JP2012045259A - Device and method for predicting blood sugar level - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for predicting blood sugar levels capable of accurately predicting changes in the blood sugar levels while reducing a load on a user.SOLUTION: The device 10 for predicting the blood sugar levels includes: a character string input unit 11, an information storing unit 12; a data base storing unit 13 in which the data base is stored, a syntactic analyzer 14; a physical condition estimator 15; and a blood sugar level output unit 17. The prediction of the blood sugar level is performed by using living information such as a diary that the user keeps as an electronic record on a daily basis.

Description

  The present invention relates to a blood sugar level prediction apparatus and a blood sugar level prediction method using life information input as electronic information, for example, text information such as a diary.

  In Patent Document 1, a prediction model for predicting a user's blood glucose level is created in advance based on the history data of the calorie intake of the user and the history data of the calorie consumption of the patient. And the technique which estimates a user's blood glucose level from the physical condition variable which shows the physical condition which the user subjectively judged using the said prediction model, a user's calorie intake, a patient's calorie consumption, etc. is disclosed.

JP 2005-328924 A

Diabetes and the like administer insulin according to the amount of insulin determined based on the actually measured blood glucose level when administering insulin, and it is a burden on the patient to measure the blood glucose level every time insulin is administered .
Further, in the technique for predicting blood glucose level described in Patent Document 1, it is necessary for a user (such as a diabetic patient) to subjectively determine his / her physical condition and to input data explicitly. As a result, it is difficult to secure a blood glucose level prediction system. Furthermore, since the input for the purpose of data input is required, there is a problem that the burden on the user increases.

SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms or application examples.
The life information described in the following application means that the user uses an electronic information input device such as a personal computer (hereinafter referred to as “PC”) or a mobile phone, for example, diary, mail, blog, mini-blog (Twitter: Shows daily text records such as tweets.

  Application Example 1 A blood sugar level prediction apparatus according to this application example includes an input unit that inputs life information as electronic character information, and an information storage unit that stores the character information input from the input unit as a character string. And database storage means for storing words used for blood sugar level prediction as a database, and syntax analysis means for extracting language information related to blood sugar levels by performing syntax analysis using the character string and the database; A physical condition estimation unit that collates the language information extracted by the syntax analysis unit with the database and predicts a blood sugar level; and an output unit that outputs the blood sugar level information predicted by the physical condition estimation unit. It is characterized by.

According to the blood sugar level predicting apparatus described in this application example, life information that the user inputs on a daily basis is stored in the information storage means as character information. Then, the syntax analysis means extracts language information from the wording database used for blood sugar level prediction stored in the database storage means and the character information stored in the information storage means. The blood glucose level is predicted by collating with the database, and the blood glucose level information is output from the output means.
In this way, since it is possible to predict blood glucose levels using life information that is routinely input by the user for blood glucose level prediction, the user does not need to perform an input operation for blood glucose level prediction. It becomes possible to reduce the load. In addition, since it becomes unnecessary for the user to subjectively judge his / her physical condition, variations in user judgment can be prevented, and the accuracy of blood sugar level prediction can be improved.

  Application Example 2 In the blood sugar level prediction apparatus according to the application example, the input unit and the output unit are information terminal devices.

  According to the blood sugar level predicting apparatus described in this application example, since the user can input information at any time, the accuracy of input character information is improved and the number of character information is easily increased. Therefore, it is possible to further improve the accuracy of blood sugar level prediction.

  [Application Example 3] In the blood sugar level prediction method according to this application example, life information is input to the input unit as character information, the input character information is stored as a character string in the information storage unit, and the character string is stored. The linguistic information related to the blood glucose level is extracted by parsing by the parsing unit, and the linguistic information extracted by the parsing unit and the database stored in the database storage unit are collated by the physical condition estimating unit. To predict the blood glucose level and output the predicted blood glucose level information from the output means.

