Method for operating a hearing aid system
The hearing aid system uses AI to monitor user behavior and respond to critical life events, enabling early detection and management of cognitive and mental health issues, addressing the limitations of existing systems in recognizing and addressing age-related health changes.
Patent Information
- Application Number
- PCT/EP2023/073558
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-15
- Filing Date
- 2023-08-28
- Publication Date
- 2024-11-21
AI Technical Summary
Hearing aid systems do not effectively monitor user behavior and respond to critical life events, leading to late recognition of cognitive and mental health issues such as depression, which can worsen due to age-related hearing loss and other life changes, requiring costly and energy-intensive constant monitoring.
A hearing aid system that uses AI techniques, including machine learning, to continuously collect user data on behavior and critical life events, providing personalized recommendations and adjustments to hearing aid settings, and offering cognitive training and behavioral tips to manage health changes.
The system enables early detection of cognitive diseases and mental health issues, reducing treatment delays and costs by providing targeted interventions and adjustments to hearing aid settings, improving user management of health changes and reducing energy consumption.
Smart Images

Figure EP2023073558_21112024_PF_FP_ABST
Abstract
Description
[0001] Description
[0002] Method for operating a hearing aid system
[0003] The invention relates to a method for operating a hearing aid system and to a hearing aid system. The hearing aid system comprises a hearing aid, which is expediently a hearing aid device.
[0004] People suffering from hearing loss typically use a hearing aid. This device typically uses a microphone, i.e. an electromechanical transducer, to convert ambient sound into an electrical (audio / sound) signal, thus creating an electrical signal. The electrical signals are processed by an amplifier circuit and fed into the person's ear canal via another electromechanical transducer in the form of a receiver. The amplifier circuit is part of a control unit of the hearing aid. The sound signals are usually also processed, for which a signal processor in the amplifier circuit is typically used. The amplification is then adjusted to suit any hearing loss of the hearing aid wearer.
[0005] Hearing loss is usually age-related. However, older people often also have other medical conditions that can lead to a deterioration in their health. It is also possible for hearing loss to cause comorbidities, such as depression. This type of depression usually sets in gradually, which is why it is often only recognized relatively late by the person themselves and by others. By the time depression is recognized, it is usually well advanced, requiring relatively extensive treatment. Illnesses and mental disorders, such as depression, can be triggered by events that drastically change a person's life situation, referred to as dramatic or critical life events.Such a dramatic event could be, for example, a worsening of (existing) hearing loss, the loss of a loved one, the transition from work to retirement, moving to another apartment or a retirement home, etc.
[0006] Artificial intelligence (AI) techniques, ranging from machine learning to deep learning, are widely used in healthcare to diagnose diseases and identify patient risks. AI-based disease detection has proven to be a powerful tool for identifying undiagnosed patients with complex common and rare diseases. However, continuous monitoring for all possible diseases is energy- and cost-intensive, e.g., when many sensors need to be active.
[0007] The long-term monitoring or detection of cognitive disorders is known from patent applications US20220345836A1 , US20200273566A1 and EP4156717A1.
[0008] The invention is based on the object of specifying a particularly suitable method for operating a hearing aid system and a particularly suitable hearing aid system, wherein in particular comfort and / or functionality are increased.
[0009] With regard to the method, this object is achieved according to the invention by the features of claim 1 and with regard to the hearing aid system by the features of claim 9. Advantageous further developments and refinements are the subject of the respective subclaims. The method serves to operate a hearing aid system which has at least one hearing aid. The hearing aid is provided and configured to be worn on the human body. In other words, in its intended state, the hearing aid is worn by a wearer, who is also referred to as the user. The hearing aid preferably comprises a holding device by means of which it can be attached to the human body. For example, the hearing aid system comprises two hearing aids, one of the hearing aids being assigned in particular to a left ear of the wearer and the other to a right ear of the wearer, such that it is a binaural hearing aid system.For example, the two hearing aids are identical in design. Alternatively, they are different, but at least similar. The two hearing aids are designed as mirror images of each other, at least externally, so that they can be adapted to different ear shapes.
