SYSTEM AND METHOD FOR CLASSIFYING AND / OR CHARACTERIZING USER INTERFACES - Patent application

Abstract

Systems and methods are disclosed for classifying and / or characterizing a user interface. The systems and methods include generating acoustic data associated with acoustic reflections of an acoustic signal, the acoustic reflections at least partially indicative of one or more characteristics of a user interface coupled to a respiratory treatment device via a conduit. The systems and methods further include analyzing the generated acoustic data. The systems and methods further include characterizing the user interface based at least in part on the analyzed acoustic data.

Description

[Technical field]

[0001] The present disclosure relates generally to systems and methods for classifying and / or characterizing user interfaces, and more particularly, to systems and methods for classifying and / or characterizing user interfaces based on acoustic reflections of acoustic signals. [Background technology]

[0002] Many people suffer from sleep-related and / or respiratory disorders such as periodic limb movement disorder (PLMD), restless legs syndrome (RLS), sleep-disordered breathing (SDB), obstructive sleep apnea (OSA), central sleep apnea (CSA), as well as other types of apnea such as mixed apnea and hypopnea, respiratory effort-related arousals (RERA), Cheyne-Stokes respiration (CSR), respiratory insufficiency, obesity hyperventilation syndrome (OHS), chronic obstructive pulmonary disease (COPD), neuromuscular disorders (NMD), rapid eye movement (REM) behavior disorder (also known as RBD), dream behavioralization (DEB), hypertension, diabetes, stroke, insomnia, and chest wall disorders. These disorders are typically treated using respiratory treatment systems. Respiratory treatment systems can be used to treat sleep-disordered breathing. Each respiratory treatment system typically has a respiratory treatment device connected to a user interface (e.g., a mask) via a conduit. A user wears the user interface and is provided with a pressurized airflow from the respiratory treatment device via a conduit.

[0003] A user interface is generally a particular category and type of user interface for a user, with categories of user interfaces including, for example, direct or indirect connection, and types of user interfaces including, for example, full face mask, partial face mask (e.g., a mask that covers the mouth but not the nose or covers only a portion of the nose), nasal mask, nasal pillows, etc. In addition to the particular category and type, a user interface is typically a specific model produced by a particular manufacturer. It is beneficial for a respiratory treatment system to know the particular category and type, and therefore the specific model, of the user interface worn by a user for various reasons, including ensuring that the user is using the correct user interface.

[0004] A variety of different user interfaces may be used, such as nasal pillows, nasal masks, nose and mouth masks (e.g., full face masks or partial face masks). In some implementations, different forms of air delivery conduits may be used. It is advantageous to know the user interface and conduits connected to the respiratory treatment device to improve control of the treatment provided to the user. For example, it may be advantageous to measure or estimate treatment parameters such as pressure in the mask and airflow. Thus, knowing which user interface is being used may enhance treatment.

[0005] Although the respiratory device may include a menu system that allows the user to enter the type of user interface being used, for example type, model, manufacturer, etc., sometimes users enter incorrect or incomplete information.

[0006] The present disclosure solves these and other problems by classifying and / or characterizing user interfaces based on analyzed acoustic data generated from acoustic reflections that are indicative of one or more characteristics of the user interface. Summary of the Invention [Means for solving the problem]

[0007] According to some implementations of the present disclosure, a method includes generating acoustic data related to an acoustic reflection of an acoustic signal. The acoustic reflection is at least partially indicative of one or more features of a user interface coupled to a respiratory treatment device via a conduit. The method further includes analyzing the generated acoustic data. The analysis includes windowing the generated acoustic data based at least in part on at least one of the one or more features of the user interface. The method further includes characterizing the user interface based at least in part on the analyzed acoustic data.

