Monitoring of sleep disorders and its system

Abstract

Monitoring of sleep disorders of a subject using a mobile device such as a smartphone or other wearable device, including placing the mobile device in juxtaposition with the subject, acquiring an audio signal indicating the subject's respiration by a microphone of the mobile device, acquiring an accelerometer signal including the subject's body position, angle, and motion information by an accelerometer of the mobile device, measuring chest movements related to respiration and respiratory sleep disorders by the accelerometer of the mobile device, identifying a plurality of silent events by the audio signal, classifying the silent events into apnea or hypopnea, classifying the classified apnea into a plurality of apnea types by chest movements recorded by an accelerometer sensor, and associating respective angle information with the classified silent events by combining the audio signal and the accelerometer signal.

Description

Technical Field

[0001] The present invention relates to a method for monitoring a subject's sleep disorder and sleep quality. More particularly, the present invention relates to corresponding methods and systems that use a mobile device such as a smartphone or other wearable device. Further, some aspects may also relate to corresponding postural therapies.

Background Art

[0002] Obstructive sleep apnea (OSA) is a common sleep disorder associated with drowsiness and accidents and is also an independent risk factor for heart, brain, and cancer diseases. However, conventional techniques for diagnosing OSA do not provide a way to reliably and cost-effectively monitor a subject's sleep, resulting in a late diagnosis for over 80% of patients. Known methods are questionnaire-based, such as the STOP-Bang and Berlin questionnaires, which are thought to quantify the quality of a patient's sleep. However, they often cannot appropriately determine various sleep postures and cannot find the relationship between sleep posture and sleep apnea.

[0003] Other techniques, such as a polysomnography (PSG) test, which is considered by some to be a representative test for diagnosing sleep-related diseases, use a plurality of sensors for recording respiratory activity, such as nasal pressure transducers, thermistors, and chest and abdominal bands, and sensors and cameras based on accelerometers for monitoring sleep posture, resulting in more objective and reliable measurements. However, PSG is cumbersome and expensive, requires the subject to sleep with multiple wires, encourages the subject to sleep in a specific supine position, and may further exacerbate existing disorders.

[0004] In addition to the above, it is well known that mobile electronic devices, especially smartphones, are now ubiquitous. The capabilities of modern smartphones are highly regarded in numerous technical fields. The fact that such mobile devices provide not only a significant amount of processing power and a high-quality user interface, but also sensors and detectors such as sound, light, position, acceleration, and force, has already led to a wide variety of applications in technical fields not directly related to telephony and communication.

[0005] However, simply having sensors and data acquisition and processing capabilities does not automatically result in a practical solution. Especially in the context of health-related issues, in the fields of teletherapy, telemedicine, remote monitoring, and mobile health, there is a need for clearly defined, predictable, and especially highly reliable solutions and the actual provision of related benefits. SUMMARY OF THE INVENTION PROBLEMS TO BE SOLVED BY THE INVENTION

[0006] Therefore, there is a need for an excellent method and system that provide reliable functions in the situation of sleep disorders. MEANS FOR SOLVING THE PROBLEMS

[0007] The above problems are solved by the subject matter of the independent claims. Further preferred embodiments are given by the subject matter of the dependent claims. Additionally, further examples are provided to facilitate the understanding of the present invention.

[0008] According to an embodiment of the present invention, a method for monitoring a subject's sleep disorder using a mobile device is provided. The method includes juxtaposing the mobile device with the subject, acquiring an audio signal indicating the subject's respiration by a microphone of the mobile device, acquiring an accelerometer signal by an accelerometer of the mobile device, and deriving the subject's body position, angle, and motion information from the accelerometer signal or a signal from the accelerometer, measuring chest movement related to respiration and sleep disorder respiration by the accelerometer, identifying a silent event based on the audio signal, classifying the silent event as apnea or hypopnea, identifying the apnea type of the classified apnea based on the chest movement, and associating respective angle information with the classified silent events by combining the audio signal and the accelerometer signal.

[0009] According to another embodiment of the present invention, a method for monitoring the quality of a subject's sleep using a mobile device is provided. The method includes juxtaposing the mobile device with the subject, acquiring an audio signal indicating the subject's respiration by a microphone of the mobile device, acquiring an accelerometer signal by an accelerometer of the mobile device, and deriving the subject's body position, angle, and motion information from the accelerometer signal, measuring chest movement related to respiration and sleep disorder respiration by the accelerometer, identifying a sleep event based on the audio signal, and identifying the body position where the sleep event occurred based on the angle information.

