EDUCATIONAL SIMULATORS - COMPLIANCE WITH THE

REQUIREMENTS OF DIABETES PATIENTS AND DIABETES

THERAPY GUIDELINES

Izworski Andrzej, Koleszynska Joanna, Tadeusiewicz Ryszard

Institute of Automatics, AGH University of Science and Technology, Al. Mickiewicza 30, Krakow, Poland

Keywords: Tele-health system, Tele-education, Web tools, glucose-insulin models, simulation, diabetes therapy.

Abstract: This paper presents renewed approach to the computer-aided diabetes educations introducing GIGISim

(Glucose-Insulin and Glycemic Index Web Simulator) e-learning tool. Together with our system, selected

solutions were summarized and their functionality compliance with diabetes therapy requirements checked.

The analysis of diabetes patients needs has been established through a series of intermediate research

surveys and literature studies. The software implementation of newly proposed innovations is presented

together with effectiveness and suitability rate of a system, prior to the identified requirements.

1 INTRODUCTION

The overall goal of computer-aided diabetes

education is to help individuals with diabetes

mellitus gain the necessary knowledge about

physiological processes, control dietetic habits and

offer support needed on daily-basis. Rules presented

below provide guidance for architects of this type of

telemedicine systems, determining a basic

functionality, which is to:

 improve or maintain the quality of life for

people with diabetes,

 assist decision-making enhancing the patient’s

personal sense of control,

 help maintaining blood glucose and

cholesterol levels as near-normal as possible,

 advice on insulin and medicaments doses,

 highlight diabetes-related complications

symptoms,

 assess the individual’s diet basing on nutrient

requirements,

 educate on physiological processes and human

carbohydrate metabolism.

Proper diet assessment is especially important as

it influences the insulin and glucose response in

time. Patients treated with insulin should adapt

insulin type (long lasting, short lasting) to the pattern

of their blood glucose response, while other patients

should consult their diet to modify eating habits,

loose weight or improve the body's sensitivity to

insulin. As the diet plays an integral role in the

therapy and as the dietetic guidelines have changed

recently and new strategies have been developed for

people with diabetes, new functionality must be thus

implemented in the dedicated software.

Our research on nutritional management of

diabetes follows the principles of Canadian (CDA,

1999) and the American Diabetes Association

which will be fully presented in Chapter 2. Chapter

3 outlines main guidelines for diabetes-advisory

systems, while chapter 4 presents the existing,

diabetes dedicated educational tools with their

innovations. Section 5 describes the GIGISim tool

main functionality, providing detailed information

on the application implementation and newly added

features, with the possible effects on the therapy,

conclusion and comments also detailed in Chapter 6.

2 NEW NUTRITIONAL

GUIDELINES FOR DIABETES

Despite the importance of healthy diet in diabetes

mellitus therapy, patients often neglect nutrition

recommendations and increasing role of the

Glycemic Index (GI) - the exact effect of GI on daily

glucose variation is difficult to identify without

dedicated methods. The Glycemic Index is a scale

that ranks carbohydrate-rich foods by how much

319

Andrzej I., Joanna K. and Ryszard T. (2007).

EDUCATIONAL SIMULATORS - COMPLIANCE WITH THE REQUIREMENTS OF DIABETES PATIENTS AND DIABETES THERAPY GUIDELINES.

In Proceedings of the Ninth International Conference on Enterprise Information Systems - HCI, pages 319-322

DOI: 10.5220/0002397203190322

 SciTePress

they raise blood glucose levels compared to glucose.

Lowering the total GI of the diet may improve blood

glucose control without raising serum triglycerides,

reduce weight and normalize appetite which was

proved for non-diabetes, type I and type II diabetes

(T.M.S. Wolever, 1992; Frost G. Wilding, 1994).

