Methodological issues in validating decision-support systems for insulin dosage adjustment

The promise of technology in the management of the patient with diabetes

The advent of technology has brought many improvements in the management of individual aspects of the care of the patient with diabetes. However, the best management requires communication between systems to enable the clinician to coordinate these various aspects. This article reviews examples of the application of technology to the individual aspects of care. It also discusses the problems and promise of technology to improve overall care management.

DIABNET: a qualitative model-based advisory system for therapy planning in gestational diabetes

An intelligent decision support system for the analysis of home monitoring data and therapy planning in gestational diabetes is presented. The paper describes the integration of qualitative and quantitative reasoning modules within the DIABNET advisory system. The system kernel is a qualitative model of the physiological processes involved in the glucose metabolism of this type of diabetic patient. A causal probabilistic network (CPN) has been used to represent the qualitative model in order to manage uncertain and missing monitoring data. The DIABNET inputs are the patient's available ambulatory monitoring data, and the output is a dietary and insulin therapy adjustment that includes initiation of insulin therapy, quantitative insulin dose changes and qualitative diet and schedule modifications. Over periods of up to seven days, monitoring data are analysed by the CPN to detect any diet or insulin therapy items which may require modification. The qualitative insulin needs are translated into a quantitative proposal in line with the characteristics of the patient and the modification strategies usually used by doctors. The first evaluation of the system has been accomplished, the encouraging results of which are also presented.

Computer assisted diabetes care: a 6-year retrospective

Over the past 6 years we have designed a number of computer-based prototypes for the provision of therapeutic advice and the generation of glycaemic predictions in insulin-dependent (type 1) diabetic patients. In this paper we provide an overview of some of this work, and describe our experiences in trying to develop such methods for clinical use. We review, as an example, a model of the glucoregulatory system which has been developed for patient and medical staff education about type 1 diabetes mellitus, as well as possibly for therapeutic use. Using individualised parameter values the predictions of the model can be applied to generate 24-h simulations of patient blood glucose profiles. Previous preliminary retrospective validation work performed with this model has revealed a mean predictive accuracy for blood glucose simulations of approximately 2 mmol/l. Conceptual limitations of such modelling approaches are considered. We comment that such "mechanistic' models may lack the necessary sophistication and flexibility to represent the complexity of the human glucoregulatory system and the challenges it has to face. Although such methodologies may therefore not be suitable for safe and effective application in routine clinical practice, we conclude that the evolution of such a system for demonstration/educational purposes could have widespread clinical utility as an interactive teaching tool.

Application of computers in diabetes care--a review. II. Computers for decision support and education

This paper concludes a review of the contribution of information technology to diabetes care. The role of decision support tools for planning insulin therapy using various clinical algorithms, knowledge-based approaches and interactive simulations is discussed, and certain telemedicine approaches to therapy are considered. The validation and clinical evaluation of these tools is reviewed, as is the likely impact of the routine clinical application of implantable/non-invasive blood glucose monitoring devices. Finally, the application of computers as teaching tools is considered, and ways in which such educational approaches might be applied for disseminating the benefits of the Diabetes Control and Complications Trial more widely are discussed.

Combining rule-based reasoning and mathematical modelling in diabetes care

A prototype computer system utilising a model of carbohydrate metabolism linked to an expert system is described. The prototype which integrates quantitative and qualitative computational methodologies can be used to predict blood glucose profiles and adjust insulin doses in insulin-dependent (type I) diabetic subjects. A feedback loop insulin-dosage optimisation procedure which allows quantitative advice to be generated is also described. Possible clinical applications for the system, which is intended for educational use and clinically as a research tool to try and attain normoglycaemia, are discussed.