Suhaimi, Fatanah M.Zin, Syatirah MatErtugrul, SenizMazlan, Mohd ZulfakarZukhi, Nur Jihan M.2023-06-162023-06-162021978-1-7281-4245-62374-3220https://doi.org/10.1109/IECBES48179.2021.9398780https://hdl.handle.net/20.500.14365/1952IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES) - Leading Modern Healthcare Technology Enhancing Wellness -- MAR 01-03, 2021 -- ELECTR NETWORKType II diabetes mellitus (T2DM) is one of severe and common chronic disease, affecting almost all populations in many countries. T2DM and its complications constitute a major worldwide public health problem and associated with high rates of diabetes-related morbidity and mortality. In this study, a retrospective clinical data was collected from three patients receiving insulin therapy in the ICU of HUSM. The auto-regressive moving average with exogenous (ARMAX) model structure techniques were used to generate a model converter that best describes the glucose and insulin relationship of the subject. Several combinations of model order were tested and simulated on the subjects. The best model fit criterion and the corresponding time-delay were identified. The estimated peak value was also compared to the real peak value. The finding shows that different patient can be represented with different model structure and different time-delay. However, there is a need to categorize the patients according to their appropriate model structure, particularly on the time-delay unit. Additional clinical parameter and a more extensive data set may be required to ensure the structure of the model precisely describe the glucose-insulin interaction of the patient.eninfo:eu-repo/semantics/closedAccesscritical careglucose-insulin systemglycemic controlARMAX modelPredictive ControlAlgorithmConference Object10.1109/IECBES48179.2021.9398780