Although diabetes canters and associations

recommend GI as the important diet factor (Jenkins

D.J, 2002; Ludwig DS 2002) and encourage

considering GI among other dietetic rules, patient’s

rarely take it into account when preparing their

meals. What is event more dangerous, but

unfortunately very common according to our

surveys, patients often ignore the importance of

basic nutritional meal assessment, despite the

necessity of the properly balanced diet in diabetes

therapy. Calculations of fats, proteins, carbohydrates

and calories are long and tedious and if the glycemic

index should be included, as the additional nutrition

factor, the process itself becomes even more

complicated.

Although the glycemic index is a numerical

value, available for nearly every food product and

can be easily found in nutrition tables together with

fats, proteins, carbohydrates and calories estimations

it may be difficult to imagine and understand how

exactly GI affects blood glucose variations. This is

why in our previous papers we introduced a new

web tool: GIGISim which is designed to educate

patients on the GI impact on their diet and therapy.

Our system visualizes the postprandial glucose

profiles correlated to user’s diet and improves

patient’s glycemic control which was proved in

clinical research. The presented examples and

research outlined advantages of the visualization

method over numeric value estimation of a meal.

The blood glucose variation graphic simulation used

in GIGISim is far more meaningful and educative

than figures and calculations.

In this paper we demonstrate the usefulness of

GIGISim, emphasize the educational role of this

internet tool and prove whether it satisfies the

requirements of the prospective diabetes users,

pointed in Chapter 1 and following Chapter.

3 DIABETES THERAPY

RECOMMENDATIONS AND

PATIENTS’ REQUIREMENTS

To improve the GIGISim tool the analysis were

carried to identify new facilities possible to

implement in web-system and to meet users’ specific

requirements. We have already pointed out the

necessity of proper nutritional diet assessment which

requires calculations of total fat, proteins,

carbohydrates and calories in patients’ diet and the

great majority of computer software dedicated for

diabetes includes simple nutritional calculators. The

role of the Glycemic Index on the daily glucose

variation is difficult to identify without specially

dedicated methods and only few systems available

on the market, including GIGISim, actually include

GI as a new factor, evaluating high-quality

carbohydrates for diabetes diet.

The additional nutritional recommendations cited

from the Canadian Diabetes Association (Guidelines

for the Nutritional Management of Diabetes Mellitus

in the New Millennium, 1999) which should be

considered to provide a sufficient meal description,

in the designed software:

 Eat at regular times

 Choose a variety of foods from all food

groups

 Limit sugar and sweets

 Reduce the amount of fat you eat

 Include foods high in fibre

 Limit salt, alcohol and caffeine.

 Choose heart healthy fats such as canola and

olive oil.

All those rules are easy to implement in the

computer algorithms, and the only information

required is the food type and the amount, which the

prospective user may provide entering consumed

products from the available database, including meal

time dependencies.

Clinical survey was carried among diabetes

patients in Krakow’s Hospital and among members

of the local Diabetes Association, to identify

additional requirements and possible application of

diabetes oriented software. Subjects were volunteers,

type I and type II diabetes, all treated with insulin.

Suggestions of patients are listed below and their

implementation in GIGISim discussed in the

following chapters. User recommendations:

 All individuals would like to receive

nutritional counselling from registered

dieticians as often as possible, preferably via

Internet.

 Specific dietary recommendations and

medications should be individualized to

accommodate the person’s preferences and

lifestyle.

 Additional tools like diet diaries and statistical

analysis of diet for longer time periods should

be provided.

ICEIS 2007 - International Conference on Enterprise Information Systems

320

 User graphical interface should be facilitated

to improve the process of data entering.

Rarely patients are capable of weighing all

consumed food products so estimations should

be possible along with hints about possible

amount consumed.

All those functionalities were analysed, and

implemented in GIGISim Internet tool.

4 COMPUTER-AIDED DIABETES

EDUCATION: A REVIEW

Although the impressive number of diabetes

dedicated software is available in WWW, there are

very few accessible solutions which would help

consumers identify the effect of the novel,

worldwide approved nutritional trend - the Glycemic

Index. Although limited number of e-lerning

diabetes tools complies with a full variety of

requirements, discussed in previous sections,

selected group of diabetes dedicated systems will be

presented in this chapter, compared to GIGISim, to

present the overview of interesting facilities and

functionalities available.

