PDF
ePub
Citation
Print
Introduction
The prevalence of diabetes mellitus is increasing in the world. Diet therapy for diabetes mellitus is very important for controling blood sugar. Food exchange is used widely for meal planning and nutrition education of diabetes mellitus. Hong et al. (2003) and Hong et al. (2004) develops calculating program for food exchange amounts of food groups. But it is very difficult to find a web-based meal planning program using food exchange.
So the program is required to be able to make meal planning by using food exchange to manage diabetes mellitus, obesity and other diseases for diet therapy.
Some off-line programs on menu planning, nutrition analysis and nutrition education counseling have been developed and used (Han, 1997a; Han, 1997b; Han & Rhee, 1993b; Hong, 1989; Hong, 1996; Kang et al., 1998; Kang et al., 1999; Kolasa & Miller, 1996; Peter et al., 1998). Similar programs on the internet are insufficient (Choi, 2000). Therefore the system based on the internet that has a friendly user interface and accepts the needs of users is required as soon as possible (Hong & Hwang, 2001).
Nutrition related programs based on the internet were developed such as the analysis of food intake (Han, 2000), nutritional counseling and diet management of diabetes mellitus (Han & Jeong, 2004; Hong & Kim, 2004), food exchange database construction and search system (Hong et al., 2003; Hong et al., 2004), menu planning and searching system: MenuGen, National Rural Living Institute (Hong et al., 2004) and meal planning and evaluation system: NutriEval (Hong, 2007). But the developed programs are insufficient in meal planning using food exchange for diabetes and obesity. Users have trouble to input food or meals and can not have nutrient and e-food exchange analysis, storing and modifying the data.
In the case of the USA, as a program based on internet, the food composition table from the USDA is commonly used. Cyberdiet services food nutrition information and menu and some web sites have meal analysis functions, but those are insufficient (Lee & Nieman, 2003). The Food Surveys Research Group of the United States Department of Agriculture has developed automated methods for collecting and processing food intake data. These methods are part of the Dietary Intake Data System designed to efficiently collect and process high quality food intake data. The foundation of the system is the Automated Multiple Pass Method (AMPM) Blaise instrument, which is used to collect 24-hour dietary recalls (McDowell, 2003; Raper et al., 2004).
In this study, we developed the internet-based food, menu and meal management expert system for nutrition management and nutrition counseling with e-food exchange, nutrient time-series analysis and data transformation. This system considers the sex, age and disease conditions and creates general meal and therapeutic meal for disease. We expect that NutriSonic will contribute to health and nutrition improvement.
Materials and Methods
The implementation of the system
Table 1 shows the development environments of server and users. The operating system for servers is Linux. The user operating system is the Microsoft Windows series. The web server is Apache. Development languages are PHP, Javascript and HTML. The database is MySQL. Users can access NutriSonic using internet browsers.
Database structures
Table 2 shows the database composition. NutriSonic expert system database is composed of food, menus and meals. The food composition table is provided from the National Rural Living Science Institution in Rural Development Administration (National Rural Living Science Institution, 2001). And we used e-food exchange database (Hong et al., 2003). The menu database is based on the results of the diet assessment system (National Rural Living Science Institute, 2000) and CAN-Pro program ver 2.0 (The Korean Nutrition Information Center, 2002). And in modifying the results, it has 24 categories from rice to others. Menu codes are divided into three: grand, middle-range, and specific classification. The meal database uses the recommended meals which are the results of the studies on developing the web-based Korean style dietary management system: MenuGen (Hong et al., 2004). These are composed of meals according to the user's characteristics and therapeutic meals.
System architecture of nutrisonic web expert system
Fig. 1 shows the system architecture of the web expert system. The database is composed of user's information such as user's meal database, user's menu database and user's information and also web expert database such as recommended meal database, recommend menu database, food nutrient database and e-food exchange database (ver 1.0 and ver 2.0). With these databases, meal nutrient time-series analysis, e-food exchange and data transition can be calculated.