According to the blood glucose level prediction method described in this application example, the life information that the user inputs on a daily basis is stored in the information storage means as character information. Then, the syntax analysis means extracts language information from the wording database used for blood sugar level prediction stored in the database storage means and the character information stored in the information storage means. The blood glucose level is predicted by collating with the database, and the blood glucose level information is output from the output means.
As described above, the blood sugar level prediction method described in this application example can predict the blood sugar level using life information that is routinely input by the user for blood sugar level prediction. An input operation for prediction is unnecessary, and the load on the user can be reduced. In addition, since it becomes unnecessary for the user to subjectively judge his / her physical condition, variations in user judgment can be prevented, and the accuracy of blood sugar level prediction can be improved.

  Application Example 4 In the blood sugar level prediction method according to the application example described above, the character information input to the input unit includes identification information for identifying character information, and the physical condition estimation unit performs the verification of the language information. In addition, the blood glucose level is predicted using the identification information.

  According to the blood glucose level prediction method described in this application example, it is possible to determine the strength of the influence of the input text information on the blood glucose level based on the identification information included in the text information, and to improve the accuracy of blood glucose level prediction. Further improvement is possible.

  Application Example 5 In the blood sugar level prediction method according to the application example described above, the physical condition estimation unit is configured to calculate a blood sugar level based on a score obtained by quantifying the language information when the physical condition is good and a score when the physical condition is bad. It is characterized by predicting.

  According to the blood sugar level prediction method described in this application example, it is possible to perform blood sugar level prediction in a balanced manner when the physical condition is good and when the physical condition is bad, and more accurate blood sugar level prediction can be performed.

The timing chart explaining the processing timing of the blood glucose level prediction concerning embodiment. The block diagram which shows the outline of the blood glucose level prediction apparatus as 1st Embodiment. The flowchart which shows the procedure of the blood glucose level prediction method of 1st Embodiment. (A)-(e) is a schematic explanatory drawing which shows an example about extraction of the language (part related to physical condition) related to physical condition. Explanatory drawing which shows the specific example which extracted the linguistic information of "the part related to a physical condition". Schematic explanatory drawing which shows the method of calculating | requiring physical condition from the score of the some character string of 1st Embodiment. (A) is a graph which shows the correlation with physical condition S and a blood glucose level, (b) is schematic which shows the example output to the information terminal. The flowchart which shows the blood glucose level prediction method as 2nd Embodiment. Schematic explanatory drawing which shows the method of calculating | requiring the physical condition S in 3rd Embodiment. The flowchart which shows the blood glucose level prediction method as 3rd Embodiment. The scatter diagram which shows the correlation with the physical condition S of an Example, and a blood glucose level.

DESCRIPTION OF EMBODIMENTS Embodiments of a blood sugar level predicting apparatus and a blood sugar level predicting method of the present invention will be described below in detail with reference to the drawings. In the following embodiment, a user (a person who is a diabetic patient and needs management of blood glucose level) predicts blood glucose level using life information such as a diary that is routinely created as an electronic record. A blood glucose level prediction apparatus and a blood glucose level prediction method to be performed will be described.
First, the processing timing of the blood glucose level prediction will be described with reference to FIG. FIG. 1 is a timing chart for explaining the processing timing of blood sugar level prediction.

<Processing timing of blood sugar level prediction>
In recent years, the blood sugar level predicting apparatus and the blood sugar level predicting method described in this embodiment use diary-like text that can be easily performed using the electronic mail function of an information device such as a personal computer or a mobile phone. In this description, an outline of processing timing of blood sugar level prediction in an example of an action pattern from when the user wakes up to sleep will be described.

As shown in FIG. 1, the user who gets up goes jogging early in the morning and then has breakfast. After that, I write a record on the blog, including the condition of jogging, breakfast menu, and quantity. After that, the user goes to work and writes an email each time. In addition, the contents of lunch, physical condition of the day, etc. are written on the blog each time. The user of this example analyzes blood glucose level prediction at the end of work in order to make reference for how to have dinner. In this case, it is possible to obtain blood sugar level prediction predicted from the language information input from the time of waking up to the analysis timing, or information related to the blood sugar level. The user can determine how to have dinner by referring to the result, and can also determine living behavior from dinner to bedtime for maintaining the blood glucose level the next day.
The analysis timing can be arbitrarily set by the user. It is also possible to acquire and analyze timing set in an electronic schedule book or the like.

