Improving the usability and satisfaction of hearing aids

Abstract

The invention relates to a method of improving the usability of a hearing aid and the satisfaction with the hearing aid. Further provided is a system comprising a hearing aid, wherein the system is configured to perform the method.

Description

Technical Field

[0001] This invention relates to improving the experience of wearing a hearing aid. A method for improving the usability and satisfaction with a hearing aid is provided. Further, a data processing system for analyzing satisfaction with a hearing aid according to the method is provided, as well as a hearing aid including at least a portion of the data processing system. Background Technology

[0002] Using hearing aids can cause discomfort or irritation due to factors such as the hearing aid's functionality and / or feel. For example, changes in auditory input caused by compensation algorithms can lead to discomfort. The result is a lack of satisfaction, which may ultimately lead users to return the hearing aids to the manufacturer.

[0003] Hearing aid returns are unpleasant for people with hearing loss, hearing healthcare professionals, and hearing device manufacturers. For users, time is spent choosing a hearing aid, having one or more fitting sessions with a hearing healthcare professional, and wearing the hearing aid, yet they are not entirely satisfied. Hearing healthcare professionals spend time helping users and fitting hearing aids. Furthermore, processing returns consumes time that could have been spent on other users. For manufacturers, the time and resources spent replacing hearing device components often translate into a higher initial cost for all hearing devices.

[0004] In some cases, hearing aids don't need to be returned because adjustments can improve the user experience. However, users may overlook seeking available help to resolve problems with their hearing aids. Instead, some users will frequently try to solve or overcome the problems they encounter. Summary of the Invention

[0005] Some embodiments are intended to address, mitigate, or eliminate at least some of the above or other disadvantages.

[0006] In a first aspect, a method is provided to improve the usability of hearing aids and satisfaction with hearing aids. The method includes the following steps:

[0007] - Data is obtained from the user's hearing aid.

[0008] -Based at least in part on the data obtained, determine a predictive score indicating the likelihood of user dissatisfaction with the hearing aid, and

[0009] - If the predicted score indicates that the user is dissatisfied, coping measures are implemented, which may include adjusting hearing aid functions, arranging human support, or a combination thereof.

[0010] Hearing aids that collect data from users typically use a portion of that data, for example, to adapt it to the user, or they may transmit the data to a processing unit outside the hearing aid. As part of the hearing aid's functionality, it includes compensation algorithms that are used to compensate for the user's hearing loss.

[0011] A predicted score is an indicator of whether a user is likely to be satisfied or dissatisfied with their hearing aid, and can be the result of a predictive model that can be built using past data.

[0012] A response is an action taken in response to a predicted score indicating user dissatisfaction, such as a predicted score exceeding a predetermined value. Implementing a response may involve arranging the implementation of specific response measures.

[0013] The data obtained, which determines the predicted scores at least in part, may include at least one of the following:

[0014] -Usage time,

[0015] - Number of times the preset / program changes,

[0016] -Number of power outages

[0017] -Number of restarts,

[0018] -Number of battery charges

[0019] - The number of times the sound environment changes.

[0020] - Patterns of sound environment changes

[0021] - The time spent in a certain sound environment

[0022] -GPS location,

[0023] -temperature,

[0024] -Pulse, or

[0025] -Oxidation saturation.

[0026] Usage time refers to the amount of time a hearing aid is used within a predetermined period, such as the number of hours in a day. Users dissatisfied with their hearing aids may tend to use them more or less.

[0027] The number of times a preset / program is changed refers to the changes between presets / programs within a predetermined time period. Users dissatisfied with their hearing aids may change the program several times to try and find settings that make them more comfortable, or they may change the program less frequently.

[0028] The number of power outages is the number of times the hearing aid is turned off within a predetermined period, such as a day. Users dissatisfied with their hearing aids may turn them off more frequently or less often.

[0029] The number of restarts is the number of times the hearing aid is turned off and then turned back on shortly afterward within a predetermined time period, such as a day. Users dissatisfied with their hearing aids may try to reset them more frequently or less often.

[0030] The number of battery charges refers to the number of times the rechargeable battery in the hearing aid is partially or fully charged within a predetermined time period.

