Patent Application
20210057078
The present invention relates to a system, a device, a method, and a program for treating a disorder treatable by behavior change.
In conventional healthcare, a physician can only treat a patient during a medical examination. Treatment provided during the medical examination includes acts such as surgery, procedures, and prescribing medicine, and many disorders are cured by such acts. On the other hand, there are also disorders treatable by changing daily behavior. For disorders and psychiatric disorders caused by lifestyle in particular, it is often the case that changing daily behavior is more effective rather than providing treatment through outpatient medical care. This is because lifestyle is not something found at a hospital which is not on a day-to-day basis, but something found at the “home” of the patient, which is on a day-to-day basis. Therefore, when it comes to treating a disorder caused by behavior in daily life, even a healthcare provider, such as a physician, cannot provide sufficient advice by merely providing explanation during medical examinations several times a month. For the patient as well, situations occur in which the patient does not know how to utilize, on a daily basis, the advice given during the medical examination.
Patent Document 1: JP 2001-92876 A
Patent Document 1 discloses a system configured to sequentially provide to an individual, on a daily basis, a behavior change message for improving a behavior detrimental to health on the basis of data collected from the individual. By using this system, the patient can receive a behavior change message once per day, and thus understand the behavior that should be adopted on that day. However, the system described in Patent Document 1 merely discloses providing a behavior change message solely on the basis of data collected from the individual, and does not provide a solution for providing effective therapy for behavior change.
The present invention has been made in view of the problems described above, and has characteristics such as the following. That is, a system according to an embodiment of the present invention is a system used for treating a disorder treatable by behavior change. The system includes a server and a user terminal, wherein medical traits, associated with a disorder, for indicating medical traits of a patient are clustered into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, the server is configured to store each of a plurality of the behavioral medical traits, a plurality of the knowledge-related medical traits, and a plurality of the cognitive medical traits in association with one or more therapies, the plurality of behavioral medical traits each being further associated with, from among the plurality of knowledge-related medical traits and the plurality of cognitive medical traits, at least a cognitive medical trait, select a therapy to be executed from among the one or more therapies associated with, from among the plurality of behavioral medical traits, a behavioral medical trait selected to be treated, and therapies associated with knowledge-related medical trait information and cognitive medical trait information associated with the behavioral medical trait thus selected, and transmit therapy information for the therapy thus selected, and the user terminal is configured to present information for the therapy on the basis of the therapy information received.
The server may be further configured to store a medical trait state indicating a state of each of the medical traits of each patient, a standard selection probability factor for each of the one or more therapies associated with each of the medical traits, and an individual selection probability factor for each of the medical traits of each patient, the individual selection probability factor for each of the medical traits may be determined on the basis of the medical trait state of each of the medical traits of the patient, and a selection probability of the therapy may be determined on the basis of the standard selection probability factor for the therapy and the individual selection probability factor for a medical trait of the medical traits associated with the therapy.
The individual selection probability factor may be further determined on the basis of a cluster factor, and the cluster factor may be determined per patient for each cluster of the medical traits.
The server may be further configured to store attributes of each patient, the attributes may include at least one of a gender, an age, or an occupation, and the cluster factor may be determined on the basis of the attributes.
The server may be further configured to acquire effectiveness information indicating whether a medical trait associated with the therapy thus selected has improved, update the medical trait state of the patient on the basis of the effectiveness information, and change the individual selection probability factor on the basis of the medical trait state thus updated.
The server may be further configured to acquire effectiveness information indicating whether a medical trait associated with the therapy thus selected has improved, and change a cluster factor of a cluster to which the medical trait belongs on the basis of the effectiveness information.
The server may be further configured to acquire effectiveness information indicating whether the medical trait associated with the therapy thus selected has improved, and change the standard selection probability factor for the therapy thus selected on the basis of the effectiveness information.
The server may be configured to, in the selection of the therapy, select two or more of the therapies, and transmit the therapy information for the two or more therapies thus selected. The user terminal may be configured to present information for the two or more therapies on the basis of the therapy information received, and transmit, to a server, user selection information indicating a therapy selected by a user from the two or more therapies of the information presented. The server may be configured to change at least the standard selection probability factor on the basis of the therapy selection information.
The standard selection probability factor thus changed may be the standard selection probability factor for each of the therapies associated with the medical traits belonging to the cluster of the medical traits associated with the therapy thus selected.
A server according to an embodiment of the present invention is a server used for treating a disorder treatable by behavior change where medical traits, associated with a disorder, for indicating medical traits of a patient are clustered into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, the server being configured to store each of a plurality of the behavioral medical traits, a plurality of the knowledge-related medical traits, and a plurality of the cognitive medical traits in association with one or more therapies, the plurality of behavioral medical traits each being further associated with, from among the plurality of knowledge-related medical traits and the plurality of cognitive medical traits, at least a cognitive medical trait, select a therapy to be executed from among the one or more therapies associated with, from among the plurality of behavioral medical traits, a behavioral medical trait selected to be treated, and therapies associated with knowledge-related medical trait information and cognitive medical trait information associated with the behavioral medical trait thus selected, and transmit therapy information for the therapy thus selected.
