CH719511A2 - Device for detecting and classifying sleep apnea. - Google Patents

Abstract

The invention relates to a device for detecting and classifying sleep apnea in a sleeper. The device can have an in-ear module (4) carried in the ear canal (8) with an optical sensor (12) for photoplethysmography. The device can also have an acoustic sensor (10, 11). Intermediate data, for example regarding the oxygen saturation of the blood, pauses in breathing and snoring, are derived from the sensor data. Based on this, the presence of sleep apnea is recognized and, if necessary, classified as obstructive or central. The device can also have the function of a hearing aid or hearing protection.

Description

Technical area

The invention relates to the field of in-the-ear devices. These can be hearing protection devices and/or hearing aids. In particular, it relates to a device for detecting and classifying sleep apnea in a sleeping individual. The invention further relates to a method in which the device is used.

State of the art

[0002] There are a variety of devices which are designed to be worn at least partially on and/or in the ear. This includes, in particular, hearing protection devices and hearing aids. Such devices can serve a second purpose by being used to monitor the health of their wearer. There are also dedicated health monitoring devices. The devices have sensors for health monitoring. A disease or disorder that needs to be recognized is so-called sleep apnea. People who suffer from sleep apnea experience pauses in breathing at night. This affects the quality of sleep and the recovery effect of sleep. It is known to monitor various biological parameters in a so-called sleep laboratory and thus diagnose sleep apnea. There are then various treatment options.

[0003] In the following patent literature references, the “et al.” is omitted for the sake of simplicity.

[0004] US 8,794,235 to Garde discloses a system for treating ventilatory instability. Photoplethysmogram is measured.

[0005] US 10,406,376 by Goodall discloses an ear stimulation system. Detection of sleep apnea and snoring is mentioned.

[0006] US 2016/324478 by Goldstein discloses biometric monitoring. Photoplethysmography and monitoring for obstructive and central sleep apnea are mentioned.

[0007] WO 2020/176414 A1 from McKinney mentions the detection of sleep apnea and the use of a PPG sensor.

[0008] EP 3 706 441 A1 by De Haan discloses a hearing aid with a PPG sensor, which allows obstructive sleep apnea to be monitored.

[0009] WO 2021/152551 A1 by Fox discloses a device for detecting sleep status for an apnea-hypopnea index calculation.

[0010] US 2017/238812 from Atlas discloses a remote physiological monitoring system that can distinguish between different types of apnea, namely central and obstructive. It can include or be a remotely programmable hearing aid.

[0011] US 2017/095202 by Facteau discloses an ear canal device with which one can monitor sleep apnea and issue warnings in this regard.

[0012] DE 10 2017 000 175 A1 from BVP GmbH discloses a method for reducing or eliminating snoring noises and preventing dangerous pauses in breathing. A disturbing or alarm signal can be generated with a hearing aid.

[0013] WO 2020/109863 A3 by Taki discloses a device for treating sleep apnea, which records snoring sound.

Presentation of the invention

[0014] The technical task underlying the patent application is to create a device and a method which allow sleep apnea to be recognized and classified in a way that is cost-effective, reliable, minimally invasive, does not require large laboratory equipment, the sleeping person does not bother the individual and/or is suitable for widespread use on a variety of individuals.

[0015] This task is solved by the subjects of claims 1 and 10. When worn, the device has sensors on and/or in the ear or the method uses sensors which are located on and/or in the ear when the device is worn.

According to claim 2, at least two different sensors are provided. This allows a reliable classification of sleep apnea.

According to claim 3, means are provided to first determine one or more intermediate sizes. This also allows a reliable classification.

According to claim 4, means are provided to recognize certain states of the individual. These allow the health status of the individual to be well characterized.

According to claim 5, the device has the purpose of hearing protection in addition to the purpose of detecting sleep apnea. This has the advantage that the user can manage with fewer devices and has additional motivation to use the device while sleeping.

[0020] According to claim 6, the device has the purpose of a hearing aid in addition to the purpose of detecting sleep apnea. This has the advantage that the solution is overall more cost-effective than two separate devices. Components such as the battery and microphones can be used for both purposes. Wearing a hearing aid at night has the advantage that the user is more aware of warning signals, such as a fire alarm or simply the alarm clock. The primary purpose of the device may be hearing aid function. However, the device can be worn sporadically or at night if sleep apnea is specifically suspected and can thus fulfill its diagnostic purpose.

[0021] According to claim 7, the device has a photoplethysmogram sensor. This has the advantage that the heart rate and the oxygen saturation of the blood can be measured.

