KR20080113307A - Providing system and method disease be or not prediction framework for service healthcare in mobile - Google Patents

Abstract

A system and a method for providing a framework to determine whether a disease exists for a mobile health management service are provided to predict a specific disease based on probability and feeds back a user's response about a prediction result, thereby improving prediction accuracy according as more bio-signals and symptoms are accumulated. A method for providing a framework to determine whether a disease exists for a mobile health management service comprises the following steps. A DCAP(Disease Combination Appearance Probability) array is generated(120). The DCAP array indicates the probability that each bio-signal and symptom of a combination array appear in a disease group. The DCAP array is formed based on an appearing frequency of a normal group and an appearing frequency of a disease group.

Description

System and method for providing disease discrimination framework for mobile health care service {PROVIDING SYSTEM AND METHOD DISEASE BE OR NOT PREDICTION FRAMEWORK FOR SERVICE HEALTHCARE IN MOBILE}

1 is a flow chart for explaining a disease providing framework providing method according to an embodiment of the present invention

2 is a VSPA arrangement of a disease group and a normal group in a disease discrimination framework providing method according to an embodiment of the present invention

3 is a VSCPA arrangement of a disease group and a normal group in a disease discrimination framework providing method according to an embodiment of the present invention

4 is a DCAP sequence obtained from the VSCPA array of the disease group and the normal population in the disease presence discrimination framework providing method according to an embodiment of the present invention

5 is a specific gravity vector array to be applied to the probability equation for determining the presence of disease with the DCAP array of FIG.

Figure 6 is a block diagram showing a system for providing a disease discrimination framework according to an embodiment of the present invention

7 is a flowchart illustrating a feedback process for an individual user according to an embodiment of the present invention.

8 is a flowchart illustrating a feedback process for a group user according to an embodiment of the present invention.

The present invention relates to a system and method for determining the presence or absence of a disease in humans. In particular, the combination of the biosignals and the symptoms is included in a correlation pattern between the biosignals and the symptoms, and the subject population is included in the disease group. The present invention relates to a system and method for providing a disease presence determination for a mobile health care service that can provide a reliable measurement by using statistical techniques by calculating a probability, learning and feedbacking the probability.

Modern man lives under a lot of stress by external physical or mental factors. People under stress feel internally nervous. These stresses are not always harmful, so moderate stress motivates people, but excessive stress causes serious morbidity. For this reason, stress has been a concern for the general public as well as patients and doctors.

Some biosignals are related to stress. For example, heart rate variability (HRV), body temperature, blood sugar levels, body fat concentrations, and cortisol hormones are known to differ significantly from those of normal people in stress. In daily life, the questionnaire that examines the question form using the causes and symptoms of stress has been mainly used.

However, there are some disadvantages to the existing methods for determining stress based on bio signals or questionnaires. First, since individual differences exist in the normal range of each biosignal, it is difficult to apply them equally to everyone. In addition, since stress causes a large number of biosignal changes, it is difficult to obtain reliable results based on a single biosignal. And the relationship between stress and biosignals used in some methods is not yet clear. For example, the relationship between cortisol hormone levels and stress is still under study. Questionnaire is also a method for determining stress, and the questionnaire that has been verified is somewhat effective in stress determination, but there is a problem that the user must correctly answer a large amount of questions. For example, in a situation such as targeting a user of a mobile terminal, the use of a common questionnaire is limited, and therefore an accurate and very simple questionnaire is required.

Therefore, the above-mentioned bio signals, which are different from those of normal people, are improved by the existing methods of determining stress, which are only limited to the determination of stress from each other, and the degree of disease prediction is determined by considering the bio signals as described above. The predicted disease prediction degree can be used as an accurate judgment data in the reference material, medical certificate or prediction system of stress, and also the user's feedback on the disease prediction method that can provide customized service for each user is required. have.

An object of the present invention for solving the above problems, VSCPA (Vital Sign-Symtom Combination Pair Appearance) indicating the frequency of appearance of the combination pattern of the bio-signals and symptoms extracted from the disease group and the normal group that has already been verified Create and store an array, generate a Dasease Combination Appearance Probability (DCAP) array that represents the probability that a particular biosignal and symptom appears in a disease population with a particular disease from the VSCPA array, and assign the DCAP array to a proposed probability equation. By applying probabilistic prediction of specific diseases in the disease population, and receiving a user response to the predicted results, through this to correct the entire system to provide a more precise disease discrimination framework.