[0008] According to some aspects of the method, the windowing of the generated acoustic data includes determining a reference point within the generated acoustic data. According to some aspects of the method, the reference point is a minimum point within a predetermined interval of a deconvolution of the generated acoustic data. According to some aspects of the method, the reference point is a maximum point within a predetermined interval of a deconvolution of the generated acoustic data. According to some aspects of the method, the reference point corresponds to a position of one or more features along a path formed at least in part by the conduit and the user interface. According to some aspects of the method, the one or more features cause a change in acoustic impedance. According to some aspects of the method, the change in acoustic impedance is based at least in part on a narrowing of the path at (i) the user interface, (ii) a junction of the conduit and the user interface, or (iii) a combination thereof. According to some aspects of the method, the change in acoustic impedance is based at least in part on a widening of the path at (i) the user interface, (ii) a junction of the conduit and the user interface, or (iii) a combination thereof. According to some aspects of the method, the windowing includes a first windowing of the generated acoustic data and a second windowing of the generated acoustic data. According to some aspects of the method, the first windowing and the second windowing vary depending on an amount of generated acoustic data selected to be before a reference point in the generated acoustic data, after a reference point in the generated acoustic data, or a combination thereof. The amount of generated data may be a predetermined number of data points / samples, a predetermined distance, a predetermined quefrency, etc. According to some aspects of the method, analyzing the generated acoustic data includes computing a deconvolution of the generated acoustic data prior to windowing the generated acoustic data. According to some aspects of the method, deconvolution of the generated acoustic data includes computing a cepstrum of the generated acoustic data.According to some aspects of the method, the cepstrum identifies distances associated with the acoustic reflections, which are indicative of a location of one or more features of the user interface relative to the acoustic sensor along an acoustic path traveled by the acoustic signal. The cepstrum can identify one or more distances associated with each of the corresponding one or more acoustic reflections, thereby indicative of a location of each of the one or more physical features along the acoustic path of the acoustic signal. According to some aspects of the method, analyzing the generated acoustic data includes calculating a derivative of the cepstrum to determine a rate of change of the cepstrum signal. According to some aspects of the method, analyzing the generated acoustic data includes normalizing the generated acoustic data. According to some aspects of the method, normalizing the generated acoustic data includes subtracting a mean value from the generated acoustic data, dividing by a standard deviation of the generated acoustic data, or a combination thereof. According to some aspects of the method, normalizing the generated acoustic data accounts for confounding conditions. According to some aspects of the method, the confounding conditions are due to microphone gain, respiration amplitude, treatment pressure, or a combination thereof. According to some aspects of the method, characterizing the user interface includes inputting the analyzed acoustic data into a machine learning model. In one or more implementations, the machine learning model can be supervised or unsupervised. In one or more implementations, the machine learning model can be a neural network. In one or more implementations, the neural network can be a deep neural network or a shallow neural network. In one or more implementations, the deep neural network can be a convolutional neural network. The machine learning model, such as a deep neural network or a convolutional neural network, can determine a form factor of the user interface, a model of the user interface, a size of one or more elements of the user interface, or a combination thereof. According to some aspects of the method, the deep neural network includes one or more convolutional layers and one or more max pooling layers.According to some aspects of the method, the convolutional neural network includes N features with a maximal pooling of M samples of the N features, where the ratio of N to M is 1:1 to 4:1. According to some aspects of the method, the method further includes emitting an acoustic signal to a conduit connected to the user interface via an acoustic transducer. According to some aspects of the method, the method further includes emitting an acoustic signal to the conduit via a motor of a respiratory treatment device connected to the conduit. According to some aspects of the method, the acoustic signal is a sound audible to a human. According to some aspects of the method, the acoustic signal is a sound inaudible to a human. According to some aspects of the method, whether the sound is audible is determined based on a frequency of the sound, an amplitude of the sound, or a combination thereof. According to some aspects of the method, the acoustic signal is an ultrasonic sound. According to some aspects of the method, the user interface is not connected to the user during generation of the acoustic data. In this configuration, certain confounding factors, such as undesirable noises generated by the user's breathing or the respiratory treatment device, may be avoided. According to some aspects of the method, the user interface is connected to the user during generation of the acoustic data. According to some aspects of the method, the method further includes providing a flow of pressurized air through the conduit and into the user interface during generation of the acoustic data. According to some aspects of the method, the method further includes emitting an acoustic signal in the conduit during the absence of a flow of pressurized air through the conduit and into the user interface. According to some aspects of the method, the generated acoustic data is generated from a plurality of acoustic reflections from a plurality of acoustic signals. According to some aspects of the method, the generated acoustic data is an average of a plurality of acoustic reflections from the plurality of acoustic signals. This averaging may improve the signal-to-noise ratio of the generated acoustic data by suppressing undesirable time-varying components or artifacts in the acoustic data.