[0022] According to claim 8, the device has means for detecting pauses in breathing. This is essential for detecting sleep apnea. The combination of acoustic and optical sensors allows for a particularly high level of reliability.

[0023] According to claim 9, the device has an external device or means for communicating with an external device. This has the advantage that additional functionalities can be offered, such as a comfortable user interface.

Claim 10 relates to a method for detecting and classifying sleep apnea, in which the device according to the invention is used. Sensors on and/or in the ear have the advantage that they are comfortable to wear, especially if the individual already wears hearing protection or a hearing aid.

Brief description of the drawings

1 shows schematically a person wearing an embodiment of the device according to the invention; Fig. 2 shows the ear of a human with an embodiment of the device according to the invention; 3 shows the ear canal of a human with an in-the-ear module of an embodiment of the device according to the invention; Fig. 4 shows a block diagram with sensors and variables and states derived from their signals.

The figures only show some possible embodiments as examples. They and their following description are not intended to limit the scope of protection of the claims.

Ways to carry out the invention and commercial applicability

1 shows a person 1 wearing an embodiment of the device according to the invention. There are two ear devices 2, which have their own power supply and can communicate wirelessly with a smartphone 3 via Bluetooth. Therefore no cables are required. The ear devices can be arranged on and/or in the ear. In particular, they can be behind the ear and/or in the ear canal. They are preferably designed in such a way that they do not disturb people while they are sleeping.

2 shows a human's ear with an embodiment of the device according to the invention. There is an in-the-ear module 4, which is arranged in the ear canal, and a behind-the-ear module 5, which is arranged behind the auricle 7. The two modules 4 and 5 are connected to a connection 6. The behind-the-ear module 5 has a behind-the-ear microphone 9.

3 shows the ear canal of a human with an in-the-ear module 4 of an embodiment of the device according to the invention. The module is worn in the ear or in the ear canal 8 of the ear. It has an in-the-ear microphone 10. This is particularly suitable for recording ambient noise. It also has an ear canal microphone 11. This is particularly suitable for recording structure-borne noise. There is also a PPG sensor 12, a pressure sensor 13 and a motion sensor 14. In order to provide the functionality of a hearing aid or alarm clock, there is a loudspeaker 15. A vent 16 is used to reduce occlusion and ventilate the ear canal. A processor module 17 has various functions. Among other things, it receives sensor data and controls the loudspeaker 15. It may include a microcontroller (MCU) and a digital signal processor (DSP). A connection 6 leads to a behind-the-ear module, not shown.

4 shows a block diagram with exemplary sensors and the variables derived from their signals. There is an acoustic sensor 21, an optical sensor 22, a pressure sensor 23, a motion sensor 24, a position sensor 25, a location sensor 26, an electrical sensor 27, a chemical sensor 28 and a temperature sensor 29. There can also be two or more sensors give a certain type. As derived intermediate variables there is a heart rate 31, a heart rate variability 32, a breathing frequency 33, a breathing intensity 34, a pause in breathing 35, an oxygen saturation 36, a snoring intensity 37 and a sleep quality measure 38. Furthermore, there are the presence of snoring 41, the presence, as output variables an apnea disorder 42, the presence of obstructive sleep apnea 43 and the presence of central sleep apnea 44. The connecting lines are for illustrative purposes only. The intermediate sizes can also be based on other intermediate sizes and/or output sizes. The output variables can also be determined directly based on the sensor signals, without determining intermediate variables. The intermediate sizes can also be used as output sizes. Not shown in the diagram are the time and personal data, such as the age and gender of the individual, which can also be included in the calculations.

Various aspects and embodiments of the invention are explained below, sorted by keywords or expressions: Health monitoring device: The health of an individual can be monitored using sensors. Based on the sensor data, diseases, especially sleep apnea, can be diagnosed and treatment recommendations can be made.

Hearing protection: Hearing protection is worn, for example, on construction sites to protect the hearing from machine noise. However, many people also wear hearing protection at night or when sleeping, for example to reduce traffic noise or to avoid being disturbed by their partner's snoring. A disadvantage of such hearing protection is that alarms, alarm signals and/or dangerous noises are less noticeable. To mitigate this problem, there is active hearing protection that detects and amplifies such sounds, signals them via another channel, such as optically, or at least does not attenuate them. The hearing protection can also provide active noise suppression, in which sound is canceled out with anti-noise. There are different types of hearing protection, in particular ear muffs that cover the ears and ear plugs that are worn in the ear canal. The latter are best suited to wearing while sleeping. The plugs can be elastic. Alternatively, an ear impression can be taken. Based on this, a so-called earmold is then made from a usually hard material. An earmold is better suited to holding electronic components and especially sensors than an elastic plug.

