Pain management system, computer-implemented method, and computer program

Abstract

The pain management system (300) is configured to: receive a target pain score (314) representing a level of pain that a user deems tolerable for a specified time period; receive one or more target input parameters (304) representing characteristics and / or activities of the user for the same specified time period, where the one or more target input parameters (314) are a subset of a complete list of available input parameters; receive one or more settings (315) representing the one or more input parameters to be increased or decreased; determine one or more calculated user parameters (316) based on the target pain score (314) and the target input parameters (304) using a neural network trained for the individual user, where at least one of the calculated user parameters (316) is set based on the settings (315); and present the one or more calculated user parameters (316) using a user interface.

Description

[Technical Field]

[0001] SUMMARY The present disclosure relates to pain management systems that can be used by a user to monitor and predict chronic pain based on activities performed by the user. Summary of the Invention [Means for solving the problem]

[0002] According to a first aspect, there is provided a pain management system comprising: receiving a target pain score representing a level of pain that the user deems tolerable for a specified period of time; receiving one or more target input parameters representative of the user's characteristics and / or activities during the same defined time period, the one or more target input parameters being a subset of the complete list of available input parameters; receiving one or more settings representing one or more input parameters to be increased or decreased; determining one or more calculated user parameters based on the target pain score and the target input parameters using a neural network trained for the individual user, wherein at least one of the calculated user parameters is set based on the settings; and presenting the one or more calculated user parameters using a user interface. A pain management system is provided.

[0003] Advantageously, such a system may guide the user to change their behavior in a way that improves their health and well-being by reducing the likelihood of exceeding their selected pain threshold, which may significantly improve the user's / patient's quality of life (QoL).

[0004] The system is examining a set of historical pain parameter log entries associated with the user to identify pain parameter log entries that match the received target pain score and target input parameters as matching pain parameter log entries, each pain parameter log entry representing a plurality of user input parameters and a user input pain score; modifying the matched pain parameter log entry based on the setting to generate a modified pain parameter log entry; applying a neural network to the modified pain parameter log entries to determine a calculated pain score; The user interface may be configured to (i) present one or more user parameters of the modified pain parameter log entry as calculated user parameters, and (ii) present the calculated pain score.

[0005] The system may be configured to modify the matched pain parameter log entry by increasing or decreasing one or more user-inputted parameters of the matched pain parameter log entry to generate a modified pain parameter log entry.

[0006] The system is examining a set of historical pain parameter log entries associated with the user to identify pain parameter log entries that match the received target pain score and target input parameters as matching pain parameter log entries, each pain parameter log entry representing a plurality of user input parameters and a user input pain score; modifying the matched pain parameter log entry based on the setting to generate a modified pain parameter log entry; applying a neural network to the modified pain parameter log entries to determine a calculated pain score; comparing the calculated pain score with a target pain score; If the calculated pain score is greater than the target pain score, adjusting one or more of the input parameters of the modified pain parameter log entry; applying a neural network to the adjusted modified pain parameter log entries to determine an adjusted calculated pain score; repeating the comparing step one or more times on the adjusted calculated pain score; If the calculated pain score is less than or equal to the target pain score, the user interface may be configured to (i) present one or more user parameters of the modified pain parameter log entry as the calculated user parameters, and (ii) present the calculated pain score.

[0007] The system may be configured to repeat the comparison step multiple times up to a predetermined maximum number, which may change over time.

[0008] The system is presenting to a user via a user interface options for modifying one or more of the calculated user parameters; receiving one or more modified calculated user parameters representing user input from a user interface; applying a neural network to the modified calculated user parameters to determine a modified calculated pain score; and presenting the modified calculated pain score using a user interface.

[0009] The system is A user interface, comprising: a plurality of user-input parameters representing characteristics and / or activities of the user during a specified period of time; and a user interface configured to receive a user-inputted pain score representing the degree of pain experienced by the user during the same defined period of time; An AI processor, The system may further include an AI processor configured to set a plurality of weighting values ​​for the neural network based on a plurality of user-input parameters and a user-input pain score of the user.

[0010] The system may be configured to store a plurality of weighting values ​​associated with a user identifier, the user identifier being uniquely associated with an individual user's profile.

[0011] The system is It may be further configured to modify the functionality of the user interface over time so that the user is presented with additional mechanisms for providing user input parameters.

[0012] The user input parameters and / or the target input parameters are: (i) a duration characteristic representing the duration for which the user performed an activity and / or exhibited a characteristic; (ii) an intensity characteristic representing the intensity with which the user performed the activity; (iii) a satisfaction attribute that describes the satisfaction experienced by the user when performing the activity; and (iv) type characteristics.

[0013] The AI ​​processor may be configured to store pain parameter log entries in memory, each pain parameter log entry comprising: Multiple user input parameters, a user-entered pain score, and It includes a date identifier associated with the corresponding user-input parameters and the user-input pain score.

[0014] Each pain parameter log entry: It may further include a user identifier uniquely associated with the individual user to whom the plurality of user-input parameters and user-input pain score are associated.

[0015] The system is processing the plurality of pain parameter log entries to identify pain parameter log entries having a user-inputted pain score that is increasing (or exceeds a high pain trigger threshold) as high pain parameter log entries, each pain parameter log entry including one or more user-inputted parameters and a user-inputted pain score; performing an analysis of the user-input parameters of the high pain parameter log entries to determine a correlation score representing the degree of correlation between values ​​of corresponding user-input parameters in the high pain parameter log entries; If the one or more user parameters have a correlation score that satisfies the correlation criterion, identifying the one or more user parameters as a pain trigger may be configured to determine the pain trigger.

[0016] The system is processing the plurality of pain parameter log entries to identify pain parameter log entries having a user-inputted pain score that is decreasing (or below a low pain trigger threshold) as low pain parameter log entries, each pain parameter log entry including one or more user-inputted parameters and a user-inputted pain score; performing an analysis of the user-inputted parameters of the low pain parameter log entries to determine a correlation score representing the degree of correlation between values ​​of corresponding user-inputted parameters in the low pain parameter log entries; If the one or more user parameters have a correlation score that satisfies the correlation criterion, identifying the one or more user parameters as a pain protector may be configured to determine the pain protector.

[0017] The plurality of user input parameters may include one or more sensed input parameters, which may be provided directly or indirectly from a sensor.

[0018] User input parameters are sleep parameters, Work parameters, physical activity parameters, Housework parameters, Entertainment parameters, Rest parameters, and pain range parameters, Heart rate parameters, blood pressure parameters, temperature parameters, Energy / fatigue level parameters, and / or stress level parameters.

[0019] 1. A computer-implemented method comprising: receiving a target pain score representing a level of pain that the user deems tolerable for a specified period of time; receiving one or more target input parameters representative of the user's characteristics and / or activities during the same defined time period, the one or more target input parameters being a subset of the complete list of available input parameters; receiving one or more settings representing one or more input parameters that a user aims to increase or decrease; determining one or more calculated user parameters based on the target pain score and the target input parameters using a neural network trained for the individual user, wherein at least one of the calculated user parameters is set based on the settings; Also provided is a computer-implemented method that includes presenting the one or more calculated user parameters using a user interface.

[0020] 1. A pain management system comprising: A user interface, comprising: a plurality of user-input parameters representing characteristics and / or activities of the user during a specified period of time; and a user interface configured to receive a user-inputted pain score representing the degree of pain experienced by the user during the same defined period of time; An AI processor, Also provided is a pain management system comprising: an AI processor configured to set a plurality of weighting values ​​for the neural network based on a plurality of user-input parameters and a user-input pain score of the user.

[0021] Such a system may include any of the features and functionality described herein.

[0022] Also, a personal management system, receiving a target score representing a score / level of a personal attribute that the user deems acceptable for a specified period of time; receiving one or more target input parameters representative of the characteristics and / or activities of the user during the same defined time period (the one or more target input parameters may be a subset of the complete list of available input parameters); receiving one or more settings representing one or more input parameters to be increased or decreased; determining, using a neural network trained for the individual user / each individual user, one or more calculated user parameters based on the target score and the target input parameters, wherein at least one of the calculated user parameters is set based on the configuration; A persona management system configured to present the one or more calculated user parameters using a user interface is also provided.

