Computer assisted diabetes care: a 6-year retrospective

Predictive modeling of multi-class diabetes mellitus using machine learning and filtering iraqi diabetes data dynamics

Diabetes is a persistent metabolic disorder linked to elevated levels of blood glucose, commonly referred to as blood sugar. This condition can have detrimental effects on the heart, blood vessels, eyes, kidneys, and nerves as time passes. It is a chronic ailment that arises when the body fails to produce enough insulin or is unable to effectively use the insulin it produces. When diabetes is not properly managed, it often leads to hyperglycemia, a condition characterized by elevated blood sugar levels or impaired glucose tolerance. This can result in significant harm to various body systems, including the nerves and blood vessels. In this paper, we propose a multiclass diabetes mellitus detection and classification approach using an extremely imbalanced Laboratory of Medical City Hospital data dynamics. We also formulate a new dataset that is moderately imbalanced based on the Laboratory of Medical City Hospital data dynamics. To correctly identify the multiclass diabetes mellitus, we employ three machine learning classifiers namely support vector machine, logistic regression, and k-nearest neighbor. We also focus on dimensionality reduction (feature selection-filter, wrapper, and embedded method) to prune the unnecessary features and to scale up the classification performance. To optimize the classification performance of classifiers, we tune the model by hyperparameter optimization with 10-fold grid search cross-validation. In the case of the original extremely imbalanced dataset with 70:30 partition and support vector machine classifier, we achieved maximum accuracy of 0.964, precision of 0.968, recall of 0.964, F1-score of 0.962, Cohen kappa of 0.835, and AUC of 0.99 by using top 4 feature according to filter method. By using the top 9 features according to wrapper-based sequential feature selection, the k-nearest neighbor provides an accuracy of 0.935 and 1.0 for the other performance metrics. For our created moderately imbalanced dataset with an 80:20 partition, the SVM classifier achieves a maximum accuracy of 0.938, and 1.0 for other performance metrics. For the multiclass diabetes mellitus detection and classification, our experiments outperformed conducted research based on the Laboratory of Medical City Hospital data dynamics.

Application of Artificial Intelligence in Discovery and Development of Anticancer and Antidiabetic Therapeutic Agents

Spectacular developments in molecular and cellular biology have led to important discoveries in cancer research. Despite cancer is one of the major causes of morbidity and mortality globally, diabetes is one of the most leading sources of group of disorders. Artificial intelligence (AI) has been considered the fourth industrial revolution machine. The most major hurdles in drug discovery and development are the time and expenditures required to sustain the drug research pipeline. Large amounts of data can be explored and generated by AI, which can then be converted into useful knowledge. Because of this, the world's largest drug companies have already begun to use AI in their drug development research. In the present era, AI has a huge amount of potential for the rapid discovery and development of new anticancer drugs. Clinical studies, electronic medical records, high-resolution medical imaging, and genomic assessments are just a few of the tools that could aid drug development. Large data sets are available to researchers in the pharmaceutical and medical fields, which can be analyzed by advanced AI systems. This review looked at how computational biology and AI technologies may be utilized in cancer precision drug development by combining knowledge of cancer medicines, drug resistance, and structural biology. This review also highlighted a realistic assessment of the potential for AI in understanding and managing diabetes.

Artificial intelligence and diabetes technology: A review

Artificial intelligence (AI) is widely discussed in the popular literature and is portrayed as impacting many aspects of human life, both in and out of the workplace. The potential for revolutionizing healthcare is significant because of the availability of increasingly powerful computational platforms and methods, along with increasingly informative sources of patient data, both in and out of clinical settings. This review aims to provide a realistic assessment of the potential for AI in understanding and managing diabetes, accounting for the state of the art in the methodology and medical devices that collect data, process data, and act accordingly. Acknowledging that many conflicting definitions of AI have been put forth, this article attempts to characterize the main elements of the field as they relate to diabetes, identifying the main perspectives and methods that can (i) affect basic understanding of the disease, (ii) affect understanding of risk factors (genetic, clinical, and behavioral) of diabetes development, (iii) improve diagnosis, (iv) improve understanding of the arc of disease (progression and personal/societal impact), and finally (v) improve treatment.