Hearing aid: Hearing aids are worn by people who are hard of hearing to compensate for their hearing loss. They have at least an ambient microphone and a loudspeaker, which is also often referred to as a listener. Hearing aids are usually only worn during the day and, if they are rechargeable, charged at night. This means they cannot be worn at night. A possible solution to this problem is to provide the user with two pairs of hearing aids so that the first pair can be charged while the second is in use. There are different designs of hearing aids. There are behind-the-ear hearing aids, in-the-ear hearing aids and ex-receiver hearing aids in which the battery and most other components are located behind the ear, while the receiver or speaker is located in the ear. Similar to hearing protection, the element in the ear canal can be designed to be elastic or as a so-called earmold, which is manufactured for the individual ear based on an ear impression or scan.

Individual: The device according to the invention is primarily for people who are suspected of having sleep apnea. Monitoring can be arranged by a doctor. One can therefore also speak of “patients”. However, it is also conceivable to carry out monitoring in general on all “users” of hearing protection or hearing aids, without there being any suspicion of an illness. This is made possible because the individual is not disturbed by the surveillance. Only the battery consumption is slightly higher with monitoring than without. Finally, it is also conceivable to use the solution in the veterinary sector for animals. Personal data such as gender, age, weight or previous illnesses can be recorded and used in the method according to the invention or in diagnosis and treatment.

Ear canal: For most people, the ear canal is large enough to accommodate a device, such as a hearing aid or a module with biometric sensors, preferably in such a way that the device sits stably in it and preferably not or hardly at all from the outside is visible. The ear canal wall contains blood vessels. The outer part is made of cartilage and the inner part is made of bone. The wall of the ear canal or the skin there secretes sweat and gum. A variety of measurement variables can therefore be determined, particularly chemical ones.

Sensors: The sensors used in the invention can in particular be acoustic, optical, mechanical, electrical, chemical or thermal. They can be used to record the environment as well as to record states and characteristics of the individual. For the latter purpose, one also speaks of biometric sensors. The sensors can deliver an analog or digital signal as a component.

Interference signal: Sensor signals can contain interference signals, for example an acoustic signal can contain an interference signal “traffic noise” in addition to the useful signal “breathing noise”. In order to mitigate this problem, means for eliminating interference signals or for source separation can be provided. This can be a simple spectral filter or an appropriately trained neural network.

Acoustic sensors: Microphones can be provided, among other things: behind the ear, at the entrance to the ear canal or in an external device. There they primarily absorb ambient noise. Microphones arranged in the ear canal, so-called ear canal microphones, primarily record structure-borne noise, such as the wearer's voice. Breathing can be monitored with the signals obtained. Furthermore, the wearer's snoring can be detected. A special type of microphone is bone conduction microphones.

Structure-borne noise: For the invention, breathing noise is the most important sound that needs to be detected. A distinction can be made here between normal breathing, inhibited breathing and snoring. There are also sounds that are caused by the wearer's heart, chewing apparatus, digestion or voice.

Ear canal microphones: Such microphones are arranged on the medial side of an in-the-ear module so that they record the sound in the space between the in-the-ear module and the eardrum. If the individual's structure-borne noise, such as breathing noises or snoring, is recorded with an environmental microphone, this has the disadvantage that the signal also contains ambient noise such as traffic noise or the body noise of another person. When sound is detected in the ear canal, these noises are less dominant. So-called acoustic emissions can also be recorded using ear canal microphones.

[0041] Bone conduction microphones: In contrast to other microphones, these are not coupled to the air, but directly to the individual's body and detect vibrations. They are particularly suitable for the interference-free detection of structure-borne noise.

Directional microphones: Hearing aids often have so-called directional microphones. These can consist of two spatially separated omnidirectional microphones. They are primarily intended to emphasize the sound of a speaker standing in front of the individual. However, the directivity can also be used to separate ambient noise and body noise. The two microphones are typically located behind the ear, one further forward and one further back.

Optical sensors: These can passively record light from the environment, or, as is the case with photoplethysmography sensors, they can emit light and detect the reflected light. Ambient light provides an indication of sleeping conditions. A distinction can be made between natural light, artificial light and darkness.