The scenario of meal planning and nutrition counseling
The NutriSonic web expert system is based on rules and cases. The recommended meal is based on pre-constructed cases. Meal properties are adjusted as user preferences. The web expert system verifies that the assessment of nutrients and calories is relevant to the properties of users. These results can be used for nutrition counseling, especially for diet therapy of diabetes mellitus patients. The dietitian or physical doctor teaches the food exchange system and meal planning using food exchange. But the dietitian, physical doctor and patient can not calculate the exact amount of food exchange because the present food exchange list is not complete. The e-food exchange can help the dietitan, physical doctor, diabetes patient, and their helpers to calculate the exact amount of food exchange of their diets to meet recommend food exchange amount. Fig. 2 shows the process of meal planning.
Fig. 2, (1) is a step to input the user's sex and age. (2) is a step to search the meal which is relevant to the user's properties from the recommended meals (8). (3) is a step to select a meal among the recommended meals. (4) assesses the nutrients and calories of the meals, the menus and foods. This step verifies that the planned meal is relevant to the user's properties. (5) completes the meal planning. By composing the menu (9) using the food database (10), a recommended meal database is made. The stored meal in (11) is included in the user's meal database. Anyone can use it. Case Base (6) is used when a user searches the recommended meal and selects the meal. (12) is a step to verify and assess e-food exchange information of a current menu or meal. In (13), users can trace his or her nutrients time-series analysis. The expert system provides users with data transition function to Excel (14).
Results
Menus of nutrisonic web expert system
Table 3 shows the meals of the NutriSonic Web Expert System. The main menu are (1) Meal Management, (2) Menu Management, (3) Food Management, (4) Nutrient and e-food exchange Analysis. Each main menu has its sub menu. And the sub meals have some node menus. As shown in Table 3, some functions are shared in each function for common purposes. For example, nutrient and e-food exchange analysis is a shared function to Meal Management and Menu Management, etc. NutriSonic web expert system provides a dynamic view, informative results and data transition function.
Introduction of home page
Fig. 3 is the main screen of NutriSonic web expert system (http://nutrition.ulsan.ac.kr/nutrisonic/). This screen is for users. The nutrient and e-food exchange database in food, menu and meal that a user can use is made by experts. The main functions of the system are meal management, menu management, food management, search, and nutrient and e-food exchange analysis. The recommended meals and menus are made by experts. The NutriSonic web expert system aims to make use of an expert's meals and menus, and to provide relevant meal planning and nutrition counseling. The system is concentrated on the informative functions.
Meal planning with nutrisonic
Fig. 4 is a screen of meal planning. The top screen is divided into 4 frames. Each frame is numbered as its operation characteristic for convenient user interface.
1) Selection of menu classification
Fig. 4, [1] is a selection of menu categories. This is classified by meals like breakfast, lunch and dinner. A user can search a menu by selecting each group from the first and the second categories. Also, a user can input the menu name directly to find it. In the menu category, the grand menu classification is classified by 14 groups, such as rice and soup. Rice has middle-range classifications like a crust of lightly scorched boiled rice, lightly scorched dried rice, plain rice, rice with glutinous rice, etc. Like this system, all menus are classified into the grand, middle-ranged, and specific categories. This classification is modified based on the "diet assessment system of National Rural Living Science Institute" or "Can Pro".
In case of a menu search, users can search the menus with the full or partial-word. Single food name searching is impossible. Retrieved food or planning menus can be added to a current meal. After meal planning and storing, the meal pattern is given automatically as a menu classification. This automatic pattern is the initiative try in internet-based meal planning. Users or experts can add the contents of recommended meals or user meals in the search meal. And they can modify and add the rate or quantity of the current meal's menus and each food.
2) Selection of menu
Fig. 4, [4] shows the results that the selected menu is added to the current meal. It is the result of searching according to the conditional input for menu selection function. In NutriSonic web expert system, experts or users can follow by adding work: (1) categorical menu selection, (2) recommended meal searching and adding, (3) meal searching and adding, (4) menu searching and adding, (5) single food searching and adding, (6) making a single food and adding it.