[0010] For example, the hearing aid is a pair of headphones or comprises a pair of headphones. These are designed, for example, as so-called in-ear, on-ear, or over-ear headphones. For example, the headphones are used to reproduce information and / or music, and / or the headphones are used to suppress background noise and are so-called noise-cancelling headphones. However, the hearing aid is particularly preferably a hearing aid. The hearing aid is used to support a person suffering from a reduced hearing ability. In other words, the hearing aid is a medical device used to compensate for, for example, partial hearing loss. The hearing aid is, for example, a receiver-in-the-canal hearing aid (RIC), an in-the-ear hearing aid, such as an in-the-ear hearing aid, an in-the-canal hearing aid (ITC), or a completely in-canal hearing aid (CIC).Alternatively, the hearing aid is a behind-the-ear hearing aid, which is worn behind an auricle. If the hearing aid is a hearing aid, the hearing aid is intended and configured to be placed, for example, behind the associated ear or within an auditory canal of the respective ear. The hearing aid preferably comprises a microphone used to capture sound. In particular, ambient sound, or at least a portion thereof, is captured during operation by means of the microphone. The microphone is, in particular, an electromechanical sound transducer. The microphone has, for example, only a single microphone unit or several microphone units that interact with one another.Each of the microphone units expediently has a membrane that is set into vibration by sound waves. The vibrations are converted into an electrical signal by a corresponding recording device, such as a magnet moved in a coil. Thus, it is possible to use the respective microphone unit to create an audio signal based on the sound impinging on the microphone unit. The microphone units are, in particular, unidirectional. The microphone is expediently arranged at least partially within a housing of the hearing aid and thus at least partially protected.
[0011] Each hearing aid suitably further comprises a control unit coupled to the respective microphone. The control unit expediently comprises a signal processing unit or is formed by it. The signal processing unit serves in particular to further process or at least analyze the audio signal(s) generated by the microphone. In particular, the control unit processes the audio signal to create an output signal that is modified compared to the audio signal. In particular, the control unit amplifies certain frequencies of the audio signal, preferably adapting to any hearing loss of the wearer. For example, the signal processing unit has several analog components. Expediently, the signal processing unit or at least the control unit comprises a digital sound processor (DSP).The (sound) processor is conveniently designed to be programmable.
[0012] The hearing aid preferably has a receiver, which serves in particular to output the respective output signal. The respective output signal is in particular an electrical signal. The receiver is expediently coupled to the control unit, in particular connected to it via signaling. Depending on the design of the hearing aid, in the intended state, the output device, for example the receiver, is arranged at least partially within the ear canal of the user of the hearing aid, i.e., a person, or is at least acoustically connected to it.
[0013] In particular, the hearing aid is wireless and designed and configured to be inserted at least partially into an ear canal. Particularly preferably, the hearing aid comprises an energy storage device, by means of which a power supply is provided. Preferably, the hearing aid has a communication device, which in particular comprises a radio system.
[0014] The method provides for the creation of information regarding the behavior of a hearing aid user. In other words, data is continuously collected in the background while the hearing aid / hearing aid system is in operation.
[0015] In addition, the method provides for receiving further information regarding a significant event for the user, also referred to as a critical life event (for the user), via an interface. In other words, the (hearing aid) system offers the option of entering critical life events, such as the occurrence of a specific incident or diagnostic data.
[0016] Depending on the information and the further information, a function of the hearing aid is determined, and the function is carried out. The information and / or the further information are used in particular to carry out the function. Two different cases can arise here. In other words, the method is used in particular in two different forms. In one case, the function is a pattern recognition method based on which a probability for the development of a (cognitive) illness is determined. In other words, a risk for the development of a (cognitive) illness of the user is preferably determined using the pattern recognition method, which expediently uses an “artificial intelligence” algorithm, preferably machine learning. If this risk is greater than a threshold value, a corresponding message or notification is expediently issued.
[0017] In the other case, once the dramatic event has occurred, for example, a specific incident has occurred or diagnostic data has been entered, the user receives a recommendation as a function on how to deal with it. The recommendation may include, for example, motivational aspects, cognitive training, advanced sensor monitoring, or adjusted hearing aid settings. Alternatively, or in combination with this, the recommendation includes behavioral tips / training, such as "CBT" ("cognitive behavioral therapy," i.e. various psychological interventions aimed at practicing positive behavior and bringing about a change in attitude / appraisal / cognition of the situation. The idea is to accept the chronic illness, in particular the hearing loss, and to practice behavioral strategies, e.g., the use of a hearing aid, in order to better cope with the situation.
[0018] Consequently, the invention provides for a data platform in which the information is stored. The data platform is, for example, local and a component of the hearing aid. Alternatively, the data platform is, for example, a server and / or cloud storage device, which thus form a component of the hearing aid system and are at least temporarily connected to the hearing aid via signaling.
[0019] According to the invention, the option of entering critical life events is also provided so that further information can be received. According to the invention, actions are also carried out, i.e. the respective function is performed. Preferably, the function is adapted to an estimated state of health of the user. Expediently, the function is determined according to the estimated state of health and / or the function is to estimate the state of health. Alternatively or in combination with this, the state of health is estimated based on the information. Preferably, the information is generated using a sensor in the hearing aid itself. For example, information about a gait is used as information. For example, this is determined using a motion sensor or gyroscope in the hearing aid. For example, information about head movements is used as information.In particular, the time elapsed between a direct response and a head movement is determined. For example, an environment is taken into account. Alternatively, or in combination with this, the time until a reaction to a specific loud event is used. Preferably, the information used is a statement about the time until a startle is triggered after a sudden loud noise. For example, the motion sensor, gyroscope, and / or microphone, if present, are used to determine this.