Pressure sensors: The contact pressure of the measuring device on the ear canal wall can be measured with a sensor. On the one hand, this makes it possible to see whether the device has been used correctly. On the other hand, the sensor signal can contain information about heart rate and blood pressure. A differential pressure sensor can also be used. This allows, for example, to determine that a pulse wave is moving in a certain direction.

[0045] Motion sensors: Acceleration sensors can be used to detect whether the person to be monitored is currently moving. The movement activity can also provide information about what sleep phase the individual is currently in. Furthermore, movements can lead to artifacts in other sensors. The movement data is therefore an indicator of the validity of other measurements. Acceleration can be both translational and rotational. The latter can be measured with a gyroscope or gyrocompass. Preferably three such sensors are provided in each case to take the three spatial dimensions into account.

Position sensors: The gravitational force can be measured using acceleration sensors. Among other things, it can be recognized whether the individual is standing or lying down, with the lying position being an indicator of sleeping. In addition to the acceleration sensors, a geomagnetic field sensor or compass can be provided.

[0047] Location sensors: The location of the device can be detected using a GPS receiver. The location can also be determined based on WiFi or cell phone signals. The location information can help to distinguish whether poor sleep quality is due to the environment or to a health impairment. Based on the location information, data about the respective location can also be retrieved from a server, for example the sleep quality of other people at that location.

Electrical sensors: Electrodes can be used to measure electrical potentials on the individual's body and in particular in the ear canal. These can provide an indication of brain activity as an EEG (electroencephalogram) and an indication of muscle activity as an EMG (electroencephalogram), with the muscles involved in breathing and eye movement being of particular interest here. An EEG can be measured in particular in and/or on the ear. The information obtained from the EEG can be used to better determine the type of sleep apnea.

Chemical sensors: The composition of sweat, cerumen or gases on the body or in the ear canal can be determined using chemical sensors. In particular, the blood sugar level can be determined.

Temperature sensors: The temperature sensor can be used to detect whether the device is inserted in the ear. It can also be determined whether the individual has a fever.

Photoplethysmography sensors: They are also referred to as PPG sensors for short. They are used to measure a so-called photoplethysmogram. They usually include an LED, which emits light, and a photodetector, which absorbs the LED light reflected from the body. Different colors of light can form channels. An entire spectrum can be detected and a variety of frequencies can be analyzed. According to the invention, a two-channel PPG sensor is preferably used, with light in the red range and in the infrared range. A PPG sensor can be used to detect heart rate, heart rate variability, blood pressure, pulse wave delay and hemoglobin oxygen saturation. The PPG sensor is preferably arranged so that it is part of an in-the-ear module and rests against the ear canal wall, with no ambient light at the measurement location if possible.

Measuring variables: The primary result sought by the invention is a statement about whether a person has sleep apnea and, if so, whether it is obstructive or central. For this it can be advantageous to determine different intermediate sizes and/or states. Such variables include, for example, heart rate, heart rate variability, breathing frequency, a measure of breathing intensity, the number of pauses in breathing, oxygen saturation, snoring intensity and a measure of sleep quality. These measured variables can only exist internally or can be output.

Oxygen saturation: The oxygen saturation of the hemoglobin contained in the blood is measured using a so-called pulse oximetry. The symbol “SPO2” is also used for this. Oxygen saturation drops during pauses in breathing. Oxygen saturation can be measured with a PPG sensor based on the fact that oxygen-saturated and oxygen-unsaturated hemoglobin have different adsorption spectra. Oxygen-saturated hemoglobin absorbs red light more strongly than oxygen-unsaturated hemoglobin. For infrared it is the other way around.

[0054] Respiratory frequency: The respiratory frequency can be recognized both based on periodic variations of the PPG signal and based on the acoustic signal. The combination allows a reliable estimate.

Heart rate: The heart rate or pulse can be determined based on the infrared PPG signal. It can be determined before and after an episode of respiratory failure. It can be compared with standard values and individual values. It makes it possible to determine the effect of a pause in breathing on the quality of sleep.

Snoring: Snoring can be detected based on the acoustic signal. Microphone signals from different microphones, e.g. on the smartphone, in the ear canal, behind the ear, can be combined. Snoring is a form of restricted breathing. Impeded breathing is a sign that apnea is obstructive.

[0057] Detection of pauses in breathing: This can be done primarily based on the breathing sounds. The time since the last breath can be compared to a threshold. The individual standard deviation of the breathing frequency can be taken into account. Furthermore, the drop in oxygen saturation can be detected. Both effects, that of breathing sounds and that of oxygen saturation, can be correlated.