[0010] According to another embodiment of the present invention, a system for monitoring a subject's sleep disorder and sleep quality is provided. The system includes a mobile device comprising a microphone configured to acquire an audio signal indicating the subject's respiration and an accelerometer configured to acquire an accelerometer signal including the subject's body position, angle, and motion information, a fixed system for juxtaposing the mobile device with the subject, and a pulse oximeter configured to measure the subject's oxygen saturation level. The fixed system is composed of an elastic band and a bag for arranging the mobile device.

[0011] Next, embodiments of the present invention, which are presented to better understand the concept of the present invention and should not be regarded as limiting the present invention, will be described with reference to the drawings.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Modes for Carrying Out the Invention

[0013] Detailed Description Figure 1 shows a schematic diagram of a configuration and an embodiment for applying the present invention. Specifically, this embodiment provides monitoring of the sleep disorder of the subject P, and this configuration considers the mobile device 10 and places the mobile device 10 in juxtaposition with the subject P. The juxtaposition can include ensuring that the mobile device 10 follows the movement by the subject P. For example, the juxtaposition can be obtained by fixing the mobile device 10 in a pocket of the clothes worn by the subject P. Further, the juxtaposition can be obtained by a fixing system 2 attached to the subject P by a strip or a belt. The fixing system 2 is described in more detail in FIG. 3. Generally, the mobile device can be implemented by any one of a mobile computing and / or communication device, a mobile phone, a smartphone, a wearable device, an electronic wristwatch, a smartwatch, etc.

[0014] Next, this embodiment includes acquiring an audio signal. An audio signal representing the breathing of the subject P is acquired by the microphone 101 of the mobile device 10. For example, the audio signal A may be acquired by the built-in microphone 101 of the mobile device 10 at a sampling rate of 48 kHz throughout the night. Examples of the audio signal A obtained by the built-in microphone 101 of the mobile device 10 at night can be seen in FIGS. 7 to 9.

[0015] This embodiment further includes obtaining an accelerometer signal by the accelerometer 102 of the mobile device 10, and deriving the body position, angle (e.g., indicated by α in FIG. 1), and motion information of the subject P from the accelerometer signal. Since the mobile device 10 is juxtaposed with the subject P, the accelerometer 102 of the mobile device can conveniently detect the movement of the subject while monitoring the possibility of acute sleep disorder or sleep disorder. Specifically, the accelerometer signal can be used to detect or distinguish the supine position (as shown in FIG. 1 for the subject P) or the lateral position shown by P' in FIG. 1. Although not shown in FIG. 1, the accelerometer signal can also be used to detect or distinguish other body positions such as the prone position or the standing position of the subject P. Further, the sleep body position of the subject P is not limited to any of the pure supine position, lateral position, prone position, or standing position where the angle information is generally a multiple of 90°, and the sleep body position of the subject P is not limited to the supine position. The angle information may be an angle, a plurality of angles, or a three-axis angle. The angle information may also be a three-axis accelerometer value derived from the accelerometer signal obtained by the accelerometer 102. The body position, angle, and motion information regarding the subject are further described in detail in relation to FIG. 2.

[0016] This embodiment further includes measuring the chest movement related to breathing and sleep disorder breathing by the accelerometer 102. This means that, for example, using the accelerometer 102 of the mobile device 10 juxtaposed above the sternum of the subject P, the repetitive vertical movement of the chest caused by the breathing of the subject P can be measured. When breathing exists in the subject, since the dimensions of the chest in the front-rear direction, vertical direction, and lateral direction increase and decrease in each of inhalation and exhalation, it is possible to recognize the presence of breathing by the chest movement (see also the acceleration measurement in FIG. 7 and the angle variation in FIG. 8). The juxtaposed mobile device 10 can be used to measure this chest movement related to breathing. Further, when breathing pauses or weakens due to sleep disorder, the chest movement may temporarily stop or weaken, so that the breathing of sleep disorder can also be detected (see also the angle variation shown in FIG. 8).