According to an aspect of the present invention to achieve the above object, in a method for providing a health care service through a wireless communication network, each element of the combination arrangement for the combination of symptoms corresponding to the combination of the bio-signals The Disease Combination Appearance Probability (DCAP) sequence representing the probability of occurrence of each biosignal and symptom of the combination array in the disease population based on the frequency of appearance of the normal population and the frequency of the disease population forming each element of the combination array. And a process of predicting the presence or absence of a disease in the disease population based on a weight vector representing a specific gravity between the bio signals and the symptoms of the combination array and the DCAP array, and the predicted disease presence. Receiving a feedback from a user and arranging the DCAP according to the feedback information And adjusting the specific gravity vector, and reflecting the adjusted DCAP array and the specific gravity vector to the DCAP array.

According to another aspect of the present invention for achieving the above object, in the system for providing a health care service through a wireless communication network, each element of the combination arrangement for the combination of symptoms corresponding to the combination of the bio-signals A learning unit for calculating a DCAP array indicating a probability of occurrence of each biosignal and symptom of the combination array in the disease population based on the frequency of appearance of the normal population and the frequency of appearance of the disease population constituting each element of the combination array; A predictor for predicting a disease degree of the disease population based on a specific gravity vector array representing a specific gravity value between each biosignal and symptoms of the combination array and the DCAP array, and information and input of a user received from a plurality of mobile terminals; Contains a user database for classifying and storing values.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, specific details such as specific components are shown, which are provided to help a more general understanding of the present invention, and the specific details may be changed or changed within the scope of the present invention. It is self-evident to those of ordinary knowledge in Esau.

In the present invention, it is assumed that a criterion for distinguishing a disease group from a normal group exists in the biosignals and symptoms and their relationships, and the biosignals of the disease group and the normal group necessary for constructing the pattern in the learning stage and It is assumed that symptom information is given in advance.

1 is a flowchart illustrating a disease providing framework providing method according to an embodiment of the present invention. Referring to Figure 1, it consists of a disease learning process (step 110 ~ 124) and disease prediction and feedback (feed back: step 126 ~ 132) process according to an embodiment of the present invention. First, the biosignal is determined by the measuring device from disease and normal groups by disease-related symptoms, such as headaches, indigestion, skin problems, chest tightness, nervousness, and forgetfulness. For example, heart rate variability (HRV), body temperature, blood sugar level, body fat concentration, etc. are measured (step 110). As in the above example, the biological signals are merely exemplary, and as the types of the biological signals vary, they are useful for accurate disease diagnosis.

Next, in step 112 and 114, a VSPA (Vital Sign-Symptom Pair Appearance) array is generated for the corresponding population. The VSPA array is a basic array for extending to a VSCPA array in steps 116 and 118, which is an array included in the VSCPA array. That is, in order to describe the VSCPA array, first, a VSPA array is generated, and the generation of the VSPA array is an optional step that does not need to be implemented separately.

In the VSPA array, a column shows a biosignal and a row shows a symptom. When there are n different biosignals and each biosignal bi has kj different kinds, the elements in the columns of the VSPA array are represented by Equation 1 below.

In the same way, when there are m different symptoms, and each symptom s has l different kinds, the elements in the rows of the VSPA array are represented by Equation 2 below.

As a result, the size of the VSPA array in the above example is expressed by Equation 3 below.

As such, given the biosignal and symptom information, it is simple and easy to generate a VSPA array. As an example, if we identify these items while generating VSPA arrays for disease and normal populations, we first assume that there are two biosignals {b1, b} and symptoms {s1, s2} for the VSPA array generation. However, there are two types of biosignals and symptoms. Therefore, at this time, as shown in FIG. 2, {b11, b12, b21, b22} and {{s11s12, s21, s22}} are composed of column elements and row elements of the VSPA array. As a result, a 4 x 4 VSPA array is created.

