APPARATUS AND METHOD FOR PROVIDING CONTEXT-BASED INTERVENTION - Patent application

Abstract

A computer-implemented method for providing a context-based healthcare intervention to a user is provided, comprising: obtaining 202 context information from a plurality of sources; updating 204 a user profile associated with the user based on the obtained context information, the user profile having an interest level value for each of a plurality of healthcare topics; selecting 206 one of the plurality of healthcare topics based on the interest level value in the user profile; and providing 208 an intervention recommendation to the user based on the selected healthcare topic, the intervention recommendation relating to suggested information related to the selected healthcare topic or a suggested adjustment to a user device.

Description

[Technical field]

[0001] The present invention relates to an apparatus and method for providing context-based healthcare interventions to a user.The present invention also relates to an apparatus and method for optimizing the provision of context-based healthcare intervention recommendations to a user. [Background technology]

[0002] Changing user behavior to achieve health benefits has always been difficult, especially if lasting health benefits are desired. User adherence to predefined, static schedules toward a goal, whether exercise or dieting, has traditionally been low. Diet coaches and personal trainers are usually more effective because they adjust their strategies based on individual characteristics, preferences, and progress.

[0003] Currently, there are several known solutions for building digital coaching solutions that tailor intervention strategies based on the user, which typically involve interaction with the user via a graphical user interface. However, these solutions typically focus on achieving one personal health goal (e.g., losing weight or running a marathon) for only a specific individual, rather than achieving multiple personal health goals simultaneously for multiple users. There are also many aspects that currently known solutions do not adequately address. Summary of the Invention [Problem to be solved by the invention]

[0004] As noted above, there are numerous shortcomings associated with currently available solutions for providing healthcare coaching and guidance interventions to a user or multiple users, for example for multiple family members in a home environment.

[0005] For example, currently known systems do not take into account the fact that the relevance of coaching content to a user may change rapidly over time due to changes in the user's health status. Also, initially, it is often unclear how to personally coach a user in an effective and acceptable way, and therefore several different strategies may need to be explored, and this need to try different strategies is not taken into account in currently available solutions. Furthermore, currently known systems do not address the fact that the coaching process requires a continuous interaction between the user and the coaching system to gather contextual information that cannot be captured by sensor data alone.

[0006] It would therefore be advantageous to provide an improved method for providing context-based healthcare interventions to users.The present invention relates to instantiating, monitoring and adapting personalized behavior-changing interventions based on personal dynamic interest profiles, particularly in the fields of personal coaching systems, connected personal care and / or health devices, conversational user interfaces and artificial intelligence. [Means for solving the problem]

[0007] To better address one or more of the above-mentioned problems, in a first aspect, a computer-implemented method for providing a context-based healthcare intervention to a user is provided, the method comprising: acquiring context information related to at least one of a user's behavior, a user's physiological state, a user's current location, a current time and / or date from a plurality of sources; updating at least one of a plurality of interest level values ​​of a user profile related to the user based on the acquired context information, the user profile having an interest level value for each of a plurality of healthcare topics, the interest level value indicating the user's degree of interest and / or requirement for receiving support for each healthcare topic; selecting one of a plurality of healthcare topics based on the interest level value in the user profile; and providing an intervention recommendation to the user based on the selected healthcare topic, the intervention recommendation relating to suggested information related to the selected healthcare topic or a suggested adjustment for a user device.

[0008] In some embodiments, the step of selecting the healthcare topic comprises the steps of selecting a healthcare topic having an associated level of interest value higher than a first predetermined threshold, determining a change in level of interest value for at least one of the plurality of healthcare topics following an update of the user profile and selecting the healthcare topic if the determined change in level of interest value is greater than a second predetermined threshold, or selecting a healthcare topic that is related to or similar to another healthcare topic having an associated level of interest value higher than a predetermined threshold.

[0009] In some embodiments, the computer-implemented method may further include obtaining a persuasive style preference of the user for the selected healthcare topic. In these embodiments, providing an intervention recommendation may be further based on the obtained persuasive style preference of the user for the selected healthcare topic.

[0010] In some embodiments, the proposed adjustment to the user device may include at least one of changing a setting on the user device, turning on the user device, and turning off the user device.

[0011] In some embodiments, the user profile may further include a performance level value for each of a plurality of healthcare topics, and in these embodiments, providing the intervention recommendation may be further based on the performance level values ​​for the selected healthcare topics in the user profile.

[0012] In some embodiments, the computer-implemented method may further include receiving input from a user indicating acceptance or rejection of the provided intervention recommendation. In these embodiments, if the received input indicates acceptance, the method may further include implementing the intervention recommendation, and if the received input indicates rejection, the method may further include providing a different intervention recommendation to the user based on the selected healthcare topic.

[0013] In some embodiments, if the received input indicates acceptance, the method further comprises determining a change in effectiveness of the performed intervention over a predetermined time period. In these embodiments, the computer-implemented method may further comprise determining a difference between the determined change in effectiveness of the performed intervention and an expected effectiveness indicator representative of an expected change in effectiveness of the performed intervention, and adjusting parameters and / or content of the performed intervention if the determined difference is greater than a predetermined threshold.

[0014] In some embodiments, if the received input indicates acceptance, the method further includes determining an overall satisfaction score indicative of the user's satisfaction with the performed intervention, where determining the overall satisfaction score comprises determining a progress satisfaction score indicative of the user's satisfaction with the progress resulting from the performed intervention, determining a progress-oriented score indicative of the user's degree of willingness to accept more aggressive persuasive measures relative to gentler persuasive measures, determining a persuasion satisfaction score indicative of the user's satisfaction with the persuasive style associated with the performed intervention, and determining the overall satisfaction score based on the progress satisfaction score, the progress-oriented score, and the persuasion satisfaction score.

[0015] In some embodiments, the method may further comprise determining a difference between the overall satisfaction score and an expected satisfaction score, the expected satisfaction score indicating the user's expected satisfaction with the performed intervention, and adjusting parameters and / or content of the performed intervention if the determined difference is greater than a predefined threshold.

[0016] In some embodiments, the computer-implemented method may further include obtaining a plurality of healthcare-related messages, determining a priority value for each of the plurality of healthcare-related messages based on at least one of the obtained contextual information and the selected healthcare topic, and providing at least one of the plurality of healthcare-related messages to a user based on the priority value.

[0017] In some embodiments, providing at least one of the plurality of healthcare-related messages may be further based on a load indicator value, the load indicator value indicating a likelihood of overloading a user with excessive information.

[0018] In some embodiments, the computer-implemented method may further include updating the strain indicator value after providing the healthcare-related message to the user.

[0019] In a second aspect, there is provided a computer program product comprising the computer readable medium, the computer readable medium having computer readable code embodied therein, the computer readable code being configured, when executed by a suitable computer or processor, to cause the computer or processor to perform a method according to the first aspect.

[0020] In a third aspect, an apparatus for providing a context-based healthcare intervention to a user is provided, the apparatus having a processor configured to: acquire context information related to at least one of a user's behavior, a user's physiological state, and a user's environment from a plurality of sources; update at least one of a plurality of interest level values ​​of a user profile associated with the user based on the acquired context information, the user profile having an interest level value for each of a plurality of healthcare topics, the interest level value indicating a user's degree of interest and / or requirement for receiving support for the respective healthcare topic; select one of a plurality of healthcare topics based on the interest level value in the user profile; and provide an intervention recommendation to the user based on the selected healthcare topic, the intervention recommendation relating to suggested information related to the selected healthcare topic or a suggested adjustment for a user device.

[0021] According to the above-mentioned aspects and embodiments, limitations of existing technologies are addressed. In particular, the above-mentioned aspects and embodiments enable medical interventions to be recommended and provided to a user based on contextual information obtained from multiple different sources. The above-mentioned embodiments provide balanced personalized coaching on multiple health goals operating in an ecosystem of connected personal care and / or health devices and in a home environment. In this way, the embodiments described in the present disclosure enable: -Building insights over time through data collection and ongoing natural language dialogue; - careful dynamic interest profiling of each user, ensuring that each user receives support for issues relevant to each user; - Tailored interactions based on context, user profile, and persuasion style preferences; -Combines device behavior with conversation-based coaching to provide personalized behavior change strategies based on efficacy and acceptance.

[0022] Thus, improved methods and apparatus are provided for providing a context-based healthcare intervention to a user. These and other aspects of the disclosure will be apparent from and elucidated with reference to the embodiments described hereinafter.

[0023] For a better understanding of the embodiments, and to show more clearly how they may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings in which: [Brief description of the drawings]

[0024] [Figure 1] FIG. 1 is a block diagram of an apparatus for providing context-based healthcare intervention to a user according to one embodiment. [Diagram 2] 1 illustrates a method for providing a context-based healthcare intervention to a user according to one embodiment. [Diagram 3] FIG. 1 illustrates a system for providing context-based healthcare intervention to a user according to one embodiment. [Figure 4] FIG. 4 shows the Caring Home module of FIG. 3. [Diagram 5] FIG. 2 is a schematic diagram of a hierarchical structure of multiple healthcare topics in a database. [Figure 6] FIG. 2 is a schematic diagram of a care module related to a healthcare topic. [Figure 7] FIG. 5 is a schematic diagram of a user profile management unit of the apparatus of FIG. 4; [Figure 8] FIG. 2 is a schematic diagram showing contextual user data in a user profile for multiple healthcare topics in a hierarchical structure. [Figure 9] 1 shows an example of a scale for each of multiple parameters in a user profile for a healthcare topic. [Figure 10] FIG. 1 is a flow diagram shown in a relevant direction for an activated care module according to one embodiment. [Figure 11] FIG. 13 is a flow diagram shown in the relevant direction of a deactivated care module according to one embodiment. [Figure 12] FIG. 5 is a schematic diagram of an interaction scheduling unit of the apparatus of FIG. 4; [Figure 13] FIG. 13 is a schematic diagram of the dialogue scheduling unit of FIG. 12. [Figure 14] FIG. 13 is a finite state diagram illustrating the behavior of the interaction handlers of the interaction scheduler of FIG. 12. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] As described above, an improved apparatus and method of operating the apparatus are provided that addresses existing problems.