[0031] If a hearing aid can detect the sound environment—such as whether it's noisy or quiet, indoor or outdoor, or whether it's a cocktail party setting or a quiet conversation setting—it can record the type and frequency of changes in the sound environment over a predetermined time period. Users dissatisfied with their hearing aids may constantly try to change their sound environment due to discomfort or inadequate function, or they may frequently switch to different types of noisy environments, such as quieter ones. Dissatisfied users may also spend more time in certain types of sound environments, such as those considered quiet.

[0032] GPS location can, for example, indicate whether a user is using their hearing aid in many different locations or in a few locations. If the hearing aid is equipped with one or more sensors, such as sensors for health monitoring, enabling it to measure one or more physical characteristics, such as temperature, pulse, or oxidation saturation, this sensor data may also be valuable in predicting user satisfaction.

[0033] Predicted scores can be determined, at least in part, based on a comparison of data recorded before the hearing aid was returned with data recorded before it was returned. In other words, data obtained during periods of user dissatisfaction can be compared with data from users who did not return their hearing aids. This allows for comparison between data from users who did not return their hearing aids and those who did, to determine parameters that help predict user satisfaction. Therefore, data recorded before and before the hearing aid was returned can be used to construct a model that forms part of the model that determines the predicted score. Users who eventually return their hearing aids may be dissatisfied with the hearing aid and its performance, and returning it indicates this dissatisfaction. Therefore, one or more of the user's motion or sensory data recorded by the hearing aid may reflect this dissatisfaction.

[0034] Patterns in data can be distinguished using artificial intelligence algorithms, such as machine learning systems. Machine learning models, such as neural networks sensitive to sequence information, like 1D ConvNets, can be trained to distinguish users who have returned their hearing aids from those who haven't by learning trends in the data parameters of those users. Therefore, the step of determining the predicted score can be performed at least in part using machine learning and / or artificial intelligence. For example, the step of determining the predicted score can be at least partially based on a model created using machine learning.

[0035] Alternatively, the predicted score may be further determined based at least in part on user-specific data. Examples of user-specific data include the type and / or model of the hearing aid, such as in-ear (ITE), behind-the-ear (BTE), in-ear receiver (RIE), and in-ear microphone and receiver (MaRIE), as well as demographic data such as age, sex, socioeconomic status, hearing loss status, and user feedback ratings provided. Other examples of user-specific data include the number of times contacted with hearing healthcare professionals and linked applications, i.e., the usage time of applications linked to the user or hearing aid via Bluetooth or Wi-Fi.

[0036] Some or all of the user-specific data can be obtained remotely, such as from one or more databases or external devices. This remotely obtained user-specific data can link information to the hearing aid ID, and thus link it to data obtained from the hearing aid.

[0037] User feedback ratings are ratings provided by users based on their use of the hearing aids. For example, a rating might be given after a user remotely adjusts the hearing aid. User feedback ratings can be given on a scale of 1-3. These ratings can be provided by users through methods such as apps or websites.

[0038] After determining the predicted score that indicates the user is dissatisfied, response measures are initiated. These measures include adjustments, such as improving or modifying hearing aid functionality, or arranging human support. Response measures may include one or more actions.

[0039] For example, adjustments to hearing aid functions can be categorized into three types: adjusting fitting parameters, firmware updates, and switching operating modes. Adjusting fitting parameters (also known as algorithm parameters) is related to individual hearing loss and can be done by a hearing healthcare professional during hearing aid fitting or fine-tuned after the initial fitting. Firmware is the software that provides the hearing aid with general operating functions (hearing compensation, wireless communication, power control, etc.). Switching operating modes is done manually or automatically by the user, for example, based on the acoustic environment or EEG sensors. Operating modes are typically determined by the firmware and customized during fitting. However, aside from switching between different modes, the individual's hearing compensation parameters generally do not change during hearing aid operation, i.e., when the hearing aid is being used normally.

[0040] For example, if the response includes adjusting the hearing aid function, the response may include one or more of the following:

[0041] - Reinstalling software on the hearing aid, such as rewriting the firmware.

[0042] - Update the software on your hearing aid.

[0043] - Change one or more algorithm parameters, i.e., compensate for the algorithm parameters.

[0044] - Perform remote automatic fine-tuning of the hearing aid, and / or

[0045] - Update one or more presets / programs on your hearing aid.

[0046] Remote automatic fine-tuning involves sending a data packet containing new settings to the hearing aid, such as adjusting the gain curve or the number of presets / programs.