Furthermore, it is assumed that {P1, P2, P3} is a group of subjects included in the disease group and {p4, p5p6} is a group of subjects included in the normal group. In addition, if each subject is represented by <biosignal, symptom> pair (Pair), the subjects of the disease group can be expressed as shown in [Equation 4].

In addition, in the same manner, the subjects of the normal group may be expressed as shown in Equation 5.

_{P1 = <b11, s11>, <b11, s22>, <b22, s11>, <b22, s22>}

_{p2 = <b11, s11>, <b11, s21>, <b21, s11>, <b21, s21>}

_{p3 = <b12, s12>, <b12, s22>, <b22, s12>, <b22, s22>}

_{p4 = <b12, s12>, <b12, s21>, <b21, s12>, <b21, s21>}

_{p5 = <b12, s12>, <b12, s21>, <b22, s12>, <b22, s21>}

_{p6  = < b11 , s11 >, < b11 , s22 >, < b21 , s11 >, < b21 , s22 >}

That is, in FIG. 2, each of the elements forming the VSPA array has a frequency of appearance for pairs of biomythological symptoms in the corresponding population. For example, in the VSPA array 210 for the subjects PP, P included in the disease population in FIG. 2, according to Equation 4, the frequency of occurrence is 2 in the pairs _{<b11, s11>} , and the pair _{<b21. , s11>} has a frequency of 1. In this manner, each element value of the VSPA array 220 of the normal population may be filled. As a result, the normal population VSPA array 220 may include the frequency of each biosignal and symptom for at least two or more symptoms corresponding to at least two or more biosignals.

Each of the VSPA arrays thus generated is expanded to the VSCPA array for the disease population in step 116 and to the VSCPA array for the normal population in step 118. In other words, because VSPA arrays treat each biosignal and symptom independently, to reflect that one biosignal can affect other biosignals and similarly, one symptom can affect other symptoms. The VSPA arrays extend to the VSCPA array.

To this end, the present invention introduces the concept of bio-signal combination and symptom combination. The biosignal combination refers to a set of combinations of biosignals, and the symptom combination refers to a set of combinations of symptoms. That is, the VSCPA array is generated by adding the biosignal combinations to columns and symptom combinations to rows.

For example, as shown in FIG. 3, the biosignal combinations {{ _{b11, b21}, {b11, b22}, {b12, b21}, {b12, b22}}} are added to the column, and the symptom combination {{s _{11, s21 },} { _{s11, s22}, {s12, s21}, {s12, s22}}} when added to a row, the frequency of each element is determined in the same way as in the VSPA array, resulting in an 8 x 8 VSCPA array. .

That is, the VSCPA array is a combination array of elements for a combination of symptoms including individual symptoms corresponding to the combination of biological signals including individual bio signals. As described above, the combination array for forming the VSCPA array may be a matrix consisting of a combination of at least two biosignals and the same number of symptoms as the combination of the biosignals corresponding to the combination of the at least two biosignals. have. However, the present invention is not limited thereto, and the number of combinations of biological signals and the number of combinations of symptoms may be different.

In the VSCPA array 310 for the subjects included in the disease group in FIG. 3, the frequency of occurrence is 2 in pairs _{<b11 and s11>} and the frequency of occurrence is 1 in pairs _{<b21 and s11>} , as well as combination pairs, In this case, the appearance frequency is 1, and in the combination pair <{ _{b12, s22} {s12, s22}>} , the appearance frequency is 1. In this manner, each element value of the VSCPA array 310 of the disease population may be filled. As a result, the VSCPA array 310 of the disease population indicates the frequency of occurrence of each biosignal and symptom of the combination array for the combination of symptoms corresponding to the combination of the biosignals. The VSCPA 310 of the disease population includes the VSPA array 210 of FIG. 2.

Similarly, in the VSCPA array 320 for the subjects included in the normal population in FIG. 3, the frequency of occurrence is 1 in pairs _{<b11 and s11> and} the frequency of occurrence is 1 in pairs _{<b21 and s11>} , as well as in combination. The pair, <{ _{b12, s22} {s12, s22}>} , has a frequency of 1 and the combination pair <{ _{b12, s22}, {s12, s22}>} has a frequency of 0.