[0023] 1. A computer-implemented method comprising: receiving a target score representing the score / level of personal attributes that the user / each user considers acceptable for a defined period of time; receiving one or more target input parameters representative of the characteristics and / or activities of the user / each user during the same defined time period (the one or more target input parameters may be a subset of the complete list of available input parameters); receiving one or more settings representing one or more input parameters that a user aims to increase or decrease; determining one or more calculated user parameters based on the target score and the target input parameters using a neural network trained for the individual user, wherein at least one of the calculated user parameters is set based on the configuration; Also provided is a computer-implemented method that includes presenting the one or more calculated user parameters using a user interface.

[0024] A personal management system, A user interface, comprising: a plurality of user-input parameters representing characteristics and / or activities of the user during a specified period of time; and a user interface configured to receive a user-entered score representing the score / level of the personal attribute experienced by the user during the same defined time period; An AI processor, Also disclosed is a personal management system comprising: an AI processor configured to set a plurality of weighting values ​​for the neural network based on a plurality of user input parameters and a user input score of the user.

[0025] Such systems and methods may provide a platform that may guide users to change their behavior in a way that improves their health and well-being by reducing the likelihood of exceeding limits in user-selected personal attributes (which may be pain, stress, etc. (further examples are described below)). This may significantly improve the user's / patient's QoL. In another example, the systems and methods disclosed herein may be used for elite training, for example, to improve the maximum performance / capacity of individual athletes in any sport. Thus, the embodiments disclosed herein may also be used to improve physical performance.

[0026] It will be appreciated that any of the features and functionality described herein with respect to pain management systems may be implemented in non-pain related systems as well.

[0027] A computer program may be provided that, when executed on a computer, causes the computer to configure any apparatus, including a circuit, controller, converter, or device disclosed herein, or to perform any method disclosed herein. The computer program may be implemented in software, and the computer may be considered as any suitable hardware, including, as non-limiting examples, implementations in a digital signal processor, a microcontroller, and a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or an electronically erasable programmable read-only memory (EEPROM). The software may also be an assembly program.

[0028] The computer program may be provided on a computer-readable medium, which may be a physical computer-readable medium such as a disk or memory device, or may be embodied as a transient signal. Such a transient signal may be a network download, including an Internet download. One or more non-transitory computer-readable storage media may be provided that store computer-executable instructions that, when executed by a computing system, cause the computing system to perform any of the methods disclosed herein. [Brief explanation of the drawings]

[0029] One or more embodiments will now be described, by way of example only, with reference to the accompanying drawings.

[0030] [Figure 1] 1 illustrates an exemplary embodiment of how a pain management system may be trained to develop an individualized pain management model for each user. [Figure 2A]1 shows one exemplary screenshot used to explain examples of user input parameters. [Figure 2B] 10 shows another exemplary screenshot used to explain examples of user input parameters. [Figure 2C] 10 shows yet another exemplary screenshot used to explain examples of user input parameters. [Figure 3] An exemplary embodiment of how a trained pain management system can be used to help each user achieve a desired level of pain in a day is shown. [Figure 4A] 4 shows exemplary screenshots used to explain how the user interface of FIG. 3 may be used to provide calculated user parameters for each user. [Figure 4B] 4 shows exemplary screenshots used to explain how the user interface of FIG. 3 may be used to provide calculated user parameters for each user. [Figure 4C] 4 shows exemplary screenshots used to explain how the user interface of FIG. 3 may be used to provide calculated user parameters for each user. [Figure 4D] 4 shows exemplary screenshots used to explain how the user interface of FIG. 3 may be used to provide calculated user parameters for each user. [Figure 5] 1 illustrates an exemplary embodiment of how a trained pain management system may be used to help each user predict expected pain levels based on planned activities. [Figure 6] Provides an overview of functionality that may be implemented by any of the AI ​​processors described herein. [Figure 7] 10A-10C illustrate exemplary screenshots that may be displayed to each user via any of the user interfaces disclosed herein. [Figure 8]1 illustrates another exemplary embodiment of how a pain management system may be trained to develop a personalized pain management model for a user. [Figure 9A] 1 shows one exemplary screenshot used to explain an example of a user input mechanism. [Figure 9B] 10 shows another exemplary screenshot used to explain an example of a user input mechanism. [Figure 9C] 10 shows yet another exemplary screenshot used to explain an example user input mechanism. [Figure 10] 1 illustrates another exemplary embodiment of a pain management system according to the present disclosure. [Figure 11] 1 illustrates a computer-implemented method according to the present disclosure. [Figure 12] To demonstrate the relationship between responding and non-responding patients regarding improvement in pain interference (QoL) in a clinical study. [Figure 13] Figure 1 shows the difference (reduction) in T-scores in pain interference (QoL) for responding patients in the first (patients 1-5) and second (patients 6-10) phases of the clinical study. [Figure 14] 1 shows the relationship between responders and non-responders regarding improvement in pain intensity in a clinical study. DETAILED DESCRIPTION OF THE INVENTION

[0031] Chronic pain represents one of the most serious challenges for European citizens and economies. Studies show that one in five Europeans suffers from chronic pain. Chronic pain is defined as pain that lasts for more than three to six months beyond the expected healing period. On average, people endure chronic pain for up to seven years. Pain significantly impacts daily life, negatively impacting patients' daily lives. Chronic pain is also associated with psychological disorders such as anxiety and depression. People with chronic pain are more likely to develop drug addiction, and the risk of suicide is at least doubled for people with chronic pain.

[0032] Numerous studies have shown that pain is commonly overlooked and therefore undertreated, and sometimes not treated at all. Most EU countries do not have specific clinical guidelines for managing chronic pain. Many patients do not encounter clinicians with sufficient knowledge of how best to manage their pain. Often, patients' first point of contact is their primary health care service, and only a lucky few encounter a pain specialist. Only one in five chronic pain patients in Europe ever encounters a pain specialist, and only 2% have ever attended a pain management program. Other studies have found that only about 1% of chronic pain patients ever encounter a pain specialist for their condition. It is important to emphasize here that primary care physicians (PCPs) often feel overwhelmed when it comes to treating chronic pain. Only 50% of PCPs are confident in managing chronic pain, and more than half (54%) of PCPs are unsure of what to do when patients still complain of pain. With very limited access to pain specialists, chronic pain is often inappropriately treated. One of the consequences of mismanaged pain is opiate abuse.

[0033] One or more of the embodiments disclosed herein may enable patients to manage the multidimensional aspects of chronic pain, thereby achieving much better outcomes for the patient without the need for face-to-face patient interaction.

[0034] As discussed in detail below, examples disclosed herein relate to advanced self-management tools for chronic pain based on artificial intelligence (AI). The systems described herein analyze data provided by individuals and can utilize machine learning capabilities to identify specific patient pain triggers and pain protectors. Patients can receive specific, actionable advice on how to reduce the negative effects of chronic pain and improve their quality of life (QoL).

[0035] 1 shows an exemplary embodiment of how a pain management system 100 may be trained to develop an individualized pain management model for each user, i.e., a truly patient-centered model. In this example, the pain management model is represented by a plurality of weighting values ​​W110 stored in computer memory 112.

[0036] The system 100 includes a user interface 102 that receives a plurality of user-input parameters 104. The plurality of user-input parameters 104 represent activities and / or characteristics of a user during a defined period of time, and more specifically, activities and / or characteristics that may affect the degree of chronic pain experienced by the user. The activities may relate to, for example, work, physical activity, and recreational time. The characteristics may relate to one or more measured or determined attributes of the user's body, such as heart rate, blood pressure, body temperature, etc.