[0009] According to some implementations, a system is disclosed that includes a control system having one or more processors and a memory having machine-readable instructions stored therein, the control system being coupled to the memory such that the machine-executable instructions in the memory, when executed by at least one of the one or more processors of the control system, can implement any of the aspects described above.

[0010] According to some implementations, a system for characterizing a user interface is disclosed, the system including a control system having one or more processors configured to implement the method of any one of the above aspects.

[0011] According to some implementations, a computer program product is disclosed that includes instructions that, when executed by a computer, cause the computer to perform the method of any one of the above aspects. According to some aspects, the computer program product is a non-transitory computer-readable medium.

[0012] According to some implementations of the present disclosure, a method includes generating acoustic data related to an acoustic reflection of an acoustic signal. The acoustic reflection is at least partially indicative of one or more characteristics of a user interface coupled to a respiratory treatment device via a conduit. The method further includes analyzing the generated acoustic data to identify one or more signatures that correlate to the one or more characteristics of the user interface. In general, the signatures can consist of one or more acoustic features in the acoustic data. The method further includes classifying the user interface based at least in part on the one or more signatures.

[0013] According to some aspects of the method, the category of the user interface is associated with a direct connection between the conduit and a cushion and / or frame of the user interface. According to some aspects of the method, the one or more features and the one or more signatures indicate a direct connection between the conduit and the cushion and / or frame. According to some aspects of the method, the category of the user interface is associated with an indirect connection between the conduit and a cushion and / or frame of the user interface. According to some aspects of the method, the one or more features and the one or more signatures indicate an indirect connection between the conduit and the cushion and / or frame. According to some aspects of the method, the indirect connection is characterized by a further conduit arranged and configured to provide a fluid connection between the conduit and the cushion and / or frame. According to some aspects of the method, the further conduit is a headgear conduit, the headgear conduit forming part of a headgear for holding the user interface on a face of a user. The headgear conduit is configured to deliver pressurized air from the conduit to the cushion and / or frame. According to some aspects of the method, the further conduit is a user interface conduit that (i) is more flexible than the conduit, (ii) has a diameter smaller than the diameter of the conduit, or has both (i) and (ii). The user interface conduit may have a length shorter than the conduit. The user interface conduit is configured to deliver pressurized air from the conduit to the cushion and / or frame. According to some aspects of the method, analyzing the generated acoustic data includes calculating a spectrum of the generated acoustic data. According to some aspects of the method, analyzing the generated acoustic data includes calculating a logarithm of the spectrum. Alternatively, non-linear operations can be used instead of a logarithmic calculation. According to some aspects of the method, analyzing the generated acoustic data includes calculating a cepstrum of the logarithmic spectrum.According to some aspects of the method, analyzing the generated acoustic data includes selecting a segment of a log spectrum of the generated acoustic data, computing a Fourier transform of the segment of the log spectrum, and associating one or more features of the user interface with one or more signatures in the Fourier transform of the segment. According to some aspects of the method, the one or more signatures include a maximum amplitude of the cepstrum, a minimum amplitude of the cepstrum, a standard deviation of the cepstrum, a skewness of the cepstrum, a kurtosis of the cepstrum, a median absolute value of the cepstrum, a sum of absolute values ​​of the cepstrum, a sum of positive areas of the cepstrum, a sum of negative areas of the cepstrum, a fundamental frequency, an energy corresponding to the fundamental frequency, an average energy, at least one resonant frequency of the combination of the conduit and the user interface, a change in at least one resonant frequency, a number of peaks in a range, a peak prominence, a distance between peaks, or a combination or change thereof. According to some aspects of the method, the generated acoustic data includes a first reflection in the acoustic reflections of the acoustic signal. According to some aspects of the method, the generated acoustic data includes a second reflection within the acoustic reflection of the acoustic signal. According to some aspects of the method, the generated acoustic data includes a third reflection within the acoustic reflection of the acoustic signal. According to some aspects of the method, the generated acoustic data includes a first reflection and a second reflection within the acoustic reflection of the acoustic signal. According to some aspects of the method, a user interface category associated with a direct connection between the conduit and the cushion and / or frame is classified based on one or more signatures having a maximum cepstrum amplitude greater than a threshold. According to some aspects of the method, a user interface category associated with an indirect connection between the conduit and the cushion and / or frame via a user interface conduit is classified based on one or more signatures meeting a threshold number of peaks. According to some aspects of the method, a user interface category associated with an indirect connection between the conduit and the cushion and / or frame via a user interface conduit is classified based on one or more signatures having a maximum cepstrum amplitude less than a threshold.According to some aspects of the method, a user interface category associated with an indirect connection between the conduit and the cushion and / or frame via the headgear conduit is classified based on one or more signatures having a mean cepstrum value below a threshold. According to some aspects of the method, classifying the user interface based at least in part on the one or more signatures includes comparing the one or more signatures to one or more known signatures of one or more user interfaces of known categories.