[0026] 1 shows a block diagram of an apparatus 100 according to one embodiment that may be used to provide a context-based healthcare intervention to a user. Although the operation of the apparatus 100 is described below in the context of a single user, it will be appreciated that the apparatus 100 is capable of providing context-based healthcare intervention to multiple users.

[0027] 1, the device includes a processor 102 that controls the operation of the device 100 and may implement the methods described herein. The processor 102 may include one or more processors, processing units, multi-core processors, or modules configured or programmed to control the device 100 in the manner described herein. In certain implementations, the processor 102 may include multiple software and / or hardware modules, each configured to or for performing individual steps or multiple steps of the methods described herein.

[0028] Briefly, the processor 102 is configured to obtain context information from multiple sources, the context information relating to at least one of a user's activity, a user's physiological state, and a user's environment. The multiple sources may include a user device. For example, the context information may be user device usage information relating to an action performed by a user using the user device. The user device may be a personal care device (e.g., electric toothbrush, hair dryer, etc.) or a personal health device (e.g., weight scale, behavior tracking device, etc.).

[0029] Based on the retrieved context information, the processor 102 is configured to update a user profile associated with the user, the user profile including an interest level value for each of a plurality of healthcare topics, the interest level value indicating the user's degree of interest and / or requirement for receiving support regarding the respective healthcare topic. The processor 102 is further configured to select one of the plurality of healthcare topics based on the interest level value in the user profile.

[0030] The processor 102 is also configured to provide an intervention recommendation to the user based on the selected healthcare topic, the intervention being associated with suggested information related to the selected healthcare topic or a suggested adjustment for the user device.

[0031] In some embodiments, the device 100 may further include at least one user interface 104. Alternatively or additionally, the at least one user interface 104 may be external to the device 100 (i.e., separate or remote from the device 100). For example, the at least one user interface 104 may be part of another device. The user interface 104 may be for use in providing information resulting from the methods described herein to a user of the device 100. Alternatively or additionally, the user interface 104 may be configured to receive user input. For example, the user interface 104 may allow a user of the device 100 to manually input commands, data, or information. In these embodiments, the processor 102 may be configured to obtain the user input from the one or more user interfaces 104.

[0032] The user interface 104 may be a user interface that may simulate and / or provide a natural language conversation with a user. Thus, in some embodiments, the user interface 104 may be referred to as a conversational user interface. A conversational user interface may be implemented as a chatbot on a mobile device (e.g., a smartphone) that may provide a graphical user interface, or as a voice-based virtual assistant in a smart speaker device. As mentioned above, in some embodiments, the device 100 may be capable of providing context-based interventions to multiple users. In some embodiments, the device 100 may have a user interface 104 for each of multiple users. Alternatively, in some embodiments, the device 100 may include a single user interface 104 for interacting with multiple users, e.g., multiple family members of a family.

[0033] User interface 104 may be any user interface that allows for the rendering (or output or display) of information to a user of device 100. Alternatively or additionally, user interface 104 may be any user interface that allows a user of device 100 to provide user input and interact with and / or control device 100. For example, user interface 104 may include one or more switches, one or more buttons, a keypad, a keyboard, a touch screen or application (e.g., on a tablet or smartphone), a display screen, a graphical user interface (GUI) or other visual rendering element, one or more speakers, one or more microphones or any other audio element, one or more lights, an element for providing haptic feedback (e.g., a vibration function), or any other user interface or combination of user interfaces.

[0034] In some embodiments, the device 100 may include a memory 106. Alternatively or additionally, one or more memories 106 may be external to the device 100 (i.e., separate or remote from the device 100). For example, one or more memories 106 may be part of another device. The memory 106 may be configured to store program code executable by the processor 102 to perform the methods described herein. The memory may be used to store information, data, signals, and measurements acquired or created by the processor 102 of the device 100. For example, the memory 106 may be used to store (e.g., in a local file) multiple user profiles. The processor 102 may be configured to control the memory 106 to store multiple user profiles. Examples of memory in the context of the present disclosure may include an interaction content storage unit (as shown in FIG. 6), a contextual user data storage unit (as shown in FIG. 7), etc.

[0035] In some embodiments, device 100 may include a communication interface (or circuitry) 108 to enable device 100 to communicate with any interface, memory, and / or device internal or external to device 100. Communication interface 108 may communicate with any interface, memory, and / or device wirelessly or via a wired connection. For example, communication interface 108 may communicate with one or more user interfaces 104 wirelessly or via a wired connection. Similarly, communication interface 108 may communicate with one or more memories 106 wirelessly or via a wired connection.

[0036] It will be understood that FIG. 1 shows only the components necessary to illustrate one embodiment of device 100, and that in an actual implementation device 100 may have components in place of or in addition to those shown.

[0037] 2 illustrates a computer-implemented method for providing a context-based healthcare intervention to a user, according to one embodiment. The illustrated method may generally be performed by or under the control of the processor 102 of the device 100.

[0038] 2, in block 202, context information is obtained from multiple sources. More specifically, the context information may be obtained by processor 102 of device 100. In some embodiments, the context information may be obtained from one or more databases in memory 106, which may be memory of device 100 or memory external to device 100, such as memory in a user device (e.g., a personal care device or personal health device, such as an electric toothbrush).

[0039] The obtained context information is related to at least one of the user's activity, the user's physiological state, and the user's environment. For example, the obtained context information may include one or more of the following: user device usage information (e.g., frequency of use and / or device settings in use for an electric toothbrush), location information (e.g., the user's current location), weather information (e.g., of the user's current location), time and / or date information (e.g., the current date and / or time), the user's physiological state (e.g., whether the user is awake or asleep), and subjective opinions based on user input (e.g., indicating that the user is interested in winding down). In some embodiments, the obtained context information may be associated with one or more other users, in particular one or more other users of the device 100 (e.g., other family members living in the same home environment). For example, the obtained context information may be related to the actions of other users.

[0040] As mentioned above, the context information is obtained from multiple sources. For example, in an embodiment, the multiple sources may include a user device, which may be a personal care device (e.g., a device related to dental care, hair care, skin care, etc.) or a personal health device (e.g., a weight scale, an activity tracker, a sleep monitor, etc.), and one or more sensors or actuators present in the user's home or worn by the user.

[0041] 2, in block 204, a user profile associated with the user is updated based on the context information obtained in block 202. The user profile may be obtained or retrieved by the processor 102 of the device 100 from one or more databases in memory 106 (which may be a memory of the device 100 or a memory external to the device 100).

[0042] The user profile has an interest level value for each of a plurality of healthcare topics, the interest level value indicating the user's interest and / or requirement for receiving support with the respective healthcare topic.

[0043] Each of the plurality of healthcare topics may be associated with a particular aspect of the user's healthcare, such as brushing teeth. Furthermore, in some embodiments, each of the plurality of healthcare topics may be organized in a hierarchical structure in the database. For example, the database may have a general healthcare topic of "personal care" at a first level (i.e., top level). At a second level, under "personal care", one or more specific healthcare topics, such as "oral healthcare" and "sleep", may be provided. Furthermore, under each of the healthcare topics at the second level, one or more healthcare subtopics may be provided at a third level. For example, under "oral healthcare", healthcare subtopics, such as "teeth brushing" and "teeth whitening" may be provided. This hierarchical structure is described in more detail with reference to FIG. 5. In this specification, either a healthcare topic or a healthcare subtopic may be generally referred to as a healthcare topic. Furthermore, as described in more detail in the following description, selection of a healthcare topic in some embodiments may correspond to the launching of a care module related to the particular selected healthcare topic.

[0044] As described in more detail below, in some embodiments, the concern level value may be expressed as one of "no problem", "worried", "concerned", "need help", etc., each of which is a different level of concern or interest in terms of severity, and which indicates at least one of the type of support to be provided to the user, the amount of support to be provided to the user, and the content of the support to be provided to the user, which support should be provided in the form of an intervention.

[0045] As described above with reference to block 202, the contextual information obtained in block 202 relates to at least one of a user's activity, a user's physiological state, and a user's environment. In some embodiments, the obtained contextual information may be associated with at least one of a plurality of healthcare topics, and in these embodiments, in block 204, a user profile may be updated with respect to the healthcare topic to which the obtained contextual information is associated.

[0046] For example, the context information obtained in block 202 may be associated with a tooth brushing session performed by a user. Thus, the obtained context information in this case may be associated with the healthcare topic "tooth brushing", and then, in this example, a user profile associated with the user may be updated with respect to the healthcare topic "tooth brushing", and in particular, the interest level value for the healthcare topic "tooth brushing" may be updated based on this obtained context information.

[0047] Additionally or alternatively, in some embodiments, the user profile may be updated with respect to a healthcare topic that has an implicit relationship with the healthcare topic to which the retrieved context information is associated. For example, the context information retrieved in block 202 may be associated with a tooth brushing session performed by the user. Thus, in this case, the retrieved context information may be associated with the healthcare topic "tooth brushing". Thereafter, in this example, the user profile associated with the user may be updated with respect to the healthcare topic "sleep" due to the implicit relationship between the healthcare topic "tooth brushing" and "sleep" (i.e., the user usually brushes his teeth before going to sleep). Specifically, the interest level value for the healthcare topic "sleep" may be updated based on this retrieved context information.