A method according to an embodiment of the present invention is a method executed by a system used for treating a disorder treatable by behavior change where medical traits, associated with a disorder, for indicating medical traits of a patient are clustered into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, the system including a server and a user terminal, and the server being configured to store each of a plurality of the behavioral medical traits, a plurality of the knowledge-related medical traits, and a plurality of the cognitive medical traits in association with one or more therapies, the plurality of behavioral medical traits each being further associated with, from among the plurality of knowledge-related medical traits and the plurality of cognitive medical traits, at least a cognitive medical trait, the method including the steps of selecting, by the server, a therapy to be executed from among the one or more therapies associated with, from among the plurality of behavioral medical traits, a behavioral medical trait selected to be treated, and therapies associated with knowledge-related medical trait information and cognitive medical trait information associated with the behavioral medical trait thus selected, and transmitting, by the server, therapy information for the therapy thus selected, and presenting, by the user terminal, information for the therapy on the basis of the therapy information received to execute the therapy.
A method according to an embodiment of the present invention is a method executed by a server used for treating a disorder treatable by behavior change where medical traits, associated with a disorder, for indicating medical traits of a patient are clustered into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, the system including a server and a user terminal, and the server being configured to store each of a plurality of the behavioral medical traits, a plurality of the knowledge-related medical traits, and a plurality of the cognitive medical traits in association with one or more therapies, the plurality of behavioral medical traits each being further associated with, from among the plurality of knowledge-related medical traits and the plurality of cognitive medical traits, at least a cognitive medical trait, the method including the steps of selecting, by the server, a therapy to be executed from among the one or more therapies associated with, from among the plurality of behavioral medical traits, a behavioral medical trait selected to be treated, and therapies associated with knowledge-related medical trait information and cognitive medical trait information associated with the behavioral medical trait thus selected, and transmitting, by the server, therapy information for the therapy thus selected.
A program according to an embodiment of the present invention is a program configured by a set of programs used for treating a disorder treatable by behavior change where medical traits, associated with a disorder, for indicating medical traits of a patient are clustered into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, the set of programs being configured to cause one or more computers to perform storing each of a plurality of the behavioral medical traits, a plurality of the knowledge-related medical traits, and a plurality of the cognitive medical traits in association with one or more therapies, the plurality of behavioral medical traits each being further associated with, from among the plurality of knowledge-related medical traits and the plurality of cognitive medical traits, at least a cognitive medical trait, selecting a therapy to be executed from among the one or more therapies associated with, from among the plurality of behavioral medical traits, a behavioral medical trait selected to be treated, and therapies associated with knowledge-related medical trait information and cognitive medical trait information associated with the behavioral medical trait thus selected, and presenting information for the therapy on the basis of therapy information for the therapy thus selected to execute the therapy.
A program according to an embodiment of the present invention is a program used for treating a disorder treatable by behavior change where medical traits, associated with a disorder, for indicating medical traits of a patient are clustered into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, the server being configured to store each of a plurality of the behavioral medical traits, a plurality of the knowledge-related medical traits, and a plurality of the cognitive medical traits in association with one or more therapies, the plurality of behavioral medical traits each being further associated with, from among the plurality of knowledge-related medical traits and the plurality of cognitive medical traits, at least a cognitive medical trait, the program being configured to cause the server to perform selecting a therapy to be executed from among the one or more therapies associated with, from among the plurality of behavioral medical traits, a behavioral medical trait selected to be treated, and therapies associated with knowledge-related medical trait information and cognitive medical trait information associated with the behavioral medical trait thus selected, and transmitting therapy information for the therapy thus selected.
Through use of the present invention, it is possible to treat a disorder by effectively modifying a behavior of a patient. According to one embodiment, a behavior of a patient is effectively treated by clustering medical traits into behavioral medical traits, knowledge-related medical traits, and cognitive medical traits, defining relationships between each medical trait and relationships between medical traits and therapies, acquiring a state of the patient for each trait, and allowing the patient to select an appropriate therapy for a cause of an undesirable behavior. Furthermore, with use of the effectiveness information of the therapies, a more effective therapy can be provided by changing the selection probability of each therapy on the basis of the effectiveness information. With use of the user terminal of the patient, an appropriate therapy based on an ever-changing state of the patient can be provided. With use of the user terminal of a healthcare provider, it is possible to select a more appropriate therapy on the basis of the selection information of the healthcare provider.