[0047] The programs on a hearing aid are predefined settings that a user can turn on or off, such as settings optimized for speech in a restaurant-like sound environment. Programs are also called presets. Typically, a hearing aid comes with a set of presets / programs.

[0048] However, if the response includes arranging human support, that support may include one or more of the following:

[0049] -Notify hearing aid users

[0050] - Notify hearing healthcare professionals, and / or

[0051] - Notify customer service staff.

[0052] When the response includes notifying the hearing aid user, the notification may be performed directly through (one or more) hearing aids and / or through one or more intermediate devices that provide services consisting of one or more of a combination of acoustic or visual signals, such as through an application / software and / or via text or email.

[0053] If the response includes notifying hearing healthcare professionals or customer service staff, the notification can be performed through at least one intermediary device that provides services consisting of one or more of an acoustic or visual signal, such as via an application / software and / or via text or email.

[0054] Intermediate devices can be computers, PDAs, mobile phones, etc.

[0055] The choice of which response to take can be based, at least in part, on at least a portion of the data obtained from the hearing aid. In other words, whether the response is, for example, reinstalling the software, updating, or arranging human support, can be selected to some extent based on one or more parameters in the obtained data.

[0056] Alternatively or additionally, responses may be selected, at least in part, based on the similarities between the obtained data or user-specific data and one or more data of the same type from one or more other hearing aid users. Specifically, this may be if one or more other hearing aid users are among those who have not returned their hearing aids. For example, a similar hearing loss profile of a hearing aid user matching one or more other users could lead to the selection of responses where one or more presets / programs on the hearing aid are updated to the settings used by one or more other hearing aid users.

[0057] The data processing system that determines the predicted score can access a cloud-based user profile database, where user-specific data such as hearing loss status are available. Another example could be a comparison of location-based information, i.e., inputs from GPS information, accelerometer readings, or specific meeting room information from a calendar. This could lead to the selection of responses where presets / programs for acoustic environments—that is, different types of sound environments known to the hearing aid—are updated to settings used by others in the same location.

[0058] The steps involved in acquiring data, determining predicted scores, and implementing responses can be fully automated, meaning they are performed without human intervention. Alternatively, one or more steps may involve human intervention. In the case of a fully automated approach, one or more steps can be optimized through human intervention, such as changing all or part of the inputs used in the machine learning model used to determine the predicted scores.

[0059] In a second aspect, a system including a hearing aid is provided, wherein the system is configured to perform the method according to the first aspect.

[0060] Additional features and advantages will become apparent from the following detailed description with reference to the accompanying drawings. Attached Figure Description

[0061] In the following, exemplary embodiments of the invention are described in more detail with reference to the accompanying drawings, wherein:

[0062] Figure 1 This is a flowchart based on an exemplary embodiment.

[0063] Figures 2 to 3 A graph showing data obtained from the user's hearing aids, and

[0064] Figures 4 to 6 A system including a hearing aid and configured to perform a method for improving the usability of the hearing aid and satisfaction with the hearing aid, according to an exemplary embodiment, is illustrated schematically.

[0065] List of reference numerals

[0066] 1 user

[0067] 3. Hearing aids

[0068] 5. Remote Server

[0069] 7 Data Transmission

[0070] 9. Data Processing System

[0071] 11 Database

[0072] 13 Wired or wireless communication

[0073] 15 External devices Detailed Implementation

[0074] Embodiments of the invention will now be described and illustrated more fully with reference to the accompanying drawings. However, the description of the invention herein may be implemented in many different forms and should not be construed as limiting it to the embodiments set forth herein. Those skilled in the art will understand that the drawings are schematic and simplified for clarity, and therefore only details essential for understanding the invention are shown, while other details are omitted. The same reference numerals always refer to the same elements. Therefore, it is not necessary to describe the same elements in detail for every drawing.

[0075] Figure 1 A flowchart illustrating an exemplary embodiment of a method for improving the availability and satisfaction with hearing aids is shown.