Several investigators have pointed out the

possibility of using the glucose and insulin plasma

levels simulation models, to help diabetes interpret

human carbohydrate metabolism (Lehmann E.D,

1997; Worthington DRL 1990) This approach to

diabetes education was proved to be effective and

numbers of mathematical models of the diabetes

mellitus metabolism have been previously reported

in literature.

The first presented in this section, well

known AIDA software (Lehmann E.D, 1997)

available since 1996, provides a simulation of

glucose and insulin levels in the blood based on the

glucose-insulin kinetics model. Authors have proved

educational advantage of visualization methods like

AIDA simulation over the traditional lectures in

clinical research.

The DiasNet software (Plougmann S., 2001)

have a clear clinical focus, however functional scope

has been extended from being used by clinicians as

advisory system to also being used by patients as an

educational tool. Users can experiment with their

insulin adjustment and meal sizes and analyse

simulations based on human carbohydrate

metabolism model.

Biermann and colleagues focused on the need

for computer-aided learning tool for the education of

patients and personal care of diabetes with insulin

therapy. DIABLOG (Biermann E., 1990) is able to

simulate glucose and insulin profiles of a 24 h period

and display them graphically as curves. The subjects

could vary the carbohydrate content of the meal, the

injection time and dose of short-acting and

intermediate acting insulin and observe simulation

results calculated from mathematical model of

glucose-insulin dynamics. This example is

especially interesting insulin dose adjustment is

considered the most difficult skill to develop for a

new diabetes patient. However, opposite to the

DiasNet solution this system doesn’t support

clinicians supervising so the results of the simulation

must be carefully interpreted by potential users.

In addition to educative software described we

should point out one which actually includes GI as a

new factor, evaluating high-quality carbohydrates

for diabetes diet. Home PC Editors' Choice (1995),

NutriGenie, provides the GI ratings classifying

commonly used food items into 3 categories: low,

medium and high. NutriGenie Nutrient Analysis

Tool controls saturated fat, cholesterol, sodium,

proteins, fibre, vitamins and minerals of the patient’s

diet and is equipped with additional tools like

DietTracker, Katogenic Meal Planner, Weight

Tracker to control the blood glucose, blood pressure

and serum triglycerides. Although the mentioned

software is well featured we believe that a simple

classification of GI doesn’t provide a sufficient meal

description. Graphic visualisation of blood glucose

level changes for all possible GI values is the biggest

GIGISim advantages over other software.

Figure 1: Example of the GIGISim meal simulation – plot

compares blood glucose variation after consumption of

50g glucose and user defined meal.

5 GIGISIM – COMPLIANCE

WITH NEW REQUIREMENTS

The main advantage of the GIGISim, emphasised in

previous chapter is that the total effect of diet on

blood glucose variations is presented on plots (Fig.

1). Graphical method illustrates metabolism reaction

more clearly and intuitively then figures in GI tables

EDUCATIONAL SIMULATORS - COMPLIANCE WITH THE REQUIREMENTS OF DIABETES PATIENTS AND

DIABETES THERAPY GUIDELINES

321

so patients treated with insulin analogues may adapt

insulin type and doses to the pattern of blood

glucose response, which is strongly correlated to GI.

What is most important, the GIGISim will be

clinician supervised system. All assessments will be

presented to the user and also sent to the physician

or dietician to approve new insulin treatment schema

or introduce new eating habits, basing on the reports

and nutrient analysis generated by the system.

Our system is an ASP.NET 2.0 web

application and has been equipped with user friendly

interface, providing help, tips and hints about food

amount estimation, which was one of the most

common user requirements. Next feature to

implement would be a system of trackers and diaries

To control patient’s progress in weight and

blood glucose control. Also, the AI algorithms will

be used to build intelligent Meal Planner and to

individualize recommendations, to satisfy the

person’s taste preferences and lifestyle. To ensure

comfort of usage the GIGISim may be easily

migrated to the mobile devices like palmtops or

mobile phones (Fig. 2).