Prediction of pauses in breathing: In particular with artificial intelligence, it is also possible to predict pauses in breathing and, based on this, control a therapy device or wake the sleeper using sound, light and/or vibration.

[0059] Sleep apnea: According to Wikipedia, sleep apnea (ICD-10 G47.3) is a sleep-related breathing disorder. During sleep, there are repeated episodes of hypopnea (insufficient ventilation of the lungs) and apnea (short-term cessation of breathing). ICD stands for “International Classification of Diseases”.

Sleep apnea classification: A distinction is made between obstructive sleep apnea (ICD-10 G47.31) and central sleep apnea (ICD-10 G47.30). Obstructive sleep apnea is usually accompanied by snoring or the sounds of restricted breathing. The device according to the invention diagnoses whether sleep apnea is present and, if so, what type it is. This information can be supplemented with a reliability value or a probability.

Severity of sleep apnea: Severity can be determined based on the number of apnea episodes and their effect on heart rate and oxygen levels.

Treatment: The invention can also be expanded to suggest consulting a doctor or directly suggesting measures for optimal patient-related treatment or alleviation of sleep apnea, for example the use of a therapy device.

[0063] Artificial intelligence (AI): When detecting, classifying and/or predicting sleep apnea, a deep neural network (DNN) can be used, which is trained with a large number of sensor data and associated classifications from a large number of individuals. Such data is also referred to as “big data” or, when used, as “data mining”. It is advantageous if all available sensor data are fed to the neural network as input variables, even if a correlation between the respective size and sleep apnea has not been proven beyond doubt. Neural networks can also be used to determine the intermediate variables from sensor data or the output variables from the intermediate variables. Neural networks can also be used to remove interference from sensor signals. See also paragraph “Interference signal”.

Designs: The invention can be implemented in such a way that the essential components of the device are completely integrated in one or two in-the-ear modules, which can communicate wirelessly with each other and with other devices. This has the advantage that no additional devices or cables are required for the user or in bed and the device can be completely or almost not perceived by other people. A design is also possible in which there is a behind-the-ear module and an in-the-ear module on each ear.

Processor: The calculations according to the invention require a processor. This can be integrated as a microcontroller (MCU) in the in-the-ear module. There may be separate additional processors, particularly a digital signal processor (DSP) and an AI processor. Calculations can also be carried out in an external processor, for example in the processor of a smartphone or a server, which is particularly advantageous for computationally intensive algorithms.

Memory: Memory is required for the programs as well as the data. This can be at least partially integrated into a microcontroller. There is preferably a non-volatile memory which retains its contents even if the power supply is interrupted.

Battery: The module worn in the ear preferably has a battery so that no cables are required. The battery can be replaceable and/or rechargeable.

Internal clock: The ear module preferably has a real-time clock, which measures the time correctly even if the battery is changed or is temporarily completely discharged. Measurement data can be given a time stamp. Alternatively, the time can be queried from an external device.

External device: The device can have or communicate with one or more external devices. This can be a smartphone, a smartwatch, a remote control, an external microphone, a computer, a server, a device for displaying diagnostic data, a therapy device and an intelligent house.

User interface: The device according to the invention is preferably operated via a smartphone, which can also be viewed as part of the device. Alternatively, the device or an in-the-ear module can also have simple operating elements such as an on-off switch. A personal computer, such as a notebook or a desktop computer, can be used to evaluate the measurement results, especially by a doctor.

Smartphone: Smartphones usually have a touchscreen, can be connected to the Internet via WLAN, support the Bluetooth protocol and have a GPS receiver. Communication between the smartphone and the in-ear module can take place via Bluetooth. The screen makes it possible to display the results graphically and offer options for forwarding them. The microphone, the motion sensor and the GPS receiver of the smartphone can provide additional sensor data for the detection and classification according to the invention.

Smartwatch: The system can also have a wristwatch. This can also be used for operation, especially if it has a touchscreen. Furthermore, further sensors can be provided on the wrist, for example another PPG sensor, another microphone and/or another motion sensor.

[0073] Remote control: Operation can be done via an app on a smartphone. However, a dedicated remote control is also possible.

External microphone: Hearing systems can have an external microphone, for example to record the voice of a teacher directly in the vicinity of the teacher, so that the signal contains less noise. In the device according to the invention, an external microphone can be used in particular to record breathing noises directly at the place where they arise, that is, near the nose and mouth or on the chest. External microphones can also be used as interference signal microphones in order to remove interference signals, such as traffic noise, from the useful signal.