[0017] This embodiment further includes identifying a silent event based on an audio signal. The silent event can correspond to an event that may represent a sleep disorder. For example, when the subject P has OSA, there may be repetitive episodes of overall or partial airflow reduction during sleep caused by upper airway obstruction (i.e., apnea and hypopnea as shown in FIGS. 7 and 8). These respiratory disturbances can be recognized as silent events in the audio signal A. For the recognition of silent events, sample entropy can be used in the calculation because snoring and breathing show more complex patterns than the silent region and thus produce higher sample entropy values. However, the calculation used for the recognition of silent events is not limited to a specific function.

[0018] When a silent event is identified, the silent event is classified as apnea or hypopnea as shown in FIG. 8. In the analysis of the audio signal A, if there is residual respiratory activity during the silent event, this silent event is classified as an event representing hypopnea, and if there is no breathing at all during the silent event, this silent event is labeled as apnea. Residual respiratory activity can be recognized by the presence of low-intensity breath sounds recorded via the microphone 101.

[0019] This embodiment further includes identifying the apnea type of apnea classified based on chest movement. The accelerometer 102 can be used to identify the classified apnea into three types: obstructive, central, and mixed. Obstructive apnea occurs due to upper airway obstruction, while central apnea is caused when the brain fails to send appropriate signals to the muscles involved in respiratory control. Mixed apnea is apnea that starts as central apnea and ends as obstructive apnea, or vice versa. Examples of obstructive apnea and central apnea detected by an embodiment of the present invention are shown in FIG. 8.

[0020] Furthermore, the present embodiment includes associating respective angle information with the classified silent events by combining the audio signal and the accelerometer signal. In each of the classified silent events, the respective angle information can be associated and analyzed to determine the occurrence due to the posture of the sleep disorder. Identification of the posture represented by the angle information in a silent event that can represent apnea or hypopnea can be extremely useful in some cases. For example, a supine sleep posture is known to promote the occurrence of apnea as compared to a lateral sleep posture and a prone sleep posture. This phenomenon is known as positional obstructive sleep apnea (pOSA) and is one of the most common diseases affecting the quality of sleep. Therefore, the posture pattern revealed for the sleep disorder can be effectively used for monitoring a subject with pOSA.

[0021] FIG. 2 schematically shows the concept of posture, angle information, and motion information regarding a subject in an embodiment of the present invention. Specifically, FIG. 2 shows a sleep angle α and a standing angle β with respect to the x-axis, y-axis, and z-axis. As shown in FIG. 2, the accelerometer 102 gives a positive value when the acceleration proceeds from the right side to the left side (x-axis), from the toe to the head (y-axis), and from the front to the back (z-axis), and thus defines an X-Z plane in which the sleep angle is calculated and a Y-Z plane in which the standing angle is calculated. A sleep angle α = 0° represents a pure left sleep posture, α = 90° represents a pure supine sleep posture, α = ±180° represents a pure right sleep posture, and α = -90° represents a pure prone sleep posture. Further, a standing angle β = 0° means that the posture of the subject P is upside down, β = ±90° means a purely lying posture, and β = ±180° means a standing posture. Further, the movement of the chest caused by inhalation and exhalation of the subject P can also be explained in FIG. 2. By measuring the vertical movement that can be represented by the z-direction of the accelerometer 102 of the mobile device 10 or the repetitive up-and-down movement in other directions when the subject P is sleeping in other postures, the movement of the chest of the subject P can be recorded and further used for classification of the apnea type or other purposes.

[0022] Figure 3 shows a schematic configuration of a fixed system 2 for juxtaposing the mobile device 10 with the subject P. Specifically, the mobile device 10 may be disposed within the bag 22 and attached to the chest of the subject P by an elastic band 21, or specifically, may be attached so as to cover the sternum of the subject P.

[0023] In a further embodiment, the method includes quantifying sleep movements and sleep posture variations based on the accelerometer signal. For example, the angular information derived from the accelerometer signal can be quantified as the angular resolution related to the sleep posture. Further, the posture and / or movement information included in the accelerometer signal can be quantified as sleep posture variations.