In this manner, each element value of the VSCPA array 320 of the normal population may be filled. As a result, the VSCPA 320 of the normal population represents the frequency of appearance of each biosignal and the symptom of the combination array for the combination of the symptoms corresponding to the combination of the biosignals. The VSCPA array 320 of the normal population includes the VSPA array 220 of FIG. 2.

As such, when the VSCPA arrays are generated for each of the normal group and the disease group, in step 120, a Disease Combination Appearance Probability (DCAP) array is calculated. The DCAP array may be calculated based on the VSCPA array 310 of the disease population and the VSCPA array 320 of the normal population, and the probability of occurrence of each biosignal and symptom of a combination array in the disease population is determined. It has an element value.

For example, as shown in FIG. 4, the average of each element of the VSCPA array 310 of the disease group and each element of the VSCPA array 320 of the normal population may be the element of the VSCPA array. For example, the value DCAP of each element of a row column in the VSCPA array may be expressed as in Equation 6 below. In Equation 6, VSCPA (disease) is the value of each element of the row column in the VSCPA 310 of the normal population, and VSCPA (normal) is the value of each element of the row column in the VSCPA 320 of the normal population. .

In the next prediction step, in order to predict the disease degree of the disease population, first, a weight vector is applied in step 122. That is, specific gravity vectors for biosignals and specific gravity vectors for symptoms are required. Here, it is assumed that the specific gravity values of the specific gravity vectors are determined by obtaining expert advice.

The specific gravity vector array has elements of rows and columns of the same combination (biosignal combination and symptom combination) as those of the DCAP array. Once the values of specific gravity vectors of each biosignal and symptom are determined, it is very simple to determine the values of the elements of the specific gravity vector. When the elements of a row and column are represented by a single biosignal or symptom, the value of the specific gravity vector is directly determined by the specific gravity of the element. In addition, when the elements of the row and column are represented by the biosignal combination or symptom combination, the average specific gravity vector value of the biosignal or symptom in the combination is determined as the specific gravity value of the element. The combination array represented by the specific gravity vector array is composed of element values representing the correlation between the biosignals and symptoms unique to each individual, and is prepared to provide customized services for each subject when the biosignals and symptoms of one subject are used. Heat.

For example, when the (x, y) element of the specific gravity vector array has the specific gravity values of weight (x) and weight (y) as elements of the columns and rows of the array, The specific gravity value is the average of weight (x) and weight (y).

FIG. 5 shows the specific gravity vector array 510 to be applied to the probability equation together with the DCAP array 410 of FIG. 4 when the specific gravity vectors 0.3 and 0.7 for the biosignal and the specific gravity vector 0.6 and 0.4 for the symptom. In other words, each element of the specific gravity vector array 510 may be an average of the specific gravity value of each biosignal of the matrix represented by the combination array and the specific gravity value of each symptom. In addition, each of the elements corresponding to a combination including two or more biosignals and a combination including two or more symptoms among the elements of the specific gravity vector array may have an average specific gravity value of the corresponding biosignals and an average specific gravity value of the corresponding symptoms. May be average.

When the specific gravity vector array is generated as described above, in step 124 and 126, the specific gravity vector array representing the specific gravity value between the respective biosignals and the symptoms of the combination array, and each of the biosignals and the symptoms forming the combination array in the disease group appear. The degree of stress of the disease population is predicted based on the DCAP array representing the probability of doing. The specific gravity vector array and the DCAP array may be applied to the following Equation 7 to predict the degree of disease in the disease population.

A predetermined average value of the elements of the DCAP array based on the elements of the specific gravity vector array may be calculated as a probability indicating a disease degree of the disease population. Alternatively, only some of the elements of the WA array may be selected, and a predetermined average value of some elements of the corresponding DCAP array based on the selected elements may be calculated as a probability indicating a disease degree of the disease population.

As described above, the system 600 for implementing the disease presence determination framework providing method for the mobile health care service may be represented as shown in FIG. 6.