[0037] Depending on the type of parameter, the user input parameters 104 may include one or more of the following characteristics: (i) a duration characteristic representing the duration for which the user performed the activity and / or exhibited the characteristic, (ii) an intensity characteristic representing the intensity with which the user performed the activity, (iii) a satisfaction characteristic representing the satisfaction experienced by the user when performing the activity, and (iv) a type characteristic. Specific examples of user input parameters 104 related to activities are described below with reference to Figures 2A-C. Specific examples of user input parameters 104 related to user characteristics are described below with reference to Figure 10.

[0038] 1, the user interface 102 also receives a user-input pain score 106, which may represent one or more pain values, representing the degree of pain experienced by the user during the same defined time period. The user-input pain score 106 may represent the degree of pain experienced by the user associated with the corresponding user-input parameter 104.

[0039] In this example, user-input pain score 106 represents the average pain value experienced by the user over a defined period of time (e.g., one day). As shown in the screenshots of Figures 2A-C, user-input parameters 104 may also be related to the user's pain, but may be different from user-input pain score 106. In this example, as again shown in Figures 2A-C, user-input parameters 104 include (i) the maximum and minimum pain values ​​experienced over the predetermined period of time, and (ii) the duration the user experienced pain (optionally, the duration the user experienced a certain level of pain).

[0040] In this example, the user-input parameters 104 represent characteristics / activities performed by the user over a 24-hour period, and the user-input pain score 106 represents one or more pain values ​​over the same 24-hour period. Optionally, each of the user-input parameters 104 and the user-input pain score 106 may be stored in memory with a date identifier. The date identifier may be associated with the corresponding user-input parameter 104 / user-input pain score 106 when the user enters the information. For example, the date identifier may be provided by the user via the user interface 102 by the user selecting a date to which the user-input parameters 104 and the user-input pain score 106 relate. In other examples, the date identifier may be implemented as a timestamp. Optionally, the date identifier may be attributed to the user-input parameter 104 / user-input pain score 106 when the information is received by the AI ​​processor 108. Using such a date identifier may beneficially allow the user-input parameter 104 / user-input pain score 106 to be later examined as being associated with a particular date. For example, it may be possible to generate plots of the user-input parameters 104 / user-input pain scores 106 to illustrate how the values ​​have changed over time, or to perform any statistical analysis that may be useful.

[0041] It will be understood from the following description that a user may provide a user-input pain score 106, thereby training a pain management model for an individual user. Depending on the data provided, the pain management model may be sufficiently trained so that it can be used to provide a useful prediction of pain (or any other output described herein) after a defined period of time, such as several days, eight days, or nine days, or several weeks. The length of time it takes to properly train the pain management model may be affected by variability in the user's daily activities; if the degree of variability is high, it may take longer to train the pain management model to an acceptable level.

[0042] The user interface 102 transmits the user input parameters 104 to the AI ​​processor 108. It will be appreciated that the AI ​​processor 108 may be co-located on the device that provides the user interface 102 to the user, or may be located remotely from the device that provides the user interface 102. In some embodiments, the device that provides the user interface 102 may be the user's smartphone, as it is portable and readily available for the user to input the user input parameters 104. The functionality of the AI ​​processor 108 may be cloud-based, whereby the user interface 102 provides the user input parameters 104 and the user input pain score 106 to the AI ​​processor 108 over the internet. In this manner, a server-side application may be provided.

[0043] The AI ​​processor 108 may implement any artificial intelligence algorithm known in the art, including a neural network. For example, the AI ​​processor 108 may train an artificial neural network (ANN) using the user-input parameters 104 as inputs and the user-input pain score 106 as ground truth (i.e., the result the ANN is intended to produce for the provided user-input parameters 104). The AI ​​processor 108 may apply multiple weighting values ​​W110 (which may also be referred to as weighting coefficients) to the multiple user-input parameters 104 to determine a calculated pain score and adjust the multiple weighting values ​​W110 based on the difference between the calculated pain score and the user-input pain score 106. As known in the art, this may include training the AI ​​engine using mathematics. For example, input and output weighting values ​​may be iteratively calculated in epochs according to a learning count, and then error-correction mathematics may be applied to minimize the difference between the average pain level input by the user (the user-input pain score 106) and the calculated average pain level suggested by the AI ​​engine (the calculated pain score). In this manner, the AI ​​processor 108 can adjust the weights W110 based on an optimization routine to train the model, thereby moving the calculated pain score toward the user-input pain score 106. More generally, the AI ​​processor 108 can set / train the weights W110 of the neural network based on the user-input parameters 104 and the user-input pain score 106 of the user.

[0044] In this manner, the AI ​​processor 108 determines a plurality of weights W110 for the ANN that represent the individual user's pain management model. The plurality of weights W110 may be stored in computer memory 112, as shown schematically in Figure 1, so that the weights W110 may be later retrieved for further training (during which they may be modified and stored back in memory 112) or for application to received information to generate an output for the user.

[0045] In some examples, the AI ​​processor 108 may also store the user-input parameters 104, the user-input pain score 106 (and optionally a date identifier) ​​in memory 113. This memory 113 may be the same as or different from the memory 112 that stores the weighting value W110. This combination of the user-input parameters 104, the user-input pain score 106, and optionally a date identifier may collectively be referred to as a pain parameter log entry, shown in FIGS. 2A-C as a "daily log." In this example, the pain parameter log entry may relate to activities performed and pain experienced over a one-day period. It will be understood that the period need not be one day, but in other applications may be, by way of non-limiting example, 1, 2, 3, 4, 6, or 12 hours. As will be appreciated from the following discussion, particularly with reference to FIG. 10, the pain parameter log entry may also include user parameters that represent characteristics of each user in addition to or in lieu of activities performed by each user.

[0046] Particularly in examples where the AI ​​processor 108 is remote from the user interface 102, the user interface 102 may also communicate a user identifier to the AI ​​processor 108. Such a user identifier may be uniquely associated with an individual user's profile. In some examples, the user identifier may be stored as part of a pain parameter log entry. The user interface 102 may associate the user identifier with the corresponding user-input parameters 104 and user-input pain score 106 when the user interface transmits this information to the AI ​​processor 108. This may enable the AI ​​processor 108 to use the received user identifier to retrieve the correct weighting values ​​110 from memory (i.e., the weighting values ​​110 associated with the same user identifier), so that the weighting values ​​110 can be adjusted for newly received information and the individual model for the particular user can be updated and further trained. In this way, the AI ​​processor 108 may ensure that the model is used only for the (single) associated user, i.e., ensures true patient-centeredness, which advantageously ensures that the model is customized for each individual. This has proven important in order to generate an accurate model for each user, as pain can be experienced differently by each individual and for different reasons.

[0047] Advantageously, each individual user's pain management model (which may be represented by weighting values ​​110) may be used to identify different pain triggers and pain protectors for a particular user.

[0048] A pain trigger may be considered an activity (or combination of activities) that causes a user to experience significant pain. In some examples, the system 100 may determine a pain trigger by processing multiple pain parameter log entries, each entry representing at least one or more of a user-input parameter 104 and a user-input pain score 106. For example, the system 100 may identify a pain parameter log entry having a user-input pain score 106 above a high pain trigger threshold (such as 8 out of 10) as a high pain parameter log entry. The system may then perform an analysis of the user-input parameters 104 of the high pain parameter log entry to determine a correlation score representing the degree of correlation between values ​​of corresponding user-input parameters 104 in the high pain parameter log entry. The analysis may be a statistical analysis, a machine learning analysis, or any other type of analysis that may determine the degree of correlation between values ​​of corresponding user-input parameters 104 in the high pain parameter log entry. For example, the system 100 may identify one or more user parameters as a pain trigger if the one or more user parameters have a correlation score that meets a correlation criterion. Examples of such correlation criteria include a correlation score that exceeds a correlation threshold, the highest correlation score for all user parameters, and a predetermined number of the highest correlation scores for all user parameters (e.g., the three user parameters with the highest correlation scores). The system may then display the pain triggers to the user (e.g., using the user interface 102). In this manner, the system 100 may beneficially determine and display to the user certain user-input parameters and their associated values ​​as "pain triggers," thereby enabling each user to modify their behavior to avoid those pain triggers and thereby reduce future levels of pain.