Dynamic Interactive Educational Diabetes Simulations Using the World Wide Web: An Experience of More Than 15 Years with AIDA Online

Background. AIDA is a widely available downloadable educational simulator of glucose-insulin interaction in diabetes. Methods. A web-based version of AIDA was developed that utilises a server-based architecture with HTML FORM commands to submit numerical data from a web-browser client to a remote web server. AIDA online, located on a remote server, passes the received data through Perl scripts which interactively produce 24 hr insulin and glucose simulations. Results. AIDA online allows users to modify the insulin regimen and diet of 40 different prestored "virtual diabetic patients" on the internet or create new "patients" with user-generated regimens. Multiple simulations can be run, with graphical results viewed via a standard web-browser window. To date, over 637,500 diabetes simulations have been run at AIDA online, from all over the world. Conclusions. AIDA online's functionality is similar to the downloadable AIDA program, but the mode of implementation and usage is different. An advantage to utilising a server-based application is the flexibility that can be offered. New modules can be added quickly to the online simulator. This has facilitated the development of refinements to AIDA online, which have instantaneously become available around the world, with no further local downloads or installations being required.

Development of AIDA v4.3b Diabetes Simulator: Technical Upgrade to Support Incorporation of Lispro, Aspart, and Glargine Insulin Analogues

Introduction. AIDA is an interactive educational diabetes simulator available on the Internet without charge since 1996 (accessible at: http://www.2aida.org/). Since the program’s original release, users have developed new requirements, with new operating systems coming into use and more complex insulin management regimens being adopted. The current work has aimed to design a comprehensive diabetes simulation system from both a clinical and information technology perspective.Methods. A collaborative development is taking place with a new generic model of subcutaneous insulin absorption, permitting the simulation of rapidly-acting and very long-acting insulin analogues, as well as insulin injections larger than 40 units. This novel, physiological insulin absorption model has been incorporated into AIDA v4. Technical work has also been undertaken to install and operate the AIDA software within a DOSBox emulator, to ensure compatibility with Windows XP, Vista and 7 operating systems as well as Apple Macintosh computers running Parallels PC emulation software.Results. Plasma insulin simulations are demonstrated following subcutaneous injections of a rapidly-acting insulin analogue, a short-acting insulin preparation, intermediate-acting insulin, and a very long-acting insulin analogue for injected insulin doses up to 60 units of insulin.Discussion.The current work extends the useful life of the existing AIDA v4 program.

Neural network based glucose - insulin metabolism models for children with Type 1 diabetes

In this paper two models for the simulation of glucose-insulin metabolism of children with Type 1 diabetes are presented. The models are based on the combined use of Compartmental Models (CMs) and artificial Neural Networks (NNs). Data from children with Type 1 diabetes, stored in a database, have been used as input to the models. The data are taken from four children with Type 1 diabetes and contain information about glucose levels taken from continuous glucose monitoring system, insulin intake and food intake, along with corresponding time. The influences of taken insulin on plasma insulin concentration, as well as the effect of food intake on glucose input into the blood from the gut, are estimated from the CMs. The outputs of CMs, along with previous glucose measurements, are fed to a NN, which provides short-term prediction of glucose values. For comparative reasons two different NN architectures have been tested: a Feed-Forward NN (FFNN) trained with the back-propagation algorithm with adaptive learning rate and momentum, and a Recurrent NN (RNN), trained with the Real Time Recurrent Learning (RTRL) algorithm. The results indicate that the best prediction performance can be achieved by the use of RNN.

A minimal model for glycemia control in critically ill patients

In this paper we propose a modified minimal model to be used for glycemia control in critically ill patients. For various reasons the Bergman minimal model is widely used to describe glucose and insulin dynamics. However, since this model is mostly valid in a rather restrictive setting, it might not be suitable to be used in a model predictive controller. Simulations show that the new model exhibits a similar glycemia behaviour but clinically more realistic insulin kinetics. Therefore it is potentially more suitable for glycemia control. The designed model is also estimated on a set of critically ill patients giving promising results.

Telemedicine and diabetes management: current challenges and future research directions

Telemedicine is lying between fading and future. Several clinical studies and critical reviews have been published recently, but the results are inconclusive and the adoption of telemedicine interventions in clinical practice is slow. This article discusses some of the current problems related to the adoption of telemedicine systems and focuses on the information technology solutions that appear to be most promising for diabetes management in the near future. Context awareness, user modeling, intelligent dialogues, and integrated information systems are presented. Some potential future scenarios for the adoption of telemedicine, which combine novel technologies and new organizational models, are also discussed. Within those scenarios, telemedicine may prove to be a good instrument to support health care providers in the effective management and prevention of diabetes mellitus.