[0048] In some embodiments, the user profile may be updated based on user input. Specifically, in some embodiments, at least one of the interest level values ​​in the user profile may be updated based on user input. For example, the interest level value for a particular healthcare topic may be changed by a user via the user interface 104 of the device 100 from "no problem" to "concerned" by, for example, deselecting "no problem" and selecting "concerned" via a touch screen input.

[0049] 2, at block 206, one of the plurality of healthcare topics is selected based on the interest level value in the user profile. Specifically, the one of the plurality of healthcare topics may be selected by the processor 102 of the device 100.

[0050] In some embodiments, selecting one of the plurality of healthcare topics may include selecting a healthcare topic having an associated level-of-interest value greater than a first predetermined threshold.

[0051] Alternatively, in some embodiments, selecting one of the plurality of healthcare topics may include determining a change in an interest-level value for at least one of the plurality of healthcare topics after updating the user profile, and selecting the healthcare topic if the respective determined change in interest-level value is higher than a second predetermined threshold.

[0052] Alternatively, in some embodiments, selecting one of the multiple healthcare topics may include selecting a healthcare topic that is related to or similar to another topic having an associated interest level value above a predetermined threshold.

[0053] Returning to Figure 2, at block 208, intervention recommendations are provided to the user based on the healthcare topic selected at block 206. The intervention recommendations are associated with suggested information related to the selected healthcare topic or suggested adjustments for the user device.

[0054] The suggested information related to the selected healthcare topic may include at least one of educational and / or guidance information related to the selected healthcare topic and a reminder related to the selected healthcare topic. The suggested adjustments to the user device may include at least one of changing a setting on the user device, turning on the user device, and turning off the user device. In some examples, the intervention recommendation may be associated with multiple simultaneous or sequential interventions to be performed.

[0055] Additionally, in some embodiments, the proposed adjustments to the user device may be based on one or more selected product behavior scripts. In these embodiments, the computer-implemented method may further include selecting one or more product behavior scripts based on the obtained context information. For example, if it is determined from the obtained context information that the user is currently sleeping or that another user in the vicinity is sleeping, the processor 102 may be configured to select a product behavior script that includes instructions for adjusting settings of the electric toothbrush to disable acoustic feedback to avoid disturbing the sleeping user.

[0056] In some embodiments, the intervention recommendation may be based on the interest level value of the selected healthcare topic. For example, if the selected healthcare topic is "sleep" and the associated interest level value is "no problem", the intervention recommendation may be associated with providing general educational information, tips and tricks regarding falling asleep, if the associated interest level value is "worried", the intervention recommendation may be associated with guidance information to help the user identify the cause of difficulty falling asleep and provide a coaching program, if the associated interest level value is "need help", the intervention recommendation may be associated with providing the user with real-time coaching, performing breathing exercises and data collection.

[0057] In some embodiments, the user profile may further include a performance level value for each of the plurality of healthcare topics, the performance level value indicating the quality of the user's performance of the behavior associated with the respective healthcare topic. In these embodiments, providing the intervention recommendation may be further based on the performance level value of the selected healthcare topic in the user profile. Furthermore, the user's performance level value for the selected healthcare topic may be updated based on the result achieved by the user with respect to the performed intervention (e.g., whether the user brushes his teeth) and / or the user's behavior with respect to the performed intervention (e.g., how often the user brushes his teeth). For example, after providing the intervention recommendation in block 208 and receiving a user input indicating acceptance of the intervention recommendation, the user profile may be updated based on the result score indicating the degree of the result achieved by the user with respect to the performed intervention.

[0058] In some embodiments, the computer-implemented method may further include obtaining a persuasion style preference of the user for the selected healthcare topic in block 206. The persuasion style preference provides guidance regarding the content of an intervention recommendation for the user and / or how the intervention should be delivered to the user. In these embodiments, providing the intervention recommendation in block 208 may be further based on the obtained persuasion style preference of the user for the selected healthcare topic (e.g., “strong intervention”, “careful intervention”, “active intervention”, or other descriptive preferences such as “kind”, “friendly”, “sociable”, “drill sergeant”, etc.).

[0059] For example, if the obtained persuasion style preference is "careful intervention," audio coaching advice provided to the user (as part of the performed intervention) may be presented in a friendly tone of voice, and if the obtained persuasion style preference is "strong intervention," audio coaching advice provided to the user may be presented in a "drill sergeant" style of tone of voice. The tone of voice (or other manner of presentation of guidance or coaching advice) may additionally or alternatively be selected based on the interest level value of the selected healthcare topic.

[0060] Furthermore, for each individual user, the persuasion style preference may be different for each of the multiple healthcare topics. For example, for an individual user, the persuasion style preference for the healthcare topic "diet" may be "kind", while the persuasion style preference for the healthcare topic "sports" may be "drill sergeant". In some embodiments, the persuasion style preference for a healthcare topic may change based on the acquired context information (e.g., time or date information). For example, for an individual user, the persuasion style preference for the healthcare topic "diet" may be "kind" if it is determined from the acquired context information that it is currently Saturday morning (e.g., by the processor 102), and the persuasion style preference for the same healthcare topic (i.e., "diet") may be "sociable" if it is determined from the acquired context information that it is currently Saturday afternoon.

[0061] In some embodiments, obtaining the user's persuasive style preference for the selected healthcare topic may include obtaining the user's persuasive style preference for the selected healthcare topic from a retrieved user profile, where the user profile already includes the persuasive style preference for the selected healthcare topic.

[0062] Alternatively, obtaining the user's persuasive style preference for the selected healthcare topic may include deriving the user's persuasive style preference for the selected healthcare topic based on the user's persuasive style preferences for healthcare topics similar or related to the selected healthcare topic.

[0063] Alternatively, obtaining the user's persuasion style preference for the selected healthcare topic may include selecting a random persuasion style preference from a plurality of persuasion style preferences in a database.

[0064] As described above, in some embodiments, the proposed adjustments for the user device (of the provided intervention recommendation) may be based on one or more selected product behavior scripts. In some embodiments where the computer-implemented method further comprises obtaining a user's persuasion style preference for the selected healthcare topic, the selection of one or more product behavior scripts may be based on the obtained persuasion style preference. For example, if the selected healthcare topic is "teeth whitening", the processor 102 may be configured to select a product behavior script including instructions for adjusting the settings of the electric toothbrush as a "strong whitening mode" if the obtained persuasion style preference is "drill sergeant", and the processor 102 may be configured to select a product behavior script including instructions for adjusting the settings of the electric toothbrush as a "mild whitening mode" if the obtained persuasion style preference is "gentle".

[0065] In some embodiments, the computer-implemented method may further include receiving input from a user indicating acceptance or rejection of the provided intervention recommendation. The input may be received via a user interface 104 of the device 100. For example, the user may indicate acceptance or rejection by clicking a virtual button displayed on the user interface 104.

[0066] In these examples, if the received input indicates acceptance, the computer-implemented method may further include implementing the intervention recommendation, and if the received input indicates rejection, the computer-implemented method may further include providing the user with a different intervention recommendation based on the selected healthcare topic. In some examples, providing the user with a different intervention recommendation may be further based on the user's obtained persuasion style preference for the selected healthcare topic.

[0067] Moreover, if the received input indicates acceptance, the method may further include determining a change in effectiveness of the implemented intervention over a predetermined time period. In these embodiments, determining a change in effectiveness of the implemented intervention over a predetermined time period may include determining a first effectiveness score at the beginning of the predetermined time period, determining a second effectiveness score at the end of the predetermined time period, and determining a change in effectiveness of the implemented intervention over the predetermined time period by calculating a difference between the first and second effectiveness scores over the predetermined time period.

[0068] In these embodiments, determining an efficacy score (e.g., the first efficacy score or the second efficacy score) may include determining an outcome score indicative of a degree of result achieved by the user with respect to the implemented intervention, determining an adherence score indicative of a degree of adherence of the user to the implemented intervention, and determining a score by dividing the outcome score by the adherence score.

[0069] The efficacy score quantifies the objective progress made by the user relative to the amount of effort the user put into adhering to the implemented intervention. More specifically, the efficacy score may be determined according to the following formula: E=R / A (1) where E represents the efficacy score, R represents the outcome score, and A represents the adherence score.

[0070] In some embodiments, the outcome score R may be a value between 0 and 1 to proportionally represent what has been achieved by the user relative to a maximum achievable goal associated with the intervention recommendation provided and / or the intervention implemented. For example, if the intervention recommendation provided in block 208 is associated with a maximum achievable goal of the user losing 10 kg in six months, and the user only loses 8 kg within six months, the processor 102 of the device 100 may assign a value of 0.8 as the outcome score R.

[0071] In some embodiments, the adherence score A may be a value between 0 and 1 to proportionally represent the degree of adherence of the user to the implemented intervention. For example, if the proposed information associated with the intervention recommendation provided in block 208 (and thus the implemented intervention) includes advice to the user regarding a selected healthcare topic, and it is determined (e.g., by the processor or based on user input) that the user acted on only 50% of the advice, then the device processor 102 may assign a value of 0.5 as the adherence score A.

[0072] The efficacy score E is a measure of the outcome relative to adherence. The lower the adherence, the lower the outcome is expected. Also, if the user is ineffective, the outcome may not be sufficient for the user's adherence. In some cases, a user may be very effective with low adherence but high outcome.

[0073] Based on the above discussion, in some embodiments, the change in effectiveness of an implemented intervention over a given period of time may be determined according to the following formula: G e (t n )=E(t n ) / T S -E(t n-1 ) / T S (2) where Ge(tn) represents the change in efficacy over a period of time, E(tn-1) represents the first efficacy score, E(tn) represents the second efficacy score, and TS represents the period of time. The first efficacy score E(tn-1) and the second efficacy score E(tn) may be determined using Equation (1) above. In some embodiments, the change in efficacy over a period of time may be referred to as "efficacy growth."