[0049] A pain protector may be considered an activity (or combination of activities) that causes the user to experience a low degree of pain. In some examples, system 100 may determine a pain protector by processing multiple pain parameter log entries, each entry representing at least one or more of user-input parameters 104 and user-input pain scores 106. For example, system 100 may identify pain parameter log entries having a user-input pain score 106 below a low pain trigger threshold (e.g., 3 out of 10) as low pain parameter log entries. The system may then perform an analysis of the user-input parameters 104 of the low pain parameter log entries to determine a correlation score and thereafter identify certain user parameters as pain protectors in a manner similar to that described above for pain triggers. Again, such analysis may be statistical analysis, machine learning analysis, or any other type of analysis. As noted above, in this manner, system 100 may beneficially determine and present to the user certain user-input parameters, and their associated values, as “pain protectors.”

[0050] In this way, the pain management model can be used to teach each individual how to be more proactive by learning their pain triggers and pain protectors. The human brain can reasonably handle only a maximum of four variables at any one time. The AI ​​engine provided by the AI ​​processor 108 can be utilized to process the more multivariate functional aspects of each individual's pain level. This can enable each chronic pain patient to have their own unique capabilities for self-management and to take an active role in their condition, which directly improves their QoL.

[0051] 2A-C show three exemplary screenshots used to explain examples of user-input parameters 204 and user-input pain scores 206, and how these screenshots may be provided by a user via a graphical user interface (GUI) to train a pain management model for an individual user. The GUI is an example of the user interface of FIG.

[0052] The first screenshot shows the GUI before the user has entered any activity or pain information. As shown in Figure 2A, the user input parameters 204 in this example are: Sleep parameters, Work parameters, Physical activity parameters, Housework parameters, Entertainment parameters, rest parameters, and · Pain range parameters.

[0053] It will be understood that any other parameters may also be used, or may alternatively be used, and in some examples, may be configured by the user. For example, the value of one or more of the parameters described herein, including the stress level parameter, may be determined by processing answers provided by the user in a questionnaire. The value of the stress level parameter may have a significant impact on the user's pain.

[0054] It will also be appreciated that because the user-input parameters 204 are used to generate an individualized pain management model, it does not matter if different users score the user-input parameters 204 differently; as long as the user scores the user-input parameters 204 in a consistent manner, the pain management model will be trained appropriately for that user. Thus, beneficially, a user does not need to worry about scoring their activities against a standard scoring scheme determined by someone else. Furthermore, each user / patient has their own AI engine trained exclusively based on their own data.

[0055] In this example, the user-entered pain score 206 is the average pain the user experienced that day.

[0056] The first screenshot also shows one way in which a user may provide a date 205 to be associated with the user-input parameters 204 and the user-input pain score 206. As discussed above, the user interface may determine a date identifier based on the date 205 provided by the user.

[0057] The second screenshot of Figure 2B shows examples of information provided by a user for user input parameters and corresponding attributes. More specifically, by way of non-limiting example, For sleep parameters: duration and satisfaction characteristics (in this example, a sliding scale between minimum and maximum values, such as 1-5), For task parameters: duration, satisfaction, and intensity attributes (in this example, a sliding scale between minimum and maximum values), Physical activity parameters: duration, satisfaction, and intensity characteristics; Housework parameters: duration characteristics, satisfaction characteristics and intensity characteristics, For entertainment parameters: duration, satisfaction, and intensity attributes; Regarding rest parameters: duration and satisfaction characteristics, and Pain range parameters: maximum pain profile, minimum pain profile, and duration profile.

[0058] Although not shown in this example, one or more of the input parameters may have a type attribute that identifies a particular type of activity. For example, a physical activity parameter may have a type attribute that can represent different types of physical activity, such as cycling, jogging, swimming, etc.

[0059] In this example, each duration characteristic corresponds to the amount of time spent performing an activity (or experiencing pain for the pain range parameter) during the day.

[0060] As discussed below, in some examples, one or more of the parameter characteristics may be automatically provided by associated software. For example, software is known in the art that can automatically classify parameters such as sleep, rest, and physical activity from signals available from wearable sensors. Such known software may also determine associated duration characteristics.

[0061] The third screenshot details how the user can provide input for a particular aspect of the job parameters.

[0062] After a short training period, the pain management model has been shown to provide suggestions as to how a user can plan their day, training them appropriately to achieve, for example, the highest level of function with the lowest possible level of pain. The use of an AI processor may enable the collection and analysis of multiple types of data so that the multifaceted causation of an individual's chronic pain can be determined.

[0063] Figure 3 shows an exemplary embodiment of how a trained pain management system 300 may be used to help a user achieve a desired level of pain in a given day. Features of Figure 3 that are also shown in Figure 1 have been given corresponding reference numerals in the 300 series and will not necessarily be described again in detail here.

[0064] In this example, the user provides a target pain score 314 to the user interface 302. The target pain score 314 represents a level of pain that the user deems tolerable when performing various activities for a specified period of time. In some examples, the user may also provide one or more target input parameters 304 (sometimes referred to as fixed activities), which typically represent a subset of the complete list of available input parameters. As described below, the system 300 may then provide one or more calculated user parameters 316 based on the target pain score 314 and the target input parameters 304. The calculated user parameters 316 may represent suggestions for values ​​of one or more of the input parameters (not provided as inputs by the user) that are expected to achieve the target pain score 314.

[0065] In this example, the user interface 302 sends the target pain score 314 and the input target input parameters 304 to a processor 303 that performs some processing on the target pain score 314 and the input target input parameters 304 before using the pain management model, as described in more detail below.

[0066] In addition to receiving the target pain score 314 and the entered target input parameters 304 from the user interface 302, the processor 303 in this implementation also receives one or more settings 315. The settings 315 may represent one or more types of input parameters (activities) to be increased or decreased. The settings 315 may include, for example, user settings, default settings, or physician settings, as described in more detail below.

[0067] The user settings may represent one or more types of input parameters (activities) that the user has indicated should be increased or decreased. The user settings 315 may be received directly from the user interface 302 or may be read from a computer memory that stores a user profile. The user settings may represent one or more types of input parameters (activities) that the user aims to increase or decrease. This may be one type of input parameter that is expected to provide a health benefit to each user. This may be an activity that is important to the user. In this particular example, assume that the user settings represent a desire to increase the duration of a physical activity parameter. It will be understood that the user settings 315 may represent a desire to increase or decrease any input parameter.

[0068] In some examples, the user settings may represent one or more types of input parameters (activities) that the system 300 determines should be increased or decreased. For example, the system may determine the user settings based on one or more information in the user's profile. In one embodiment, the system 300 may process one or more pain location settings that the user included in their profile to determine the appropriate user settings. This may include, as a non-limiting example, determining that if the pain location setting indicates lower back pain, the user setting is to increase the duration attribute for the rest parameter. Such a determination may be made by the system 300 according to any suitable algorithm or by accessing any database / lookup table associated with the system 300.

[0069] A default setting may also represent one or more types of input parameters (activities) that are increased or decreased, but not necessarily user-defined. This may include an indication that all types of input parameters are equally important and therefore should be increased or decreased equally. That is, in some examples, the setting 315 may indicate that all input parameters are to be increased or decreased by the same percentage.

[0070] The physician settings may represent one or more types of input parameters (activities) that the physician has indicated should be increased or decreased. Thus, the physician may advantageously provide information that may cause the user to change their behavior in a way that the physician has identified as likely to improve the user's quality of life.

[0071] The processor 303 first examines a set of historical pain parameter log entries (daily logs) stored in memory 313 associated with the same user to identify pain parameter log entries that match the received target pain score 314 and the input target input parameters 304. For example, the processor 303 may determine (i) the degree of correlation between the received target pain score 314 and the input target input parameters 304, and (ii) each of the pain parameter log entries stored in memory 313. The processor 303 may then select the pain parameter log entry with the highest correlation as the matched pain parameter log entry, or may select at least one with a sufficiently high correlation that it is expected to provide a reasonable starting point for subsequent processing.