An insulin infusion advisory system for type 1 diabetes patients based on non-linear model predictive control methods

In this paper, an Insulin Infusion Advisory System (IIAS) for Type 1 diabetes patients, which use insulin pumps for the Continuous Subcutaneous Insulin Infusion (CSII) is presented. The purpose of the system is to estimate the appropriate insulin infusion rates. The system is based on a Non-Linear Model Predictive Controller (NMPC) which uses a hybrid model. The model comprises a Compartmental Model (CM), which simulates the absorption of the glucose to the blood due to meal intakes, and a Neural Network (NN), which simulates the glucose-insulin kinetics. The NN is a Recurrent NN (RNN) trained with the Real Time Recurrent Learning (RTRL) algorithm. The output of the model consists of short term glucose predictions and provides input to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. For the development and the evaluation of the IIAS, data generated from a Mathematical Model (MM) of a Type 1 diabetes patient have been used. The proposed control strategy is evaluated at multiple meal disturbances, various noise levels and additional time delays. The results indicate that the implemented IIAS is capable of handling multiple meals, which correspond to realistic meal profiles, large noise levels and time delays.

A real time simulation model of glucose-insulin metabolism for type 1 diabetes patients

In this paper, a simulation model of glucose-insulin metabolism for Type 1 diabetes patients is presented. The proposed system is based on the combination of Compartmental Models (CMs) and artificial Neural Networks (NNs). This model aims at the development of an accurate system, in order to assist Type 1 diabetes patients to handle their blood glucose profile and recognize dangerous metabolic states. Data from a Type 1 diabetes patient, stored in a database, have been used as input to the hybrid system. The data contain information about measured blood glucose levels, insulin intake, and description of food intake, along with the corresponding time. The data are passed to three separate CMs, which produce estimations about (i) the effect of Short Acting (SA) insulin intake on blood insulin concentration, (ii) the effect of Intermediate Acting (IA) insulin intake on blood insulin concentration, and (iii) the effect of carbohydrate intake on blood glucose absorption from the gut. The outputs of the three CMs are passed to a Recurrent NN (RNN) in order to predict subsequent blood glucose levels. The RNN is trained with the Real Time Recurrent Learning (RTRL) algorithm. The resulted blood glucose predictions are promising for the use of the proposed model for blood glucose level estimation for Type 1 diabetes patients.

Retrospective pilot feedback survey of 200 users of the AIDA Version 4 Educational Diabetes Program. 1--Quantitative Survey Data

This column reports a detailed, questionnaire-based, post-release feedback survey of 200 users of the AIDA version 4 educational diabetes simulator. AIDA is a freeware computer program that permits the interactive simulation of plasma insulin and blood glucose profiles for educational, demonstration, self-learning, and research purposes. Since its Internet launch in 1996 over 700,000 visits have been logged to the AIDA Websites-including www.2aida.org-and over 200,000 program copies have been downloaded free-of-charge. The main goals of the current study were: (1) to establish what people have thought about the AIDA program, (2) to assess the utility of the software, and (3) to ascertain how much people have actually used it. An analysis was therefore undertaken of the first 200 feedback forms that were returned by AIDA users. The questionnaire-based survey methodology was found to be robust and reliable. Feedback forms were received from participants in 21 countries. One hundred six of 209 responses (50.7%) were received from people with diabetes, and 36 of 209 (17.2%) from relatives of patients, with lesser numbers from doctors, students, diabetes educators, nurses, pharmacists, and other end users. Please note some respondents fulfilled more than one end-user category, hence the denominator <200; for example, someone with diabetes who was also a doctor. This study has established the feasibility of using a simple feedback form to survey a substantial number of diabetes software users. In addition, it has yielded interesting data in terms of who are the main users of the AIDA program, and has also provided technical (computer) information that has aided the release of a freeware upgrade to the software. In general, users reported finding the program to be of educational value. The majority also felt it would be of interest to diabetes educators and people with diabetes. Most were clear about its limitations as a simulator-based learning tool. The implications of these findings will be discussed.

Interactive Health Communication Applications for people with chronic disease

BACKGROUND

Interactive Health Communication Applications (IHCAs) are computer-based, usually web-based, information packages for patients that combine health information with at least one of social support, decision support, or behaviour change support. These are innovations in health care and their effects on health are uncertain.

OBJECTIVES

To assess the effects of IHCAs for people with chronic disease.

SEARCH STRATEGY

We designed a four-part search strategy. First, we searched electronic bibliographic databases for published work; second, we searched the grey literature; and third, we searched for ongoing and recently completed clinical trials in the appropriate databases. Finally, researchers of included studies were contacted, and reference lists from relevant primary and review articles were followed up. As IHCAs require relatively new technology, the search time period commenced at 1990, where possible, and ran until 31 December 2003.

SELECTION CRITERIA

Randomised controlled trials (RCTs) of IHCAs for adults and children with chronic disease.