[0024] Furthermore, the method further includes detecting the breathing pathway during sleep from the audio signal. The breathing pathway during sleep can be classified into mouth breathing or nasal breathing (see also Figure 9). Mouth breathing indicates that the subject is breathing through the mouth, while nasal breathing indicates that the subject is breathing through the nose. Considering that mouth breathing can become more dominant in subjects with OSA having nasal congestion leading to an increase in breathing through the mouth, detecting the breathing pathway by the audio signal A acquired by the mobile device 10 can be advantageous for monitoring sleep disorders.

[0025] When an audio signal is acquired, the audio signal is downsampled to 5 kHz, band-pass filtering is performed between 70 and 2000 Hz, and power line noise can be removed. If there is too much noise in the signal, an optimal filter or spectral subtraction can be applied to reduce interference and improve signal quality. Then, apnea and hypopnea events can be identified using time-domain representation, spectral representation, and time-frequency representation. The spectrogram can be used as a time-frequency representation with a window of 0.1 s, 90% overlap, and NFFT = 1024 to distinguish whether breathing is present during an event and thus classify them as apnea or hypopnea. As shown in FIG. 9, a fast Fourier transform is calculated, a linear envelope is extracted using a window of 15 Hz, and it can be checked whether there is a prominent peak between 950 Hz and 2 kHz indicating mouth breathing. The information extracted in this calculation can be used for further classification of hypopnea into nasal breathing or mouth breathing, and can even be used for detection of the breathing pathway at any point during the night. The percentage of time the subject is breathing through the mouth can be calculated.

[0026] In yet another embodiment, the method further includes obtaining the oxygen saturation level of the subject P by the pulse oximeter 3 via a wireless link to the mobile device 10. The pulse oximeter is a non-invasive device for measuring blood oxygen saturation (SpO2). Apnea and hypopnea may be associated with a local decrease in SpO2, which is a decrease in saturation followed by re-oxygenation as shown in FIG. 7. For this reason, the oxygen saturation level can be used as a measure for identifying OSA.

[0027] Furthermore, this embodiment further includes selecting a region of interest, where the region of interest is a region in the voice signal A that precedes a decrease in oxygen saturation measured by the wireless pulse oximeter 3. When the oxygen saturation level drops by more than 3%, this is recognized as an event of saturation decrease and can be used for the analysis of sleep disorders such as apnea, hypopnea, and / or snoring. For higher accuracy, the silent event may be identified only when a saturation decrease event measured by the wireless pulse oximeter 3 connected to the mobile device 10 follows.

[0028] In one embodiment, the method further includes removing artifacts by considering the accelerometer signal during the identification of silent events. When the subject P moves during sleep, the noise generated during the movement may be recorded in the voice signal A together with the breathing sound. Therefore, the accelerometer 102 can be used to identify the movement of the subject P and remove artifacts in the identification of silent events from the voice signal A.

[0029] In one embodiment, the method further includes classifying a silent event as apnea or hypopnea according to the presence of low-intensity breathing sounds in the voice signal. When low-intensity breathing sounds are present, it can be assumed that residual breathing activity exists and the corresponding silent event should be classified as hypopnea. On the other hand, when there is no breathing at all, the silent event should be labeled as apnea. Examples of apnea with no breathing at all and hypopnea with low-intensity breathing sounds can be seen in FIGS. 7 and 8.

[0030] In another embodiment, the method further includes providing the classified silent events on the mobile device 10 together with the respective associated angle information. The classified silent events with their respective associated angle information can be provided on the mobile device 10. Specifically, the silent events recorded and classified at night can be shown in a smartphone application for easier sleep monitoring of the subject P.

[0031] In another embodiment, the method further includes outputting vibration and / or sound from the mobile device 10 based on the angle information and the oxygen saturation level and / or the audio signal. For example, when the subject P uses the mobile device 10 with the method provided in this embodiment, the method can provide a warning to the subject P in a specific position, such as the supine position, which may promote the occurrence of apnea compared to the sleeping positions of the lateral recumbent position and the prone position. For this postural therapy, the mobile device 10 is configured to output vibration and / or sound to warn the subject P. By prompting the subject P to change the sleeping position, it is possible to avoid sleeping in the supine position by postural therapy.

[0032] In one embodiment, the method further includes detecting heart activity from the accelerometer 102, as shown in FIG. 10. For example, the heart activity detected from the accelerometer 102 can include detecting changes in the heart rate of the subject. Then, the method further includes associating the relationship between the change in the heart rate of the subject and the silent event.

