Fuzzy decision support system for body fluid replacement

Surgical fluid replacement is a critical issue in medicine as the fluid volume excess or deficit can both complicate the patient's condition. Currently, the administration of fluid volume is carried out primarily based on the experience and expertise of the anaesthetist as there is no analytical method available to estimate the patient's fluid level. The development of a decision support system (DSS) to assist the anaesthetist in estimating the required fluid infusion rate fora particular patient has been the focus of the research work reported in this thesis. The DSS is developed based on Fuzzy Logic Control (FLC) technique which is ideal for developing input/output models in an unstructured and imprecise environment. The fuzzy rules used in the DSS are derived automatically from the clinical data produced in surgical operations. The DSS employs a Multi Rule Base (MRB) learning scheme to adapt its model according to the significant variation in the physiological parameters of a patient. The performance of the developed algorithms is validated through experimental work using clinical data. The results obtained so far are encouraging.