(First embodiment)
First, a blood sugar level prediction apparatus as a first embodiment according to the present invention will be described with reference to FIGS. FIG. 2 is a block diagram showing an outline of the blood sugar level predicting apparatus as the first embodiment, and FIG. 3 is a flowchart showing the procedure of the blood sugar level predicting method.

<Device configuration>
As shown in FIG. 2, the blood sugar level prediction apparatus 10 of the first embodiment includes a character string input means 11, an information storage means 12, a database storage means 13 in which a database is stored, a syntax analysis means 14, a physical condition estimation means 15, And blood glucose level output means 17. The information storage means 12, the database storage means 13 in which the database is stored, the syntax analysis means 14, and the physical condition estimation means 15 form a blood glucose level prediction means 16.

The character string input means 11 is an electronic information input device for a user to input life information such as daily events as diary-style electronic information. The life information input from the character string input unit 11 is sent to the information storage unit 12. As the character string input means 11, for example, an input device provided in a device itself such as a personal computer, a mobile phone, and other electronic information devices, or a terminal device can be used. Then, Web services such as SNS, Twitter, and blog may be used through input devices provided in the devices themselves or terminal devices.
The above-mentioned “diary-like electronic information” refers to daily text records such as diaries, e-mails, the above-mentioned Twitter, and blogs.

  The information storage unit 12 stores electronic information input (written) from the character string input unit 11, and has a function of reading out the stored electronic information according to an instruction from the syntax analysis unit 14, for example, a server. .

  The database storage means 13 stores a database necessary for collation of words, languages, etc. with the syntax analysis means 14 or the physical condition estimation means 15. More specifically, the database storage means 13 is preliminarily input as a database, a word, language, or the like, which is determined to be related to the blood glucose level, or a word, language, etc., which is determined to be related to the physical condition.

  The syntax analysis unit 14 parses the character string extracted from the information storage unit 12 while collating it with the database, and determines the blood glucose level or the words determined to be related to the physical condition, the number of languages used, the frequency, etc. Ask for.

  The physical condition estimation means 15 uses a calculation formula described later from language information including a blood sugar level obtained by the syntax analysis of the syntax analysis means 14, a wording determined to be related to the physical condition, the number and frequency of use of the language, and the like. Quantify the user's physical condition and predict blood sugar level. Then, electronic information including information related to the blood glucose level from the predicted blood glucose level is output to the blood glucose level output means 17.

  The blood glucose level output unit 17 has an image display unit (not shown), and displays the blood glucose level sent from the physical condition estimation unit 15 or information related to the blood glucose level on the image display unit. Examples of the blood glucose level output means 17 include a display unit of a personal computer and a display screen of a mobile phone. The blood sugar level output means 17 may have a voice output unit (not shown), and informs the user of blood sugar level or information related to the blood sugar level by emitting a voice or a sound effect. Also good. In addition, the image display unit and the audio output unit may be provided together, and the image display and the audio notification can be performed at the same time, and the user's blood glucose level or the degree of recognition of information related to the blood glucose level can be increased. Can be increased.

<Glucose level prediction method>
The blood glucose level prediction method will be described with reference to the flowchart of FIG.
First, the blood sugar level prediction apparatus 10 starts a blood sugar level prediction flow on the assumption that the block i configured with the input character string is 1 (i = 1) (step S101).

  Next, the syntax analysis unit 14 acquires a character string related to the blood glucose level such as blood glucose level and physical condition from the electronic information within the analysis target period stored in the information storage unit 12 (step S103).

Next, the syntax analysis unit 14 collates the wording and language constituting the acquired character string with the database stored in the database storage unit 13, and selects a language related to physical condition (portion related to physical condition). Extract (step S105).
Here, extraction of a language related to physical condition (portion related to physical condition) in step S105 will be described with reference to FIG. FIGS. 4A to 4E are explanatory views of the abbreviation showing an example of extraction of a language related to physical condition (portion related to physical condition).