DATA COLLECTION AND ANALYSIS

One reviewer screened abstracts for relevance. Two reviewers screened all candidate studies to determine eligibility, apply quality criteria, and extract data from included studies. Authors of included RCTs were contacted for missing data. Results of RCTs were pooled using random-effects model with standardised mean differences (SMDs) for continuous outcomes and odds ratios for binary outcomes; heterogeneity was assessed using the I(2 )statistic.

MAIN RESULTS

We identified 24 RCTs involving 3739 participants which were included in the review.IHCAs had a significant positive effect on knowledge (SMD 0.46; 95% confidence interval (CI) 0.22 to 0.69), social support (SMD 0.35; 95% CI 0.18 to 0.52) and clinical outcomes (SMD 0.18; 95% CI 0.01 to 0.35). Results suggest it is more likely than not that IHCAs have a positive effect on self-efficacy (a person's belief in their capacity to carry out a specific action) (SMD 0.24; 95% CI 0.00 to 0.48). IHCAs had a significant positive effect on continuous behavioural outcomes (SMD 0.20; 95% CI 0.01 to 0.40). Binary behavioural outcomes also showed a positive effect for IHCAs, although this result was not statistically significant (OR 1.66; 95% CI 0.71 to 3.87). It was not possible to determine the effects of IHCAs on emotional or economic outcomes.

AUTHORS' CONCLUSIONS

IHCAs appear to have largely positive effects on users, in that users tend to become more knowledgeable, feel better socially supported, and may have improved behavioural and clinical outcomes compared to non-users. There is a need for more high quality studies with large sample sizes to confirm these preliminary findings, to determine the best type and best way to deliver IHCAs, and to establish how IHCAs have their effects for different groups of people with chronic illness.

Diabetes website review: www.2aida.org

Previous Diabetes Information Technology & WebWatch columns have addressed the use of diabetes simulators, and, in particular, aspects of the AIDA software. AIDA is a freeware computer program, which simulates the interaction of carbohydrates and insulin administered in people with insulin-dependent (type 1) diabetes mellitus. The program is intended to be used as an educational support tool, and is available via the Internet without charge from www.2aida.org. In this article, the AIDA Website is described and reviewed in terms of both content and functionality. This popular non-commercial Internet site provides free access to a downloadable PC version of AIDA, as well as access to a Web-based version of the simulator that can be run online (accessible directly at: www.2aida.net). User feedback suggests that the Website and the AIDA software have been of significant interest and value to many patients, their relatives and carers, students, and a variety of health-care professionals and researchers. The interactive and dynamic nature of the simulations adds a real-life dimension to the Web-based educational material, and the software is complemented by a substantial amount of supporting information at the Website. The on-going collection of subjective feedback continues to provide anecdotal evidence of the utility of the software, and this will hopefully be corroborated by results from more formal and objective evaluations. The future potential of diabetes simulators, in both education and research, is becoming increasingly apparent, and the AIDA Website is evolving accordingly.

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.

Derivative weighted active insulin control modelling and clinical trials for ICU patients

Close control of blood glucose levels significantly reduces vascular complications in Type 1 and Type 2 diabetic individuals. Heavy derivative controllers using the data density available from emerging biosensors are developed to provide tight, optimal control of elevated blood glucose levels, while robustly handling variation in patient response. A two-compartment glucose regulatory system model is developed for intravenous infusion from physiologically verified subcutaneous infusion models enabling a proof-of-concept clinical trial at the Christchurch Hospital Department of Intensive Care Medicine. This clinical trial is the first of its kind to test a high sample rate feedback control algorithm for tight glucose regulation. The clinical trial results show tight control with reductions of 79-89% in blood glucose excursions for an oral glucose tolerance test. Experimental performance is very similar to modelled behaviour. Results include a clear need for an additional accumulator dynamic for insulin behaviour in transport to the blood and strong correlation of 10% or less between modelled insulin infused and the amounts used in clinical trials. Finally, the heavy derivative PD control approach is seen to be able to bring blood glucose levels below the (elevated) basal level, showing the potential for truly tight control.

Computerised decision-support tools in diabetes care: hurdles to implementation

In this Diabetes Information Technology & WebWatch column hurdles to the use of computerised decision-support tools in clinical diabetes care will be considered. The clinical background with respect to insulin-dependent (type 1) diabetes mellitus and the Diabetes Control and Complications Trial is reviewed, and an overview is given of various computer applications. The use of decision-support tools is discussed, and the importance of identifying the proposed user, e.g., health-care professional, student, or patient, is highlighted. Validation/evaluation issues are considered as important topics that remain to be properly addressed for many decision-support prototypes. The column concludes by highlighting that in this era of evidence-based medicine well-conducted, rigorous evaluation and validation studies are required to inform decisions about whether or not to make use of current computerised decision-support prototypes.