Server: The device according to the invention can communicate with a server via the Internet. The server can be part of a cloud system. The server can provide a website for querying the diagnostic data. Computationally intensive operations, such as evaluating sensor data with artificial intelligence, can be carried out on the server.

Device for displaying diagnostic data: The data obtained is preferably displayed graphically on a large display.

Therapy device: The data collected by the device according to the invention can be used to control a therapy device. This can be a CPAP breathing therapy device (continuous positive airway pressure). This has a mask and can create slight excess pressure in the airways. It can also be an oxygen therapy device that can be controlled by the device and supplies oxygen to the patient. The therapy device can be switched completely on or off. But the intensity of the therapy can also be controlled. Furthermore, information regarding the breathing status, such as inhalation or exhalation, can be transmitted to the therapy device.

Health data: The device produces health data which should be treated confidentially. An encrypted transfer, for example to a doctor or his computer, is preferably provided. The transfer must preferably be authorized by the user and/or their carer.

[0079] Consent to data use: The user can preferably choose whether his usage data may be used to improve the device, and in particular to train a neural network. If the user agrees, the usage data will be transferred to a server. This can be done once, regularly or depending on the occasion. Preferably, data is also recorded as to whether the diagnosis made by the device turns out to be correct, for example if it was confirmed in a sleep laboratory.

Although the claims relate to a device and a method, it is hereby pointed out that the invention can also be claimed in the form of a computer program product, system, data carrier or a signal.

Claims (10)

1. A device for detecting and classifying sleep apnea in a sleeping individual (1), characterized in that, when worn, it has one or more sensors (9, 10, 11, 12) arranged on and/or in the ear (7, 8). , 13, 21, 22, 23, 24, 25, 26, 27, 28, 29). 2. The device according to claim 1, wherein it has at least two different sensors (9, 10, 11, 12, 13, 21, 22, 23, 24, 25, 26, 27, 28, 29) from the following list: • An acoustic sensor (9, 10, 11, 21); • An optical sensor (12, 22), in particular a PPG sensor (12); • A pressure sensor (13, 23); • A motion sensor (24); • A position sensor (25); • A location sensor (26); • An electrical sensor (27); • A chemical sensor (28); • A temperature sensor (29). 3. The device according to one of the preceding claims, wherein it has means for determining one or more of the following intermediate sizes: • A heart rate (31); • A heart rate variability (32); • A respiratory rate (33); • A breathing intensity (34); • A pause in breathing count (35); • An oxygen saturation (36); • A snoring intensity (37); • A sleep quality measure (38). 4. The device according to one of the preceding claims, wherein it has means for recognizing one or more of the following states of the individual: • Presence of snoring (41); • Presence of sleep apnea (42); • Presence of obstructive sleep apnea (43); • Presence of central sleep apnea (44). 5. The device according to one of the preceding claims, wherein it has an ear canal module (4) which has a hearing-protecting effect, or it is built into hearing protection. 6. The device according to one of the preceding claims, wherein it has a loudspeaker (15) and it is designed for use as a hearing aid or it is built into a hearing aid. 7. The device according to one of the preceding claims, wherein it has a photoplethysmogram sensor (12), which is arranged in the ear canal (8) when the device is worn and in particular has at least one of the following properties: • Having an LED; • Having a photodetector; • Emitting red light; • Emitting infrared light; • Using two optical channels; • Detecting an optical spectrum. 8. The device according to one of the preceding claims, wherein it has means for detecting a pause in breathing of the individual (1) based on • Data obtained with an acoustic sensor (9, 10, 11, 21) and • Data obtained with an optical sensor (12, 22). 9. The device according to one of the preceding claims, wherein it has at least one external device (3) or means for communicating with at least one external device (3), the at least one external device (3) being in particular: • A smartphone (3) • A smartwatch • A remote control • An external microphone • A computer • A server • A device for displaying diagnostic data • A therapy device • A smart house 10. A method for detecting and classifying sleep apnea in a sleeping individual (1), characterized in that it involves use of the device according to one of claims 1 to 9 and the method has the following steps: • Inserting a module (4) into an ear canal (8) of the individual (1) • Acquisition of data with an acoustic sensor (9, 10, 11); • Acquisition of data with an optical sensor (12, 22), which is part of the module (4); • Evaluate the data; • Recognition of sleep apnea and, if necessary, classification of sleep apnea as obstructive or central.