The stored text such as a blog is a block. For example, when there is a message block as shown in FIG. 4 (a), a part with a description related to blood sugar level such as a meal or physical condition is shown as "part related to physical condition" as shown in FIG. 4 (b). "(The part indicated by a two-dot chain line i in the figure, hereinafter referred to as" part (block) i related to physical condition "). Further, the other parts are “parts not related to physical condition” (parts indicated by P in the figure, hereinafter referred to as “parts P not related to physical condition”).
The “portion (block) i related to physical condition” is indicated by hatching in FIG. Here, among the total number of characters Ni of the obtained character string, the number of characters of “part i related to physical condition” is defined as the number of characters Ai. It should be noted that the “part related to physical condition i” is a case where there are many “parts related to physical condition” in the obtained character string as shown in FIG. Q1, Q2, and Q3) shown are collectively treated as one as shown in FIG.

FIG. 5 shows a specific example in which “portions related to physical condition” are extracted. Recognize sentence breaks using punctuation marks, punctuation marks, spaces, etc., and check whether each recognized sentence has words or phrases related to physical condition or meals, and extract sentences with words or phrases. The following can be illustrated as examples of words and phrases related to physical condition and meals.
In this example, “tired”, “eat”, “waffle”, and “eat” are extracted and character strings related to physical condition are extracted.

(Examples of words related to physical condition)
Tired, sluggish, fever, feverish, heavy body, lack of appetite, not motivated ...
(Examples of words related to meals)
Food name, product name: waffle, spaghetti, sushi, sukiyaki, etc.
Actions and actions related to eating and drinking: eating, drinking, tasting, delicious ...

  These illustrated words and phrases are stored as a database in the database storage means 13 and used when collating with sentences. A dictionary will also be prepared for changes such as case changes and linkages.

  Next, the total number of characters (Ni) and the number of characters related to physical condition / meal (Ai) are counted from the extracted portions related to physical condition (step S107). Incidentally, in the example shown in FIG. 5, the total number of characters Ni of the character string of the portion related to the physical condition is 29 characters, and the number of characters Ai related to the physical condition is 13 characters.

  Then, using the counted total number of characters (Ni) and the number of appearance of characters related to physical condition / meal (character number Ai), the score Si (related to physical condition) is converted into a physical condition by the following equation (1) Is obtained (step S111). The score Si calculated | required here will have a bad physical condition, so that a numerical value is large.

  Next, it is determined whether or not the block i composed of character strings is n (step S113). If the block i composed of character strings is not n (step S113: No), for example, if there are character string blocks input by blogs or emails several times a day, each character string block I = i + 1 (i = 1, 2, 3,..., N) (step S117), the process returns to step S103, and the same flow as described above is performed to extract all the scores Si.

After all the scores Si are obtained (step S113: Yes), the physical condition S is obtained from the obtained scores Si (step S115).
An example of how to obtain the physical condition S will be described with reference to FIG. FIG. 6 is a schematic explanatory diagram showing a method for obtaining physical condition S from scores Si of a plurality of character strings.
Here, the block of each character string is i (i = 1, 2, 3,..., N). Then, Ai is the number of characters in “part i related to physical condition” in each character string (i = 1, i = 2,..., I = n), and Ni is the total number of characters in block i of the character string. . And the average value of score Si is calculated | required by following Formula (2). The obtained average value of Si is the physical condition S.