[0076] Modern hearing aids are complex electronic devices that can record various data, such as the time, manner, and location of hearing aid use, as well as any sensor data from onboard sensors. "Time" can include, but is not limited to, date, time since the last restart, time since the user's first activation, and usage time. "Management" can include, but is not limited to, whether the hearing aid is on or off, whether presets / programs are used or changed, whether specific parts of the hearing loss compensation software are active, such as ambient sound compensation, for example, focusing on a single speaker or conversation in a noisy environment (cocktail party effect). "Location" can include, but is not limited to, location, such as input based on specific meeting room information from GPS, accelerometer, or calendar, and also includes the type of sound environment the user is in. Sensor data can include, but is not limited to, temperature, pulse, and oxidation saturation. Data obtained from hearing aids can be used to analyze the user and their actions, thereby providing ways to improve hearing aid usability and user satisfaction.

[0077] For example, if the sound environment is used as a parameter, one can often see people's dissatisfaction with sound environments, such as which environments they linger in and which they try to avoid. For example, this might mean: increased time spent in quiet environments, decreased time spent in noisy environments, and / or decreased time spent in environments with both speech and noise.

[0078] exist Figure 1 In step S10, data is obtained from the user's hearing aid. If the data is to be analyzed on a data processing system external to the hearing aid, the data can be transmitted from the hearing aid to the data processing system, for example, via the Internet or wireless protocols such as Bluetooth, Wi-Fi, NFC, etc. The data processing system can also be included within the hearing aid and obtain data through a communication path within the hearing aid.

[0079] After acquiring data from the hearing aid, step S20 determines a predicted score indicating the likelihood of the user's dissatisfaction with the hearing aid, which is determined at least in part based on the acquired data. The predicted score is an indicator of whether the user is likely to be satisfied or dissatisfied with their hearing aid, and may be the result of a predictive model built based on past data.

[0080] If historical data is used for prediction, the likelihood of dissatisfaction can be given by the probability that a customer will return their device, and this probability can be indicated by a numerical value returned by a machine learning model. The machine learning model is trained on a training set from a data lake (i.e., a data repository) or a database, and it creates internal representations of users who return their hearing aids and those who do not, based on pre-determined interaction parameters. The model compares user behavior patterns with their trained internal representations and assigns probabilities based on the closeness of the comparison.

[0081] Using past data to build predictive models can be achieved, for example, by comparing data recorded over a period of time from hearing aids used by users who have returned their hearing aids with data recorded over a period of time from hearing aids used by users who have not returned their hearing aids. Differences and / or trends from a large amount of user-recorded data can be used to build models that form a specific component of the predicted score.

[0082] Machine learning models are designed for specific tasks, and their algorithms learn and improve with the input of new data. As more data is added, the model becomes more refined. The model can use raw data—data obtained directly from the hearing aid—or processed data. Data from the hearing aid can be processed in a variety of known ways, such that the processed data is used to determine the predicted score, rather than the raw data. For example, simple calculations such as adding or subtracting data can be performed on the raw data. As another example, raw data can be combined to obtain new types of data that are not obtained directly from the hearing aid but are generated using the raw data.

[0083] exist Figure 2 and Figure 3 The image shows an example of data that can be used to determine predicted scores (see below). Figure 2 and Figure 3 (Further description). Figure 2 The five types of interaction parameters shown and Figure 3 The parameters shown appear to exhibit high confidence in predicting whether a hearing aid user will return to using the device. An example could be monitoring sequence patterns in data obtained from the hearing aid, such as... Figure 2 The system uses one or more of the data types shown in ae and determines a predicted score based on the obtained data, where the predicted score then gives an indication of whether the user is likely to return the hearing aid, thus indicating the user's dissatisfaction.

[0084] A machine learning model was implemented using parameters such as usage time, number of volume changes, number of restarts, number of preset changes, and number of power outages. This model correctly identified users who returned their hearing aids in 77% of cases and those who did not return them in 70% of cases. The setup used average sequential data from several weeks prior to the return to the time of return, thus incorporating the dynamic behavior of the parameters.

[0085] Data obtained from hearing aids can be acquired over a period of time, such as a short to medium term, like during a 90-day trial period. It can also be acquired long after the initial use of the hearing aid to continuously ensure satisfaction. Even if a user is unable to return the hearing aid after months or years of use, monitoring the data from the hearing aid and the resulting predicted scores can still continue to provide an indication of user satisfaction. The data can also be collected over a very short timeframe, such as a week, a day, or even hours, minutes, or seconds, before being used to determine the predicted score.