[0014] According to some implementations, a system is disclosed that includes a memory storing machine-readable instructions and a control system including one or more processors configured to execute the machine-readable instructions to perform any one or more of the methods and / or method implementations described above.

[0015] The above summary is not intended to describe each embodiment or every aspect of the present disclosure. Rather, the foregoing summary merely provides examples of some of the novel aspects and features described herein. The above and other features and advantages of the present disclosure will be readily apparent from the following detailed description of exemplary embodiments and modes for carrying out the invention, taken in conjunction with the accompanying drawings and the appended claims. Additional aspects of the present disclosure will be apparent to those skilled in the art in view of the detailed description of various embodiments, which are provided below with reference to the drawings, a brief description of which is provided below. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a functional block diagram of a system for classifying and / or characterizing user interfaces according to some implementations of the present disclosure. [Diagram 2] FIG. 2 is a perspective view of at least a portion of the system of FIG. 1 according to some implementations of the present disclosure. [Diagram 3] FIG. 3 illustrates an example timeline of a sleep session, according to some implementations of the disclosure. [Figure 4]FIG. 4 shows an example hypnogram associated with the sleep session of FIG. 3, according to some implementations of the present disclosure. [Diagram 5] FIG. 5 illustrates generation of acoustic data responsive to acoustic reflections indicative of one or more features of a user interface according to some implementations of the present disclosure. [Figure 6] FIG. 6 is a flow diagram of a method for characterizing a user interface according to some implementations of the present disclosure. [Figure 7] FIG. 7 is a plot of a calculated cepstrum of generated acoustic data according to some implementations of the present disclosure. [Figure 8] FIG. 8 is a plot of a windowed cepstrum of generated acoustic data according to some implementations of the present disclosure. [Figure 9] FIG. 9 is a plot of an aligned windowed cepstrum of generated acoustic data according to some implementations of the present disclosure. [Figure 10] FIG. 10 is a plot of a normalized windowed cepstrum of generated acoustic data according to some implementations of the present disclosure. [Figure 11A] FIG. 11A is a perspective view of one category of user interfaces according to some implementations of the present disclosure. [Figure 11B] FIG. 11B is an exploded view of the user interface of FIG. 11A according to some implementations of the present disclosure. [Figure 12A] FIG. 12A is a perspective view of another category of user interface according to some implementations of the present disclosure. [Figure 12B] FIG. 12B is an exploded view of the user interface of FIG. 12A according to some implementations of the present disclosure. [Figure 13A] FIG. 13A is a perspective view of another category of user interface according to some implementations of the present disclosure. [Figure 13B] FIG. 13B is an exploded view of the user interface of FIG. 13A according to some implementations of the present disclosure. [Figure 14A] FIG. 14A is a flow diagram of a method for characterizing a user interface according to some implementations of the present disclosure. [Figure 14B] FIG. 14B is a flow diagram of a method for analyzing generated acoustic data according to some implementations of the present disclosure. [Figure 15] FIG. 15 is a plot of a cepstrum computed for a user interface according to some implementations of the present disclosure. [Figure 16] FIG. 16 is a plot of an additionally calculated cepstrum for a user interface according to some implementations of the present disclosure. [Figure 17] FIG. 17 is a plot of an additionally calculated cepstrum for a user interface according to some implementations of the present disclosure. [Figure 18] FIG. 18 is a plot of an additionally calculated cepstrum for a user interface according to some implementations of the present disclosure. [Figure 19] FIG. 19 is a plot of an additionally calculated cepstrum for a user interface according to some implementations of the present disclosure. [Figure 20] FIG. 20 is a plot of an additionally calculated cepstrum for a user interface according to some implementations of the present disclosure. [Figure 21] FIG. 21 is a plot of a calculated cepstrum showing first and second order reflections within the cepstrum in accordance with some implementations of the present disclosure. [Figure 22] FIG. 22 is a plot of a log spectrum of generated acoustic data in accordance with some implementations of the present disclosure. [Diagram 23] FIG. 23 is a plot of a transform of a segment of the log spectrum shown in FIG. 22 in accordance with some implementations of the present disclosure. [Figure 24A] FIG. 24A is a plot of a frequency domain intensity signal based on generated acoustic data in accordance with some implementations of the present disclosure. [Figure 24B]FIG. 24B is a plot of a wave period intensity signal based on the frequency domain intensity signal of FIG. 24A in accordance with some implementations of the present disclosure. [Figure 25A] FIG. 25A shows a first plot of a cepstrum calculated from acoustic data generated using a first conduit and a second plot of a cepstrum calculated from acoustic data generated using a second conduit, according to some implementations of the present disclosure. [Figure 25B] FIG. 25B shows a first plot of a corrected cepstrum based on the first cepstrum of FIG. 25A and a second plot of a corrected cepstrum based on the second cepstrum of FIG. 25A according to some implementations of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] The present disclosure, as well as its advantages and advantages, will be better understood from the following description of exemplary embodiments with reference to the accompanying drawings, which depict only exemplary embodiments and therefore should not be considered limiting of the various embodiments or claims.