[0020] For example, information about participation in conversations and / or the user's voice and / or a listening environment is used. For example, the microphone, if present, is used to determine this. Suitably, the manner in which a conversation is conducted is used as information. For example, communication patterns are used for this. In particular, a so-called turn-taking parameter, back-channeling or the like is used for this. Suitably, the information used includes how often a participant in the conversation takes the floor and / or responds to requests / makes requests. For example, information about movement activity, a movement pattern and / or a fall is used as information. For example, the motion sensor, gyroscope and / or microphone, if present, is used to determine this.For example, information about how the user uses an app to control the hearing aid is used. The app is run in particular using a smartphone, which is in particular a component of the hearing aid system and is connected to the hearing aid, for example via a radio signal connection. In this case, it is suitably possible to use the app to select one of several hearing programs and / or to change hearing aid settings, such as in particular an amplification factor. For example, information about a change in settings and / or a change in personal preferences is used as information. These changes are made by the user, for example, if their hearing loss has worsened or their concentration span has shortened. For example, the respective change is made using the app.
[0021] For example, information about a pause in the middle of the movement and / or about a repetition of actions is used as information.
[0022] This is determined, for example, using the accelerometer or gyroscope. If the attention span is shortened or brain performance declines, it is possible that the user will perform an action multiple times or that the user may temporarily lose memory of what they are currently doing. Thus, the information corresponds at least partially to impaired brain performance. For example, information about the use of a personal assistant is used as information. The app and / or a hearing aid microphone, for example, are used to determine this.
[0023] In summary, the (hearing aid) system preferably collects user data essentially continuously, which expediently forms the information and / or from which, in particular, the information is derived or represents it. At a minimum, the user data preferably includes the type of walking (gait), head movements (time to respond when the user is spoken to in quiet; time to turn the head when the user is spoken to in quiet; time to startle after a sudden loud noise), participation in conversations, the user's own voice (voice biometer), the listening environment (e.g., speech in noise, dB level in a noisy environment, etc.).), movement activities or movement patterns, fall detections (in particular per time interval), the way in which the user uses any app to control the hearing aid (for example, the number of interactions per day and / or which settings are used and / or which functionalities are used most), the use of personal assistants (for example Signia Assistant, Google Assistant, Siri, etc.), setting changes by the user, changes in personal preferences (e.g.Use of listening programs, especially when the user stops using a previously frequently used program), pausing mid-movement (the user does this, for example, to remember the last activity), repeating actions more often than in the past, time to respond when spoken to in quiet, time to turn the head when spoken to in quiet, and / or time to startle after a sudden loud noise (when this is an indication of aging...).
[0024] In a further training, for example, information about a location is used as information. In other words, geolocation data (e.g., GPS) is preferably used as user data / information.
[0025] The user data, i.e. the details / information, is suitably generated mostly by the hearing aid and its sensors, but can also be obtained from networked devices such as smartwatches, smartphones or other devices that form a component of the hearing aid system, at least temporarily. It is also expedient for the details, i.e. the data, to be entered. Here, for example, a user account is used that is stored on a server. The data (information) is entered and saved by the user or a specialist, for example an audiologist. Alternatively, or in combination with this, the data (information) comes from third parties, e.g.
[0026] B. from relatives, digital patient data or other healthcare providers, etc. For data collection and transmission, a data transmission device such as a smartphone is expediently used, which is connected to a cloud-based data system via the user account. Such data / information is preferably categorized or tagged; alternatively, this is omitted, and the data can remain in "raw format" and be analyzed as needed. In summary, the information is thus received via another interface. The further interface is thus formed, for example, via the smartphone and / or the user account. For example, the information is created based on a single and / or short-term observation. However, it is particularly preferred that the information be created for a time window.This means that short-term changes, particularly those due to current influences and / or circumstances that are independent of the user's current state of health, are not taken into account. The information is expediently created for successive time windows, allowing a change to be detected. In particular, the information is created multiple times per time window, and preferably the average of the respective information for the time window is used. In other words, the sensor data and / or human interaction are expediently monitored over time. This is particularly helpful in understanding medium- to long-term developments and reducing misinterpretations of data due to normalization over time. The time windows, and in particular their length, are selected depending on what is to be analyzed.For example, it is intended that during the time windows the user should perform certain "tasks" / activities in order to increase the meaningfulness / comparability of the collected sensor data.