[0072] As a next step, processor 303 may modify the matched pain parameter log entry based on settings 315. This may include increasing or decreasing one or more user-input parameters of the matched pain parameter log entry. In this example, settings 315 represent a desire to increase the duration of the physical activity parameter, so processor 303 increases the duration attribute of the physical activity parameter in the matched pain parameter log entry. Processor 303 may temporarily save this modified entry as a modified pain parameter log entry. In some examples, processor 303 may increase the duration attribute by a predetermined amount that may be coded (e.g., hard-coded) into software or provided as part of settings 315. For example, the predetermined amount may be a predetermined period of time or a predetermined percentage increase. Optionally, processor 303 may include the original target input parameters 304 in the pain parameter log entry without modification. That is, processor 303 may prevent the original target input parameters 304 from being modified (at least initially) based on the user indicating that the input parameters are fixed.

[0073] Processor 303 then sends the complete set of input parameters for modified pain parameter log entry 317 to AI processor 308. AI processor 308 applies a neural network using each user's individual pain management model (defined by weighting values ​​310 stored in memory 312) to modified pain parameter log entry 317 to determine a calculated pain score 319. In this manner, the system uses the neural network trained for the individual user to determine one or more calculated user parameters 316 based on target pain score 314 and target input parameters 304, at least one of which is set based on user settings 315. AI processor 308 then returns calculated pain score 319 to processor 303. In some examples, user interface 302 may then present one or more calculated user parameters 316 (and optionally calculated pain score 319) to each user.

[0074] In this example, processor 303 compares calculated pain score 319 with target pain score 314. If calculated pain score 319 is greater than target pain score 314, modified pain parameter log entry 317 may be deemed unacceptable for causing excessive pain to the user. (It will be appreciated that calculated pain score 319 may be greater than target pain score 314 because the duration of physical activity has increased in this example.) In this case, processor 303 may adjust one or more of the input parameters of modified pain parameter log entry 317 (e.g., input parameters that are not related to target input parameters 304 or physical activity parameters identified by settings 315). For example, processor 303 may, by way of non-limiting example, increase the duration characteristic of a rest parameter or decrease the intensity characteristic of a work parameter. Processor 303 may adjust one or more other input parameters based on a predetermined rule set, which may or may not be provided as part of user settings 315, to indicate which activities are less important to the user.

[0075] Processor 303 then sends the complete set of corrected input parameters for modified pain parameter log entry 317 to AI processor 308. AI processor 308 applies a neural network using the user's individual pain management model (defined by weighting values ​​310 stored in memory 312) to adjusted modified pain parameter log entry 317 to determine a corrected calculated pain score 319. Processor 303 compares corrected calculated pain score 319 to target pain score 314 and continues the loop of adjusting one or more other input parameters and correcting calculated pain score 319 until calculated pain score 319 is less than or equal to target pain score 314 or until a predetermined number of iterations have been performed. Once either of these requirements is met, processor 330 sends the final iteration of modified pain parameter log entry 317 to user interface 302, which may cause the processor to display calculated user parameters 316 to the user. If processor 303 performs a predetermined number of iterations without achieving target pain score 314, then processor 303 may direct user interface 302 to display an appropriate message to the user, such as: "Activity combinations cannot be identified without exceeding target pain level." In this manner, processor 303 may repeat the comparison step on the adjusted calculated pain score one or more times, and in some applications, multiple times up to a predetermined maximum.

[0076] Assuming the predetermined number of iterations has not been reached, the calculated user parameters 316 represent a set of activities the user should be able to perform without exceeding the target pain score 314 while increasing or decreasing one of the user parameters according to the settings 315. In this way, the system may advantageously guide the user to change their behavior in a way that improves their health and well-being by reducing the likelihood of exceeding the user-selected pain limit. This may significantly improve the user / patient's quality of life (QoL). Advantageously, the system 300 may attempt to increase the selected activity that the user indicated was high in the settings until the set pain threshold is reached. Furthermore, because the human brain may be able to calculate too many variables to identify a correlation between an undertaken activity and associated pain, the patient / user may not recognize what changes to make to their behavior without the system disclosed herein. The system described herein may be used to improve the patient / user's QoL regardless of a suboptimal current activity balance, for example, whether the patient / user is overly or underactive in their daily activities. This can be achieved through the use of appropriate settings 315. It will be appreciated that a user may be overly active or underactive with respect to a particular activity. For example, a person who is afraid to move may be considered overly active with a certain amount of rest per day, while a person who refuses to accept permanent damage and continues as they have been may be considered overly active in physical activity, work, etc., but underactive with rest. In some implementations, a pain management model may be used to calculate the optimal exercise duration and intensity for any one patient. The systems described herein may guide a user to find an appropriate activity balance, e.g., a balance between activity time and rest time for that particular user.

[0077] In some examples, processor 303 may compare the value of each input parameter in modified pain parameter log entry 317 with pain parameter log entries stored in memory 313. If processor 303 determines that the value of an input parameter in modified pain parameter log entry 317 is outside the range of the corresponding input parameter in the pain parameter log entry, processor 303 may cause an error message to be displayed to the user. Such an error message may be, "AI engine is expecting values ​​it is not trained on. You have never performed that activity!" Also, processor 303 may not pass modified pain parameter log entry 317 to AI processor 308 in determining calculated pain score 319. This is because the resulting calculated pain score 319 may be unreliable because the AI ​​engine has not been trained on such values.

[0078] 4A-D show example screenshots used to explain how the user interface of FIG. 3 can be used to provide calculated user parameters 416 to a user.

[0079] Referring to the first screenshot, the user may, in this example, use a slider to provide a target pain score 414 as an average pain level. The user also provides information about two of the input parameters in this implementation, which are then considered target input parameters 404. The target input parameters 404 in the first screenshot of FIG. 4A are a sleep parameter and a work parameter. These target input parameters 404 may be referred to as fixed activities. The user may also, or instead, provide information about any of the other input parameters 419, in which case the system moves those input parameters to a list of fixed activities and the input parameters are treated as target input parameters. In this way, the activities (input parameters) for which the user provided input are considered fixed activities.

[0080] Once the user is satisfied with the information they have entered, they can have the system calculate and display calculated user parameters 416, in this example by selecting the "Suggest Activities" button 418. The system performs the process described in detail with reference to Figure 3 to display the calculated user parameters 416 to the user, as shown in the second screenshot.

[0081] As also shown in the second screenshot, in this example, the user is provided with an "Explore Pain" button 423, which the user can press to explore how modifying one or more input parameters will affect the expected pain level. Optionally, the user may modify one or more of the calculated user parameters 416 (and, in some examples, one or more of the target input parameters 404) before pressing the "Explore Pain" button 423. For example, the user may interact with a duration value or slider shown in the second screenshot to adjust a characteristic associated with one or more of the input parameters.

[0082] After the user selects the "Investigate Pain" button 423, the user is presented with a third screenshot. This screenshot may be used to adjust the level of any of the input parameters 420 (here, both the calculated user parameters and the target input parameters). After adjusting the input parameters 420, the user may select the "Pain Suggestion" button 424 to determine a new calculated pain score. The fourth screenshot in FIG. 4D shows the calculated pain score 421 along with the associated input parameters 420. In this example, the user extended the duration of the break from 0.45 hours to 5.45 hours, thereby reducing the user's expected pain level from a level of 5 (shown as the target pain score 414) to a level of 4 (shown as the calculated pain score 421).

[0083] In this manner, the system may present the user, via the user interface, with an option ("Explore Activities" button 423) to modify one or more of the calculated user parameters 416. Upon receiving one or more modified calculated user parameters (representing user input) from the user interface, the system may apply a neural network trained for the individual user to the modified calculated user parameters to determine a modified calculated pain score and present the modified calculated pain score using the user interface. Advantageously, this may provide the user with the ability to fine-tune the set of suggested daily activities to better meet the user's requirements, yet not be expected to exceed the user's target pain score.