In the present embodiment, the disorder treatable by behavior change is fatty liver, but may be any disorder treatable by behavior change, such as a so-called lifestyle disease such as hypertension, or a psychiatric disorder. The disorder need only be a physically undesirable state, and need not be a disorder in a medical sense. Treatment by behavior change includes preventive medicine. The term “patient” refers to a person who attempts to treat a disorder by behavior change using the present invention, and does not necessarily need to treat the disorder under the guidance of a healthcare provider.
Medical traits related to the disorder of the patient are categorized and clustered into behavioral medical traits (MTs), knowledge-related medical traits (MTs), and cognitive medical traits (MTs). The behavioral MTs are traits pertaining to the behavior of the patient associated with the disorder. The knowledge-related MTs are traits pertaining to the knowledge of the patient associated with the disorder to be treated, and the cognitive MTs are traits pertaining to the cognition of the patient associated with the disorder. Knowledge pertains to objective facts, whereas cognition is the way of thinking of the patient and is subjective. These states of each patient are referred to as behavioral medical trait states (MSs), knowledge-related medical trait states (MSs), and cognitive medical trait states (MSs).
The behavioral MTs, the knowledge-related MTs, the cognitive MTs, and the therapies are given the correlation illustrated in
The present invention treats a disorder by behavior change, and therefore is intended to modify an undesirable behavior of the patient to a desirable one. The cause of the undesirable behavior of the patient is considered to be one or both of not having the correct knowledge associated with the behavior and not having the correct cognition associated with the behavior. Accordingly, an appropriate treatment for eliminating the cause of the undesirable behavior of the patient is carried out, thereby modifying the undesirable behavior of the patient to the desirable one. Further, therapies that directly modify a behavior itself to the desirable one, without modifying knowledge or cognition also exist. In the present invention, it is possible to define the correlation between each MT and therapy as illustrated in
In order to start the information processing in the present embodiment, a Behavioral MT Table, a Knowledge-Related MT Table, a Cognitive MT Table, and a Therapy Table are generated. Furthermore, in order to define the correlation between each MT and therapy illustrated in
The Behavioral MT Table, the Knowledge-Related MT Table, and the Cognitive MT Table include MT IDs, descriptions, and possible states. The MT IDs are each an identifier for referencing an MT, and the descriptions are each a detailed description of the MT identified by the MT ID. The possible states each indicate a possible state of the MT. For example, the MT having the behavioral MT ID “eatAtOnce” is a trait pertaining to the behavior “May binge on opened sweets until all are gone” of the patient, and is indicated as having five possible states on a scale of “1: Strongly disagree” to “5: Strongly agree”. When the patient binges on opened sweets until all are gone, the caloric intake is likely to be excessive, and thus the behavior is not desirable for the treatment of fatty liver. Accordingly, in a case where the patient is in the state “5: Strongly agree”, this indicates that modification is required.
The knowledge-related MT is a trait pertaining to knowledge in association with the disorder. For example, the MT having the knowledge-related MT ID “EatAtOnceIntent” indicates the trait pertaining to the knowledge “To binge on opened sweets is not a bad thing.” That is, the MT indicates a trait of the patient pertaining to whether he or she has the correct knowledge that bingeing on sweets is not a desirable behavior for the treatment of fatty liver, and the possible states indicate how accurately the patient has that knowledge. The state “5: Strongly agree” indicates that the patient lacks the knowledge that bingeing on sweets is a bad thing, and is a state in which modification is required.
The cognitive MT is a trait pertaining to the cognition of the patient in association with the disorder. For example, the MT of the cognitive MT ID “noLeave” indicates a trait pertaining to the cognition “To leave or throw away food is a bad thing” and the possible states indicate how strongly the patient has the cognition “To leave or throw away food is a bad thing”. The cognition that not cleaning your plate is a bad thing is likely to lead to behavior resulting in excessive caloric intake and is not a desirable cognition for the treatment of fatty liver. The state “5: Strongly agree” indicates that the patient has a strong cognition that to leave or throw away food is a bad thing, and is a state in which modification is required.
The Therapy Table includes therapy IDs and descriptions. The therapy IDs are each an identifier for referencing a therapy, and the descriptions are each a detailed description of the therapy. The therapy IDs should be associated with the MT IDs, and information for modifying the associated medical trait to a desirable state is included as the description. Here, the description is information that serves as the basis of the message presented to the patient, and is, for example, “Just throw them away!” for the therapy ID “trush”.