6 is a block diagram illustrating a system for providing a disease discrimination framework according to an embodiment of the present invention. As shown in FIG. 6, the disease presence determination framework providing system 600 includes a user database (DB) 616, a learning unit 610, a predicting unit 620, a monitoring unit 602, and a DCAP array control unit ( 604). The learner 610 includes a VSCPA array generator 611 and a DCAP array generator 612, and the predictor 620 includes a specific gravity vector generator 621 and a disease presence determiner 622. do.

The learner 610 performs the learning process of FIG. 1 (see FIG. 1, steps 110 to 126). In particular, when the bio-signals for disease-related symptoms are measured and obtained from each of the normal group and the disease group, the learning unit 610 determines a probability of each biosignal and the symptom of the combination array appearing in the stress group. Calculate the array (see 410 of FIG. 4). The learning unit 610 is based on the frequency of appearance of the normal population constituting each element of the combination arrangement for the combination of symptoms corresponding to the combination of the bio-signals and the frequency of appearance of the disease population constituting each element of the combination arrangement The DCAP array (see 410 of FIG. 4) is calculated.

To this end, the VSCPA array generating unit 611 is a VSCPA array of the normal group (see 320 in FIG. 3) and the VSCPA array of the stress group (310 of FIG. ). In the VSCPA array of the normal population (see 320 of FIG. 3), a VSPA array representing each biosignal and frequency of appearance of at least two symptoms corresponding to at least two or more biosignals measured from the normal population (See 220 in FIG. 2). In addition, the VSPA array of the disease population (see 310 of FIG. 3) indicates the frequency of each biosignal and symptom for at least two or more symptoms corresponding to at least two or more biosignals measured from the disease group. A VSPA array (see 210 in FIG. 2) is included.

The DCAP array generation unit 612 included in the learning unit 610 may include each element of the VSCPA array (see 320 in FIG. 3) of the normal population and the VSCPA array of the disease population (Equation 6). The average of each element of 310 of FIG. 3 is calculated as each element of the DCAP array (see 410 of FIG. 4).

The prediction unit 620 performs the learning process (steps 122 to 126) of FIG. 1. In particular, a specific gravity vector (see 510 of FIG. 5) is generated according to the specific gravity value of the biosignals prepared by an expert and the like, and thus the specific gravity vector and the DCAP array (see 410 of FIG. 4) are generated. To predict the degree of stress in the disease population. The specific gravity vector generation (refer to 510 of FIG. 5) has a specific gravity value between each biosignal and a symptom in a combination array of combinations of symptoms corresponding to the combination of the biosignals.

The specific gravity vector generator 621 included in the predictor 620 receives the specific gravity value of each of the biological signals and the specific gravity value of each symptom to calculate the specific gravity vector (see 410 of FIG. 4). As illustrated in FIG. 5, each element of the specific gravity vector (see 410 of FIG. 4) may be an average of the specific gravity value of each biosignal of the matrix represented by the combination array and the specific gravity value of each symptom. In addition, among the elements of the specific gravity vector (see 410 of FIG. 4), each element corresponding to a combination including two or more biosignals and a combination including two or more symptoms may have an average specific gravity value and a corresponding value. It can be the average of the mean specific gravity values of symptoms.

The disease determination unit 622 included in the prediction unit 620 may determine a predetermined average value of elements of the DCAP array (see 410 of FIG. 4) based on the elements of the DCAP array (see 410 of FIG. 4). It is calculated as the probability of indicating the degree of stress in the disease population. In addition, only some of the elements of the specific gravity vector array (see 410 of FIG. 4) are selected, and a predetermined average value of some elements of the corresponding DCAP array (see 410 of FIG. 4) based on the selected elements is determined by the stress group. It can be calculated as a probability of indicating the degree of stress.

As described above, the disease probability prediction result of the disease group output through the learning unit 610 and the prediction unit 620 is transmitted to the corresponding user, and if the satisfaction of the disease probability prediction result is input from the user, that is, the feedback is input. The feedback information is reconstructed from the DCAP array and the specific gravity vector by receiving a weight of a specific biosignal from a user and reflecting the weight of a specific biosignal from the user to the specific gravity vector generator 621 for a customized service for an individual user. Probably predict the presence of a disease by performing a probabilistic calculation again, and provides more advanced information compared to the disease presence prediction result before receiving feedback from the user. The evolved information is transferred from the corresponding user by recognizing the corresponding user through the user DB 616 that classifies and stores the information of the plurality of users and the input value for each user when performing the prediction of the disease presence of the same user. The weight is set to the specific gravity vector generation unit 621, and based on this to predict the presence of a customized disease only for the user.