[0074] In some embodiments, the change in effectiveness (i.e., "effectiveness growth") over a predetermined time period may be determined based on the first and second efficacy scores, which are determined based on the outcome score and the adherence score, while in alternative embodiments, the change in effectiveness over a predetermined time period may be determined, at least in part, based on user input. In these alternative embodiments, the computer-implemented method may further include receiving user input indicating a degree of perceived change in the effectiveness of the implemented intervention over a predetermined time period and determining the change in effectiveness over a predetermined time period based on the received user input.

[0075] In some embodiments, the computer-implemented method may further include acquiring sensor data associated with at least one of an activity performed by a user and a user device. In these embodiments, at least one of an outcome score and an adherence score may be determined based on the acquired sensor data. In particular, at least one of an outcome score and an adherence score may be determined based on comparing the acquired sensor data to a predefined sensor data pattern. Further details regarding the sensor data patterns are described below with reference to FIG. 6.

[0076] In some embodiments, the determined adherence score may be used to update an interest level value of the selected healthcare topic. For example, if the determined adherence score is low, the processor 102 may be configured to update the interest level value (e.g., change the value from "no problem" to "need help"). Similarly, in some embodiments, the determined outcome score may be used to update an interest level value of the selected healthcare topic. For example, if the determined outcome score is high, the processor 102 may be configured to update the interest level value (e.g., change the value from "having a problem" to "no problem").

[0077] In some embodiments, the computer-implemented method further comprises determining a difference between the determined change in effectiveness of the performed intervention (within a predefined time period) and an expected effectiveness index, the expected effectiveness index representing an expected change in effectiveness of the performed intervention. After determining the difference, the computer-implemented method may further comprise adjusting parameters and / or content of the performed intervention if the determined difference is greater than a predefined threshold.

[0078] In some embodiments, the difference between the determined change in efficacy of the performed intervention and the expected efficacy index may be compared to a predefined threshold according to the following formula: |G e (t n )-C e (t n )|>M e (3) Here, G e represents the change in efficacy, and C e (t n ) represents the expected efficacy index, and M e represents the margin of error, which represents the maximum allowable deviation from the expected change in efficacy without requiring further action, such as providing a different intervention recommendation to the user.

[0079] Therefore, in these embodiments, as long as the change in efficacy does not substantially differ from the expected efficacy indicator (i.e., remains within the margin of error), no action is required with respect to intervention recommendations and intervention instruction execution. However, if the difference between the change in efficacy and the expected efficacy indicator is greater than the margin of error (i.e., a predefined threshold), the parameters and / or content of the intervention to be performed may be adjusted (e.g., further modifying the settings of the user device or adapting the content of the coaching / guidance information provided to the user) or a different intervention recommendation may be provided. The provision of a different intervention recommendation may be based on the user's persuasion style preference for the selected healthcare topic.

[0080] In some embodiments, if the input received from the user indicates acceptance of the intervention recommendation provided in block 208, the computer-implemented method may further include determining an overall satisfaction score indicative of the user's satisfaction with the intervention performed.

[0081] In these embodiments, determining the overall satisfaction score may include determining a progress satisfaction score indicative of the user's satisfaction with the progress resulting from the performed intervention, determining a progress-oriented score indicative of the user's willingness to tolerate more aggressive persuasive measures versus gentler persuasive measures, determining a persuasion satisfaction score indicative of the user's satisfaction with the persuasive style associated with the performed intervention, and determining an overall satisfaction score based on the progress satisfaction score, the progress-oriented score, and the persuasion satisfaction score.

[0082] More specifically, in some embodiments, the overall satisfaction score may be determined according to the following formula: S t =(S prog V prog +S pers (1-V prog )) / twenty four) Here, S t represents the overall satisfaction score, and S prog represents the progress satisfaction score, and V prog represents the progress orientation score, and S pers represents the persuasion satisfaction score.

[0083] In some embodiments, a progress satisfaction score S prog may be a value between 0 and 1 to proportionally represent the user's satisfaction with the progress resulting from the performed intervention. The value assigned as the progress satisfaction score may be based on an interaction between the user and the device 100. For example, the processor 102 may be configured to present a query to the user via the user interface 104 regarding their satisfaction with the progress of the result, and the value assigned as the progress satisfaction score may be based on an input by the user in response to this query.

[0084] In some embodiments, the progress-oriented score V prog may be a value between 0 and 1, representing the degree to which a user is willing to accept more drastic persuasive measures versus gentler persuasive measures. Some users are willing to accept more drastic persuasive measures to achieve progress, and for these users the progress-oriented score will be relatively high. In some embodiments, the progress-oriented score may be stored in a context user data storage unit.

[0085] In some embodiments, the persuasion satisfaction score S pers may be a value between 0 and 1 to represent the user's satisfaction with the persuasive style associated with the performed intervention. For example, if the persuasive style associated with the intervention recommendation provided in block 208 is deemed too aggressive by the user (which may be indicated in the user feedback via the user interface 104), this may be reflected by a lower value of the persuasive satisfaction score. As another example, if the performed intervention is associated with suggested adjustments for the user device and the user subsequently changes or modifies the settings of the user device from the suggested adjustments, this may be reflected by a decrease in the value of the persuasive satisfaction score.

[0086] The computer-implemented method may further comprise determining a difference between an overall satisfaction score and an expected satisfaction score, the expected satisfaction score being indicative of the user's expected satisfaction with the performed intervention and may be determined based on the progress-oriented score. Furthermore, after determining the difference between the overall satisfaction score and the expected satisfaction score, the computer-implemented method may further comprise adjusting parameters and / or content of the performed intervention if the determined difference is greater than a predefined threshold.

[0087] In some embodiments, intelligent agent technology may be used to learn and / or infer at least one of persuasion style preferences, parameters, and / or content associated with an intervention recommendation that contribute to an increase in the user's satisfaction with the implemented intervention (and thus an increase in the overall satisfaction score). In these embodiments, the intelligent agent technology may include a model-based agent or a goal-based agent having a reward function to determine an effect of at least one of persuasion style preferences, parameters, and / or content associated with an intervention recommendation on the overall satisfaction score. Further, providing an intervention recommendation in block 208 may be based on this determined effect.

[0088] It will be appreciated that in alternative embodiments, the overall satisfaction score may be determined based on different factors, for example, the overall satisfaction score may be determined based on user input indicating a general level of satisfaction with the intervention recommendations provided. In some embodiments, if the input received from the user indicates a rejection of the intervention recommendation provided in block 208, the computer-implemented method may further include updating a user profile regarding persuasion style preferences.

[0089] In some embodiments, the computer-implemented method may further include obtaining a plurality of healthcare-related messages, determining a priority value for each of the plurality of healthcare-related messages based on at least one of the obtained contextual information and the selected healthcare topic, and providing at least one of the plurality of healthcare-related messages to a user based on the priority value.

[0090] In these embodiments, at least one of the plurality of healthcare-related messages may be retrieved from the device's memory 106. Also, in these embodiments, at least one of the plurality of healthcare-related messages may be associated with the healthcare topic selected in block 206.

[0091] In some embodiments, providing at least one of the plurality of healthcare-related messages may be further based on a load indicator value. The load indicator value may indicate a possibility of overloading the user with excessive information. In these embodiments, the computer-implemented method may further include updating the load indicator value after providing the healthcare-related message (i.e., at least one of the plurality of healthcare-related messages) to the user. Specifically, the processor 102 of the device 100 may be configured to update the load indicator value.

[0092] In some embodiments, the burden indicator value may be used by a dialogue processing unit in the system to determine whether healthcare related messages should be presented to the user. A more detailed description of the dialogue processing unit is provided below with reference to Figures 12 and 13.

[0093] In some embodiments, updating the burden indicator value includes increasing the burden indicator value each time a healthcare related message is presented to the user (e.g., +1 unit for any message, or other value depending on the healthcare topic to which the message is associated and / or the length of the message) and / or decreasing the burden indicator value according to a predetermined rate or scale (e.g., -1 unit / minute).

[0094] In some embodiments, updating the burden indicator value may include updating the burden indicator value based on the obtained context information. For example, if the obtained context information (e.g., based on data received from the user's electric toothbrush) determines that the user is brushing his / her teeth, the burden indicator value may be decreased to indicate that the user is less likely to be overloaded with excessive information when presented with multiple healthcare-related messages. Similarly, if the obtained context information determines that the user is wearing makeup, for example, the burden indicator value may be increased to indicate that the user is more likely to be overloaded with excessive information. Furthermore, depending on the current date or current time (which may be determined from the obtained context information), the burden indicator value may be increased or decreased to a different extent, e.g., by −5 units if the toothbrushing session is in the morning, −10 units if the toothbrushing session is in the evening, etc.

[0095] In some embodiments, a respective burden indicator value may be provided for each of the plurality of healthcare topics. In these embodiments, providing the at least one of the plurality of healthcare related messages may be based on a respective burden indicator value with which the at least one of the plurality of healthcare related messages is associated.

[0096] FIG. 3 is a diagram illustrating a system 300 for providing context-based healthcare interventions to a user according to one embodiment, and FIG. 4 is a diagram illustrating a Caring Home module of the system of FIG.

[0097] As shown in FIG. 3, the device 300 includes a caring home module 310, an interactive user interface 320, one or more personal care devices 330, one or more health devices 340, one or more sensors 350, and one or more actuators 360.

[0098] In this embodiment, the caring home module 310 may be implemented in the system 300 in a home environment and / or in a computing cloud. The caring home module 310 is configured to interact with one or more users via a conversational user interface 320, which may be embodied in the form of a chatbot or a voice-based virtual assistant on a mobile device, in a smart speaker device (e.g., via a graphical user interface), as described in relation to Fig. 1. Furthermore, in some embodiments, the caring home module 310 may be embodied in a device, such as the device 100 as described with reference to Fig. 1.