Next, the Behavioral MT-Knowledge-Related MT Relationship Table, the Behavioral-Cognitive MT Relationship Table, the Behavioral MT-Therapy Relationship Table, the Knowledge-Related MT-Therapy Relationship Table, and the Cognitive MT-Therapy Relationship Table are tables indicating the correlation between the MTs and the correlation between the MTs and the therapies illustrated in
For example, the behavioral MT ID “eatAtOnce” in the Behavioral MT-Knowledge-Related MT Relationship Table is associated with the knowledge-related MT ID “EatAtOnceIntent”. This indicates that the behavioral MT ID “eatAtOnce” indicates a trait pertaining to the behavior of whether the patient “May binge on opened sweets until all are gone”, and this behavioral MT is associated with the presence or absence of the knowledge “To binge on sweets is not a bad thing” identified by the knowledge-related MT ID “EatAtOnceIntent”. A patient in the state “5: Strongly agree” for the trait of whether he or she “May binge on opened sweets until all are gone” is thought to binge on sweets until all are gone due to the mistaken knowledge that to binge on sweets is not a bad thing. On the other hand, a patient in the state “1: Strongly disagree” is thought to binge on sweets for another reason. These behavioral MTs and knowledge-related MTs are associated with each other to illustrate such relationships.
Each MT-Therapy Relationship Table associates MT IDs with therapy IDs to identify the therapies for each MT. For example, the therapy ID “trush” is associated with the behavioral MT ID “eatAtOnce” in the Behavioral MT-Therapy Relationship Table, indicating that the behavioral therapy “Just throw them away!” is applicable as a therapy for improving the trait “May binge on opened sweets until all are gone” of the behavioral MT ID “eatAtOnce”. The therapy ID “BeyondGoodandEvil” is associated with the cognitive MT ID “noLeave” in the Cognitive MT-Therapy Relationship Table, indicating that the cognitive therapy “It is more important to treat the disorder that you yourself are facing than assume blind values of good and evil”, which indicates the correct way of thinking, is applicable as a therapy for improving the trait “To leave or throw away food is a bad thing” of the cognitive MT ID “noLeave”.
Furthermore, each MT-Therapy Relationship Table includes a standard selection probability factor. This is a factor for determining the probability of selection of a therapy ID associated with the MT ID, and is a standard applied to all patients. The standard selection probability factor can be set in advance by a healthcare provider, a system provider, or the like, and can be subsequently updated on the basis of the actual effectiveness in all patients, or the like. Highly effective therapies are set to be more likely selected.
In the present embodiment, each of the tables described above is stored in the storage unit 504 of the server 130. These tables are then used to execute the processing of selecting the appropriate therapy for each patient. The operation of the user terminal 120-1 and the server 130 according to the present embodiment is described below using
First, the control unit 401 of the smartphone 120 acquires attribute information and the medical trait states (MSs) of the patient on the basis of input by the patient via the input unit 403 (S701). The attribute information indicates patient information such as a gender, an age, and an occupation. The medical trait states indicate the individual states of the patient for each MT. The medical trait states can be acquired through interaction with a bot incorporated into the application, for example. According to one preferred aspect, at the timing when the application is installed and treatment is initiated, the control unit 401 displays predetermined questions on the display unit 402 and receives patient responses to the questions from the input unit 403, thereby acquiring the medical trait states of the patient at that point in time. For example, the application presents to patient A the question, “Mr. A, do you agree that ‘To leave or throw away food is a bad thing’?” along with the response options “5: Strongly agree, 4: Somewhat agree, 3: Not sure, 2: Somewhat disagree, 1: Strongly disagree”, prompting a response from patient A. In a case where patient A strongly agrees, patient A enters “5” via the input unit 403 in response. Further, because each MT of the patient is modified by practice of the present invention, preferably the patient interacts with the bot again to update the medical trait states of the patient.
The control unit 401 of the user terminal 120-1 transmits the attribute information and the medical trait states of the patient input via the communication unit 405 to the server 130 via the network 110 (S702). The attribute information and the medical trait states may be input via the user terminal 120 of the healthcare provider and transmitted to the server 130. A portion of the medical trait states may be input by the user terminal 120 of the healthcare provider and transmitted to the server 130 while the other portion is input from the user terminal 120 of the patient and transmitted to the server 130.
The server 130 generates a Cluster Factor Table and a Medical Trait State Table for the patient on the basis of the received medical trait states of the patient (S704). In a case where not all states pertaining to the medical traits have been acquired, the medical trait state “3: Not sure” can be input by default, for example, for those medical traits of which states are not acquired. In the present embodiment, the Medical Trait State Table of the patient includes an individual selection probability factor. The individual selection probability factor is one of the factors used in calculating the selection probability of each therapy, and is calculated here by Equation (1) below.
INDIVIDUAL SELECTION PROBABILITY FACTOR=MEDICAL TRAIT STATE×CLUSTER FACTOR Equation 1
Here, cluster factors are set for each patient. A cluster factor is determined on the basis of which cluster, behavioral MTs, knowledge-related MTs, or cognitive MTs, results in effective treatment for the patient. Depending on the patient, therapies for behavioral MTs may exhibit more effectiveness while therapies for cognitive MTs may not be very effective. In such a case, the cluster factor is set so that therapies for behavioral MTs are more likely selected. This cluster factor may be set in advance by the healthcare provider or the like, or may be automatically set on the basis of the attributes. For example, in a case where, for males, therapies for behavioral MTs exhibit more effectiveness, then the cluster factor is set high for a patient having an attribute of male. Examples of the Cluster Factor Table and the Medical Trait State Table for patient A are illustrated in the tables below.