Meanwhile, a plurality of pieces of information obtained through the process of predicting the presence of the plurality of users and re-executing the presence of the disease through user feedback on the predicted result, that is, probability information to appear in a group having a specific disease The DCAP array indicating a is stored in the DCAP array control unit 604. In order to establish a framework for a disease prediction service for a group user, the stored information monitors a plurality of user feedbacks through a series of preset monitoring through the monitoring unit 602, and extracts reliable feedback to normal group. And disease groups.

In addition, the DCAP array control unit 604 compares a plurality of user feedback DCAP arrays generated from the preset reference DCAP array, and detects an error. The error value is measured based on the reference DCAP array, and is applied to the existing DCAP array. If the error value is improved, the new DCAP array is created by combining two DCAP arrays. Here, the process of integrating the two DCAP arrays simply consists of summing the corresponding element values of each DCAP array. If the error value is not improved, the generated feedback DCAP array is discarded.

As described above, the system for determining a disease presence framework 600 directly receives feedback on a disease prediction result from a user at the same time as a group or individual disease presence prediction service, thereby accumulating information of many biosignals and symptoms. According to the present invention, the possibility of improvement of the prediction accuracy was confirmed, and the system for determining the disease presence framework 600 may be further improved by feeding back the bio signals and symptoms obtained from one user in the prediction step to the learning step.

7 is a flowchart illustrating a feedback process for an individual user according to an embodiment of the present invention. Referring to FIG. 7, first, a corresponding user inputs feedback on a disease presence determination result predicted in the learning step (see FIG. 1, steps 110 to 124) (step 610). If the result input from the corresponding user is affirmative, the information applied to the corresponding user is stored to reflect the entire disease presence determination framework providing system (step 614). This means that the existing disease presence discrimination framework is corrected by reflecting the DCAP arrangement in the learning process.

However, if the user is negative about the result, the user adjusts the weight of the biosignals and symptoms characteristically present only to the user (step 616). Here, the adjusting of the biosignal and the weight is performed in the application of the specific gravity vector reflecting the previously formed specific gravity, and can be directly adjusted by the corresponding user.

For example, a specific gravity vector is generated using the weight-adjusted data (step 618), and a stochastic calculation is performed from the DCAP array and the specific gravity vector (step 620). Probability is predicted (step 622). The user again inputs feedback on the predicted result (step 624). If the result input from the corresponding user is affirmative, it is checked whether the weight adjusted by the corresponding user has an effect on the determination of his / her disease, and selects a specific gravity vector value most suitable to the user (step 628). In this case, the process of changing the specific gravity vector value may be repeated at least one or more times according to the satisfaction with the result of determining the presence or absence of a disease of a specific user. Then, the adjusted specific gravity vector value selected in step 628 is set only the specific gravity vector value of the user (step 630). The disease presence determination result that the user satisfies is determined in the next disease state by determining the presence or absence of the disease based on the biosignal specific gravity vector and the symptom vector of the user adjusted as described above. Decision service is available.

8 is a flowchart illustrating a feedback process for a group user according to an embodiment of the present invention. Referring to FIG. 8, first, when feedback for each user is input (step 710), a plurality of responses are identified for each user through a preset set of monitoring to confirm consistency, and then a normal group and a disease group are separated. (Step 714).

In this case, when separating the normal group and the disease group, four types for each user to be considered are as follows.

1) the judgment framework determines that there is a disease and the user accepts it;

2) the judging framework determines that there is a disease, and the user denies it,

3) the judging framework determines that there is no disease, and the user accepts it;

4) The judgment framework determines that there is no disease, and the user denies it.

In order to be reflected in the DCAP array in the learning process, as in the case of 1) and 3), when the user accepts a system disease determination, as long as the system decision and the user's response match, the user's data Shall be included in the learning group. However, in this case, since there is a possibility that the user's unresponsive response may be reflected, after monitoring, several reactions are identified for each user to confirm consistency (step 712), and the normal group and the disease group are separated (step 714). .