[0099] The caring home module 310 is also configured to interact with one or more users via one or more personal care devices 330 (e.g., devices related to dental care, hair care, skin care, etc.), and / or one or more personal health devices 340 (e.g., weight scales, activity trackers, sleep monitors, etc.), and / or one or more sensors 350, and / or one or more actuators 360. In some embodiments, the one or more sensors 350 and the one or more actuators 360 may be present in the home environment of the one or more users or may be worn by the one or more users.

[0100] 4, a caring home module 410, which corresponds to the caring home module 310 shown in FIG 3, includes a context modeling unit 411, a dialogue scheduling unit 412, a healthcare topic database 413, and a user profile management unit 414. In some examples, the context modeler unit 411 may be referred to as a "context modeler", the dialogue scheduling unit 412 may be referred to as a "dialogue scheduler", the medical topic database 413 may be referred to as a "library of care modules", and the user profile management unit 414 may be referred to as a "user profile manager".

[0101] The context modeling unit 411 is configured to determine and / or monitor a context associated with each of one or more users. This may be performed with respect to a plurality of healthcare topics included in the healthcare topic database 413. Examples of contexts may include brushing teeth, sleeping, waking up, etc. In some embodiments, the operations of the context modeling unit 411 may be implemented in a processor of the device (e.g., processor 102 of device 100 of FIG. 1 ).

[0102] The dialogue scheduling unit 412 is configured to receive conversation content (e.g., one or more healthcare-related messages) from the healthcare topic database 413 and prioritize and balance the received conversation content before providing the conversation content to one or more users. The dialogue scheduling unit 412 may be configured to perform such prioritization and balancing based on the context determined by the context modeling unit 411 and the interests and / or requirements of one or more users. The interests and / or requirements of a user may be indicated in a user profile associated with the user. A more detailed description of the dialogue scheduling unit 412 is provided below with respect to Figures 12 to 14.

[0103] The healthcare topic database 413 includes a plurality of healthcare topics organized in a hierarchical structure. Examples of healthcare topics may include brushing teeth, whitening teeth, improving sleep, oral healthcare, etc. In some embodiments, a selection of one of the plurality of healthcare topics may correspond to the launching of a care module associated with the selected healthcare topic. When the care module associated with the selected healthcare topic is activated, coaching support for the respective healthcare topic may be provided. This will be described in more detail below with reference to FIG. 5.

[0104] The user profile management unit 414 is configured to manage and maintain one or more user profiles associated with one or more users, respectively. The user profile management unit 414 can be considered as providing a function of maintaining a model of interests and preferences of one or more users regarding a plurality of healthcare topics included in the healthcare topic database 413. In some embodiments, the user profile management unit 414 may further comprise a user profile control unit and a context user storage unit. This will be described in more detail with reference to FIG. 7.

[0105] It will be understood that Figures 3 and 4 show only the components necessary to illustrate one aspect of the system 300 and the caring home module 410, and that in an actual implementation, the system 300 and / or the caring home module 410 may have components in place of or in addition to the components shown.

[0106] 5 is a schematic diagram of a hierarchical structure of a plurality of healthcare topics in a database, such as the healthcare topic database 413 of FIG. 4. As described above, each of the healthcare topics can be selected, and the selection of a healthcare topic may correspond to the launch of a care module related to the selected healthcare topic. Thus, in some cases, for example in FIG. 5, the plurality of healthcare topics may be referred to as a "library of care modules." It will be understood that the plurality of healthcare topics shown in FIG. 5 are provided by way of example, and that in alternative embodiments, more or fewer healthcare topics may be shown.

[0107] In this embodiment, each of the multiple healthcare topics may be associated with a particular aspect of the user's healthcare. As shown in Figure 5, the database includes, at a first level (i.e., top level), a general healthcare topic, "personal care" 510, and at a second level, under "personal care," one or more additional, more specific healthcare topics may be provided, such as "oral healthcare" 522 and "sleep" 524, both of which fall under the more general topic of "personal care" 510. Additional healthcare topics, represented by a box labeled 526, may also be provided at the second level.

[0108] Additionally, under each of the second level health care topics 522, 524, 526, one or more health care subtopics may be provided at a third level. This is illustrated in FIG. 5, where health care subtopics such as "teeth brushing" 522A and "teeth whitening" 522B may be provided under "oral health care." Additional health care subtopics may also be provided under "oral health care" 522, as represented by the box labeled 522C. Similarly, although not shown in the drawing, health care subtopics such as "effective diet" and "behavioral routines" may be provided at a third level under the health care topic "sleep" 524.

[0109] Therefore, in an embodiment of multiple healthcare topics organized in a hierarchical structure as shown in FIG. 5, the selection of a healthcare topic may be further optimized by enabling the selection of a healthcare topic related to the current context. For example, during a user's teeth brushing session, it may be determined that the current context is associated with "tooth brushing" 522A. Based on this, the healthcare topic "tooth brushing" 522A may be selected, or a related healthcare topic (e.g., "teeth whitening" 522B) may be selected. Additionally or alternatively, the healthcare topic "oral healthcare" 522 may be selected, as it may also be considered a related healthcare topic at a directly higher level in the hierarchical structure. It should be understood that for purposes of this description, any healthcare topic or healthcare subtopic may be generally referred to as a healthcare topic.

[0110] FIG. 6 is a schematic diagram of a care module 600 related to healthcare topics, which may be contextually activated by selecting a corresponding healthcare topic.

[0111] As shown in FIG. 6, the care module 600 includes a care engine 610, a dialogue content storage unit 620, a product behavior script storage unit 630, a training content storage unit 640, and a sensor data pattern storage unit 650.

[0112] The dialogue content storage unit 620 is configured to store dialogue content related to each healthcare topic, which is used to generate a conversation to enable collection of a user's subjective opinion and interest level on each healthcare topic and to provide coaching to the user on each healthcare topic.

[0113] The product behavior script storage unit 630 is configured to store product behavior scripts associated with each healthcare topic that enable product behavior generation for intervention instruction execution. In some embodiments, the product behavior scripts may include instructions for adjusting settings of a user device. For example, the product behavior scripts may instruct certain settings of the user device, such as automatically setting a toothbrush to a teeth whitening mode and / or setting a warning sound effect if a user stops a tooth brushing session too soon.

[0114] The educational content storage unit 640 is configured to store educational content associated with each healthcare topic, and the sensor data pattern storage unit 650 is configured to store one or more sensor data patterns associated with each topic. The sensor data patterns may enable meaningful data collection from sensors related to care modules related to the healthcare topic. The sensor data patterns may be predefined based on domain knowledge. For example, domain knowledge for the healthcare topic "tooth brushing" may include a desired end result of "no cavities and no pain" and a corresponding desired user behavior of "cover all teeth and brush teeth for 2 minutes at least twice a day". Also, the domain knowledge may be derived at least in part by deep learning (or other artificial intelligence) techniques using the acquired sensor data.

[0115] In some embodiments, the sensor data pattern may comprise at least one of a context data pattern and a behavioral data pattern. By comparing the acquired sensor data with the context data pattern, an action performed by the user (e.g., whether the user has started or finished a tooth brushing session) or a physiological state of the user (e.g., whether the user has just fallen asleep or just woken up) may be determined. An output of the determination may be sent to the context modeling unit. Also, by comparing the acquired sensor data with the behavioral data pattern, a quality and / or duration of the activity performed by the user (e.g., whether the session is short, medium, long, suitable, or unsuitable) may be determined.

[0116] The care engine 610 is configured to manage the behavior of the care module 600. In particular, the care engine 610 is configured to activate and / or deactivate the care module 600, evaluate the interest level (which in some embodiments may be expressed as an interest level value), and determine the content and product behavior scripts presented to the user. The operation of the care engine 610 may be based on user input received via a user interface and / or on received sensor data. The user's progress may be monitored and intervention recommendations may be adjusted based on performance and experience metrics defined in the care module 600. The operation of the care engine 610 may be configured in terms of multiple phases: - User interest: A desired level of care (interest level value) may be determined (e.g. "no problem", "worried" or "need help"). The care engine 610 may be configured to determine what interventions to recommend to the user and what outcomes may be expected by the user. The desired level of care (interest level value) may be stored in the user profile management unit. - Information: Educational content related to healthcare topics may be provided to users based on their desired level of care (level of interest value). - Suggestions: Behavioral changes may be suggested in the form of intervention recommendations. This may be based on the desired level of care (interest level value), persuasion style preferences, and / or the user's current physiological state. - Execute: A recommended intervention may be executed (e.g. product behavior, coaching dialogue). - Progress assessment: Based on the health goals set and how the coaching is evaluated by the user, it is decided whether different intervention strategies should be offered. - Final result evaluation: An objective final result and the user's satisfaction with the final result are determined; the final result may be evaluated based on acquired sensor data or by querying the user.

[0117] It will be understood that Fig. 6 shows only the components necessary to illustrate one embodiment of the care module 600, and that in an actual implementation, the care module 600 may have components instead of or in addition to the components shown, for example, in an alternative embodiment, the care module 600 may not include a dialogue content storage unit or a teaching content storage unit.

[0118] Figure 7 is a schematic diagram of a user profile management section of the apparatus of Figure 4. As previously described with reference to Figure 4, in some embodiments the user profile management unit may further comprise a user profile control unit and a context user data storage unit. These components are shown in Figure 7 as a user profile control unit 710 (which may be referred to as a "user profile controller") and a context user data storage unit 720 (denoted as "context user data" in Figure 7).