The cluster factors of each cluster of behavioral MTs, knowledge-related MTs, and cognitive MTs of patient A are set to 1.2, 1.0 and 0.8 on the basis of attributes, as illustrated in the Cluster Factor Table (Table 10). The Medical Trait State Table includes MT IDs, cluster types, states, cluster factors, and individual selection probability factors. The MT IDs are each an ID of a medical trait, and the cluster types each indicate the cluster type to which the MT ID belongs. The states are each a state of the medical trait and are determined on the basis of the medical trait states transmitted from the user terminal 120. The cluster factors are each extracted from the Cluster Factor Table on the basis of the cluster type of the MT ID, and the individual selection probability factors are each calculated from the medical trait state and the cluster factor on the basis of Equation (1).
For example, for the MT ID “eatAtOnce”, the table indicates that the cluster type is “behavioral MT” and the medical trait state is “5”, that is, “5: Strongly agree”, on the basis of input by the patient. Then, the table indicates that the cluster factor is input as “1.2” on the basis of the Cluster Factor Table of Patient A, and the individual selection probability factor “6.0” is calculated by multiplying the cluster factor 1.2 by the medical trait state 5.
Next, in the server 130, from among the behavioral medical traits (MTs), a goal behavioral medical trait to be treated is selected (S706). For example, the behavioral MT ID “eatAtOnce” is a trait pertaining to whether the patient “May binge on opened sweets until all are gone.” Selection as the goal behavioral MT means that this behavioral MT is selected as the MT to be treated with the goal of achieving the desirable state of not bingeing on opened sweets until all are gone.
The goal behavioral MT can be selected by various methods. In one suitable working example, the patient selects the goal behavioral MT from behavioral MTs that are easily achievable. This is because gaining a successful experience can increase motivation to improve lifestyle habits. For example, the behavioral MT having the least number of cognitive MTs to be modified can be easily achieved. In a case where one of the two behavioral MTs of eatAtOnce and cookTooMuch is to be selected, the number of cognitive MTs associated with each MT is determined with reference to the Behavioral MT-Cognitive MT Relationship Table. Here, the number of cognitive MTs associated with eatAtOnce is two and the number of cognitive MTs associated with cookTooMuch is three, and therefore eatAtOnce, which is associated with less cognitive MTs, can be selected first as the goal behavioral MT. Alternatively, while the state of the behavioral MT of the patient is in a more desirable state, the MT for which the value of the state (1 to 5) is lowest can be selected, or the MT for which the average value of the medical trait states of the knowledge-related MT and the cognitive MT is lowest can be selected, for example.
Next, the knowledge-related MTs, the cognitive MTs, the therapies, and the medical trait states of patient A associated with the selected goal behavioral MT are acquired from each table to generate a Therapy Selection Probability Table for patient A (S708). To generate the Therapy Selection Probability Table, the selection probabilities of the therapies are determined. The selection probabilities of the therapies are each determined on the basis of the standard selection probability factor and the individual selection probability factor specific to the patient, for each therapy. An example of the Therapy Selection Probability Table for patient A for eatAtOnce with eatAtOnce selected as the goal behavioral MT in the present embodiment is illustrated below.
The Therapy Selection Probability Table includes knowledge-related MT/cognitive MT IDs, therapy IDs, standard selection probability factors, individual selection probability factors, comprehensive selection probability factors, and selection probabilities. The knowledge-related MT/cognitive MT IDs each indicate the knowledge-related MT/cognitive MT ID associated with the goal behavioral MT “eatAtOnce” for extracting the therapy ID. A hyphen (“-”) entered for the knowledge-related MT/cognitive MT ID means that the therapy ID is a therapy ID directly associated with the behavioral MT “eatAtOnce”. The therapy IDs are each a therapy ID directly associated with the goal behavioral MT, or a therapy ID associated with a knowledge-related MT or cognitive MT ID associated with the goal behavioral MT. The comprehensive selection probability factors are each determined on the basis of the standard selection probability factor and the individual selection probability factor, and the selection probability is determined on the basis of the determined comprehensive selection probability factor. The therapy ID for execution is selected from the therapy IDs included in the Therapy Selection Probability Table on the basis of the selection probabilities (S710).
