P. D., Kamaryati, N. P., & Wulandari, S. K. (2021). Health literacy is strongest determinant on self-monitoring blood glucose (SMBG) type 2 DM patients during COVID-19 pandemic at public health centre in Tabanan Regency. International Journal of Health & Medical Sciences, 4(3), 288-297. https://doi.org/10.31295/ijhms.v4n3.1752

Marling C and Bunescu R. The OhioT1DM dataset for blood glucose level prediction. In: Proceedings of the 3rd international workshop on knowledge discovery in healthcare data, Stockholm, 13–14 July 2018, pp. 60–63. CEUR

Oviedo S, Vehi J, Calm R, et al. A review of personalized blood glucose prediction strategies for T1DM patients. Int J Numer Method Biomed Eng 2017; 33(6): e2833

Ozaslan B, Patek S and Breton M. Quantifying the effect of antecedent physical activity on prandial glucose control in type 1 diabetes: defining exercise on board. In: Abstracts from ATTD 2017 10th international conference on advanced technologies & treatments for diabetes, Paris, 15–18 February 2017, pp. A24–A25. New Rochelle: Mary Ann Liebert, Inc.

Schnell O, Barnard K, Bergenstal R, et al. Role of continuous glucose monitoring in clinical trials: recommendations on reporting. Diabetes Technol Ther 2017; 19(7): 391–399

Sudharsan B, Peeples M and Shomali M. Hypoglycemia prediction using machine learning models for patients with type 2 diabetes. J Diabetes Sci Technol 2015; 9(1): 86–90.

Suryasa, I. W., Rodríguez-Gámez, M., & Koldoris, T. (2021). Health and treatment of diabetes mellitus. International Journal of Health Sciences, 5(1), i-v. https://doi.org/10.53730/ijhs.v5n1.2864

Zarkogianni K, Mitsis K, Litsa E, et al. Comparative assessment of glucose prediction models for patients with type 1 diabetes mellitus applying sensors for glucose and physical activity monitoring. Med Biol Eng Comput 2015; 53(12): 1333–1343.