3) Detailed information of meal
Fig. 4, [3] is a detailed view of menus in a current meal. Users can change or modify the menu name, foodstuffs of a menu and quantity of foodstuffs. And a user can analyze nutrients of menus and food. In case of changing the total weight of a current meal, an expert is able to change the total weight to the weight, percentage and user's degree. Adding or deleting the menus in a current meal is possible. Also, the rates of calories, fat and proteins are presented. And the sum for a current meal, for meals and for menus is calculated. The rates and percentages of carbohydrates, proteins and fats in a whole meal are shown by a bar graph. And the amount of e-food exchange amount of a meal is calculated. In [4], the name of the current planning meal is created automatically with composition of user-name and total calories. When a menu is added to a current meal, the meal name is changed simultaneously.
4) Information of meal
Fig. 4, [4] shows the meal properties. There is a meal planning date and a meal applying date. Users or experts can select each property. Age properties have general age groups with infants. In case of women, the age groups are divided into pregnant or not and the first or second half of pregnancy. The planned meal is stored in the database. The properties of recommended meals is not sufficient to provide the meals which a user needs. For a more intellectual meal planning, we consider the following factors: user's activity, a clinical history and a propensity of meal intake.
Nutrient and e-food exchange analysis and storing of meal
The function of meal planning has a view of recommended energy, nutrient, e-food exchange. If a user does not select the meal properties, the criterion of nutrient analysis is based on the 20 ~ 29 years of age and a male. Fig. 4 shows the view of recommended energy and nutrients.
1) Nutrients analysis and storing meal
This is comparing and analyzing a daily nutrient recommended quantity of planning or of a planned meal. For a user's age and sex, the web expert system calculates the sum of each meal such as breakfast, lunch and dinner. Then, the web expert system compares the total energy with a daily nutrient recommended quantity. There is a nutrient analysis of foodstuffs in each meal. Also, it provides information of foodstuffs in a current meal. If needed, it is possible to modify and delete the current meals, menus, and foodstuffs. When the meal planning is completed, the web expert system stores the meal as meal properties. Stored meals can be used in meal searching, meal modification and new meal planning.
2) Analysis of nutrient changes
After making a meal for users, or recommending nutritionists or users can review the nutrient changes and check the status. The system provides the changes of meal nutrients, which a user selects among registered meals. Stored meals can be used in meal analysis or changes analysis at any time. Fig. 5 shows the sample meal for changes analysis.
3) View of e-food exchange
In making a meal, nutritionists or users are able to review the nutrient information with food exchange. Fig. 5 shows the e-food exchange information of a current meal or selected meal.
4) Nutrient time-series analysis of user meals
Users or experts can select any meal which they want. Fig. 6. shows the selection of 3 meals data (June 5, 19, 29) for nutrient time-series analysis information among the meals which are selected by a user. Fig 7 shows nutrient time-series analysis bar grape among the 3 meals data which are selected by user. The change information consists of the comparing for each nutrient and ratio of major energy nutrient like carbohydrates, fat and protein of selected days.
Meal search
Fig. 8 is a screen of a meal search. The conditional items are menu name, meal name, sex, age, calories of meal, and the patterns of a meal. In this search process, recommended meals and user's meals are available.
Fig. 9 is a comparison screen of a current meal with the Korean Recommended Dietary Allowance (2000). The screen shows the menu names, quantities and nutrient analysis data. Also, it shows the sum of breakfast, lunch, dinner and the comparison of daily recommended nutrient quantity. The rate is presented as a percentage. On the bottom of the screen is the calorie rates of three main nutrients. Fig. 10 is a screen of nutrient analysis of a current meal. In a detailed view, there are menu names, food names, quantities of foods, menus and meals, calories and three main nutrients. Also, the web expert system shows the rates of the three main nutrients as a bar graph. Besides, the system provides a comparison analysis with the recommended nutrient quantity and a view of nutrient analysis.
Menu search and management
Fig. 11 is a screen of menu search and management. It is a search result in order of cooked rice, boiled rice with assorted mixtures, fried rice and fried rice with kimchi. It is possible for a user to search menus by menu name or menu classification. So a searched menu provides food ingredients with each nutrient, menu nutrients and pictures (Fig. 12). The web expert system provides a screen print-out and view of a portion.
