Application of physicians' predictions of meal and exercise effects on blood glucose control to a computer simulation

Artificial Intelligence in Decision Support Systems for Type 1 Diabetes

Type 1 diabetes (T1D) is a chronic health condition resulting from pancreatic beta cell dysfunction and insulin depletion. While automated insulin delivery systems are now available, many people choose to manage insulin delivery manually through insulin pumps or through multiple daily injections. Frequent insulin titrations are needed to adequately manage glucose, however, provider adjustments are typically made every several months. Recent automated decision support systems incorporate artificial intelligence algorithms to deliver personalized recommendations regarding insulin doses and daily behaviors. This paper presents a comprehensive review of computational and artificial intelligence-based decision support systems to manage T1D. Articles were obtained from PubMed, IEEE Xplore, and ScienceDirect databases. No time period restrictions were imposed on the search. After removing off-topic articles and duplicates, 562 articles were left to review. Of those articles, we identified 61 articles for comprehensive review based on algorithm evaluation using real-world human data, in silico trials, or clinical studies. We grouped decision support systems into general categories of (1) those which recommend adjustments to insulin and (2) those which predict and help avoid hypoglycemia. We review the artificial intelligence methods used for each type of decision support system, and discuss the performance and potential applications of these systems.

Home blood glucose prediction: clinical feasibility and validation in islet cell transplantation candidates

AIMS/HYPOTHESIS

Diabetic subjects do home monitoring to substantiate their success (or failure) in meeting blood glucose targets set by their providers. To succeed, patients require decision support, which, until now, has not included knowledge of future blood glucose levels or of hypoglycaemia. To remedy this, we devised a glucose prediction engine. This study validates its predictions.

METHODS

The prediction engine is a computer program that accesses a central database in which daily records of self-monitored blood glucose data and life-style parameters are stored. New data are captured by an interactive voice response server on-line 24 h a day, 7 days a week. Study subjects included 24 patients with debilitating hypoglycaemia (unawareness), which qualified them for islet cell transplantation. Comparison of each prediction with the actually observed data was done using a Clarke Error Grid (CEG). Patients and providers were blinded as to the predictions.

RESULTS

Prior to transplantation, a total of 31,878 blood glucose levels were reported by the study subjects. Some 31,353 blood glucose predictions were made by the engine on a total of 8,733 days-used. Of these, 79.4% were in the clinically acceptable Zones of the CEG. Of 728 observed episodes of hypoglycaemia, 384 were predicted. After transplantation, a total of 45,529 glucose measurements were reported on a total of 12,906 days-used. Some 42,316 glucose predictions were made, of which 97.5% were in the acceptable CEG Zones A and B. Successful transplantation eliminated hypoglycaemia, improved glycaemic control, lowered HbA(1)c and freed 10 of 24 patients from daily insulin therapy.

CONCLUSIONS/INTERPRETATION

It is clinically feasible to generate valid predictions of future blood glucose levels. Prediction accuracy is related to glycaemic stability. Risk of hypoglycaemia can be predicted. Such knowledge may be useful in self-management.

Home blood glucose prediction: validation, safety, and efficacy testing in clinical diabetes

BACKGROUND

Patients with diabetes do daily self-monitoring of blood glucose (SMBG). For such patients, we devised an engine that predicts not only the expected blood glucose level at the next meal but also the pending risks of hypoglycemia. The purpose of this study was to validate the predictions and provide evidence of the safety and efficacy of using predicted data in dosing decision support for routine patient care.

RESEARCH DESIGN AND METHODS

The prediction engine is a computer program that accesses a central database into which daily records of self-monitored blood glucose data are captured by direct access either across the WWW or by an interactive voice response service on-line 24/7. Validation was done by comparison of predicted values to the subsequently observed data using a Clarke Error Grid. Safety focused on body weight and the frequency of hypoglycemia. Efficacy was judged according to glycated hemoglobin and daily insulin dosages. The experimental design contrasted patients in the tight control (TC) group who had been recently converted to intensified (basal-bolus) therapy with patients in the poor control (PC) group on conventional therapy and who were referred to begin intensified therapy. Both groups accessed the remote database to report their daily SMBG. Predicted glucose values were used in dosing decision support for the PC but not the TC group.

RESULTS

Over the 6-month study period a total of 30,129 blood glucose levels were reported by the 54 study patients, and some 24,953 blood glucose predictions were made. Of these, 83% were in the clinically acceptable zones of the Clarke Error Grid. When these data were used for dosing decision support in the PC group, glycated hemoglobin levels fell significantly from 9.7 +/- 1.7% to 7.9 +/- 1.2%, and hypoglycemia dropped fourfold. Total daily insulin doses declined 22%, while body weight remained constant. In the parallel TC group (n = 24), glycated hemoglobin also fell, but only slightly from 7.6 +/- 0.9% to 7.2 +/- 1.1%, while daily insulin doses, rates of hypoglycemia and body weight remained constant.

CONCLUSIONS

A novel engine is capable of generating meaningful predictions of blood glucose levels. Use of these validated predictions in decision support for managing medication doses proved safe and efficacious.

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.