[0084] In some examples, the system may automatically determine and present the calculated user parameters 416 to the user in response to one or more predetermined triggers. For example, immediately after the user inputs information about their sleep parameters (which may be expected to be immediately after the user wakes up in the morning), the system may determine and present the calculated user parameters 416 according to one or more settings (as discussed above). In this way, the system may proactively suggest daily logs for the user intended to change the user's behavior according to one or more settings.

[0085] Figure 5 shows an exemplary embodiment of how a trained pain management system 500 can be used to help predict expected pain levels based on a user's planned activities. Features of Figure 5 that are also shown in either Figure 1 or Figure 3 are given corresponding reference numerals in the 500 series, and these features will not necessarily be described again in detail here.

[0086] In this example, the user provides target user parameters 520 to the user interface 502. The target user parameters 520 represent a set of activities that the user wishes to perform in a day and for which the user wishes to have a prediction of their expected pain level.

[0087] The system 500 may then provide a calculated pain score 522 based on the target user parameters 520. This may be accomplished by transmitting the target user parameters 520 to the AI ​​processor 508 via the user interface 502. The AI ​​processor 508 then applies the weighting value W 510 to the target user parameters 520 to determine the calculated pain score 522.

[0088] It will be appreciated that the functionality described with reference to Figure 5 is similar to some of the functionality described with reference to Figures 4A-D in that it provides the user with the opportunity to modify target user parameters 520 and see how that modification affects the calculated pain score 522. Thus, the user may plan their day to achieve an appropriate balance between being able to perform the activities they desire and reducing the likelihood that they will exceed what they have set as their maximum pain level.

[0089] Figure 6 shows an overview of functionality that may be implemented by any of the AI ​​processors described herein. Figure 6 shows a schematic of a neural network having an input layer, in this example one hidden layer (although it will be understood that there may be any number of hidden layers), and an output layer. The output layer may provide a prognosis (in this example, a calculated pain score).

[0090] Figure 7 shows an example screenshot that may be displayed to a user via any of the user interfaces disclosed herein. Figure 7 shows how a work parameter 726 may be expanded by a user to show historical information about the work parameter 726 along with an associated pain score. This historical information may be provided to the user interface from a memory that stores pain parameter log entries, such as the memory shown in Figure 1. In this manner, a user may view the stored historical information to look for correlations between particular activities and increases or decreases in pain.

[0091] Figure 8 shows another exemplary embodiment of how a pain management system 800 may be trained to develop a personalized pain management model for a user. Features of Figure 8 that are also shown in Figure 1 have been given corresponding reference numerals in the 800 series and will not necessarily be described again in detail here.

[0092] 1, the system 800 of FIG. 8 includes a UI controller 826. As discussed in detail below, the UI controller 826 is used to modify the functionality of the user interface 802 over time so that the user is presented with additional mechanisms for providing user input parameters 804.

[0093] Initially, the user interface 802 may provide the user with a display screen that allows the user to input user input parameters 804 using sliders in the same manner as shown in Figures 2A-C. This may be referred to as a first user input mechanism. The UI controller 826 provides UI control signals to the user interface 802 that instruct the user interface 802 to activate the first user input mechanism so that the user can input the user input parameters 804 using the first user input mechanism.

[0094] After the user has entered user input parameters 804 for a predetermined period of time (e.g., a predetermined number of days), or after a predetermined period of time, UI controller 826 may provide UI control signals to user interface 802 that instruct user interface 802 to additionally or alternatively activate further user input mechanisms.

[0095] 9A-C show three example screenshots used to illustrate different examples of user input mechanisms.

[0096] The first screenshot in FIG. 9A is the same as the second screenshot in FIG. 2B and illustrates how a user may use sliders to enter information for user input parameters. As noted above, this may be referred to as the first user input mechanism. The first screenshot also illustrates a “Timer” button 928 associated with each of the user input parameters. Selecting the “Timer” button 928 may allow the user to access additional input mechanisms. In this example, the “Timer” button 928 may be deactivated (so that it cannot be selected by the user) until the user has provided user input parameters for at least a predetermined number of days (e.g., at least three, four, or seven days). Such deactivation may be implemented according to a UI control signal received from the UI controller 826. In this manner, the system 800 may ensure that the user is satisfied with providing information using the first user input mechanism before additional mechanisms are made available to the user. This may enable improved ongoing use of the user interface, as the user may be able to reliably and accurately provide the necessary information to the system so that the pain management model can be effectively trained.

[0097] The second screenshot of FIG. 9B shows a display that may be presented to a user after the "Timer" button 928 is selected. The display in the second screenshot is an example of an additional user input mechanism. In this example, the timer may count down from a target time, which is set by the user as the goal in this implementation. The user interface 802 may then convert the target time into the duration characteristic of the associated user input parameter when the timer counts down to zero.

[0098] The third screenshot of FIG. 9C shows an alternative display that may be presented to the user after the "Timer" button 928 is selected. The display of the third screenshot is another example of a further user input mechanism. The display of the third screenshot may also be accessed by pressing the "Stopwatch" button 930 shown in the second screenshot. In this example, the stopwatch may count up from zero to measure the duration of the associated activity. After the user stops and saves the stopwatch, the user interface may convert the stopwatch duration into the duration characteristic of the associated user input parameter.

[0099] A further example of a further user input mechanism may be accessed by the user selecting a “Guide” button 932. As shown in the first screenshot, there may be a “Guide” button 932 associated with each of the user input parameters. Alternatively, there may be a single “Guide” button that applies to all or a subset of the user input parameters. When the user selects the “Guide” button 932, the user interface may pre-populate information about the associated user input parameter (or parameters) based on historical information. For example, the system may extract historical information about the user input parameter (or parameters) from memory (which may be stored as part of the pain parameter log entries, as discussed above) and then perform statistical operations on these historical user input parameters to calculate guide input parameter values. One example of a simple statistical operation is an average operation, such as an average. Another example of a statistical operation is a mode in which the most frequently provided user input parameter values ​​are provided to the user as a guide. The user may log the guide input parameter values ​​as is if they do not need to be changed, or they may modify the guide input parameter values ​​before sending them to the AI ​​processor 808 for further processing.

[0100] Figure 10 illustrates another exemplary embodiment of a pain management system 1000 according to the present disclosure. The pain management system 1000 may be used to train a personalized pain management model for a user in a manner similar to the system of Figure 1 and / or to help a user predict expected pain levels based on planned activities in a manner similar to the system of Figure 5. Features of Figure 10 that are also illustrated in the previous figures are given corresponding reference numerals in the 1000 series.

[0101] 10 includes one or more sensors 1036 that can provide one or more sensed input parameters 1034 (directly or indirectly after preprocessing 1038) to the user interface 1002. The user interface 1002 can then process the sensed input parameters 1034 in the same manner as the user input parameters 1004 in any of the examples described herein. In some examples, as described herein, the preprocessing 1038 can include known algorithms that classify activities based on the sensor signals. For example, algorithms are known that can identify sleep, rest, and physical activity from the sensor signals and, therefore, determine at least the duration associated with those activities to provide to the user interface 1002 as the sensed input parameters 1034.

[0102] The sensed input parameters 1034 may be used as additional or alternative inputs by the AI ​​processor 1008 to train a pain management model together with the user-input pain score 1006 and / or use the trained model to determine a calculated pain score (not shown in FIG. 10) or calculated user parameters (also not shown).

[0103] Advantageously, the sensors 1036 may be used to automatically gather information about the input parameters, such that the user does not have to manually enter the information themselves. In this manner, the sensed input parameters 1034 may be considered a subset of the user input parameters 1004.