While there are a variety of techniques for the method for generating the Therapy Selection Probability Table, herein first the therapy ID “trush” directly associated with the goal behavioral MT “eatAtOnce” and the standard selection probability factor thereof (1.2) are acquired with reference to the Behavioral MT-Therapy Table (Table 7), and the individual selection probability factor (6.0) of the behavioral MT “eatAtOnce” is acquired with reference to the Medical Trait State Table of Patient A (Table 11). Furthermore, the knowledge-related MT “EatAtOnceIntent” associated with the goal behavioral MT “eatAtOnce” is acquired from the Behavioral MT-Knowledge-Related MT Association Table (Table 5), the therapies “calorieEstimate” and “decisionFatigue” associated with the acquired knowledge-related MT and the standard selection probability factors thereof (1.0 and 0.8) are acquired from the Knowledge-Related MT-Therapy Relationship Table (Table 8), and the individual selection probability factors (5.0) are acquired with reference to the MT ID “EatAtOnceIntent” in the Medical Trait State Table of Patient A (Table 11). Similarly, the therapies for cognitive MTs associated with the goal behavioral MT and the standard selection probability factors thereof are acquired from the Behavioral MT-Cognitive MT Association Table (Table 6) and the Cognitive MT-Therapy Relationship Table (Table 9), and the individual selection probability factors are acquired with reference to the MT IDs in the Medical Trait State Table of Patient A (Table 11).
A comprehensive selection probability factor Fn and a selection probability Pn are calculated by the equations below.
Here, given therapy numbers are assigned to therapies starting from the top of the Therapy Selection Probability Table (Table 12), n is the therapy number of the therapy for which the selection probability is to be calculated. The denominator of the right side of Equation (3) is the sum of the comprehensive selection probability factors of all therapies, and N is the number of selectable therapies (6 in the present embodiment).
For example, in the Therapy Selection Probability Table of Patient A (Behavioral MT ID=eatAtOnce) (Table 12), the comprehensive selection probability of the therapy ID “trush” is 7.20, which is calculated by multiplying the individual selection probability factor 6.0 by the standard selection probability factor 1.2. Then, the selection probability 0.335 is calculated by dividing the comprehensive selection probability 7.20 of the therapy ID “trush” by the sum of the comprehensive selection probabilities for all therapies in the Therapy Selection Probability Table.
Next, the control unit 501 of the server 130 selects a therapy for execution on the basis of the selection probabilities in the Therapy Selection Probability Table (S710), and transmits the therapy information for the selected therapy to the user terminal 120 via the communication unit 505 (S712). Here, the selection of the therapy is made by selecting a therapy ID on the basis of the selection probabilities of the Therapy Selection Probability Table of the patient. The therapy information indicates the information presented to the user for the selected therapy, and here includes the description for the selected therapy ID.
For example, in a case where “trush” is selected as the therapy ID for the behavioral MT ID “eatAtOnce,” the control unit 501 of the server 130 acquires the description “May binge on opened sweets until all are gone” for the behavioral MT ID “eatAtOnce” with reference to the Behavioral MT Table (Table 1) stored in the storage unit 504, and further acquires the information of the description “Just throw them away!” of the therapy “trush” with reference to the Therapy Table (Table 4). Then, the therapy information is generated on the basis of this information. For example, the message “To ensure that you do not binge on opened sweets until all are gone, just throw them away!” is generated and included in the therapy information.
When the control unit 401 of the user terminal 120-1 receives the therapy information, the information for the therapy is presented on the display unit 402 on the basis of the therapy information (S714). Here, the message “To ensure that you do not binge on opened sweets until all are gone, just throw them away!” is displayed on the display unit 402. The therapy information can also be presented to the patient by audio by using an output unit such as a speaker, or by another method.
A patient presented with the therapy executes the presented therapy to modify the behavior. For example, the behavior of patient A for which the state pertaining to the behavioral MT “Binges on opened sweets until all are gone” is “5: Strongly agree” and to whom the message “To ensure that you do not binge on opened sweets until all are gone, just throw them away!” is presented is expected to be modified to discarding remaining sweets without bingeing even if the sweets are opened. This makes it possible to suppress caloric intake.
Subsequently, the control unit 401 of the user terminal 120-1 executes an effectiveness information acquisition step (S716), and transmits the acquired effectiveness information to the server 130 (S718). For example, after a predetermined period has elapsed following presentation of the information for the therapy, the query message “Do you still binge on opened sweets until all are gone?” to confirm effectiveness, along with response options “5: Strongly agree, 4: Somewhat agree, 3: Not sure, 2: Somewhat disagree, 1: Strongly disagree” are presented on the display unit 402, and the effectiveness information is acquired by reception of a selection input from the patient.