[0033] The present invention can also be used to monitor the sleep quality of the subject P. Monitoring the sleep quality can be identified in the same way as monitoring sleep disorders. However, in this embodiment, sleep events may be identified instead of silent events.

[0034] For example, the method according to this embodiment includes the same steps as the method for monitoring sleep disorders until the identification of sleep events. Then, the method according to this embodiment identifies sleep events including snoring episodes based on the audio signal A acquired by the microphone 101 of the mobile device 10. For example, the arrow in FIG. 7 indicates a snoring episode.

[0035] When a sleep event is identified, based on the angle information of the subject P derived from the accelerometer signal, the body position where the sleep event occurred is identified.

[0036] This embodiment further includes automatic detection of multimodal features and labeling of sleep events based on the multimodal features. The multimodal features are derived from an audio signal A, an accelerometer signal, and an oxygen saturation level.

[0037] This embodiment further includes detecting heart activity from the accelerometer 102 of the mobile device 10. The accelerometer signal automatically detected as part of the multimodal features can include higher frequency components compared to the movement from the accelerometer 102. Therefore, as shown in FIG. 10, it is possible to detect heart activity from the accelerometer 102 when monitoring the sleep quality of the subject P. Further, this embodiment can associate the relationship between the change in the heart rate of the subject and the sleep events. Further, the detection of heart activity may be performed in the same manner as in the method for monitoring sleep disorders of the subject P. In this case, an association can be made between the change in heart rate and the silent event. However, the detection of heart activity may be performed in the same manner in both the monitoring of sleep disorders and the monitoring of sleep quality.

[0038] FIG. 4 shows a schematic diagram of the configuration of a system for monitoring the sleep disorders and sleep quality of a subject. In this embodiment described in FIG. 4, the system 4 includes a mobile device 10, a fixed system 2, and a pulse oximeter 3. Although shown as a wireless pulse oximeter in FIG. 4, the pulse oximeter 3 may be wired or wireless and is configured to measure the oxygen saturation level of the subject P. The mobile device 10 of the system 4 includes a microphone 101 configured to acquire an audio signal A indicating the respiration of the subject P, and an accelerometer 102 configured to acquire an accelerometer signal including the body position, angle, and movement information of the subject P. The system 4 further includes a fixed system 2 as described in FIG. 3.

[0039] In a further embodiment of the present invention, various multimodal features can be extracted from audio, accelerometer, and pulse oximeter signals. In such an embodiment, a method for monitoring the quality of a subject's sleep is provided, the method comprising using a mobile device and co-locating the mobile device with the subject, obtaining an audio signal indicative of the subject's breathing by a microphone of the mobile device, obtaining an accelerometer signal including the subject's body position, angle, and motion information by an accelerometer of the mobile device, measuring chest motion related to breathing and sleep disordered breathing by the accelerometer of the mobile device, identifying a plurality of sleep events from the audio signal, identifying the body position at which a sleep event occurred based on the angle information, labeling the sleep events with the multimodal features, and automatically detecting multimodal sleep features, the multimodal features being the acquired audio signal, the acquired accelerometer signal, and the oxygen saturation level acquired by a wireless pulse oximeter via a wireless link to the mobile device.

[0040] In such an embodiment, the detection of multimodal features can be automatic since it can be automatically calculated by a custom-made computational algorithm that does not require manual review or annotation of the signals. Several features can be extracted, including features that describe desaturation events (e.g., depth or duration of desaturation), features related to apnea and hypopnea (metrics that count the number of apneas and / or hypopneas, duration, whether snoring is included, etc.), snoring (from vibrations in the audio signal or accelerometer signal), features related to movement, sleep position, or sleep stage, features related to cardiac activity, etc. It should be noted that the "labeling" of sleep events is to be understood as being annotated in the sense of identification and classification to a sleep event, for example, there is an event that starts at time = 1100 seconds and ends at time = 1122 seconds and it is indicated that it corresponds to an obstructive apnea. Further, "multimodal features" should be understood to be more general, and "multimodal sleep features" can indicate that they are related to aspects of sleep.