[0027] The interface through which the additional information is received is, for example, also the user account and / or a smartphone, through which the user or another person can enter the additional information. For example, the user, a relative, medical staff, caregiver, hearing aid acoustician, and / or the like has entered the additional information into a pre-programmed app. Alternatively, or in combination with this, the additional information was generated based on a speech or text message from the user, the relative, medical staff, caregiver, hearing aid acoustician, and / or the like using a speech-to-text assistant, which, in particular, interpreted the conversation / text and based on which the additional information was generated. A "generative AI" approach is expediently chosen for this purpose.Alternatively or in combination with this, the additional information was identified from the context of the significant events by means of an AI (“Artificial Intelligence”)-supported assistant and, in particular, the additional information was created.
[0028] Preferably, the decisive event is the diagnosis of an illness. The illness could be, for example, a cognitive illness or a psychological disorder / illness. Alternatively, or in combination with this, a move and / or the loss of a close person is used as a decisive event. The person could be, for example, a person, in particular a relative, or an animal. In particular, a presumption or suspicion or diagnostic data regarding the decisive event, in particular regarding the presence of the cognitive illness, is received via the interface. At the very least, the decisive event is expediently a significant personal change in the user's life. This could be, for example, a doctor's diagnosis of an advanced cognitive illness (e.g., Alzheimer's) or a significant change in private life, such as a recently deceased relative, etc.In one alternative, the life-changing event is positive or changes the wearer's life in a positive way. For example, the birth of a grandchild or something similar is used as a life-changing event. In other words, the life-changing event at least drastically changes the wearer's life.
[0029] For example, the (hearing aid) system, for example, offers a platform, for example the user account, for entering further information, for example by uploading a photo of a doctor's letter. Using suitable algorithms, such as a text recognition algorithm, the diagnosis is recognized. Alternatively, or in combination with this, it is provided that freely formulated content can be entered. In particular, the platform, preferably the user account, also provides the opportunity to track the further development of any illness or to establish contact with a specialist, therapist, or doctor. The platform / user account is expediently designed to encourage the user to enter private data.The user is particularly motivated to enter personal data if a service is offered that is only provided when needed, at a time / place where the service is appreciated, and where / when the user is able to "digest" and apply the solution. This also increases usage, leading to compliance with the solution. Furthermore, there is less burden, and the solution / service does not lead to frustration and additional disruption. The user thus feels valued because the hearing aid provides an additional service.
[0030] In continuing education, the interface serves to connect to other platforms such as health or fitness apps, communication platforms with other people or health insurance companies, or healthcare systems such as digital patient records. This allows caregivers, relatives, or medical professionals to enter data about the user into the platform, i.e., the user account, and thus, for example, provide or at least adapt further information.
[0031] Some examples of significant events include the user developing Parkinson's disease, Alzheimer's disease, diabetes, audiological disorders such as tinnitus, hyperacusis, or changes in the degree of hearing loss. Alternatively, or in combination with this, the significant event is another relevant comorbidity for the user, for example, (signs of) depression / mood swings, (signs of) dementia, memory / disorientation loss, but also changes in diet and physical activity that deviate from the "normal," i.e., in particular, an "ecological momentary intervention." Non-medical effects can also be relevant, e.g., if a relative or important person (e.g., a partner) has died, if the user has moved to another apartment or a retirement home, etc.In summary, the impact event is, in particular, a worsening of (pre-existing) hearing loss, the loss of a loved one, the transition from work to retirement, the user's move to another home or a retirement home, etc. In addition to the sensor data, geolocation data (e.g., GPS) and key observations from relatives can be provided via the interface as additional information, as these may identify the impact event. In this case, the impact event is, for example, a major local event, such as a disaster. Such third parties could also have the opportunity to enter data via a corresponding interface, e.g., via a "Help Angel" app on their phones, to support people in need.
[0032] When determining the function, a weighting factor for the additional information is preferably taken into account, which is selected depending on the person providing the additional information. This makes it useful to distinguish between a well-founded assumption or suspicion of a layperson and a diagnosis by a professional physician who created the additional information. This introduces a quality dimension that helps to weight inputs differently depending on who entered them or the type of input. For example, if a user moves to another city and a relative provides / enters a new address as additional information, this can be considered correct.However, if a relative provides further information that the user is developing cognitive diseases, such as dementia or Alzheimer's, this will only be considered a reasonable guess or suspicion, but not at the same level of seriousness as a qualified diagnosis from a doctor.
[0033] In summary, the information and / or additional information is conveniently collected in the following ways: via an app for the user or a caregiver, via a cloud-based hearing aid fitting (e.g., Fit XP), or a user account management system. Alternatively, or in combination with this, the information and / or additional information is collected via an interface of another program (e.g., Connexx, e.g., in a separate field such as "additional comments"), a digital patient record (particularly with GDPR consent), or the relevant data can be entered by the clinical expert. For this purpose, a medical expert conveniently has the option of "scanning" a QR code.