The obtained physical condition S has a correlation with the blood sugar level as shown in FIG. 7A, and the blood sugar level is predicted based on this correlation (step S119). This correlation is obtained in advance by actual measurement or the like.
The blood sugar level prediction data obtained from the physical condition S is transmitted to the blood sugar level output means 17 shown in FIG. 2, and the blood sugar level sent from the physical condition estimation means 15 to the image display unit of the blood sugar level output means 17 or the blood sugar level Information related to the value, action information to be output, and the like are output and displayed (step S121). The image display unit of the blood glucose level output means 17 may be a display unit of a portable information device terminal (for example, a mobile phone, a wristwatch, etc.) as shown in FIG.
The information related to the blood glucose level to be displayed can be performed as a rough expression such as the blood glucose level itself, as shown in FIG. 7A, such as low (lower), normal, higher (higher). . This display method can also be selected by the user.

The user can predict the current blood glucose level state by checking the displayed data.
According to the blood sugar level predicting apparatus and the blood sugar level predicting method as described in the above embodiment, life information that is regularly input by the user is stored in the information storage unit 12 as character information. Then, the syntax analysis unit 14 extracts language information from the database of words used for blood sugar level prediction stored in the database storage unit 13 and the character information stored in the information storage unit 12, and the extracted language information Is compared with the above-mentioned database to predict the blood glucose level and output the blood glucose level information from the blood glucose level output means 17.

Thus, since it is possible to predict a blood glucose level using life information that is routinely input by the user for blood glucose level prediction, an input operation performed by the user for blood glucose level prediction is unnecessary. The load can be reduced.
In addition, since it becomes unnecessary for the user to subjectively judge his / her physical condition, variations in user judgment can be prevented, and the accuracy of blood sugar level prediction can be improved.

(Second Embodiment)
A blood glucose level prediction apparatus 10 and a blood glucose level prediction method as a second embodiment according to the present invention will be described with reference to FIG. FIG. 8 is a flowchart showing a blood sugar level prediction method as the second embodiment. In the present embodiment, the same configurations and the same methods as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

  Since the apparatus configuration of the blood sugar level predicting apparatus 10 of the second embodiment is the same as that of the first embodiment, a description thereof will be omitted, and different portions of the processing flow in the blood sugar level predicting method will be described. The second embodiment is different from the flow of the blood sugar level prediction method described in the first embodiment in that the method for obtaining the score Si is different. Therefore, the method for obtaining the score Si will be mainly described.

  In the second embodiment, when a user inputs life information such as daily events as diary-like electronic information from the character string input means 11, words that are considered to have a large influence, words that are considered important, etc. Select and enter identifiable identification information to the target word. The identification information can be selected from, for example, the size of the input character (font size), the thickness of the character, the character color, and the addition of a pictograph. That is, the intention of the user can be reflected in the input electronic information.

In the processing flow shown in FIG. 8, the block i configured with the input character string is 1 (i = 1), and the blood sugar level prediction process is started. Since the steps up to step S107 are the same, description thereof will be omitted, and step S208 for counting the number of modified characters, which is the next step, will be described.
In step S208 for counting the number of modified characters, the presence / absence of character modification, which is identification information, is confirmed, and the number of characters Bi with character modification is counted. In counting the number of characters Bi, for example, it is possible to calculate by setting a coefficient depending on the degree of influence such as x1 for bold and x2 for coloring.

Then, using the number of characters Ai obtained in step S107 and the number of characters modified by character Bi, a score Si is obtained by the following equation (3) (step S111).
Si = Ke · Ai + Kb · Bi (3)
Ke and Kb are constants that are adjusted so that Si is in an appropriate range.
Ai and Bi are positive integers.

  After step S111, through the same processing flow as in the first embodiment, the physical condition S is obtained and the blood sugar level is predicted (step S119). The blood sugar level prediction data obtained from the physical condition S is transmitted to the blood sugar level output means 17 shown in FIG. 1, and the blood sugar level sent from the physical condition estimating means 15 to the image display unit of the blood sugar level output means 17, Alternatively, information related to blood glucose level, action information to be performed, and the like are output and displayed (step S121).

  According to the blood sugar level predicting apparatus 10 and the blood sugar level predicting method of the second embodiment, in addition to the effects of the first embodiment, in addition to electronic information that can reflect the intention of the user, blood sugar level prediction is performed. It becomes possible to perform prediction with higher accuracy.