[0119] In this embodiment, the user profile control unit 710 is configured to manage a user interest model that may be updated over time for each of one or more users, where the user interest model may be represented by a number of interest level values ​​for each of a number of healthcare topics in a database (e.g., the healthcare topics database 413 of FIG. 4). The contextual user data storage unit 720 is configured to store contextual user data.

[0120] In some embodiments, the contextual user data stored in the contextual user data storage unit 720 may be structured in the same manner as the multiple healthcare topics structure as shown in FIG. 5. This structure groups together healthcare topics with similar purposes. This approach also allows a user's interest in a broader topic to trigger the selection of multiple healthcare topics (and thus the activation of multiple care modules), making the database / library more easily searchable and expandable. Under each of the multiple healthcare topics, one or more of the following parameters may be stored: - a level of concern value (which in some embodiments may be called "level of concern" or "level of care") -Performance level value -Activation Level Value -Preference for persuasion style - Load indicator value

[0121] In some embodiments, the user profile control unit 710 may be configured to receive updated values ​​for at least one of the interest level value, the performance level value, the activation level value, and the persuasion style preference for the user, and control the context user data storage unit 720 to store the received updated values. Furthermore, the user profile control unit 710 may be configured to determine or estimate at least one value of the interest level, the performance level, the activation level, and the persuasion style preference for each healthcare topic based on the values ​​of the corresponding parameters of related or similar healthcare topics. The determination or estimation may be based on an inference algorithm. In particular, in some embodiments, the user profile control unit 710 may be configured to determine or estimate the performance level and / or the persuasion style preference for each healthcare topic based on the values ​​of the corresponding parameters of at least one related healthcare topic that is lower in the hierarchical structure.

[0122] For example, if it is determined that the level of interest in the healthcare topic "tooth brushing" has a value of 3 (on a scale of 1 to 10) and the level of interest in the healthcare topic "tooth whitening" has a value of 1 (on a scale of 1 to 10), the user profile control unit 710 may be configured to determine or estimate the level of interest value for the healthcare topic "oral healthcare" (one level higher in the hierarchical structure than "tooth brushing" and "tooth whitening") as having a value of 2 (i.e. the average value of the level of interest values ​​of the two related healthcare topics). Other parameters such as persuasion style preferences may be estimated or determined in a similar manner.

[0123] The interest level values ​​of the healthcare topics may be used by a dialogue scheduling unit (e.g., a dialogue scheduling unit as described with reference to Figures 12 and 13) to determine priority values ​​of healthcare related messages to be presented to the user.

[0124] FIG. 8 is a schematic diagram illustrating contextual user data in a user profile relating to multiple healthcare topics in a hierarchical structure, and FIG. 9 illustrates exemplary scales for each of multiple parameters in a user profile relating to healthcare topics.

[0125] The hierarchical structure of healthcare topics from the example of Fig. 5 is used in Fig. 8 to illustrate information or data that may be associated with each of the multiple healthcare topics. Similar to Fig. 5, the hierarchical structure of this embodiment also has "personal care" 810, "oral health" 822, "sleep" 824, "teeth brushing" 822A, and "teeth whitening" 822B. Other healthcare topics under "personal care" are represented as 826, and other healthcare topics under "oral health" 822 are represented as 822C. In this embodiment, the context user data (which may be stored in the context user data storage unit) includes an interest level value (which may be referred to as "care level"), a performance level value, an activation level value, and a persuasion style preference for each of the multiple healthcare topics.

[0126] For example, referring to the healthcare topic "brushing teeth" 822A, the interest level value is 3, the performance level value is 6.9, the activation level value is "strong", and the persuasion style preference is 2.

[0127] To illustrate the importance of each of the parameters (e.g., performance level values) related to healthcare topics, a number of exemplary scales are provided in Fig. 9. The number of exemplary scales include an interest level value scale (which may be referred to as an "interest level" scale as shown in the figure) 910, a performance level value scale 920, an activation level value scale 930, and a persuasion style preference scale 940.

[0128] In this embodiment, the interest level values ​​are provided on a scale of 1 to 3 according to the user's interest and / or requirements for receiving support for the respective healthcare topic, with a value of 1 corresponding to "no problem", a value of 2 corresponding to "concerned", and a value of 3 corresponding to "needs help", thus an interest level value of 3 for "brushing teeth" 822A corresponds to "needs help".

[0129] In this embodiment, the performance level values ​​are provided on a scale of 1 to 10 according to the quality of the user's performance of the behavior associated with each healthcare topic. As shown in Figure 9, the performance level value of 6.9 for "brushing teeth" 822A can be interpreted as slightly less serious than "doing OK", with the performance level value of 9 corresponding to "doing better", the performance level values ​​of 5 and 6 corresponding to "doing OK", and the performance level value of 2 corresponding to "doing not OK".

[0130] In this embodiment, the activation level value is either 1 or 0, as shown in the activation level value scale 930, with a value of 1 corresponding to "active" and a value of 0 corresponding to "inactive, checking performance". Although not shown in Fig. 9, in some embodiments, the activation level may take other values, for example, the activation level value for "teeth brushing 822A" is "strong", which is a higher level than "active", indicating that a stronger intervention should be provided to the user for this particular healthcare topic. In some other embodiments, a possible activation level value may be "under investigation", indicating that the care module associated with the respective healthcare topic is in the process of determining whether it should be activated.

[0131] In this embodiment, a persuasion style preference is provided on a scale of 1 to 3 depending on the degree of strength of the intervention recommendation provided to the user, with a value of 1 corresponding to "cautious intervention", a value of 2 corresponding to "aggressive intervention", and a value of 3 corresponding to "strong intervention".

[0132] It will be appreciated that in alternative embodiments, each of the interest level value scale 910, performance level value scale 920, activation level value scale 930, and persuasion style preference scale 940 may be, for example, a scale of 1 to 20 rather than a scale of 1 to 10. Furthermore, while each of the interest level value scale, performance level value scale, activation level value scale, and persuasion style preference scale are described above as having a plurality of discrete values, it will be appreciated that in alternative embodiments, one or more of these scales may be a continuous scale instead of a discrete scale to allow for more granular monitoring and action.

[0133] Figure 10 is a flow diagram illustrating a method associated with an activated care module, according to one embodiment. In some embodiments, the method illustrated in Figure 10 may be performed by a processor 102 of the device 100, as described with reference to Figure 1. In some embodiments, the method may be performed by a care engine of an activated care module, such as the care engine 610 illustrated in Figure 6.

[0134] In this embodiment, the method begins in step 1001, where a care module that can be triggered by the selection of a particular healthcare topic is launched. For example, if the healthcare topic "teeth brushing" is selected, a care module associated with the healthcare topic "teeth brushing" can be activated. Once the care module is activated, the method proceeds to step 1002, where it is determined whether the interest level value (i.e., "interest (level of care)") of the selected healthcare topic is one of "needs help" or "worried".

[0135] If in step 1002 it is determined that the interest level value is one of "needs help" or "concerned", the method proceeds to step 1003 where tailored information (e.g., educational content) is provided to the user based on the interest level value. In some embodiments, the tailored information may be provided via an interactive user interface, and in some embodiments, the tailored information may be retrieved from a dialogue content storage unit.

[0136] On the other hand, if in step 1002 it is determined that the interest level is not "needs help" or "worried", the method proceeds to step 1004 where the care module is deactivated. Alternatively, if in step 1002 it is determined that the interest level value is unknown, the method proceeds to step 1005 where an interest level value for the healthcare topic associated with the activated care module is determined (e.g., by querying a user via a user interface) before returning to step 1002.

[0137] After determining the concern level value at 1005, the method returns to step 1002 where it is determined whether the concern level value is "needs help" or "worried."

[0138] Following providing the tailored information in step 1003, the method proceeds to step 1006, where it is determined whether there is an available persuasion style preference related to the healthcare topic corresponding to the activated care module.

[0139] If at step 1006 it is determined that there is an available persuasion style preference associated with the healthcare topic, the method proceeds directly to step 1009. Otherwise, if at step 1006 it is determined that there is no available persuasion style preference associated with the healthcare topic, the method proceeds to step 1007, where a persuasion style preference is either derived based on at least one of big data, a user profile of the user, and persuasion style preferences of similar healthcare topics, or is randomly selected, before proceeding to step 1008.

[0140] In step 1008, an intervention recommendation is provided based on the available persuasion style preferences. The intervention recommendation may comprise dialogue content (e.g., reminders, coaching, and guidance) retrieved from a dialogue content storage unit, and / or a product behavior script (e.g., specific settings of the user device, such as automatically setting the toothbrush to a teeth whitening mode and / or setting an alert sound effect if the user stops a tooth brushing session too early) retrieved from a product behavior script storage unit. As mentioned above, the product behavior script may include instructions for adjusting settings of the user device.

[0141] If the intervention recommendation is accepted in step 1009, the method proceeds to step 1010. Otherwise, the method proceeds to step 1011, where the next best persuasion style is selected, followed by step 1008, where a (different) intervention recommendation is provided based on the updated persuasion style preference.

[0142] In step 1010, an intervention is performed based on the intervention recommendation, and progress resulting from the performed intervention is tracked. In some embodiments, tracking of progress may be based on determining changes in effectiveness, as described with reference to FIG. 2. If in step 1012 the outcome progress is determined to be satisfactory, the method proceeds to step 1013, where the persuasive style preference for the healthcare topic associated with the activated care module is updated (e.g., by replacing the currently selected persuasive style preference in the contextual user data stored in the contextual user data storage unit). If not, the method returns to step 1011, where a (different) next best persuasive style preference is selected, followed by step 1008, where another (different) intervention is provided based on the updated persuasive style preference.