[0018] While the invention is susceptible to various modifications and alternative forms, specific implementations of the invention have been illustrated in the drawings and are described in further detail herein. It should be understood, however, that it is not intended to be limited to the particular forms disclosed. Rather, the invention covers all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.

[0019] Various embodiments are described with reference to the accompanying drawings, in which the same or equivalent components are provided with the same reference numerals in each figure. The figures are not drawn to scale and are provided merely to illustrate the invention. Some aspects of the invention are described below with reference to illustrative example applications. It should be understood that numerous specific details, relationships and methods are described to provide a thorough understanding of the invention. However, one of ordinary skill in the art will readily recognize that the invention can be practiced without one or more of the specific details, or with other methods. In other instances, well-known structures or operations have not been shown in detail to avoid obscuring the invention. Various embodiments are not limited by the described order of acts or events, as some acts may occur in different orders and / or simultaneously with other acts or events. Moreover, not all described acts or events are required to implement a methodology in accordance with the present invention.

[0020] For example, elements and limitations described in the Abstract, Summary of the Invention, and Detailed Description of the Invention that are not explicitly stated in the claims should not be incorporated into the claims singularly, comprehensively, implicitly, inferentially, or otherwise. For the sake of detailed description herein, the singular includes the plural, and vice versa, unless otherwise expressly stated. The word "including" means "including, but not limited to." Furthermore, approximation words such as "about," "approximately," "substantially," "approximately," and "generally" can be used herein to mean, for example, "at," "near," "near," "3-5%," "within manufacturing tolerance," or logical combinations thereof.

[0021] Many people suffer from sleep-related and / or breathing-related disorders such as Periodic Limb Movement Disorder (PLMD), Restless Legs Syndrome (RLS), Sleep-Disordered Breathing (SDB), Obstructive Sleep Apnea (OSA), Central Sleep Apnea (CSA), and other types of apnea such as mixed apneas and hypopneas, Respiratory Effort Related Arousals (RERA), Cheyne-Stokes Respiration (CSR), Respiratory Insufficiency, Obesity Hyperventilation Syndrome (OHS), Chronic Obstructive Pulmonary Disease (COPD), Neuromuscular Disorders (NMD), Rapid Eye Movement (REM) Behavior Disorder (also known as RBD), Dream Acting Out (DEB), Hypertension, Diabetes, Stroke, Insomnia, and Chest Wall Disorders.