[0034] Depending on the information and / or additional information, preferably the input data, an information cluster is preferably created that describes the current data and the latest developments regarding the diagnosed disease based on the previously collected data. It may be important to identify and aggregate only relevant data for analysis to avoid creating unnecessary data collections and increasing power consumption on the cloud servers.
[0035] Using artificial intelligence (AI) methods, the system creates data clusters based on the information and / or additional information and identifies other users who are in similar problem situations or who could develop or have already developed the same disease. In other words, the probability of developing a (cognitive) disease is determined. As soon as the predicted value (probability) exceeds a predefined threshold, appropriate functions are performed. Thus, the function is determined depending on the information and the additional information, preferably using the artificial intelligence algorithm. In particular, the information / additional information that has been created regarding other users is already used when determining the function.The functions performed during these are also taken into account.
[0036] The artificial intelligence algorithm conveniently compares clusters and attempts to identify similar cases or patterns. The result is a pattern, a multidimensional vector, or a vector cloud that can be compared with the available information of other similar or relevant users, or even all users. If the similarity exceeds a certain threshold, i.e., the similarity between the two cases / users and their recent developments is relatively high, the function used in the similar cases is conveniently selected and thus determined / performed. The artificial intelligence algorithm includes, for example, machine learning or deep learning, "random forest," and / or a neural network (e.g., CNN).
[0037] When implementing the procedure, two cases (main scenarios or use cases) can arise. In one, the system uses pattern recognition methods with machine learning to identify and predict users at risk for developing a (cognitive) disease. In the other, the user who has experienced a significant event receives a recommendation on how to cope. This can include motivational aspects, cognitive training, advanced sensor monitoring, or customized hearing aid settings.
[0038] For example, an information output is used as a function. In other words, the function represents a notification. The user is notified, for example. Alternatively, relatives, caregivers or doctors can also be informed. Who should be informed is best determined by the hearing system wearer themselves or at least with their confirmation in advance. Since this is critical information, the thresholds defined above must be chosen very carefully. The thresholds are best chosen depending on the quality standard mentioned above, i.e. the difference between whether the data was entered by a private individual or by a professional, or the type of significant event (critical life event). In some cases, for example, the user is not informed directly, but rather another party, such as a doctor or professional.As an information output, it may also be suggested, for example, to switch to other "earpieces" such as an External Receiver Unit (ERU) with additional sensors that allow for more precise monitoring of the current situation. Such additional sensors could be PPG sensors, temperature sensors, EEG or EMG electrodes, additional microphones, sweat sensors, motion sensors, other optical sensors, etc. When the procedure is subsequently performed, it is thus possible to determine the information in a different way and / or with the other sensors, so that it is comparatively accurate. Thus, for example, the function is determined in a different way.
[0039] For example, a setting change, i.e., adjusting the hearing aid settings, is used as a function. This involves appropriately changing the hearing aid settings or preparing for the near-future occurrence of a significant event. There is also the potential to delay or prevent a specific health deterioration (e.g., hearing loss) through appropriate preventive measures such as information or preventative therapy settings (e.g., an attempt is made to protect the user from stress if this could lead to a certain physical reaction in the future, as has already happened in similar users).
[0040] In a further development, for example, the hearing aid includes an accelerometer, which requires less power, and a gyroscope, which requires more power from the battery. The default operating mode may be to monitor data via the accelerometer, and the information is generated specifically based on this. If data provided by a similar device indicates that additional information from the gyroscope would be beneficial to the user (for readjustment, verification, etc.), the gyroscope is activated. Thus, the function corresponds to activating the gyroscope.
[0041] Subsequently, the procedure is repeated and a new function is determined. Thus, the user's hearing aid functionality is modified, especially when the user's data set is analyzed "in light" of data / information donated by another (similar) user. On average, power consumption is lower because the gyroscope is only used (and draws power from the battery) when justified, particularly based on other donated data used by the artificial intelligence algorithm. This enhanced monitoring should be designed to use only sensors / data relevant for accurate and efficient monitoring of the assumption resulting from the data. This should avoid unnecessary data waste / energy consumption while maintaining a high level of quality and care.
[0042] If desired, the (hearing aid) system can support the user in managing their new condition. For example, the user can receive further background information to increase health literacy regarding the "discovered" problem and inform the user about treatment options. Furthermore, the user can learn new skills or apply and adapt existing skills to cope with the new condition. Depending on the identified problem, the system, preferably as a respective function, offers training programs to specifically delay further deterioration. For example, the system monitors how the user develops or how they utilize the training offers. This data is also usefully analyzed and shared with the relevant experts to improve their services.