In a case where the presented therapy is associated with a knowledge-related MT and a cognitive MT, the effectiveness can be confirmed by confirmation that the knowledge-related MT and the cognitive MT have been modified. Because it is expected that the goal behavioral MT is modified by modification of the knowledge-related MT and the cognitive MT, preferably the effectiveness information of the goal behavioral MT is acquired. As illustrated in
The control unit 501 of the server 130, upon acquisition of the effectiveness information, updates the Medical Trait State Table on the basis of the effectiveness information (S720). For example, for the effectiveness of the therapy ID “trush” as a treatment for the behavioral MT ID “eatAtOnce”, the state of the MT ID “eatAtOnce” in the Behavioral Medical Trait State Table of Patient A is changed to “1” in a case where the response to the question “Do you still binge on opened sweets until all are gone?” is “1: Strongly disagree”.
Furthermore, the standard selection probability factor for the therapy is increased in a case where the medical trait state is improved, and decreased in a case where there is no effect (S722). With the standard selection probability factor modified on the basis of the actual treatment results of all users, the probability that a more effective therapy will be selected is increased, enabling more effective treatment.
The standard selection probability factor may be changed by changing the standard selection probability factors for all therapies, or by changing only the relationship with the MT confirmed as effective. For example, the therapy ID “postSatisfactionOverEating” is associated with both cognitive MT IDs “noRestriction” and “worryAboutShortness” in Table 9. In a case where the therapy ID “postSatisfactionOverEating” is selected in relationship to the cognitive MT ID “noRestriction” of the two cognitive MT IDs, and is confirmed as effective, only the standard selection probability factor “0.6” of the therapy ID “postSatisfactionOverEating” associated with the cognitive MT ID “noRestriction” may be increased, or the standard selection probability factor “0.7” associated with the cognitive MT ID “worryAboutShortness” may also be increased.
The standard selection probability factor can also be changed on a per cluster basis. That is, the standard selection probabilities for all therapies associated with MTs belonging to the cluster to which the MT associated with the therapy confirmed as effective belongs can be increased by a predetermined amount. Other therapies belonging to the cluster to which the effective therapy belongs are also similarly considered highly effective, and therefore the selection probability of the entire cluster is increased. For example, in a case where, as a treatment for the behavioral MT ID “eatAtOnce”, the effectiveness of treatment by the therapy ID “trush” is confirmed, then the selection probabilities of all therapies associated with behavioral MTs are increased.
Further, a cluster factor for an individual patient can also be modified on the basis of the effectiveness information. For example, the cluster factor of a cluster to which a medical trait associated with a therapy confirmed as effective belongs is increased and, in a case where there is no effect, is decreased. Whether a treatment for any cluster is effective may differ according to the patient. Because the cluster to which the medical trait associated with an effective therapy belongs is likely a cluster effective for that patient, increasing the cluster factor increases the selection probability of a therapy for that cluster, enabling a more effective treatment.
Next, the control unit 501 of the server 130 determines whether the goal behavioral MT has been sufficiently modified (S724) and, in a case where it is determined that the goal behavioral MT has been sufficiently modified, ends the treatment for this goal behavioral MT, returns to the goal behavioral medical trait selection step (S706), and selects a new goal behavioral MT, and the subsequent processing is repeated. For example, in a case where the state of the goal behavioral MT becomes “1: Strongly disagree”, it can be determined that the goal behavioral MT has been sufficiently modified.
In a case where it is determined that the goal behavioral MT has not been sufficiently modified, the control unit 501 returns to the Therapy Selection Probability Table generation step (S708) and updates the Therapy Selection Probability Table, and then the subsequent processing is repeated. Since the Medical Trait State Table of the patient and the selection probabilities have been updated according to the therapies already executed by the patient (S720 to S722), the Therapy Selection Probability Table is updated on the basis of the updated Medical Trait State Table and selection probabilities. The medical trait states for the knowledge-related MT and the cognitive MT sufficiently modified by therapies already executed indicate favorable states, and thus the selection probabilities of therapies for these are changed to lower values, and a therapy for an MT still in an undesirable state is preferentially selected.
With use of the present embodiment, medical traits are clustered into behavioral MTs, knowledge-related MTs, and cognitive MTs and stored in association with therapies suitable for the MTs, and the state of the patient with respect to each trait is stored, making it possible to select a therapy appropriate for the cause of the undesirable behavior of the patient and effectively modify the behavior of the patient to a desired behavior. Furthermore, the effectiveness information of treatments is acquired and the selection probability of the therapy is changed on the basis of this effectiveness information, making it possible to provide a more effective therapy. Furthermore, because the user terminal of the patient is used, an appropriate therapy based on an ever-changing state of the patient can be provided.
The second embodiment differs from the first embodiment in that the user of the user terminal 120 is the healthcare provider and the embodiment includes S801, S802, and S811 as illustrated in
In the present embodiment, a user terminal 120-2 is a tablet used by the healthcare provider, but may be another electronic device such as a smartphone or a computer. The healthcare provider inputs, via the user terminal 120-2 of the healthcare provider, the attributes and medical trait states of the patient acquired during interaction with the patient (S701), and transmits the information to the server 130 (S702).