Automated insulin infusion trials in the intensive care unit

The objective is to demonstrate the effectiveness of a simple automated insulin infusion for controlling the rise and duration of blood glucose excursion following a glucose challenge in critically ill patients with impaired glucose tolerance. A two-compartment model of the glucose regulatory system was developed for intravenous infusion control design. On two subsequent days a critically ill patient with impaired glucose tolerance was given a 75 g oral glucose tolerance test (OGTT), and the glucose level was measured every 15 min. The first day's data were used to design a heavy-derivative insulin infusion controller for the second day. Ethics approval was granted for this test. Five patients were studied. In four patients, the magnitude and duration of blood glucose excursion were reduced over 50%. Fasting level was reduced 15%, from an average of 7.2 mmol/L to 6.1 mmol/L. The fifth patient's results showed a diminished response due to the antagonistic effects of hydrocortisone on insulin, a data point not provided prior to testing. Modeling to account for this effect yielded better correlation with the test. The automated algorithm provided rapid, effective control of the blood glucose rise in response to an OGTT input. These results highlight the effectiveness of automated infusions for regulating blood glucose rise and excursions, and the potential of this approach for non-hospitalized individuals.

Research use of the AIDA www.2aida.org diabetes software simulation program: a review--part 2. Generating simulated blood glucose data for prototype validation

The purpose of this review is to describe research applications of the AIDA diabetes software simulator. AIDA is a computer program that permits the interactive simulation of insulin and glucose profiles for teaching, demonstration, and self-learning purposes. Since March/April 1996 it has been made freely available on the Internet as a noncommercial contribution to continuing diabetes education. Up to May 2003 well over 320,000 visits have been logged at the main AIDA Website--www.2aida.org--and over 65,000 copies of the AIDA program have been downloaded free-of-charge. This review (the second of two parts) overviews research projects and ventures, undertaken for the most part by other research workers in the diabetes computing field, that have made use of the freeware AIDA program. As with Part 1 of the review (Diabetes Technol Ther 2003;5:425-438) relevant research work was identified in three main ways: (i) by personal (e-mail/written) communications from researchers, (ii) via the ISI Web of Science citation database to identify published articles which referred to AIDA-related papers, and (iii) via searches on the Internet. Also, in a number of cases research students who had sought advice about AIDA, and diabetes computing in general, provided copies of their research dissertations/theses upon the completion of their projects. Part 2 of this review highlights some more of the research projects that have made use of the AIDA diabetes simulation program to date. A wide variety of diabetes computing topics are addressed. These range from learning about parameter interactions using simulated blood glucose data, to considerations of dietary assessments, developing new diabetes models, and performance monitoring of closed-loop insulin delivery devices. Other topics include evaluation/validation research usage of such software, applying simulated blood glucose data for prototype training/validation, and other research uses of placing technical information on the Web. This review confirms an unexpected but useful benefit of distributing a medical program, like AIDA, for free via the Internet--demonstrating how it is possible to have a synergistic benefit with other researchers--facilitating their own research projects in related medical fields. A common theme that emerges from the research ventures that have been reviewed is the use of simulated blood glucose data from the AIDA software for preliminary computer lab-based testing of other decision support prototypes. Issues surrounding such use of simulated data for separate computer prototype testing are considered further.

Active insulin infusion using optimal and derivative-weighted control

Close control of blood glucose levels significantly reduces vascular complications in Type I diabetes. A control method for the automation of insulin infusion that utilizes emerging technologies in blood glucose biosensors is presented. The controller developed provides tighter, more optimal control of blood glucose levels, while accounting for variation in patient response, insulin employed and sensor bandwidth. Particular emphasis is placed on controller simplicity and robustness necessary for medical devices and implants.A PD controller with heavy emphasis on the derivative term is found to outperform the typically used proportional-weighted controllers in glucose tolerance and multi-meal tests. Simulation results show reductions of over 50% in the magnitude and duration of blood glucose excursions from basal levels. A closed-form steady state optimal solution is also developed as a benchmark, and results in a flat glucose response. The impact and trade-offs associated with sensor bandwidth, sensor lag and proportional versus derivative-based control methods are illustrated. Overall, emerging blood glucose sensor technologies that enable frequent measurement are shown to enable more effective, automated control of blood glucose levels within a tight, acceptable range for Type I and II diabetic individuals.

