An integrated approach for the computer-assisted treatment of diabetic patients on insulin

A prototype computer system has been developed to provide advice on the day-to-day adjustment of carbohydrate intake and insulin regimen in the insulin-dependent diabetic patient. The system also produces a 24-h simulation of the patient's blood glucose profile based on these adjustments. Advice is generated by a qualitative knowledge-based system which suggests what the next step in improving glycaemic control might be for a given patient, e.g. 'decrease morning short-acting insulin by 2 units'. The quantitative simulator module contains two different mathematical models. The first is a non-linear model in differential equation form which consists of a one-compartment glucose model linked to a model with free and bound insulin compartments. This physiological model is solved by a general-purpose simulation engine. The second is a linear systems model which uses a transfer function to describe the insulin input/blood glucose response relationship for individual diabetic patients. Results of a preliminary medical validation are presented.