Food management and search
Table 4 shows the food management and search contents. Food management and search has input food composition data and conditional search items such as food group, food code, Korean food name, English food name, calorie and nutrient. And food is divided into key foods and total foods in the database. Fig. 13 is a screen of food management. Food data modifying, deleting and saving with new food name are available on food management. Fig. 14 shows nutrients of food as a result of search including e-food exchange.
Discussion
NutriSonic is a nutrition counseling and meal management web expert system with time-series analysis, e-food exchange and data transition based on the internet. The friendly user interface was considered. It is composed of recommended meals and user-created meals. The food, menus and meals are the fundamental data to assess the nutrient analysis. Experts or users can search, add, modify and delete the fundamental data. Also nutrition experts are able to verify food exchange and trace their nutrient time-series analysis. Using the database is extensible and is expandable. We add searching conditions to verify the detailed needs of experts. We have transformed the extremely small quantities and source of data to machine-readable data like Excel files. Experts or nutritionists can use the downloaded files for analysis or statistics.
USDA has a long history of methodology research related to dietary surveys. A program based on internet, the food composition table from the USDA is commonly used. Automated methods for collecting and processing food intake data have been developed by the Food Surveys Research Group at USDA to increase the quality and efficiency of food intake surveys and other dietary research studies. These automated methods are part of a Dietary Intake Data System, which consists of three computer systems and an extensive food and nutrient database. Computer systems included are the Automated Multiple Pass Method (AMPM) for collecting food intakes, the Post-Interview Processing System (PIPS) for reformatting data and assigning food codes, and Survey Net for final coding, quality review, and nutrient analysis. Features addressing data quality were prominent in the design of each of these systems (Raper et al., 2004). The AMPM and other components of the Dietary Intake Data System are currently being used for the dietary interview portion of the National Health and Nutrition Examination Survey, conducted by the US Department of Health and Human Services in collaboration with the USDA (McDowell, 2003). During the interview, individuals recall the foods and beverages that were consumed the day before the interview. Details about each food and beverage are collected as well as a description of the amount consumed. Information is also collected about the time of day the food was eaten, the name of the eating occasion, and where the food was obtained (Raper et al., 2004).
In Korea, some internet programs such as MenuGen (Hong et al., 2004) and Nutrieval (Hong, 2007) provide food composition tables with a searching engine. And web-based programs for nutrition were developed (Han & Jeong, 2004; Her & Lee, 2002; Hong & Kim, 2005; Hong et al., 2004; Hong, 2007; Lee et al., 2002), but web-based food exchange calculating programs for diet therapy partly. Nutrition related programs based on the internet were developed such as the analysis of a food intake and nutrition screening system (Han, 2000), a web-based internet program for nutritional counseling and diet management of patient with diabetes mellitus (Han & Jeong 2004), food exchange database construction and search system (ENECC/E-Food Exchange) based on the internet (Hong et al., 2003), e-food exchange database construction of commonly used foods and search system (Hong et al., 2004) and menu planning and recommended menu search system (MenuGen) of the National Rural Living Science Institution in Rural Development Administration (Hong et al., 2004). In addition, there is the status of eating habits, analysis of eating habits, obesity, calorie expenditure and nutrient analysis of menus in the internet program of the nutrition computing (Hong & Kim, 2004). And NutriEval (KFDA, Korea Food and Drug Adminstation) was developed with drag and drop for meal planning and nutrient evaluation for children (Hong, 2007). But the developed programs are insufficient in meal planning and nutrient analysis. Especially on diet therapy using food exchange for diabetes and obesity, users have trouble to input food or meals and can not have nutrient and e-food exchange analysis, storing and modifying the data.
We expect that NutriSonic will contribute to meal planning using e-food exchange for meal management of diabetes and obesity and nutrition improvement. And we hope NutriSonic with e-food exchange, nutrient time-series analysis and data transition could be widely used for diet therapy and nutrition education.
XML Download