Why are people downloading the freeware AIDA diabetes computing software program: a pilot study

The purpose of this paper is to report a pilot survey about why people are downloading the AIDA interactive educational diabetes simulator. AIDA is a diabetes computer program that permits the interactive simulation of plasma insulin and blood glucose profiles for teaching, demonstration, and self-learning purposes. It has been made freely available, without charge, on the Internet as a noncommercial contribution to continuing diabetes education. Since its launch in 1996 well over 200,000 visits have been logged at the main AIDA Website--www.2aida.org--and over 40,000 copies of the AIDA program have been downloaded free-of-charge. This article documents a pilot survey of comments left by Website visitors while they were downloading the AIDA software, before they had a chance to actually use the program. The overall paradigm adopted for this study has endeavored to establish why people are resorting to the Internet to obtain diabetes information. Specific intended goals of the study were: (1) to demonstrate ongoing use of the World Wide Web for surveying diabetes software users by obtaining their free-text comments; (2) to identify what sort of things people were planning to do with the AIDA software simulator; and (3) to more generally gain some insight into why people are turning to the Web for healthcare-related information. The Internet-based survey methodology was found to be robust and reliable. Over an 8-month period (from February 2, 2001 to October 1, 2001) 642 responses were received. During the corresponding period 2,248 actual visits were made to the Website survey page--giving a response rate to this pilot study of 28.6%. Responses were received from participants in over 56 countries--although over half of these (n = 343; 53.4%) originated from the United States and United Kingdom. Two hundred forty-four responses (38.0%) were received from patients with diabetes, and 73 (11.4%) from relatives of patients, with fewer responses from doctors, students, diabetes educators, nurses, pharmacists, and other end users. This pilot survey has confirmed the feasibility of using the Internet to obtain free-text comments, at no real cost, from a large number of medical software downloaders/users. The survey has also offered a valuable insight into why members of the public are turning to the Internet for medical information. Furthermore it has provided useful information about why people are actually downloading the AIDA interactive educational "virtual diabetes patient" simulator.

A randomized trial of a computerized versus an audio-booklet decision aid for women considering post-menopausal hormone replacement therapy

Decision aids (DAs) are increasingly being developed to help patients make shared health care decisions with their practitioners. There are no formal comparisons of the efficacy of different delivery methods. Interactive computerized delivery methods have the advantage of allowing patients control over flow of information and to receive feedback on their comprehension. The purpose of this study was to compare the efficacy of an interactive computerized DA for women considering long-term hormone replacement therapy, to that of a validated audio-booklet version of the same intervention. Fifty-one peri-menopausal women were randomized to use either the computerized or the standard audio-booklet version of the DA. The computerized version presented identical information with the addition of feedback modules to reinforce the participant's understanding. The patients were interviewed with a pre- and post-intervention questionnaire. The computerized DA improved realistic expectations by 52.7% over baseline versus 27.6% with the audio-booklet (P=0.015). Knowledge (Kn) scores improved by 17.5 and 8.4% for the computer and standard DA groups, respectively (P=0.019). The results of this study have implications for future DA design, and other areas where patient Kn and understanding are important, such as in the setting of informed consent.

Using the AIDA--www.2aida.org--diabetes simulator. Part 2: recommended training requirements for health-carers planning to teach with the software

The purpose of this article is to document some recommended training requirements for health-carers planning to teach using the AIDA interactive educational diabetes simulator. AIDA is a diabetes computer program that permits the interactive simulation of plasma insulin and blood glucose profiles for teaching, demonstration, and self-learning purposes. It has been made freely available, without charge, on the Internet as a noncommercial contribution to continuing diabetes education. Since its launch in 1996 over 200,000 visits have been logged at the AIDA Website--www.2aida.org--and over 40,000 copies of the AIDA program have been downloaded free-of-charge. This report describes various training requirements that are recommended for health-care professionals who are interested in teaching with the software. Intended goals of this article are to answer possible questions from teachers using the program, highlight some minimum recommended training requirements for the software, suggest some "hints and tips" for teaching ideas, explain the importance of performing more studies/trials with the program, overview randomised controlled trial usage of the software, and highlight the importance of obtaining feedback from lesson participants. The recommendations seem to be straightforward and should help in formalising training with the program, as well as in the development of a network of teachers "accredited" to give lessons using the software. This report, together with the previous article (Part 1, Diabetes Technol Ther 2002;4:401-414), highlights the utility of providing guidelines and suggesting recommended training requirements for health-carers planning to make use of educational medical/diabetes software.