(Third embodiment)
A blood sugar level prediction apparatus and blood sugar level prediction method as a third embodiment according to the present invention will be described with reference to FIGS. 9 and 10. FIG. 9 is a schematic explanatory view showing a method for obtaining physical condition S in the third embodiment. FIG. 10 is a flowchart showing a blood sugar level prediction method according to the third embodiment. In the present embodiment, the same reference numerals are assigned to the same configurations and the same methods as those in the first embodiment and the second embodiment described above, and the description thereof is omitted.

  Since the apparatus configuration of the blood sugar level predicting apparatus 10 of the third embodiment is the same as that of the first embodiment, description thereof will be omitted, and different parts of the processing flow in the blood sugar level predicting method will be described. The third embodiment is different from the flow of the blood sugar level prediction method described in the first embodiment and the second embodiment in the way of obtaining the physical condition S. .

  In the third embodiment, as in the second embodiment, when the user inputs life information such as daily events as diary-style electronic information from the character string input means 11, words that are considered to have a great influence, Select the words that you think are important, and enter the target words with identifying information that you can identify.

  In the processing flow shown in FIG. 10, the block i configured with the input character string is 1 (i = 1), and the blood glucose level prediction processing is started. Step S <b> 101-Counting the number of modified characters Steps up to S208 are the same as those in the second embodiment, and a description thereof will be omitted.

  In the third embodiment, the number of characters and the number of lines in the portion related to physical condition are also counted and reflected when the score Si is obtained. In addition, among the electronic information input by the user, the number of characters not related to the physical condition is counted and reflected when the physical condition S is obtained.

First, how to obtain the score Si will be described.
Subsequent to step S208 for counting the number of modified characters, the number of characters and the number of lines Wi in the portion related to the physical condition are counted (step S109).
Then, using the number of characters Ai obtained in step S107, the number of characters modified with characters Bi, and the number of characters / number of lines Wi in the portion related to the physical condition, a score Si is obtained by the following equation (4) (step S111).
Si = Ke / Ai + Kb / Bi + Kw / Wi (4)
Ke, Kb, and Kw are constants that are adjusted so that Si is in an appropriate range.
Ai, Bi, and Kw are positive integers.

Further, following step S105 for extracting a language related to physical condition (portion related to physical condition), the number of characters not related to physical condition (non-physical part) is counted from the words and languages constituting the acquired character string (step S300). ).
Subsequently, by using the acquired number of characters not related to the physical condition, a negative score Ki not related to the physical condition is obtained corresponding to all the related character strings (step S302).

When the physical condition is good, the user tends to write many characters that are not related to the physical condition. Using this tendency, a negative score Ki is obtained by the following equation (5).
Ki = k · (Ni−Ai) / Ni (5)
k is a negative constant. Ni is the total number of characters in the character string. Ai is the number of characters related to physical condition.
It can be seen from the following equations (6) and (7) that the absolute value of Ki increases as the number of characters unrelated to physical condition increases.
Ki = 0 at Ni-Ai = 0 (6)
Ki <0 at Ni-Ai> 0 (7)

Next, a physical condition S is obtained using the obtained score Si and negative score Ki (step S115).
An example of how to obtain the physical condition S will be described with reference to FIG. Here, the block of each character string is i (i = 1, 2, 3,..., N). In each character string (i = 1, i = 2,..., I = n), Ai is the number of characters in “part i related to physical condition”, and Ni is the total number of characters in the block of character strings. Then, in addition to the score Si obtained in the same manner as in the first embodiment, the physical condition S is obtained by the following equation (8) using the minus score Ki obtained corresponding to each character string.

  In the processing flow after step S115, the blood sugar level is predicted from the physical condition S through the same processing flow as in the first and second embodiments (step S119). The blood sugar level prediction data obtained from the physical condition S is transmitted to the blood sugar level output means 17 shown in FIG. 1, and the blood sugar level sent from the physical condition estimating means 15 to the image display unit of the blood sugar level output means 17, Alternatively, information related to blood glucose level, action information to be performed, and the like are output and displayed (step S121).