If the user denies the system disease determination as in the case of 2) and 4) above, it is not appropriate to use the data immediately for the diagnosis of learning because the system judgment and the user's response are inconsistent. not. Therefore, even in this case, as described above, after checking several responses for each user through monitoring (step 712), the consistency and the disease group are separated (step 714).

After a certain period of time in the disease determination process, that is, a feedback DCAP array is generated for each corresponding group of DCAP arrays in a learning process generated through a plurality of user's disease presence prediction and feedback processes (step 716). The error is detected (step 718) by comparison with the DCAP array generated in step i.e. generated prior to the user feedback process. As a result of the comparison, when the error value is measured, when the error value is improved compared to the existing DCAP array, a new DCAP array in which two DCAP arrays are integrated is generated. If the error value has not improved, the generated feedback DCAP array is discarded.

The disease presence determination framework system 600 may be implemented by a network server, a computer, or other dedicated device. In addition, the functions used in the methods and apparatus disclosed herein can be embodied as computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). do. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

As described above, the configuration and operation of a system and method for providing a disease discrimination framework for a mobile health care service according to an embodiment of the present invention can be made, and the description of the present invention has been described with reference to specific embodiments. Branch modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be defined by the described embodiments, but by the claims and equivalents of the claims.

As described above, the present invention includes a combination of biosignals and a combination of symptoms in a correlation pattern between the biosignals and symptoms, and calculates a reliable probability of whether the target group belongs to a stress group. Reliable data on the presence or absence of a disease with superior sensitivity and specificity can be presented.

In addition, the present invention probabilistically predicts a specific disease, and by receiving and feeding back a user response to the predicted result, it is possible to improve prediction accuracy as more biosignals and symptoms are accumulated, and at the same time, a specific target By learning the biosignals and symptoms obtained from the feedback process, the system can be further improved.

Claims (19)

In the method for providing a health care service through a wireless communication network, Each of the combination arrays in the disease population based on the frequency of appearance of the normal population constituting each element of the combination array for the combination of symptoms corresponding to the combination of vital signs and the frequency of occurrence of the disease population constituting each element of the combination array. Generating a Disease Combination Appearance Probability (DCAP) array indicating the probability of the biosignal and symptoms appearing, Predicting the presence or absence of a disease in the disease population based on a weight vector representing a specific gravity between the bio signals and the symptoms of the combination array and the DCAP array Receiving feedback from the user regarding the predicted disease, and Adjusting the DCAP array and specific gravity vector according to the feedback information; A method for providing a disease discrimination framework for a mobile healthcare service, comprising the step of reflecting the adjusted DCAP array and specific gravity vector to the DCAP array. The method of claim 1, wherein the specific gravity vector and the DCAP array are applied to a predetermined probability equation to probably predict a disease level of the disease group. Way. The method of claim 1, wherein the combination array is a matrix composed of a combination of at least two biosignals and the same number of symptoms as the combination of the biosignals corresponding to the combination of the at least two biosignals. Method for providing a disease discrimination framework for a mobile health care service characterized by. The method of claim 1, wherein the generating of the DCAP array comprises: Generating a first VSPA (Vital Sign-Symptom Pair Appearance) array of the normal population indicating each biosignal of the combination array and the frequency of appearance of each symptom. Generating a second VSCPA array of the disease population representing each biosignal of the combination array and the frequency of appearance by symptom; And And generating a VSCPA array based on the generated first and second arrays. 5. The first VSPA array of claim 4, wherein the first VSCPA array indicates each biosignal and frequency of symptoms for at least two or more symptoms corresponding to at least two or more biosignals measured from the normal population. Including, Wherein the second VSCPA array comprises a second VSPA array representing each biosignal and frequency of appearance for at least two symptoms corresponding to at least two or more biosignals measured from the disease population How to provide a disease discrimination framework for mobile health care services. The method of claim 4, wherein each element of the DCAP array is an average of each element of the first VSPA array and each element of the second VSPA array. The mobile health of claim 1, wherein each element of the specific gravity vector is obtained through calculation and normalization of an average value of specific gravity values of respective biological signals and specific gravity values of each symptom of the matrix represented by the combination array. How to provide a disease discrimination framework for managed services. The mobile health of claim 1, wherein each of the elements of the specific gravity vector array corresponding to a combination of biosignals and a combination of symptoms is an average of an average specific gravity value of the corresponding biosignals and an average specific gravity value of the corresponding symptoms. How to provide a disease discrimination framework for managed services. The disease of claim 1, wherein a predetermined average value of some elements of the DCAP array based on some elements of the specific gravity vector array is calculated as a probability indicating a degree of disease in the disease population. How to provide a presence determination framework. According to claim 1, The probability of indicating the presence of disease in the disease population,