The server 130 generates a Cluster Factor Table and a Medical Trait State Table (S704), and subsequently transmits goal behavioral medical trait input instructions to the user terminal 120-2 (S801). A message prompting input of a goal behavioral MT for the patient is displayed on the display unit 402 of the user terminal 120-2 that receives the instructions, a goal behavioral MT is then selected by the healthcare provider, and the goal behavioral medical trait selection information is transmitted to the server 130 (S802). The server 130, selects a goal behavioral medical trait on the basis of the information (S706), and generates a Therapy Selection Probability Table on the basis of the selected goal behavioral MT (S708).
The control unit 501 of the server 130 selects a therapy on the basis of the Therapy Selection Probability Table in the same way as in the first embodiment (S710), and transmits the therapy information (S712). The user terminal 120-2 that receives the therapy information presents the information for the therapy on the display unit 402 (S714). In a case where a plurality of therapies are selected by the server 130 and the plurality of therapies are presented on the display unit 402, the healthcare provider selects the therapy to be actually applied to the patient. The user terminal 120-2 acquires user selection information indicating the selected therapy and transmits the user selection information to the server 130 (S811). The healthcare provider then provides, to the patient, guidance based on the therapy selected to be applied to the patient.
After a predetermined period elapses, another medical examination is carried out with the patient to inquire about the effectiveness of the applied treatment, and the effectiveness information is input to the user terminal 120-2 (S716) and transmitted to the server 130 (S718). The server 130 updates the Medical Trait State Table and the selection probabilities on the basis of the user selection information and the effectiveness information (S720, S722). The therapy selected by the healthcare provider on the basis of the user selection information is presumed to be an appropriate therapy, and therefore the standard selection probability of the therapy is increased. That is, the standard selection probability is varied by using the selection by the healthcare provider as instructor information. The standard selection probabilities associated with all therapy IDs of the therapy may be changed, or only the relationship with the MT to be treated may be changed. The standard selection probabilities of all therapies associated with the MTs of the cluster of the MT associated with the user-selected therapy can also be increased.
Then, the control unit 501 of the server 130 determines whether modification of the goal behavioral MT is completed (S724) and, in a case where it is determined that modification is completed, returns to the input instruction transmission step (S801) for determining the next goal behavioral medical trait and, in a case where modification is not completed, returns to the Therapy Selection Probability Table generation step (S708) and updates the Therapy Selection Probability Table, and then the subsequent processing is repeated.
In the present embodiment, because the selection probability of the therapy is changed with selection of the therapy by the healthcare provider serving as instructor information, the selection probability of the therapy considered to be highly effective is increased, making it possible to carry out more effective treatment.
The third embodiment differs from the first and second embodiments in that the disorder to be treated is hypertension. Hereinafter, the differences from the first embodiment and the second embodiment will be mainly described.
The information processing flow in the present embodiment is similar to that of
Furthermore, the following table is generated on the basis of these tables as well as the cluster factors and the acquired medical trait states of patient A.
An example of the Therapy Selection Probability Table for patient A for tooMuchSoySauce with tooMuchSoySauce selected as the goal behavioral MT in the present embodiment is illustrated below.
In the same way as in the first and second embodiments, a therapy is selected on the basis of the Therapy Selection Probability Table, and the therapy information associated with the treatment is presented on the user information terminal 120.
The fourth embodiment differs from the first to third embodiments in that the disorder to be treated is a psychiatric disorder (depression). Hereinafter, the differences from the first to third embodiments will be mainly described.
The information processing flow in the present embodiment is similar to that of
Furthermore, the following table is generated on the basis of these tables as well as the cluster factors and the acquired medical trait states of patient A.
An example of the Therapy Selection Probability Table for Patient A for homePrison with homePrison selected as the goal behavioral MT in the present embodiment is illustrated below.
In the same way as in the first to third embodiments, a therapy is selected on the basis of the Therapy Selection Probability Table, and the therapy information associated with the treatment is presented on the user information terminal 120.
Similarly, for other disorders treatable by behavior change as well, the present invention can be implemented using a similar information processing flow by preparing a table associated with the disorder.
Further, while the functions of the user terminal 120-1 of the patient, the user terminal 120-2 of the healthcare provider, and the server 130 have been described in the embodiments described above, these functions can be provided and implemented by any of the devices included in the system according to the present invention. For example, in the first embodiment, each table stored in the storage unit 504 of the server 130 can be stored in the storage unit 404 of the user terminal 120-1, and all functions of the server 130 can be performed by the user terminal 120-1.
The embodiments of the present invention have been described for illustrative purposes, but the present invention is not limited to these embodiments. The present invention can be implemented in various forms without departing from the spirit thereof.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2018/001407 | 1/18/2018 | WO | 00 |