[0041] Figure 7 shows signals recorded by a system according to an embodiment of the present invention. Specifically, Figure 7 shows a comparison between signals recorded from a classical sleep polysomnography and signals recorded from this system. Among the various signals that can be recorded from a classical sleep polysomnography, nasal airflow, thoracic activity, and the SpO2 channel are shown in the upper plots of Figure 7. In contrast, in this system, voice, acceleration measurement, and SpO2 are measured (see the lower plots of Figure 7). As shown in the measured acceleration measurement, triaxial accelerometer values may be recorded, and the angular information may be derived from the triaxial accelerometer values of the measured acceleration measurement. As shown in Figure 7, apnea episodes detected and recorded in a sleep polysomnography can also be identified by this system. The dotted squares represent examples of apnea. Similarly, hypopnea and snoring episodes can be identified. The gray squares represent hypopnea, and the arrows represent snoring.

[0042] Figure 8 shows the types of apnea and hypopnea detected by a system according to an embodiment of the present invention. When a silent event is detected, the silent event can be classified as either apnea or hypopnea. If there is a breathing sound or a snoring sound in the voice signal, the event can be classified as hypopnea, while if there is only silence or an artifact, the event can be classified as apnea.

[0043] As shown in Figures 7 and 8 and as can be understood from the above description, the expression "silent event" may not be limited to an event where no sound is detected at all. Since a silent event can be recognized based on the sample entropy value, an event where the sample entropy value is smaller than a threshold value can be identified as a silent event. A silent event with a sample entropy smaller than the threshold value may mean the presence of low-intensity breathing sounds in the voice signal.

[0044] In the episode of the silent event shown in FIGS. 7 and 8, there is an audio signal indicating the presence of breath sound or snoring sound. Based on this sound and the respective angle information, the event can be classified into different types. For example, in hypopnea and obstructive apnea, there is a decrease in chest wall movement as indicated by the angular variation derived from the triaxial accelerometer data. In central apnea, there is no chest wall movement as indicated by the flat region of the angle information.

[0045] FIG. 9 shows the difference between nasal breathing and mouth breathing. According to an embodiment of the present invention, different breathing paths can be distinguished based on the spectral content of the audio signal. The upper panel of FIG. 9 shows the signal in the time domain, and the lower plot shows the respective fast Fourier transform (FFT) of them with an envelope line in dotted line. Different from nasal breathing, mouth breathing has a prominent peak at a higher frequency, i.e., about 1.5 kHz, so nasal breathing and mouth breathing can be distinguished by this system.

[0046] FIG. 10 shows a comparison of the heart activity measured by the accelerometer of the system according to an embodiment of the present invention with the electrocardiogram (ECG) recorded by a classical sleep polysomnogram. The upper plot of FIG. 10 shows the ECG from the classical sleep polysomnogram, while the lower plot shows the movement along the z-axis measured by the accelerometer of the system including the peaks corresponding to the heart activity. Therefore, the accelerometer of this system can identify the heart activity.

[0047] FIG. 11 shows an example of a report that can be created by the system according to an embodiment of the present invention. For example, the report may be shown in the smartphone application of the mobile device or may be created so as to be downloaded therefrom.

[0048] In the report shown in FIG. 11, the AHI represents the apnea-hypopnea index, which is the number of silent events, i.e., the number of episodes of absence or significant reduction of respiratory flow per sleep time. The absence of respiratory flow can mean apnea, and a significant reduction in respiratory flow can mean hypopnea.

[0049] The method according to the present invention can be used in various use cases such as monitoring the quality of healthy sleep in the general population, early detection of sleep disorders, monitoring of sleep apnea, long-term follow-up, and providing personalized postural therapy.

[0050] The present invention is considered to be beneficial to various types of users including healthy subjects, i.e., the general population, and patients with sleep apnea. In addition, the present invention is considered to be particularly beneficial to patients with coexisting diseases that lead to an increased risk of sleep disorders, such as patients with spinal cord injuries, patients after stroke, patients with chronic obstructive pulmonary disease (COPD), and patients with other respiratory diseases, neuropathies, or mental diseases.

[0051] Although detailed embodiments have been described, these only serve to provide a better understanding of the present invention as defined by the independent claims and should not be regarded as limiting.