[0104] The sensors 1036 may be provided as part of a wearable device (such as a smartwatch or fitness / activity tracker) or a user's smartphone. Such devices are known to be able to provide information related to various activities, such as a user's sleep, physical activity, and rest. This activity information (at least some of which may be considered pre-classified) may be provided to the user interface as sensory input parameters 1034 (possibly after pre-processing operations 1038). In some examples, upon receiving the sensory input parameters 1034, the user interface 1002 may automatically push a message to the user (e.g., using the user interface 1002) requesting the user to provide information to input any aspects of the activity that are not provided as part of the sensory input parameters 1034. As an example, if the sensory input parameters 1034 relate to physical activity parameters, then the sensors 1036 may provide a duration aspect for the activity but not a satisfaction aspect. In that case, the user interface 1002 may provide the user with an opportunity to manually input the satisfaction aspect as part of the user input parameters 1004.

[0105] Sensors 1036 may also be used to provide sensed input parameters 1034 related to characteristics of the user, such as one or more measured or determined attributes of the user's body. By way of non-limiting example, these may include a heart rate parameter, a blood pressure parameter, a temperature parameter, etc.

[0106] In some examples, the sensed input parameters 1034 may include a timestamp. The timestamp may be associated with a start time of an activity / measured characteristic and / or an end time of an activity / measured characteristic. Optionally, the timestamp may be stored in memory 1013 as part of a pain parameter log entry. In this manner, the stored timestamp may be used as part of the process of identifying pain triggers and protectors described above. Furthermore, in some examples, the timestamp may be part of the input parameters provided as input to the ANN. In this manner, the timestamp may influence the training of the pain management model and thereby also be included in subsequent processing to determine the pain score or calculated user parameters calculated using the trained model.

[0107] Additionally, in some examples, the AI ​​processor 1008 may receive one or more environmental input parameters (not shown) that may potentially affect how the user experiences pain. For example, the environmental input parameters may represent one or more of the environmental air temperature, weather conditions, altitude, and the user's location. Such environmental input parameters may be provided to the AI ​​processor 1008 by an appropriate sensor. In some examples, one or more of the environmental input parameters may be retrieved from an online web service; for example, a GPS sensor on a device associated with the user (such as the user's smartphone) may provide the user's location to the web service, which may provide one or more environmental input parameters (weather, body temperature, etc.) to the user interface 1002 (or directly to the AI ​​processor 1008) based on the location. Again, the user interface 1002 / AI processor 1008 may process such environmental input parameters in the same manner as the user input parameters 1004 in any of the examples described herein.

[0108] 11 illustrates a computer-implemented method according to the present disclosure. The method of FIG. 11 generally corresponds to at least some of the functionality described above with reference to FIGS.

[0109] In step 1150, the method includes receiving a target pain score representing a level of pain that the user deems tolerable for a specified period of time. In step 1152, the method receives one or more target input parameters representing the user's characteristics and / or activities for the same specified period of time. As described above, the one or more target input parameters are a subset of the complete list of available input parameters. In step 1154, the method receives one or more user settings representing one or more input parameters that the user aims to increase or decrease. It will be understood that the order in which steps 1150, 1152, and 1154 are performed is not important. In fact, one or more of these steps may be performed simultaneously.

[0110] In step 1156, the method continues by determining one or more calculated user parameters based on the target pain score and the target input parameters using the neural network trained for the individual user. At least one of the calculated user parameters is set based on the user settings. Then, in step 1158, the method includes presenting the one or more calculated user parameters using a user interface. Presenting the calculated user parameters in this manner may enable the user to change their behavior / activity to improve their health and well-being, while simultaneously controlling the amount of pain they experience.

[0111] In other examples, the general principles discussed in detail herein may be applied to applications not necessarily related to pain. Such systems may be referred to as personal management systems, in that they may be trained to be customized for an individual. In such examples, a target score (which is a more general version of the target pain score described above) may be received that represents the score / level of a personal attribute that the user deems acceptable for a specified period of time. Non-limiting examples of personal attributes may include stress, anxiety, depression, burnout, fatigue, quality of life, long-term effects of COVID-19 / COVID-19 (or any other disease or condition), mental health well-being, and personal performance (such as one or more specific activities, including, but not limited to, elite sports).

[0112] Clinical trials Introduction Chronic pain is a major global health issue, affecting more than 100 million individuals in the United States (Pitcher, MH, et al. (2019) Prevalence and profile of high-impact chronic pain in the United States. J. Pain 20, 146-160), and even current treatments are inadequate for the majority of patients due to lack of efficacy or significant side effects. PainDrainer™ is an artificial intelligence-driven digital pain self-management coach designed to improve quality of life (QoL). (The PainDrainer™ tool corresponds to the system and method described above.) This tool utilizes the concept of Acceptance and Commitment Therapy, which is considered to constitute a fundamental element of evidence-based treatment for chronic pain (Twohig MP (2012) Introduction: The Basics of Acceptance and Commitment Therapy, Cognitive and Behavioral Practice 19, 499-507). This study examines key components, such as patient acceptance and improvement of QoL.

[0113] method A single-arm, open-label study was conducted at UC San Diego Health's Koman Family Outpatient Pavilion. Fifteen eligible patients (67% women) suffering from neck, shoulder, and / or low back pain were included after signing informed consent. The pilot study was conducted in two phases representing different user experiences: nine patients in the first phase and six patients in the second phase self-managed pain using PainDrainer™. The PROMIS Pain Interference 6a validated questionnaire was used to measure changes in pain interference / quality of life and pain intensity. T-scores were calculated from the PROMIS questionnaire. Statistical significance (p-values) between T-scores was then estimated using a one-tailed, paired T-test. We also compared the differences in T scores with the minimally important difference (MID) observed in pain management (Chen CX, et al., (2018) Estimating minimally important differences for the PROMIS pain interference scales: results from 3 randomized clinical trials, Pain 159, 775-782).

[0114] result PainDrainer™ was developed and evaluated in this study as a tool for pain self-management in collaboration with healthcare providers, pain specialists, and artificial intelligence experts. The frequency of patients who showed a positive response in the first phase was 56% (5 / 9 patients), and in the second phase, it was 83% (5 / 6 patients). The strength of the response was analyzed using a one-tailed, paired t-test, and statistical differences between pre- and post-treatment t-scores were calculated. For the first phase, the p-value was 0.0086, and for the second phase, the p-value was 0.0014. The primary outcome—patients who experienced an improvement in QoL (10 / 15)—is shown in Figure 12 (improvement in QoL is indicated by reference number 1260, deterioration in QoL is indicated by reference number 1262, and no change is indicated by reference number 1264). The MID refers to the smallest significant difference in T-score that affects patients and typically ranges from 2 to 34. The difference in T-scores after treatment with PainDrainer™ exceeded the MID and was 3.0 and 4.8 in the first and second phases, respectively (FIG. 13). Secondary outcomes - Patients (10 / 15) in the two phases who experienced a decrease in pain intensity (PI) are shown in FIG. 14 (decreased pain is indicated by reference number 1470, increased pain is indicated by reference number 1472, and no change is indicated by reference number 1474). The mean decrease in pain intensity was 1.6 units (range 1-4 units).

[0115] conclusion PainDrainer™ is the first truly patient-centric device, as it is equipped with AI that adapts to each patient's needs. PainDrainer™ achieves clinically significant improvements in quality of life and reductions in pain intensity for chronic pain patients. PainDrainer™ enables chronic pain sufferers to better manage the relationship between daily activities and their pain.

Claims

1. 1. A pain management system that alters a user's behavior in a manner that improves the user's health and well-being, comprising: receiving a target pain score (314) representing a level of pain that the user deems tolerable for a specified period of time; receiving one or more target input parameters (304) representative of characteristics and / or activities of the user during the same defined period of time, wherein the one or more target input parameters (304) are a subset of a complete list of available input parameters; receiving one or more settings (315) representing one or more input parameters to be increased or decreased, wherein the one or more input parameters represent one or more physical activities performed by the user; determining one or more calculated user parameters (316) based on the target pain score (314) and the target input parameters (304) using a neural network trained for the individual user, wherein at least one of the calculated user parameters (316) is set based on the settings (315), and the calculated user parameters (316) include suggestions for values ​​of the one or more settings (315) that are expected to achieve the target pain score (314), wherein: The one or more calculated user parameters (316) may be: (i) a duration characteristic representing the duration of the physical activity performed by the user and associated with the setting (315); (ii) an intensity characteristic representing the intensity of the physical activity performed by the user and associated with the setting (315); Equipped with presenting the one or more calculated user parameters (316) using a user interface (102, 302, 502, 802, 1002) to enable the user to change their behavior / activity to improve their health and well-being while enabling them to control the amount of pain they experience.