[0086] Factors used to determine the predicted score can be simple numbers, such as usage time in hours, or more complex interactions between the user and the hearing aid, such as changing presets / programs or activating volume control within a specific time pattern. Such complex interactions facilitate analysis using machine learning methods, where patterns in the data are identified by artificial intelligence algorithms. Machine learning models, such as neural networks sensitive to sequence information, like 1DConvNets, can be trained to distinguish users who have returned their hearing aids from those who haven't by learning trends in the data parameters of those users. Therefore, the step of determining the predicted score can be performed at least in part using machine learning and / or artificial intelligence. For example, the step of determining the predicted score can be at least partially based on a model created using machine learning.

[0087] The predicted score can be a number, and its value can be compared to a predetermined threshold that separates an indication of satisfaction from an indication of dissatisfaction. For example, if the predicted score is higher than a predetermined value, the user can be classified as dissatisfied. The predicted score can alternatively be expressed in a more complex way than a single number, such as as a series of digits or as a combination of letters and numbers. Any marker that allows for a decision indicating whether the user is satisfied or dissatisfied can be used.

[0088] If the predicted score indicates that the user is dissatisfied, then in Figure 1 In step S30, a response is implemented. Responses will include adjusting hearing aid functionality or scheduling human support. The choice of which response to take may be based, at least in part, on some or all of the data obtained from the hearing aid. For example, if the user frequently changes the presets / programs, this may, along with other data, indicate user dissatisfaction with the program, and updating one or more presets / programs may be chosen as a response to try and improve the user experience. If the hearing aid volume frequently changes, this may again, along with other data, indicate that the hearing aid is not properly calibrated for the user's hearing loss, and an appropriate response may be to notify a hearing healthcare professional so that a new calibration can be performed.

[0089] In this way, the data collected from users and their interactions with hearing aids provides a data-driven approach to predict whether users are dissatisfied with their hearing aids, thus allowing for steps to be taken to improve hearing aid availability and satisfaction without having to directly contact users to find out if they are satisfied with their hearing aids.

[0090] Figure 2 A graph shows the average sequence data of five parameters from the 12 weeks prior to the last data recording before hearing aid return, compared to the same type of data from non-returned hearing aids. These five parameters are (a) usage time [h]; (b) number of preset / program changes; (c) number of power outages; (d) percentage of users with at least one volume change; and (e) number of restarts, all of which are functions of weeks. This data is based on 4000 non-returned hearing aids and 2000 returned hearing aids. For Figure 2 All parameters in AE can determine the trend of returned and non-returned cases, thus providing the possibility of creating predictive models.

[0091] exist Figure 3 Another example of data is shown, which can be used to determine the predicted score. It shows the percentage (%) of hearing aids used in the 8 weeks prior to the last data recording before return, compared to the number of daily preset switches from the same type of data from non-returned hearing aids. The data shown is based on 2300 returns and 11000 non-returned hearing aids. In the graph, returned data is shown in black, while non-returned data is shown in gray. The results show that returned hearing aids had a higher number of daily preset switches compared to non-returned hearing aids. Using this data, a machine learning model was implemented that correctly identified users who returned hearing aids in 72% of cases and users who did not return hearing aids in 96% of cases.

[0092] Figure 4 A system including a hearing aid and configured to perform methods for improving the usability and satisfaction with the hearing aid, according to an exemplary embodiment, is schematically illustrated. A user 1 wears a hearing aid 3, which collects data about the user and user behavior, such as usage time, number of preset / program changes, number of power outages, number of restarts, number of battery charges, number of changes in the sound environment, patterns of sound environment changes, time spent in a certain type of sound environment, location, temperature, pulse, and oxidation saturation. This data can be used in various ways and can be used by a data processing system 9 configured to obtain data from the hearing aid, determine a predicted score, and perform countermeasures.

[0093] exist Figure 4In the illustrated embodiment, the data processing system 9 is included in the remote server 5, and the hearing aid 3 is configured to communicate with the remote server 5, enabling data transmission 7 between the hearing aid 3 and the remote server 5. The data transmission 7 between the hearing aid 3 and the remote server 5 can be performed by software, such as an application, running on an external device (e.g., a mobile phone).