[0043] Because the system continuously monitors changes in the user and the environment, it can only offer additional services if they are appreciated by the user. For example, if the suggestion, whose output represents the function in particular, includes breathing control exercises (to reduce stress), this will not be displayed if the user is in a heated discussion. However, if the user is alone at home, the suggestion will be displayed. For example, the suggestion includes specific brain exercises in the user's app to recognize cognitive patterns. A "partner" app for caregivers / relatives can request additional data / information via specific questionnaires (questionnaire for partners / caregivers via a "partner app"). In summary, the function is thus also determined depending on environmental parameters.
[0044] The system can be used to check whether other users have already been successfully treated by making adjustments or data collection modes using additional (expensive) sensors, which can also be applied to the user who has entered their diagnosis / reasonable suspicion into the system. This would provide further motivation to enter data into the system. A repeat diagnosis / needs assessment (audiology and beyond) or connection to support groups or therapists in the neighborhood is also suitable. In other words, a therapy recommendation is used as a function in particular.
[0045] In summary, the invention thus relates in particular to a (hearing aid) system for recording critical life events (persistent events) with AI-supported adjustment of hearing aids and early detection of diseases. According to the invention, data is expediently collected continuously in the background. The hearing aid system further offers the possibility of entering critical life events or diagnostic data in order to adjust hearing aids and accessories and / or predict a (cognitive) disease. The hearing aid system is also capable of assisting in the adjustment of other users' hearing aids and learning from similar "cases." Such a persistent event is, for example, a worsening of a (pre-existing) hearing loss, the loss of a loved one, the transition from work to retirement, the relocation of the HIW to another apartment or a retirement home, etc.
[0046] Constant monitoring for all possible diseases, which is energy- and cost-intensive because many sensors must be active, is unnecessary. If early signs are already detected with "baseline data," sensors can be deployed more specifically and only when needed or to monitor a reasonable suspicion, so that the information is conveniently generated more accurately, more specifically, and / or using other sensors.
[0047] In contrast to the disclosure of US 2022 / 0345836 A1, the invention provides a simple way to enter suspicions, reasonable assumptions about critical life events, or even diagnostic data. The data is used in particular for adjusting the hearing aid to the respective user who enters the data, but also to check whether other users could benefit from this knowledge, to adjust the settings of their hearing aids, and / or to warn them or their relatives, etc.
[0048] EP 4 156 717 A1 aims to detect predefined data indicating health problems, particularly for early detection. The invention specifically does not detect such health problems, but rather provides a system with which such data (information / further information) can be easily input. There is no need to measure or interpret measured data, which can have a high rate of misinterpretation.
[0049] The hearing aid system comprises a hearing aid, which is particularly designed as a hearing aid. The hearing aid system is operated according to a method in which information relating to a behavior of the hearing aid is generated. Further information relating to a significant event for the user is received via an interface. Depending on the information and the further information, a function of the hearing aid is determined, and the function is executed. To determine the function, a KL algorithm is used, in particular. In this case, a comparison is made, at least in part, with already existing information / further information assigned to other users. In particular, a so-called cluster analysis is carried out.
[0050] The further developments and advantages described in connection with the procedure can also be transferred to the hearing aid system and vice versa.
[0051] An embodiment of the invention is explained in more detail below with reference to a drawing. In the drawings:
[0052] Fig. 1 schematically shows a wearer of a hearing aid system comprising two hearing aids and a portable device,
[0053] Fig. 2 shows a schematic section of the hearing aid system, Fig. 3 shows a method for operating the hearing aid system, Fig. 4-6 shows further explanations of the method.
[0054] Corresponding parts are provided with the same reference numerals in all figures.
[0055] Figure 1 shows a simplified schematic representation of a user 2 of a hearing aid system 4. User 2 is a person suffering from partial hearing loss. To remedy this, user 2 wears two hearing aids 6 of the hearing aid system 4. One of the hearing aids 6 is assigned to the left ear of user 2, and the other hearing aid 6 to the right ear. Thus, the hearing aid system 4 is designed to be binaural. The hearing aid system 4 also has a portable device 8 in the form of a smartphone, which the user 2 also carries with them.