[0143] After updating the persuasion style preference in step 1013, the method proceeds to step 1014, where it is determined whether the desired end result has been achieved. In some embodiments, the determination of whether the desired end result has been achieved may be based on determining an outcome score, as described with reference to FIG. 2. If it is determined in step 1014 that the desired end result has been achieved, the method proceeds to step 1005, where a (new) interest level value for the healthcare topic associated with the activated care module is determined. If it is determined in step 1014 that the desired end result has not been achieved, the method proceeds to step 1011, where a (further) intervention is implemented and progress resulting from the (further) intervention is tracked.

[0144] It will be appreciated that in alternative embodiments, the method may not include all of the described steps and may omit one or more steps shown in Figure 10. It will also be appreciated that in alternative embodiments, the steps shown in Figure 10 may be performed in a different order.

[0145] Figure 11 is a flow diagram illustrating a method associated with a deactivated care module, according to one embodiment. In some embodiments, the method illustrated in Figure 11 may be performed by the processor 102 of the device 100, as described with reference to Figure 1. In some embodiments, the method may be performed by the caring home module 310, as shown in Figure 3.

[0146] As described in more detail below, while deactivated, the care module continues to track the user's performance (outcomes and / or behavior) with respect to the healthcare topic associated with the care module. Unless the user has not previously indicated that they are doing well with the healthcare topic, a careful investigation of the user's interest level (represented by an interest level value) with respect to the healthcare topic may begin.

[0147] In this embodiment, the method starts at step 1101, in which a care module that may be triggered due to a low associated interest level value is deactivated. The method proceeds to step 1102, in which it is determined whether there is an inappropriate performance of an action related to the deactivated care module, or whether the interest level value (level of care) is high (e.g., exceeds a certain severity threshold). This determination can be based on context information acquired by the processor 102 related to the user's action, which is associated with the healthcare topic corresponding to the deactivated care module. If it is determined in step 1102 that there is no inappropriate performance and the interest level value is not high, the method returns to step 1101, in which the care module remains deactivated, and if it is determined in step 1102 that there is an inappropriate performance or the interest level value is high, the method proceeds to step 1103, in which it is determined whether the user has indicated "no problem" as the interest level value associated with the healthcare topic corresponding to the deactivated care module.

[0148] If, in step 1103, it is determined that the user indicated "no problem" as the interest level value, the method returns to step 1104, where it is determined whether the last user query ("under investigation") regarding the level of care (interest level value) was a long time ago. This may be determined by comparing the time period between the last user query and the current time to a predefined threshold. If, in step 1104, it is determined that the last user query regarding the level of care was not a long time ago, the method returns to step 1101 where the care module remains deactivated. However, if it is determined that the last user query regarding the level of care was a long time ago, the method proceeds to step 1105.

[0149] Further, if in step 1103 it is determined that the user has not indicated “no problem” as the interest level value, the method also proceeds to step 1105, where an interest level value for a healthcare topic associated with the deactivated care module is obtained by querying the user (e.g., via a user interface).

[0150] After obtaining the interest level value in step 1005, it is determined in step 1106 whether the interest level value is not "no problem". If the determination is negative, i.e. the interest level value is "no problem", the method proceeds to step 1101, where the care module remains deactivated. If the determination is positive, i.e. the interest level value is not "no problem", the method proceeds to step 1107, where the context user data is updated with respect to the newly obtained interest level value, and then proceeds to step 1108, where the care module is activated.

[0151] It will be appreciated that in alternative embodiments, the method may not have all of the steps described and may omit one or more of the steps shown in Figure 11. It will also be appreciated that in alternative embodiments, the steps shown in Figure 11 may be performed in a different order.

[0152] 12 and 13 are schematic diagrams of the interaction scheduling unit 1200 of the device of FIG 4. Specifically, FIG 13 illustrates an exemplary implementation of the interaction scheduling unit 1200 of FIG 12.

[0153] 12 and 13, the dialogue scheduling unit 1200 (which may be referred to as a "dialogue scheduler" as described above) comprises a dialogue processing unit 1210 (which may be referred to as a "dialogue handler" as shown) and a message queue 1220. The dialogue scheduling unit 1200 may be configured to receive context information from dialogue content from a context modeling unit, a user profile management unit, and one or more care modules.

[0154] In this embodiment, the message queue 1220 includes a plurality of healthcare-related messages 1221, 1222, 1223, 1224, 1225, and 1226. Each of the plurality of healthcare-related messages may include interaction content (e.g., related to a particular care module) to be presented to the user. In particular, each of the plurality of healthcare-related messages may include information or questions (e.g., user interests, information, suggestions, instruction execution, progress evaluation, and final result evaluation) related to one of a plurality of phases, as described with reference to the operation of the care engine 610 in FIG. 6. For example, the healthcare-related message may be a question regarding the degree of teeth whitening desired by the user. The user may be able to provide a response via the user interface (e.g., by selecting one of two images showing different degrees of teeth whitening).

[0155] The interaction processing unit 1210 is configured to manage the selection and timing of healthcare related messages in the message queue 1220 and to track the following aspects: - the current behavior of the user (obtained from the context modeling unit); - The current topic of the conversation, which becomes unset after a period of no conversation - A load indicator value that indicates the amount of recent interaction with the user and thus the possibility of overloading the user with too much information.

[0156] The load indicator value in this embodiment may operate in a similar manner as described with reference to Fig. 2. For example, the load indicator value may be updated based on the obtained context information. Therefore, for the sake of brevity, the description of the load indicator value is omitted.

[0157] In some embodiments, the interaction processing unit 1210 may be configured to determine a priority value (which may be referred to as a "relevance score") for each of the plurality of healthcare-related messages based on at least one of the acquired context information and the selected healthcare topic. The interaction processing unit 1210 may also be configured to provide at least one of the plurality of healthcare-related messages in the message queue 1220 to a user based on the priority value. At least one of the plurality of healthcare-related messages may be provided via a user interface (e.g., a conversational user interface). In some embodiments, providing at least one of the plurality of healthcare-related messages may be further based on a timing of each healthcare-related message in the message queue 1220.

[0158] FIG. 14 is a finite state diagram illustrating the behavior of the dialogue processing units of the dialogue scheduling portion of FIGS.

[0159] In this embodiment, a conversation may be initiated by a system for providing context-based interventions to a user, e.g., a system 300 as shown with reference to FIG. 3, at any time via a user interface (e.g., a conversation user interface 310 of the system 300 of FIG. 3). The types of healthcare-related messages that may be presented to a user may be limited depending on the action being performed by the user. As described in more detail below, the types of healthcare-related messages that may be presented to a user may depend on the selected healthcare topic (i.e., the "care module" that is activated in some embodiments). In some embodiments, at least one healthcare-related message may be at least a part of an intervention recommendation provided to the user.

[0160] The finite state diagram of FIG. 14 includes a number of exemplary states that may be set by the system, each of the multiple states being associated with one of “conversation topic not set” and “conversation topic set”, and one of “start action”, “end action”, and “end action”, where “start action” corresponds to when a user starts an action related to a care module, “start action” corresponds to when a user is in an action related to a care module, and “end action” corresponds to when a user is not in an action related to a care module.

[0161] Thus, the states include "no topic set, action in progress" (NS-SA) 1401, "no topic set, action in progress" (NS-IA), "no topic set, action completed" (NS-EA) 1403, "no topic set, no action" (NS-NA) 1404, "topic set, action in progress" (TS-SA) 1405, "topic set, action in progress" (TS-IA) 1406, "topic set, action completed" (TS-EA) 1407, "topic set, no action" (TS-NA) 1408. Further description of some of the states is provided below.

[0162] "No Topic, Start Action" (NS-SA) 1401: When a user starts an action (e.g., brushing teeth), only healthcare related messages associated with this action (e.g., healthcare related messages associated with the care module "Brushing Teeth" and / or healthcare related messages associated with the care module "Oral Health") may be presented. Also, only healthcare related messages associated with the user's interests, information and suggestions are presented. If such healthcare related messages are in the message queue, the state of the system may be set as "No Topic, Start Action" (TS-SA) 1401 and may be further defined by assigning "Brushing Teeth" or "Oral Health" as a healthcare topic to start a related conversation and present the healthcare messages to the user.

[0163] "Topic not set (set), in action" (NS-IA / TS-IA) 1402, 1406: During an action, it is preferable not to distract the user with healthcare related messages other than those that support the current action. Therefore, during an action, the system should only allow the initiation or continuation of an action-related conversation with healthcare related messages related to the instruction execution of the action, in other words, when the system is in the "Topic not set (set), in action" state, it may only present the user with healthcare related messages related to coaching or guidance of the action (e.g., messages related to coaching or guidance of the tooth brushing action).

[0164] "No Topic, Action Ended" (NS-EA) 1403: The end of an action (e.g., brushing teeth) marks an important moment: (1) Initiate a conversation about the just-completed action, e.g., about the user's status, the user's interest in receiving coaching, evaluating the progress of the coaching, or asking the user for subjective input into the action session. In this case, after setting the state to "topic not set, action completed" (NS-EA) 1403, the state may be set to "topic set, action completed" (TS-EA) 1407. (2) Initiating a conversation about an action related to the just-completed action (e.g., by presenting a message related to the healthcare topic “tongue cleaning”) or within the same domain as the just-completed action (e.g., by presenting a message related to the healthcare topic “oral health”). This may result in (a) the user becoming engaged in a conversation about the new healthcare topic, thus triggering the state of the system to be set as “topic set, no action” (TS-NA) 1408, or (b) initiating an action related to the new healthcare topic during the conversation, thus triggering the state of the system to be set as “topic set, action started” (TS-SA) 1405.