Who is downloading the freeware AIDA v43 interactive educational diabetes simulator? An audit of 2437 downloads

The purpose of this paper is to report an audit of 2437 downloads of the AIDA interactive educational diabetes simulator. AIDA is a diabetes computer program that permits the interactive simulation of plasma insulin and blood glucose profiles for educational, demonstration, and self-learning purposes. It has been made freely available, without charge, on the Internet as a noncommercial contribution to continuing diabetes education. Since its launch in 1996 over 200000 visits have been logged at the AIDA Website - www.2aida.org - and over 37000 copies of the AIDA program have been downloaded free-of-charge. This report documents an audit of downloaders of the software, with the intended goals of the study being to demonstrate the use of the Internet for auditing and surveying diabetes software users and to confirm the proportion of patients with diabetes and their relatives who are actually making use of the AIDA v4.3 program. The Internet-based survey methodology was confirmed to be robust and reliable. Over a 7(1/2)-month period (from mid-July 2000 to early March 2001) 2437 responses were received. During the corresponding period 4100 actual downloads of the software were independently logged via the same route at the AIDA Website - giving a response rate to this audit of 59.4%. Responses were received from participants in 61 countries - although over half of these (n = 1533; 62.9%) originated from the United States and United Kingdom. Of these responses 1,361 (55.8%) were received from patients with diabetes and 303 (12.4%) from relatives of patients, with fewer responses from doctors, diabetes educators, students, nurses, pharmacists, and other end users. This study has confirmed the feasibility of using the Internet to survey, at no real cost, a large number of medical software downloaders/users. In addition, it has yielded up-to-date and interesting data about who are the main downloaders of the AIDA program.

Further user comments regarding usage of an interactive educational diabetes simulator (AIDA)

This "Diabetes Information Technology & WebWatch" column continues the diabetes simulation theme from previous issues and overviews various user experience with the AIDA v4 interactive educational freeware diabetes simulator. AIDA is a diabetes computer program that permits the interactive simulation of plasma insulin and blood glucose (BG) profiles for educational, demonstration, and self-learning purposes. It has been made freely available, without charge, via the Web as a noncommercial contribution to continuing diabetes education. Since its Internet launch in 1996, over 145,000 visits have been logged at the AIDA Website--www.2aida.org--and over 29,000 copies of the program have been downloaded, free of charge. While these statistics may appear impressive, they do not tell the personal story of how people have been making use of the software, and what they actually think about the program. In this respect, this column documents some of the independent user comments about AIDA sent in spontaneously via electronic mail (email) by patients with diabetes and their relatives, as well as by health-care professionals. Comments posted to diabetes newsgroups and diabetes email lists, as well as a selection of those which have been found at other, linked, diabetes Websites are also highlighted.

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.

Computer-assisted instructions for patients with bronchial asthma

We produced computer-assisted instruction (CAI) software for bronchial asthma patients (asthma educational system with computer-assisted instruction; ASTCAI) to assist in self-management and avoid asthmatic attacks and death. ASTCAI is a question-and-answer program operating in a multimedia environment, and was evaluated from questionnaires which 33 patients were asked. Thirty-two patients could perform ASTCAI without any assistance. The responses of 31 patients (94%) indicated that they had no difficulty with manipulation, and 29 patients (88%) stated that the program was beneficial to control of their asthma. Elderly patients (over 65) required more time than younger adults. Emergency visits or admissions of at least 1 year after the first CAI trial decreased in eight out of 26 patients, while only two patients deteriorated compared to the previous year. Our results show that CAI is feasible for most patients, and through active self-learning CAI can improve motivation for self-management as well as supplement the physician's instructions.

Experience with the Internet release of AIDA v4.0--http://www.diabetic.org.uk.aida.htm--an interactive educational diabetes simulator

AIDA v4.0 is a freeware computer program that permits the interactive simulation of plasma insulin and blood glucose profiles for demonstration and teaching purposes. It has been made freely available, without charge, on the World Wide Web as a noncommercial contribution to continuing diabetes education. Since its Internet launch in 1996 over 23,000 people have visited the AIDA Web site (http://www.diabetic.org.uk/aida.htm) and over 7,750 copies of the program have been downloaded gratis. This report overviews the Internet release of AIDA v4.0 and provides examples of the simulator in operation. The concept of a "virtual diabetic patient" is introduced. This provides an electronic representation of a patient with diabetes that can be used for self-learning/teaching/demonstration purposes.

Compartmental models for glycaemic prediction and decision-support in clinical diabetes care: promise and reality

This paper reviews and critically appraises the application of compartmental models for generating glycaemic predictions and offering clinical decision support in diabetes care. Comparisons are made with alternative algorithmic-based approaches. Unresolved issues raised for model-based techniques include the relative lack of input data necessary for generating reasonable blood glucose predictions, and the high level of uncertainty associated with such predictions which limits their use as guides for therapeutic insulin-dosage adjustments. It is concluded that compartmental model-based approaches, while not offering much benefit for clinical/therapeutic application, will have a role to play as research tools and for educational use. By contrast it is proposed that algorithmic-based approaches, especially in conjunction with telemedicine and Internet applications, are likely to see growing use for day-to-day therapeutic decision support. Randomised controlled clinical trials however will be required, together with other evaluation efforts, before algorithmic-based approaches-like any other clinical technique-can be widely adopted into routine medical practice.