[0056] The hearing aid system 4 is shown in Figure 2 in a simplified schematic, partially illustrated form. Only one of the two hearing aids 6 is shown, which are mirror images of each other but otherwise identical. The hearing aid 6 is designed as a hearing aid device that is intended and configured to be inserted into the ear canal of the respective ear of the wearer (user, hearing aid wearer) 2. In other words, it is an "in-the-canal" hearing aid (ITC hearing aid device). The hearing aid 6 comprises a housing 10 made of a plastic and adapted to the shape of the respective associated ear canal. Arranged within the housing 10 is a microphone 12 in the form of an electromechanical sound transducer with an omnidirectional design. The microphone 12 is signal-coupled to a control unit 14, which comprises a signal processing unit with an amplifier circuit (not shown in detail) and a signal processor.The control unit 14 is further formed by circuit elements, such as electrical and / or electronic components. The signal processor is a digital signal processor (DSP) and is signal-connected to the microphone 12 via an A / D converter (not shown in detail). A receiver 16 is coupled, i.e., signal-connected, to the control unit 14. By means of the receiver 16, which is an electromechanical sound transducer, an output signal provided by the control unit 14 is converted into an output sound, i.e., into sound waves. During intended use, these are introduced into the auditory canal of the respective ear in the direction of the eardrum located there.
[0057] A sensor 18 is also arranged in the housing 10. The sensor 18 comprises an acceleration sensor, by means of which it is possible to measure the acceleration of the hearing aid 6 and thus also of the head of the wearer 2. From this, it is possible to derive a movement of the user 2 themselves. Furthermore, a communication device 20 is arranged in the housing 10, which serves to establish a radio connection with the portable device 8. The communication device 20 and thus also the radio connection comply with a Bluetooth standard. The sensor 18 and the communication device 20 are controlled by the control unit 14 and are powered via the same. This also supplies the microphone 12 and the receiver 16 with electrical energy. The control unit 14 is powered by a battery (not shown in detail) arranged in the housing 10.
[0058] The portable device 8, configured as a smartphone, has a further communication device 22, which is connected to the communication devices 20 of the two hearing aids 6 via the radio link during operation. Therefore, the further communication device 22 also complies with the Bluetooth standard. The portable device 8 includes a further sensor 24, by means of which a position of the portable device 8 can be determined. The further sensor 24 is configured for this purpose, for example, as a GPS sensor. The further communication device 22 and the further sensor 24 are signal-connected to a further control unit 26 of the portable device 8 and are operated by means of the latter.
[0059] The control units 14 and the further control unit 26 are provided and configured to carry out a method 28 shown in Figure 3 for operating the hearing aid system 4. In other words, the hearing aid system 4 is operated according to the method 28, and the method 28 is at least partially carried out by means of the respective control units 14, 26.
[0060] In a first work step 30, information 32 relating to the behavior of the user 2 of the hearing aid 6 is created. Information 32 includes details about the gait of the user 2, head movements, a time period until a response from the user 2 in a conversation, participation in conversations, the voice of the user 2, a listening environment, movement activity and / or movement patterns of the user 2, a fall of the user 2, a pause in mid-movement, a repetition of actions, and / or a time period until startling after a sudden loud noise. The details are determined at least partially by means of the microphone 12 and the sensor 18 and / or the further sensor 24. Alternatively, or in combination with this, information 32 includes details about the manner in which the user 2 uses an app to control the hearing aid 4.The app is executed on the portable device 8. Alternatively, or in combination with this, a setting change and / or a change in personal preferences is used as information 32, with the app being used in particular to implement the changes. Alternatively, or in combination with this, information 32 is used as information 32, which includes information about the use of a personal assistant via the portable device 8.
[0061] The details and thus also the information 32 are created for a time window, which is, for example, one day. Within the time window, the information is determined several times, for example, and the average thereof is subsequently used as information 32. Thus, short-term fluctuations are not taken into account / compensated for. It is also possible for at least part of the information 32 to be received via a further interface, namely the further communication device 22. This makes it possible for a person to provide the information 32. If the method 28 is carried out at least partially by means of a server, the further interface represents, in particular, a user account into which another person or the user 2 logs.
[0062] In a second step 34, further information 36 relating to a significant event for the user is received via an interface. The further communication device 22 is also used as the interface. If the method 28 is carried out at least partially using a server, the interface represents, in particular, a user account into which another person or the user 2 logs in. The significant event used is the diagnosis of a cognitive disorder, a move, and / or the loss of a loved one. In this case, it is particularly possible for the user 2, a relative, a treating physician, and / or the like to provide the additional information 36. A weighting factor is selected depending on the type of significant event and the person providing the additional information 36.If the additional information 36 is the diagnosis of the disease, the weighting factor is increased if the person is a doctor.
[0063] In a third work step 38, a function of the hearing aid 6 is determined depending on the information 32 and the further information 36, and the function is executed. An "artificial intelligence" algorithm, for example machine learning or a neural network, is used to determine the function. In addition, the weighting factor of the further information 36, which was selected depending on the person providing the further information 36, is taken into account when determining the function. Depending on the information 32 and the further information 36, an information output, a setting change, and / or a therapy recommendation is used as a function, whereby these are adapted in particular depending on the information 32 and / or the further information 36.