  As described above, according to the third embodiment described above, the number of characters and the number of lines Wi in the portion related to the physical condition are also counted and reflected when the score Si is obtained. Further, the negative score Ki calculated by counting the number of characters not related to the physical condition among the electronic information input by the user is reflected when the physical condition S is obtained. By adding these, the physical condition S can take the sum of the score Si and the negative score Ki and balance the good physical condition and the poor physical condition. Therefore, the blood sugar level predicting apparatus and the blood sugar level of the third embodiment In addition to the above-described embodiment, the prediction method can further increase the accuracy of blood glucose level prediction.

(Modification 1)
The predicted value of the blood glucose level acquired as described above can be used other than the terminal of the information device owned by the user.
For example, data written on the Web, such as SNS and Twitter, is sent to a third party organization via the network for use. Examples of third-party organizations include
(1) Employee cafeteria: A warning that the menu you are going to eat or drink is likely to affect blood glucose levels.
(2) Restaurant: Guidance related to calorie acquisition at the time of menu selection (calling attention such as being discreet).
(3) Medical institution: auxiliary data for deciding the contents of examinations for patients with high blood sugar levels.
Etc.
In addition, blood sugar level prediction can be performed regularly by linking the blood sugar level prediction timing, frequency, and display contents with the user's own electronic schedule book or the like.
It is also possible to reflect the description contents of the electronic schedule book, for example, the number of business trips, the number of entertainments, etc. in the physical condition S.

FIG. 11 shows a daily diary for about one month using the blood sugar level predicting apparatus and the blood sugar level predicting method according to the above-described embodiment, and measured the physical condition S and the physical condition S expressed in the above formula (8) in the early morning. It is an example of the scatter diagram which showed the correlation with a blood glucose level. Here, Kc, Kb, and Kw are each set to 1. According to this, a positive correlation with a correlation coefficient of 0.44 is recognized between the physical condition S and the blood sugar level. It should be noted that the correlation can be further increased by adjusting the values of Kc, Kb, and Kw.
As shown in the above example, the blood sugar level can be predicted from the physical condition S according to the blood sugar level predicting apparatus and the blood sugar level predicting method of the embodiment described above.

  DESCRIPTION OF SYMBOLS 10 ... Blood glucose level prediction apparatus, 11 ... Character string input means, 12 ... Information storage means, 13 ... Database storage means, 14 ... Syntax analysis means, 15 ... Physical condition estimation means, 16 ... Blood glucose level prediction means, 17 ... Blood glucose level output means.

Claims (5)

Input means for inputting life information as electronic character information;
Information storage means for storing the character information input from the input means as a character string;
Database storage means for storing the wording used for blood glucose level prediction as a database;
Syntax analysis means for extracting linguistic information related to blood glucose level by performing syntax analysis using the character string and the database;
Physical condition estimation means for collating the language information extracted by the syntax analysis means with the database and predicting blood glucose level;
An output means for outputting blood sugar level information predicted by the physical condition estimating means. In the blood glucose level prediction apparatus according to claim 1,
The blood glucose level prediction apparatus, wherein the input means and the output means are information terminal devices. Enter life information into the input means as text information,
Storing the input character information as a character string in an information storage means;
The language information related to the blood glucose level is extracted by parsing the character string by the parsing means,
Predicting blood glucose level by collating the language information extracted by the syntax analysis means and a database stored in the database storage means in advance by the physical condition estimation means,
A blood glucose level prediction method, wherein the predicted blood glucose level information is output from an output means. In the blood glucose level prediction method according to claim 3,
The character information input to the input means includes identification information for identifying character information,
The blood sugar level predicting method, wherein the physical condition estimating means predicts a blood sugar level using the identification information in addition to the collation of the language information. In the blood glucose level prediction method according to claim 3 or 4,
The blood sugar level predicting method, wherein the physical condition estimating means predicts a blood sugar level based on a score obtained by digitizing the language information when the physical condition is good and a score when the physical condition is bad.

Images