[0165] "No Topic Set, No Action" (NS-NA) 1404: If the user is not involved in a care module related activity (i.e., an activity related to a selected healthcare topic or an activity that directly / indirectly causes a healthcare topic to be selected) and is not currently engaged in a conversation, any care module related conversation can be started and the system state is set as "Topic Set, No Action" (TS-NA) 1408. In this state, only healthcare related messages that are not related to the execution of a behavioral command may be presented to the user. In some embodiments, "No Topic Set, No Action" (NS-NA) 1404 may be triggered when a conversation topic is released (indicated by the arrow "Topic Released") during the "Topic Set, No Action" (NS-NA) 1408 state.

[0166] In some embodiments, in "topic not set (set), no action" (NS-NA, TS-NA), a dialogue processing unit of the system, e.g. a dialogue handler 1210 as illustrated with respect to FIG. 12, may be configured to organize and / or prioritize healthcare related messages in a message queue as follows: -Healthcare-related messages associated with higher levels of the hierarchy of health care topics (care modules) are prioritized; and / or All healthcare-related messages associated with each healthcare topic (care module) are presented in a bundled manner, e.g., the messages are presented together at the same time or in a sequence, and / or the load indicator value remains below a predefined threshold and / or no healthcare topic is selected after a predefined period of no interaction between the user and the system; and / or - If a conversation is ongoing (i.e. "topic set"), the dialogue processing unit is configured to prevent undesired topic switching while the user is engaged in a conversation about an activity. Thus, depending on the starting point of the conversation, the state of the system may be triggered from "topic set, action started" (TS-SA) 1405 to "topic set, action in progress" (TS-IA) 1406 and then to "topic set, action finished" (TS-EA) 1407. Only when the action is finished and the conversation about the healthcare topic associated with the action is finished, a topic switch is allowed ("topic set, no action" (TS-NA) 1408).

[0167] By controlling the operation of the dialogue processor in accordance with the finite state diagram as described above, the following aspects can be ensured: -The conversation is natural and the topic doesn't change in an unnatural way. - Assuming the user is in the current context (based on information obtained from the context modeling unit), the user is approached at a natural and convenient moment. Topics that the user is currently interested in are prioritized over topics that he is less interested in (based on information obtained from the user profile management unit).

[0168] Thus, an improved method and apparatus for providing context-based healthcare interventions to a user is provided that overcomes existing problems.

[0169] There is also provided a computer program product comprising a computer readable medium, the computer readable medium having computer readable code embodied therein, which, when executed by a suitable computer or processor, causes the computer or processor to instruct the computer or processor to carry out the method or methods described herein. It will therefore be understood that the invention also applies to computer programs, in particular computer programs on or in a carrier adapted to carry out the invention. The program may be in the form of source code, object code, code intermediate source and object code such as partially compiled form, or in any other form suitable for use in implementing the method according to the invention.

[0170] It will be understood that such a program may have many different structural designs. For example, the program code implementing the functionality of the method or system according to the invention may be divided into one or more subroutines. Many ways of distributing functionality among these subroutines will be apparent to those skilled in the art. These subroutines may be stored together in an executable file to form a self-contained program. Such an executable file may comprise computer executable instructions, such as processor instructions and / or interpreter instructions (e.g. Java interpreter instructions). Alternatively, one or more or all of these subroutines may be stored in at least one external library file and linked, for example at run-time, statically or dynamically with the main program. The main program includes at least one call to at least one of these subroutines. These subroutines may also comprise function calls to each other.

[0171] Other variations to the disclosed embodiments can be understood and effected by a person skilled in the art in implementing the claimed invention, from a study of the drawings, the description and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may perform the functions of several items recited in the claims. The mere fact that certain means are recited in mutually different dependent claims does not indicate that a combination of these means cannot be used to advantage. The computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium provided together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless communication systems. Any reference signs in the claims should not be interpreted as limiting the scope of the claims.

Claims

1. 1. A computer-implemented method for providing a context-based healthcare intervention to a user, comprising: obtaining context information from a plurality of sources, the plurality of sources including at least one of a personal care device, a personal health device, and a sensor or actuator present in the user's home or worn by the user, the context information including at least one of a subjective opinion based on user input, usage information of a user device based on sensor data, location information of the user based on sensor data, weather information associated with the user's current location, time and / or date information, and a physiological condition of the user based on sensor data; updating at least one of a plurality of interest level values ​​of a user profile associated with the user based on the obtained context information, each of the plurality of interest level values ​​being related to one of a plurality of healthcare topics, the interest level value indicating a degree of interest of the user in receiving support for the respective healthcare topic and a requirement related to at least one of usage information of the user device, location information of the user, the weather information, the time and / or date information, and a physiological condition of the user; selecting one of a plurality of health care topics based on the interest level value in the user profile; providing an intervention recommendation to a user based on the selected healthcare topic, the intervention recommendation relating to suggested information related to the selected healthcare topic or a suggested adjustment to a user device, the suggested adjustment to the user device comprising at least one of changing a setting of the user device, switching on a user device, and switching off a user device; The method according to claim 1,

2. The step of selecting a healthcare topic includes: selecting health care topics having an associated interest level value greater than a first predetermined threshold; determining a change in an interest level value for at least one of the plurality of healthcare topics following updating of the user profile and selecting the healthcare topic if the respective determined change in interest level value is greater than a second predetermined threshold; or selecting a healthcare topic related or similar to another healthcare topic based on an inference algorithm or a hierarchical structure of the plurality of healthcare topics, the selected related or similar healthcare topic being associated with an interest level value higher than a predefined threshold; 2. The computer-implemented method of claim 1, comprising:

3. 3. The computer-implemented method of claim 1, further comprising the step of acquiring a persuasive style preference of the user for the selected healthcare topic, the acquired persuasive style preference of the user providing guidance on the content of an intervention recommendation for the user or how an intervention should be provided to the user, and the step of providing the intervention recommendation is further based on the acquired persuasive style preference of the user for the selected healthcare topic.

4. 4. The computer-implemented method of claim 1, wherein the user profile further comprises a performance level value for each of the plurality of healthcare topics, the performance level value indicating a quality of the user's performance of behaviors related to each healthcare topic, and the step of providing the intervention recommendation is further based on the performance level value for the selected healthcare topic in the user profile.

5. receiving input from a user indicating acceptance or rejection of the provided intervention recommendation; If the received input indicates acceptance, the method further comprises implementing the intervention recommendation; If the received input indicates a rejection, the method further comprises providing a different intervention recommendation to the user based on the selected healthcare topic. A computer-implemented method according to any one of claims 1 to 4.

6. 6. The computer-implemented method of claim 5, wherein if the received input indicates acceptance, the method further comprises determining a change in effectiveness of the implemented intervention over a predetermined period of time, the effectiveness indicating a degree of compliance of the user with the implemented intervention.

7. determining a difference between the determined change in effectiveness of the implemented intervention and an expected effectiveness index, the expected effectiveness index representing an expected change in effectiveness of the implemented intervention; - adjusting parameters and / or content of the performed intervention if the determined difference is greater than a predefined threshold. The computer-implemented method of claim 6 , further comprising:

8. If the received input indicates acceptance, the method further comprises determining an overall satisfaction score indicative of a user's degree of satisfaction with the performed intervention, the determining of the overall satisfaction score comprising: determining a progress satisfaction score indicative of a user's degree of satisfaction with the progress resulting from the performed intervention; determining a progress orientation score indicative of the user's degree of willingness to accept more radical persuasive measures versus gentler persuasive measures; determining a persuasion satisfaction score indicative of a user's degree of satisfaction with a persuasion style associated with the performed intervention; determining an overall satisfaction score based on the progress satisfaction score, the progress orientation score, and the persuasion satisfaction score; 8. A computer implemented method according to any one of claims 5 to 7, comprising:

9. determining a difference between the overall satisfaction score and an expected satisfaction score, the expected satisfaction score indicating a user's expected degree of satisfaction with the performed intervention; - adjusting parameters and / or content of the performed intervention if the determined difference is greater than a predefined threshold. The computer-implemented method of claim 8 , further comprising:

10. retrieving a plurality of healthcare-related messages from a database; determining a priority value for each of the plurality of healthcare related messages based on at least one of the obtained context information and the selected healthcare topic; providing at least one of the plurality of healthcare related messages to the user based on the priority value; 10. The computer implemented method of claim 1, further comprising:

11. The computer-implemented method of claim 10 , wherein providing at least one of the plurality of healthcare-related messages is further based on a load indicator value, the load indicator value indicating an amount of recent interaction with the user.

12. 12. A computer implemented method according to claim 10 or 11, further comprising the step of updating a stress indicator value following the provision of a healthcare related message to the user.

13. A computer program configured, when executed by a suitable computer or processor, to cause the computer or processor to carry out a method according to any one of claims 1 to 12.

14. 1. An apparatus for providing a context-based healthcare intervention to a user, the apparatus comprising: Context information is obtained from a plurality of sources, the plurality of sources including at least one of a personal care device, a personal health device, and a sensor or actuator present in the user's home or worn by the user, the context information including a subjective opinion based on user input, and at least one of usage information of a user device based on sensor data, location information of the user based on sensor data, weather information associated with the user's current location, time and / or date information, and a physiological condition of the user based on sensor data; updating at least one of a plurality of interest level values ​​of a user profile associated with the user based on the obtained context information, where each of the plurality of interest level values ​​is associated with one of a plurality of healthcare topics, and the interest level value indicates a degree of interest of the user in receiving support for a respective healthcare topic and a requirement related to at least one of usage information of the user device, location information of the user, the weather information, the time and / or date information, and a physiological condition of the user; selecting one of a plurality of health care topics based on the interest level value in the user profile; providing an intervention recommendation to a user based on the selected healthcare topic, wherein the intervention recommendation relates to suggested information related to the selected healthcare topic or a suggested adjustment to a user device, the suggested adjustment to the user device comprising at least one of changing a setting of the user device, switching on the user device, and switching off the user device; 23. An apparatus having a processor configured to:

Images