[0064] Figure 4 partially illustrates the method 28 in blocks. This shows which people are or can be involved in the method 28, namely the wearer 2, a doctor, relatives and / or healthcare professionals (HCP), who, for example, adjust the hearing aids 6. Various inputs are used for the method 28, namely assumptions or suspicions, a diagnosis, inputs made by the hearing aid system 4, for example by the hearing aid system itself or by operating it, or for example by means of a digital assistant. Based on the inputs, in particular the information 32 and the further information 36 are provided. These are stored in particular in a database, which is provided, for example, by a cloud server. This is in particular a component of the hearing aid system 4 and is preferably used by multiple users 2 of a respective hearing aid 6.Alternatively, each user 2 is expediently assigned a user account ("personal profile"), which expediently forms a component of the respective hearing aid system 4. Several points are provided to motivate the user 2 (HAW "hearing aid wearer") 2 of the hearing aid ("hearing instrument" "hearing aid").
[0065] Figures 5 and 6 show the two possible cases that can arise in method 28. In the first case, the data available in the database is used to identify those users 2 who have an increased risk of also developing a certain disease. Machine learning (ML) is used as an artificial intelligence algorithm for this purpose. Different clusters are identified based on the entered data, thus identifying a pattern with regard to the different diseases. In this case, a corresponding message is issued via the wearable device 8 of user 2 and can also be sent to the smartphone of a person caring for user 2 or a relative.
[0066] Figure 6 illustrates the other application scenario. In this case, method 28 uses the input data to provide user 2 with a therapy recommendation to manage an existing condition. Machine learning is used as an "artificial intelligence" algorithm to determine the output of the respective therapy recommendation. In particular, the invention relates to a corresponding server system and / or a system in which multiple corresponding users / persons interact.
[0067] The invention is not limited to the exemplary embodiment described above. Rather, other variants of the invention can also be derived therefrom by those skilled in the art without departing from the subject matter of the invention. In particular, all individual features described in connection with the exemplary embodiment can also be combined with one another in other ways without departing from the subject matter of the invention.
[0068] List of reference symbols
[0069] 2 users
[0070] 4 Hearing aid system
[0071] 6 Hearing aid
[0072] 8 portable device
[0073] 10 housings
[0074] 12 Microphone
[0075] 14 Control unit
[0076] 16 listeners
[0077] 18 Sensor
[0078] 20 Communication device
[0079] 22 additional communication devices
[0080] 24 additional sensors
[0081] 26 additional control units
[0082] 28 procedures
[0083] 30 first step
[0084] 32 Information
[0085] 34 second step
[0086] 36 further information
Claims
Claims 1. Method (28) for operating a hearing aid system (4) having a hearing aid (6), in which - information (32) concerning a behavior of a user (2) of the hearing aid (4) is created, - further information (36) concerning a significant event for the user (2) is received via an interface, - a function of the hearing aid (6) is determined as a function of the information (32) and the further information (36), and - the function is executed.
2. Method (28) according to claim 1, characterized in that as information (32) an indication about a gait, about head movements, about a time period until a response, about participation in conversations, about the voice, about a listening environment, about a movement activity, about a movement pattern, about a fall, about a way in which the user (2) uses an app to control the hearing aid (6), about use of a personal assistant, about a change in settings, about a change in personal preferences, about pausing in the middle of a movement, about repetition of actions, and / or about a time period until startling after a sudden loud noise is used.
3. Method (28) according to claim 1 or 2, characterized in that the information (32) is received via a further interface.
4. Method (28) according to one of claims 1 to 3, characterized in that the information (32) is created for a time window.
5. Method (28) according to one of claims 1 to 4, characterized in that as a drastic event - the diagnosis of a disease, - a move and / or - a loss of a close living being is taken into account.
6. Method (28) according to one of claims 1 to 5, characterized in that when determining the function, a weighting factor of the further information (36) is taken into account, which is selected as a function of a person providing the further information (36).
7. Method (28) according to one of claims 1 to 6, characterized in that the function is determined by means of an “artificial intelligence” algorithm.
8. Method (28) according to one of claims 1 to 7, characterized in that as a function - an information output, - a change in attitude, and / or - a therapy recommendation is used.
9. Hearing aid system (4) comprising a hearing aid (6) and operated according to a method (28) according to one of claims 1 to 8.
Citation Information
Patent Citations
Sharing of health-related data based on data exported by ear-wearable device
US20200273566A1
Health monitoring with ear-wearable devices and accessory devices
US20220345836A1
Fall prediction system including a beacon and method of using same
EP3579752B1
Method and computer program for early detection of a health issue, controller for a hearing device, hearing device, and hearing system comprising a hearing device
EP4156717A1
Method for transmitting a processing state in an audiological adaptation application for a hearing device
WO2019233602A1