Description of Reference Numerals

[0052] A Audio signal P, P’ Subject 10 Mobile device 101 Microphone 102 Accelerometer 2 Fixing system 21 Elastic band 22 Bag 3 Pulse oximeter 4 System

Claims

1. A method for monitoring sleep disorders in subjects using mobile devices, Placing the aforementioned mobile device next to the subject, The microphone of the mobile device acquires an audio signal indicating the subject's breathing, The accelerometer signal is acquired by the accelerometer of the mobile device, and the subject's body position, angle, and motion information is derived from the accelerometer signal. The accelerometer is used to measure thoracic movement related to respiration in respiratory and sleep disorders, To identify multiple silent events based on the aforementioned audio signal, The aforementioned silent events are classified as apnea or hypopnea, Based on the aforementioned chest wall movement, the classification of the apnea type is identified. By combining the aforementioned audio signal and the aforementioned accelerometer signal, the angle information is associated with the classified silent events. A method for monitoring sleep disorders in subjects, including the following.

2. Based on the accelerometer signal, the sleep movements and changes in sleep position are quantified, To detect the respiratory pathway during sleep from the aforementioned audio signal. A method for monitoring sleep disorders in a subject according to claim 1, further comprising:

3. The oxygen saturation level of the subject is obtained by a wireless pulse oximeter via a wireless link to the aforementioned mobile device, Selecting a region of interest from the aforementioned audio signal that precedes the decrease in oxygen saturation measured by the wireless pulse oximeter. A method for monitoring sleep disorders in a subject according to claim 1 or 2, further comprising:

4. Artifacts are removed by considering the accelerometer signal during the identification of the aforementioned silent event. A method for monitoring sleep disorders in a subject according to claim 1 or 2, further comprising:

5. The silent event is classified as either apnea or hypopnea depending on the presence of low-intensity breathing sounds in the audio signal. A method for monitoring sleep disorders in a subject according to claim 1 or 2, further comprising:

6. The classified silent events, along with their associated angle information, are provided on the mobile device. A method for monitoring sleep disorders in a subject according to claim 1 or 2, further comprising:

7. To output vibration and / or sound from the mobile device based on the angle information and the acidity saturation level and / or the audio signal for postural therapy. A method for monitoring sleep disorders in a subject according to claim 1 or 2, further comprising:

8. The aforementioned apnea types are obstructive, central, or mixed. The aforementioned respiratory pathway is the mouth or nose. The angle information is one angle, multiple angles, or three-axis accelerometer values. A method for monitoring sleep disorders in a subject according to claim 1 or 2.

9. To detect cardiac activity from the aforementioned accelerometer, To correlate the changes in the heart rate of the aforementioned subjects with the silent events. A method for monitoring sleep disorders in a subject according to claim 1 or 2, further comprising:

10. A method for monitoring the sleep quality of subjects using mobile devices, Placing the aforementioned mobile device next to the subject, The microphone of the mobile device acquires an audio signal indicating the subject's breathing, The accelerometer signal is acquired by the accelerometer of the mobile device, and the subject's body position, angle, and motion information is derived from the accelerometer signal. The accelerometer is used to measure thoracic movement related to respiration in respiratory and sleep disorders, Based on the aforementioned audio signals, identify sleep events including snoring episodes, Based on the angle information, the body position in which the sleep event occurred is identified. A method for monitoring the sleep quality of subjects, including the following.

11. A method for monitoring the sleep quality of a subject according to claim 10, wherein identifying the aforementioned sleep event includes identifying a snoring episode.

12. Automatic detection of multimodal features in sleep episodes, and Labeling of sleep events based on the multimodal features It further includes, The method for monitoring the sleep quality of a subject according to claim 10 or 11, wherein the multimodal features are derived from the audio signal, the accelerometer signal, and the oxygen saturation level acquired by a wireless pulse oximeter via a wireless link to the mobile device.

13. To detect cardiac activity from the aforementioned accelerometer, To correlate the changes in the heart rate of the aforementioned subjects with sleep events. A method for monitoring the sleep quality of a subject according to claim 10 or 11, further comprising:

14. A system for monitoring sleep disorders and sleep quality in subjects, A microphone configured to acquire audio signals indicating the breathing of the subject, An accelerometer configured to acquire an accelerometer signal including the subject's body position, angle, and motion information, Mobile devices equipped with, A fixed system for placing the aforementioned mobile device alongside the subject, A pulse oximeter configured to measure the oxygen saturation level of the subject, It is equipped with, The aforementioned fixing system comprises an elastic band and a bag for holding the mobile device.

Images