[0094] Data processing system 9 acquires data and determines predicted scores via data transmission 7, based at least in part on the acquired data, but also in part on user-specific data. User-specific data may include, for example, the type of hearing aid, hearing aid model, age, gender, socioeconomic status, hearing loss status, provided user feedback ratings, number of contacts with hearing healthcare professionals, number of days since the last contact with a hearing healthcare professional, and usage time of linked applications. This type of user-specific data can be acquired remotely, i.e., from outside the hearing aid, such as from one or more databases or external devices. Figure 4 In the illustrated embodiment, user-specific data can be obtained from a remote server 5. This remotely obtained user-specific data can link information to a hearing aid ID, thereby linking it to data obtained from the hearing aid 3.

[0095] In addition, data generated during the testing and / or manufacturing of hearing aids can also be used to determine the predicted score. The predicted score indicates the likelihood of user dissatisfaction with the hearing aid, and if the predicted score indicates dissatisfaction, corresponding measures are taken.

[0096] Data transmission 7 can be performed regularly or intermittently. When using a predictive model based on past data, for example, by comparing data recorded over a period of time from the hearing aids of users who have returned their hearing aids with data recorded over a period of time from the hearing aids of users who have not returned their hearing aids to determine a predicted score, the predictive model can be updated continuously or periodically. The remote server 5 can connect to multiple hearing aid users, receiving data from them, allowing the predictive model to improve over time. The remote server 5 can include machine learning algorithms that analyze the data, for example, by looking for trends in data parameters between users who have returned their hearing aids and those who have not. Alternatively, the remote server 5 can connect to a system that includes machine learning algorithms.

[0097] Figure 5 Another system, including a hearing aid and configured to perform methods for improving the usability of the hearing aid and satisfaction with the hearing aid, is illustrated schematically according to other exemplary embodiments. Figure 4 and Figure 6 As shown, user 1 wears hearing aid 3, which collects data about the user and their behavior. Figure 5In the illustrated embodiment, the data processing system 9 is included in the hearing aid 3, and the data processing system 9 obtains data through a communication path within the hearing aid 3. Figure 5 In the illustrated embodiment, the data processing system 9 is configured to acquire data from the hearing aid 3, determine a predicted score based at least in part on the acquired data, and execute a response. Executing a response may mean that the hearing aid schedules the implementation of a response.

[0098] The predicted score can be the result of a predictive model based on past data, such as a model obtained by comparing data recorded over a period of time from the hearing aids of users who have returned their hearing aids with data recorded over a period of time from the hearing aids of users who have not returned their hearing aids. The data processing system 9 within the hearing aid 3 may include software that executes the predictive model. The predictive model can be updated regularly or periodically through software updates or through machine learning algorithms included in the data processing system 9.

[0099] To update software or machine learning algorithms, or to collect data from other hearing aid users, such as for creating predictive models, the hearing aid 3 may have a device for wired or wireless communication 13 with an external system, such as with a remote server 5 (e.g., Figure 4 (as shown) or wireless communication with an application running on an external device 15, wherein the external device can communicate with another system, such as a remote server 5.

[0100] Figure 6 Another system, including a hearing aid and configured to perform a method for improving the usability of the hearing aid and satisfaction with the hearing aid, is illustrated schematically according to other exemplary embodiments. Figure 4 and Figure 5 As shown, user 1 wears hearing aid 3, which collects data about the user and their behavior. Figure 6 In the illustrated embodiments, as in Figure 5 In this device, the data processing system 9 is included in the hearing aid 3 and the data processing system 9 obtains data through the communication path within the hearing aid 3.

[0101] In order to obtain data from other hearing aid users, hearing aid 3 has a means of wired or wireless communication 13 with an external system, such as wireless communication with a remote server 5. The remote server 5 has a database 11 that includes data from other hearing aid users, which the data processing system 9 can use when determining its predicted score.

[0102] Data from other hearing aid users can be categorized into, or can be categorized into, data from users who returned their hearing aids and data from users who did not return their hearing aids.

[0103] In all embodiments, if the predicted score indicates that the user is dissatisfied, countermeasures are performed, and these countermeasures include adjusting hearing aid functions or arranging human support.

[0104] For example, interventions to adjust hearing aid functions may include one or more of the following: reinstalling software on the hearing aid, updating the software on the hearing aid, changing one or more algorithm parameters, performing remote automatic fine-tuning of the hearing aid, and / or updating one or more presets / programs on the hearing aid.