(Where P is a probability value, X is a biosignal specific gravity vector, Y is a symptom specific gravity vector, and CP is a set of pairs of <biosignal combinations, symptom combinations> of the subject to be tested for disease) Method for providing a disease discrimination framework for a mobile health care service, characterized in that determined by. The method of claim 1, wherein the feedback is, A method for providing a disease discrimination framework for a mobile health care service, comprising a feedback process for an individual user and a feedback process for a group user. The method of claim 11, wherein the feedback process for the individual user is as follows. The user inputting feedback on the predicted disease determination result; If the user is negative about the result, the user adjusts the weights of the biosignals and symptoms characteristically present only to the user; Generating a specific gravity vector using the weight-adjusted data; Performing a stochastic calculation from the generated specific gravity vector and the DCAP array; A method for providing a disease discrimination framework for a mobile health care service, characterized in that it comprises a probable process of predicting the presence or absence of a disease based on the DCAP array and specific gravity vector. The mobile healthcare service of claim 12, wherein the feedback process for the individual user is stored in order to reflect the information applied to the user in the DCAP arrangement when the result input from the user is positive. How to provide disease discrimination framework for children. The method of claim 12, wherein the feedback process for the individual user is The user receives the feedback on the predicted result once again, If the result input from the corresponding user is positive, confirming the presence or absence of a disease change prediction due to the weight adjusted by the corresponding user, and selecting a weight vector value most suitable for the user; The selected adjusted specific gravity vector value further comprises the step of setting the specific gravity vector value of the user only disease identification framework providing method for a mobile healthcare service. A computer-readable recording medium in which a program for executing the method of any one of claims 1 to 14 is recorded. In the system for providing a health care service through a wireless communication network, Based on the frequency of appearance of the normal population constituting each element of the combination array for the combination of symptoms corresponding to the combination of vital signs and the frequency of occurrence of the disease population constituting each element of the combination arrangement A learning unit that calculates a DCAP array indicating a probability of each biosignal and symptom appearing, A prediction unit for predicting a disease degree of the disease population based on a specific gravity vector array representing the specific gravity value between each biosignal and the symptom of the combination array and the DCAP array; System for providing disease determination framework for mobile health care services, characterized in that it comprises a user database for classifying and storing the user information and input values received from a plurality of mobile terminals. The method of claim 16, wherein the learning unit, VSCPA (Vital Signs) indicating the frequency of each bio-signal and symptom for at least two or more symptoms corresponding to the at least two or more bio-signals measured from the normal group and the disease group -VSCPA array generation unit for generating a symptom pair approach (array) and And a DCAP array generation unit for calculating an average of each element of the VSCPA array as each element of the DCAP array. The method of claim 16, wherein the prediction unit, A specific gravity vector generator which receives the specific gravity value of each of the biological signals and the specific gravity value of each of the symptoms and calculates the specific gravity vector array; A disease presence determining framework for a mobile health care service, comprising: a disease presence predictor configured to calculate a predetermined mean value of elements of the DCAP array based on the elements of the specific gravity vector as a probability indicating a disease degree of the disease group; Provide system. 17. The method of claim 16, The monitoring unit for monitoring the consistency of feedback input from the user with respect to the result of the disease prediction; And a DCAP arrangement control unit for collecting and storing data when a positive feedback is input from the corresponding user with respect to a disease prediction result output from the disease prediction unit, and reflecting the data to the DCAP arrangement. System for providing disease discrimination framework for health care services.

Images