TIME-SERIES ANALYSIS AND CONTROL OF BLOOD-GLUCOSE LEVELS IN DIABETIC-PATIENTS

This paper describes features of a computer-based decision support sys tem which is being developed to assist in the management of insulin-de pendent diabetic patients. The clinical context is the provision of ad vice on the adjustment of the basic insulin regimen such as occurs at regular visits to the clinician. The integrated system combines data p rocessing and interpretation, generation of qualitative advice and tes ting the implications of that advice using a glucose/insulin dynamic s imulator. The two major features described in this paper are time seri es analysis of blood glucose data, and their interpretation in relatio n to the provision of advice for controlling the patient's blood gluco se level. It is demonstrated that two approaches may be adopted in suc h time series analysis: an intuitive approach, manipulating symbolic r epresentations of the data, and formal time series methods which decom pose the series into clinically related components.