[0105] Alternatively, the data processing system 9 may be partially included within the hearing aid 3 and partially located outside the hearing aid, for example, within a remote server 5, such that one or more of the method steps are performed by the circuitry within the hearing aid 3 and the remainder of the circuitry included outside the hearing aid.

[0106] If the data processing system 9 or a portion thereof is included in the remote server 5, it can perform one or more adjustments to the hearing aid function by pushing it to the hearing aid 3, or it can wait for a request. For example, the hearing aid 3 can periodically request updates and / or fine-tuning.

[0107] If the response involves arranging human support, it can include, for example, notifying the hearing aid user, notifying hearing healthcare professionals, and / or notifying customer service staff. For instance, the user 1 of the hearing aid 3 can be notified via an app or through a communication device included in the hearing aid 3.

Claims

1. A method for improving the usability and / or satisfaction with hearing aids, the method comprising: Data is obtained from the hearing aid; Based at least in part on the data, a predicted score is determined, which indicates the likelihood that the user of the hearing aid will be dissatisfied with the hearing aid, and If the predicted score indicates that the user is dissatisfied with the hearing aid, then coping measures are implemented, wherein the coping measures include adjusting the functionality of the hearing aid, arranging human support, or a combination thereof. The predicted score is based at least in part on data recorded before the hearing aid was returned, data from records that were not returned, or a combination thereof.

2. The method according to claim 1, wherein, The adjustment of the hearing aid's function includes one or more of the following: Reinstall the software on the hearing aid. Update the software on the hearing aid. Change one or more algorithm parameters, Perform remote automatic fine-tuning of the hearing aid, and / or Update one or more presets / programs on the hearing aid.

3. The method according to claim 1, wherein, The arrangement of human support includes one or more of the following: notifying the user of the hearing aid, notifying a hearing healthcare professional, notifying a customer service employee, or any combination thereof.

4. The method of claim 1, further comprising selecting a response measure before the step of performing the response measure is performed, wherein the response measure is selected based at least in part on the data from the hearing aid.

5. The method according to claim 1, wherein, The predicted score is based, at least in part, on a comparison between data recorded before the hearing aid was returned and data from records that were not returned.

6. The method according to claim 1, wherein, The steps of determining the predicted score are performed using machine learning and / or artificial intelligence, at least in part.

7. The method according to claim 1, wherein, The model is used to determine the predicted score.

8. The method according to claim 7, wherein, The model is constructed based on data recorded before the hearing aid was returned, data from records that were not returned, or a combination thereof.

9. The method according to claim 7, wherein, The model includes a neural network.

10. The method according to claim 1, wherein, The steps of obtaining the data, determining the predicted score, and implementing the response measures are performed automatically.

11. The method according to claim 1, wherein, The data obtained includes: usage time, number of preset / program changes, number of power outages, number of restarts, number of sound environment changes, sound environment change patterns, time spent in a certain type of sound environment, GPS location, temperature, pulse, oxidation saturation, or any combination thereof.

12. The method according to claim 1, wherein, The predicted scores are also based, at least in part, on user data.

13. The method according to claim 12, wherein, The user data includes: the type of hearing aid, the model of the hearing aid, age, gender, socioeconomic status, hearing loss status, user rating, number of times contacted a hearing healthcare professional, number of days since the last contact with the hearing healthcare professional, application usage time, or any combination thereof.

14. The method according to claim 12 further includes remotely acquiring the user data.

15. The method of claim 1, further comprising selecting the response based on the similarity between the data and other data of one or more other hearing aid users.

16. The method of claim 15, further comprising determining the similarity between the data and other data of the one or more other hearing aid users.

17. The method of claim 1, further comprising selecting the response measures based on user data.

18. The method according to claim 17, wherein, The user data includes: the type of hearing aid, the model of the hearing aid, age, gender, socioeconomic status, hearing loss status, user rating, number of times contacted a hearing healthcare professional, number of days since the last contact with the hearing healthcare professional, application usage time, or any combination thereof.

19. The method of claim 17, wherein, The response is selected based on the similarity between the user data and other user data of one or more other hearing aid users.

20. A system configured to perform the method of claim 1, wherein, The system includes a hearing aid and a data processing system.

Images