Diabetes is one of the most prevalent cardio-metabolic conditions that can lead to debilitating complications. For type 2 diabetic patients in whom lifestyle modification strategies and oral pharmacotherapy are not adequate for glycemic control, insulin injection is the final life-saving intervention. This selected subgroup of patients is at much higher risks of adverse cardiovascular events compared to those who are not receiving insulin, partly owing to their poorer glycemic control. Moreover, the use of insulin itself increases the risk of potentially lethal, hypoglycemic events. The application of machine learning allowed more accurate prediction of arrhythmic and cause-specific mortality outcomes in type 2 diabetes patients receiving insulin therapy, by using measures of glycemic and lipid variability as well as those of chronic inflammation as input variables. Recently, such efforts have been extended to type 2 patients who are not receiving any medical therapy or those managed pharmacologically with oral medications. In doing so, the aim is to produce accurate yet computationally efficient models that can be generalized to all diabetic patients.

Publications

1. Lee, S., Zhou, J., Leung, K.S.K., Wu, W.K.K., Wong, W.T., Liu, T., Wong, I.C.K., Jeevaratnam, K., Zhang, Q.*, Tse, G.* (2021) Development of a predictive risk model for all-cause mortality in diabetic patients in Hong Kong. BMJ Open Diabetes Research & Care. https://doi.org/10.1136/bmjdrc-2020-001950. Impact factor: 3.4.

2. Lee, S., Zhou, J., Wong, W.T., Liu, T., Wu, W.K.K., Wong, I.C.K., Zhang, Q.*, Tse, G.* (2021) Glycemic and lipid variability for predicting complications and mortality in diabetes mellitus using machine learning. BMC Endocrine Disorders. 21(1):94. PMID: 33947391. http://dx.doi.org/10.1186/s12902-021-00751-4. 5-year impact factor: 2.8.

3. Lee, S., Zhou, J., Guo, C.L., Wong, W.T., Liu, T., Wong, I.C.K., Jeevaratnam, K., Zhang, Q.*, Tse, G.* (2021) Predictive scores for identifying type 2 diabetes mellitus patients at risk of acute myocardial infarction and sudden cardiac death. Endocrinology, Diabetes & Metabolism. https://doi.org/10.1002/edm2.240.

4. Lee, S., Liu, T., Zhou, J., Zhang, Q., Wong, W.T., Tse, G.* (2020) Predictions of diabetes complications and mortality using hba1c variability: a 10-year observational cohort study. Acta Diabetologica. 58(2): 171-180. PMID: 32939583. https://doi.org/10.1007/s00592-020-01605-6. 5-year impact factor: 4.3.