2. The system comprises: examining a set of historical pain parameter log entries associated with the user to identify pain parameter log entries that match the received target pain score (314) and target input parameters (304) as matching pain parameter log entries, wherein each pain parameter log entry represents a plurality of user input parameters (104, 204, 804, 1004) and a user input pain score; modifying the matched pain parameter log entry based on the settings (315) to generate a modified pain parameter log entry; applying the neural network to the modified pain parameter log entries to determine a calculated pain score; 2. The system of claim 1, wherein the user interface (102, 302, 502, 802, 1002) is configured to: (i) present one or more user-input parameters of the modified pain parameter log entry as the calculated user parameters (316); and (ii) present the calculated pain score.

3. 3. The system of claim 2, wherein the system is configured to modify the matched pain parameter log entry by increasing or decreasing one or more user-input parameters (104, 204, 804, 1004) of the matched pain parameter log entry to generate the modified pain parameter log entry.

4. The system comprises: examining a set of historical pain parameter log entries associated with the user to identify pain parameter log entries that match the received target pain score (314) and target input parameters (304) as matching pain parameter log entries, wherein each pain parameter log entry represents a plurality of user input parameters (104, 204, 804, 1004) and a user input pain score; modifying the matched pain parameter log entry based on the settings (315) to generate a modified pain parameter log entry; applying the neural network to the modified pain parameter log entries to determine a calculated pain score; comparing the calculated pain score with the target pain score (314); If the calculated pain score is greater than the target pain score (314), adjusting one or more of the input parameters of the modified pain parameter log entry; applying the neural network to the adjusted modified pain parameter log entries to determine an adjusted calculated pain score; repeating the step of comparing the calculated pain score with the target pain score (314) one or more times for the adjusted calculated pain score; if the calculated pain score is less than or equal to the target pain score (314), using the user interface (102, 302, 502, 802, 1002): (i) present one or more user-input parameters of the modified pain parameter log entry as the calculated user parameters (316); and (ii) present the calculated pain score; 2. The system of claim 1, wherein the system is optionally configured to repeat the step of comparing the calculated pain score with the target pain score (314) multiple times, up to a predetermined maximum number of times.

5. The system comprises: presenting the user via the user interface (102, 302, 502, 802, 1002) with options to modify one or more of the calculated user parameters (316); receiving one or more modified calculated user parameters (316) representing user input from the user interface (102, 302, 502, 802, 1002); applying the neural network to the modified calculated user parameters (316) to determine a modified calculated pain score; 10. The system of claim 1, further configured to: present the calculated pain score after modification using the user interface.

6. A user interface (102, 302, 502, 802, 1002), a plurality of user-input parameters (104, 204, 804, 1004) representing characteristics and / or activities of the user during the defined period; and a user interface (102, 302, 502, 802, 1002) configured to receive a user-inputted pain score representing the degree of pain experienced by the user during the same defined period of time; An AI processor, 2. The system of claim 1, further comprising: an AI processor configured to set a plurality of weighting values ​​of the neural network based on the plurality of user-input parameters (104, 204, 804, 1004) and the user-input pain score of the user.

7. the system is configured to store the plurality of weighting values ​​associated with a user identifier, the user identifier being uniquely associated with a profile of an individual user; and / or The system comprises:

7. The system of claim 6, further configured to modify functionality of the user interface (102, 302, 502, 802, 1002) over time such that the user is presented with additional mechanisms for providing the user input parameters (104, 204, 804, 1004).

8. The user input parameters (104, 204, 804, 1004) and the target input parameters (304) (i) a duration characteristic representing the duration for which the user performed the activity and / or exhibited the characteristic; (ii) an intensity characteristic representing the intensity with which the user performed the activity; (iii) a satisfaction characteristic that describes the satisfaction experienced by the user when performing the activity; and (iv) a type attribute.

9. The AI ​​processor is configured to store pain parameter log entries in a memory, each pain parameter log entry comprising: the plurality of user input parameters (104, 204, 804, 1004); the user-entered pain score; and a date identifier associated with the corresponding user-input parameters (104, 204, 804, 1004) and user-input pain score; Optionally, The system of claim 6 , further comprising a user identifier uniquely associated with the individual user to which the plurality of user-input parameters (104, 204, 804, 1004) and the user-input pain score relate.

10. The system comprises: processing the plurality of pain parameter log entries to identify pain parameter log entries having an increasing user-input pain score as high pain parameter log entries, wherein each pain parameter log entry includes one or more user-input parameters (104, 204, 804, 1004) and a user-input pain score; performing an analysis of the user-input parameters (104, 204, 804, 1004) of the high pain parameter log entries to determine a correlation score representing the degree of correlation between values ​​of corresponding user-input parameters (104, 204, 804, 1004) in the high pain parameter log entries; 10. The system of claim 9, wherein the system is configured to determine a pain trigger by: if the one or more user-input parameters (104, 204, 804, 1004) have a correlation score that satisfies a correlation criterion, identifying the one or more user-input parameters (104, 204, 804, 1004) having a correlation score that satisfies a correlation criterion as a pain trigger.

11. The system comprises: processing the plurality of pain parameter log entries to identify pain parameter log entries having a decreasing user-input pain score as low pain parameter log entries, wherein each pain parameter log entry includes one or more user-input parameters (104, 204, 804, 1004) and a user-input pain score; performing an analysis of the user-input parameters (104, 204, 804, 1004) of the low pain parameter log entries to determine a correlation score representing the degree of correlation between values ​​of corresponding user-input parameters (104, 204, 804, 1004) in the low pain parameter log entries; 10. The system of claim 9, wherein the system is configured to determine a pain protector by: if one or more user input parameters have a correlation score that satisfies a correlation criterion, identifying the one or more user input parameters having a correlation score that satisfies a correlation criterion as a pain protector.

12. The system of claim 6 , wherein the plurality of user input parameters (104, 204, 804, 1004) comprises one or more sensed input parameters, the sensed input parameters being provided directly or indirectly from a sensor.

13. The user input parameters (104, 204, 804, 1004) sleep parameters, Work parameters, physical activity parameters, Housework parameters, Entertainment parameters, Rest parameters, and pain range parameters, Heart rate parameters, blood pressure parameters, temperature parameters, Energy levels / fatigue parameters, and / or A stress level parameter.

14. 1. A computer-implemented method comprising: receiving a target pain score (314) representing a level of pain that the user deems tolerable for a specified period of time; receiving one or more target input parameters (304) representative of characteristics and / or activities of the user during the same defined period of time, wherein the one or more target input parameters (304) are a subset of a complete list of available input parameters; receiving one or more settings (315) representing one or more input parameters that the user aims to increase or decrease, wherein the one or more input parameters represent one or more physical activities performed by the user; determining one or more calculated user parameters (316) based on the target pain score (314) and the target input parameters (304) using a neural network trained for the individual user, wherein at least one of the calculated user parameters (316) is set based on the settings (315), and the calculated user parameters (316) include suggestions for values ​​of the one or more settings (315) that are expected to achieve the target pain score (314), wherein: The one or more calculated user parameters (316) may be: (i) a duration characteristic representing the duration of the physical activity performed by the user and associated with the setting (315); (ii) an intensity characteristic representing the intensity of the physical activity performed by the user and associated with the setting (315); Equipped with and using a user interface (102, 302, 502, 802, 1002) to present one or more of the calculated user parameters (316) and enable the user to change their behavior / activity to improve their health and well-being while enabling them to control the amount of pain they experience.

15. A computer program configured to cause a computer to perform the method of claim 14.

Images