Figure 2: The ways of accessing the GIGISim on-line.

6 CONCLUSION

Nutrition management is a key component for the

health and quality of life for people with diabetes but

it is also probably one of the most complicated and

complex long-term therapy. Many factors should be

taken into account, like the individual’s micro- and

macronutrient, physical activity, lifestyle and

medical needs. A meal and insulin dose planning

requires experience from patients, their families and

physicians. We believe the computer aided-diabetes

therapy has a promising perspective if only the

designed system will be able to assure and facilitate

contact between the patient and his physician via

web application, like GIGISim.

REFERENCES

CDA, 1999. Guidelines for the Nutritional Management of

Diabetes Mellitus in the New Millennium. A position

statement by the Canadian diabetes association.

Canadian Journal of Diabetes Care, 23(3): 56–69.

Willett W, Manson J, Liu S, 2002. Glycemic index,

glycemic load, and risk of type 2 diabetes. American

Journal of Clinical Nutrition 76:274S-280S.v

Ludwig DS, 2002. The glycemic index: physiological

mechanisms relating to obesity, diabetes, and

cardiovascular disease. Journal of the American

Medical Association 287: 2414-2423.

Jenkins, D.J., Kendall, C.W., Augustin, L.S., Franceschi,

S., Hamidi, M., Marchie, A., Jenkins, A.L., &

Axelsen, 2002. Glycemic index: overview of

implications in health and disease. American Journal

of Clinical Nutrition, 76 (suppl): 290S-298S.

T.M.S. Wolever, D.J.A. Jenkins, V. Vuksan, A.L. Jenkins,

G.S. Wong, R.G. Josse, 1992. Beneficial effect of low-

glycemic index diet in overweight NIDDM subjects.

Diab. Care,15, 562–566.

Frost G,Wilding J, Beecham J., 1994. Dietary advice

based on the glycemic index improves dietary profile

and metabolic control in type 2 diabetic patients.

Diabetic Med. 11:397–401.

A. Izworski, J. Koleszynska, R. Tadeusiewicz, J. Bulka,

and I. Wochlik, 2005. GIGISM (Glucose-Insulin and

Glycemic Index Web Simulator) - The Online System

Supporting Diabetes Therapy. The IASTED

International Conference on Telehealth Banff, Canada

A. Izworski, J. Koleszynska, R. Tadeusiewicz, J. Bulka,

and I. Wochlik, 2006. Internet Tools and Computer-

Aided Diabetes Education: Introducing GIGISim

Online. Proceeding of Communications, Internet, and

Information Technology, St. Thomas, USVI, USA

Lehmann E.D, 1997. Interactive educational simulators in

diabetes care. Med. Inform. 22 (1997) 47–76.

Worthington DRL, 1990. The use of models in the self-

management of insulin-dependent diabetes mellitus.

Computer Methods & Programs in Biomedicine,

vol.32, no.3-4, pp.233-9. Netherlands.

Lehmann E.D, 2001. The freeware AIDA interactive

educational diabetes simulator. A download survey for

AIDA v4.0. Med Sci Monit. May-Jun;7(3):504-15

Plougmann S, Hejlesen OK, Cavan DA., 2001. DiasNet—

a diabetes advisory system for communication and

education via the internet. Int J Med InformDec; 64(2-

3): 319-30.

Biermann E., Mehnert H., 1990. DIABLOG: a simulation

program of insulin-glucose dynamics for education of

diabetics. Computer Methods and Programs in

Biomedicine, Volume 32, Issues 3-4, July-August

1990, Pages 311-318.

This paper was supported by AGH grant

10.10.120.39

ICEIS 2007 - International Conference on Enterprise Information Systems

322