Preliminary experience with the Internet release of AIDA--an interactive educational diabetes simulator

This paper overviews the Internet release of AIDA, a freeware interactive educational diabetes simulator. Since its release on the World Wide Web as a non-commercial contribution to continuing diabetes education over 14,000 people have visited the AIDA Web site--http:/(/)www.diabetic.org.uk/aida.htm--and over 5000 copies of the program have been downloaded, without charge. User responses thus far have been very encouraging. Example feedback and clinical experience reported by two insulin-dependent (type 1) diabetic patients, a patient's carer, the father of a diabetic teenager, a diabetes doctor and nurse educator, an endocrinologist and a postgraduate educator are presented. While such anecdotal, qualitative assessments are worthwhile and form a necessary step in the overall evaluation process--they are clearly subjective in nature and fully recognised as such. Given this, definitive outcome measures are highlighted as being required for the next stage in the evaluation process, and various objective evaluation criteria are proposed. A general protocol for the evaluation of interactive educational simulation tools, like AIDA, with patients is described and the concept of applying this in multiple centres--as a way of increasing study sample sizes--is discussed. It is highlighted that such a protocol could also be used to objectively compare a number of different interactive educational diabetes simulators. Clinicians who are interested in collaborating by enrolling patients into such a study are invited to contact the author, by email, at aida@globalnet.co.uk

Controlling blood glucose: insights from an engineering control systems perspective

In order to discern areas of potential improvement in various aspects of glycaemic control for patients with insulin dependent (type 1) diabetes mellitus, blood glucose control is analysed in the light of general engineering feedback control systems theory. This approach is based on models of the system being controlled, using appropriate control strategies. The models presently used for glycaemic control are analysed from this perspective, revealing certain limitations that they impose on the control strategies that use them. Current type 1 diabetes regimens are evaluated for the ease with which they may be analysed mathematically, suggesting areas where improvements in control may be effected by simplifying calculation of appropriate insulin quantities. A new model of undergraded insulin action, derived from established insulin action profiles, along with a control strategy which flexibly extends the basal/bolus regimen using patient-specific parameters, is proposed. This may provide the information needed to enable prediction of expected glycaemia several hours into the future, thereby enabling earlier corrective action to be taken should it fail outside the target range, and in turn potentially reduce the degree and frequency of both hyperglycaemia and hypoglycaemia.

Interactive educational simulators in diabetes care

Since the Diabetes Control and Complications Trial demonstrated the substantial benefits of tight glycaemic control there has been renewed interest in the application of information technology (IT) based techniques for improving the day-to-day care of patients with diabetes mellitus. Computer-based educational approaches have a great deal of potential for patients use, and may offer a means of training more health-care professionals to deliver such improved care. In this article the potential role of IT in diabetes education is reviewed, focusing in particular on the application of compartmental models in both computer-based interactive simulators and educational video games. Close attention is devoted to practical applications-available today-for use by patients, their relatives, students and health-care professionals. The novel features and potential benefits of such methodologies are highlighted and some of the limitations of currently available software are discussed. The need for improved graphical user interfaces, and for further efforts to evaluate such programs and demonstrate an educational benefit from their use are identified as hurdles to their more widespread application. The review concludes with a look to the future and the type of modelling features which should be provided in the next generation of interactive diabetes simulators and educational video games.

Insulin algorithms in the self-management of insulin-dependent diabetes: the interactive 'Apple Juice' program

The 'Apple Juice' program is an interactive diabetes self-management program which runs on a lap-top Macintosh Powerbook 100 computer. The dose-by-dose insulin advisory program was initially designed for children with insulin-dependent (type 1) diabetes mellitus. It utilizes several different insulin algorithms, measurement formulae, and compensation factors for meals, activity, medication and the dawn phenomenon. It was developed to assist the individual with diabetes and/or care providers, in determining specific insulin dosage recommendations throughout a 24 h period. Information technology functions include, but are not limited to automated record keeping, data recall, event reminders, data trend/pattern analyses and education. This paper highlights issues, observations and recommendations surrounding the use of the current version of the software, along with a detailed description of the insulin algorithms and measurement formulae applied successfully with the author's daughter over a six year period.