[0022] Obstructive sleep apnea (OSA) is a form of sleep-disordered breathing (SDB) that is characterized by events such as obstruction or blockage of the upper airway during sleep due to an abnormally small upper airway in combination with normal loss of muscle tone in the region of the tongue, soft palate, and posterior oropharyngeal wall.

[0023] Central sleep apnea (CSA) is another form of SDB that occurs when the brain temporarily stops sending signals to the muscles that control breathing. More generally, apnea generally refers to the cessation of breathing or the cessation of respiratory function caused by air obstruction. Typically, an individual stops breathing for about 15 to about 30 seconds during an obstructive sleep apnea event. Mixed sleep apnea is another form of SDB that is a combination of OSA and CSA.

[0024] Other types of apnea include hypopnea, hyperpnea, and hypercapnia. Hypopnea is generally characterized by slow or shallow breathing caused by a narrowed rather than obstructed airway. Hyperpnea is generally characterized by an increase in the depth and / or rate of breathing. Hypercapnia is generally characterized by a sudden or excessive amount of carbon dioxide in the bloodstream, usually resulting from insufficient breathing.

[0025] Respiratory Effort-Related Arousals (RERA) events are typically characterized by increased respiratory effort for 10 seconds or more that results in arousal from sleep, and do not meet the criteria for apnea or hypopnea events. In 1999, the AASM Task Force defined RERA as "a series of breaths characterized by increased respiratory effort that results in arousal from sleep, but does not meet the criteria for apnea or hypopnea." These events must meet the following criteria: 1. A pattern of gradually increasing negative esophageal pressure terminated by an abrupt change in the negative pressure level and arousal, and 2. The event must last for 10 seconds or more. In 2000, a study conducted at the New York University School of Medicine and published in Sleep, vol. 23, No. 6, pp. 763-771, entitled "Non-Invasive Detection of Respiratory Effort-Related Arousals (RERAs) by a Nasal Cannula / Pressure Transducer System," demonstrated that the nasal cannula / pressure transducer system was adequate and reliable in detecting RERAs. The RERA detector may be based on an actual flow signal derived from a respiratory therapy (e.g., PAP) device. For example, a measure of flow limitation may be determined based on the flow signal. A measure of arousal may then be derived as a function of the measure of flow limitation and the measure of ventilation surge. Such methods are described in WO 2008 / 138040, U.S. Pat. No. 9,358,353, U.S. Pat. No. 10,549,053, and U.S. Pat. App. Pub. No. 2020 / 0197640, each of which is fully incorporated herein by reference.

[0026] Cheyne-Stokes respiration (CSR) is another form of SDB. CSR is a disorder of the patient's respiratory control that results in alternating periods of waxing and waning ventilation, known as CSR cycles. CSR is characterized by repeated deoxygenation and reoxygenation of arterial blood.

[0027] Obesity hyperventilation syndrome (OHS) is defined as the combination of severe obesity and chronic awake hypercapnia in the absence of any other clear cause of hypoventilation. Symptoms include dyspnea, morning headache, and excessive daytime sleepiness.

[0028] Chronic obstructive pulmonary disease (COPD) encompasses any of a group of diseases of the lower respiratory tract that share certain characteristics, such as increased resistance to air movement, a prolonged expiratory phase of breathing, and loss of the normal elasticity of the lungs.

[0029] Neuromuscular disorders (NMDs) encompass a number of diseases and illnesses that impair muscle function directly through intrinsic muscle pathology or indirectly through neuropathology. Chest wall disorders are a group of thoracic deformities that result in inefficient coupling between the respiratory muscles and the rib cage.

[0030] These and other disorders are characterized by specific events that occur while an individual is sleeping (e.g., snoring, apnea, hypopnea, restless legs, sleep disturbances, choking, increased heart rate, labored breathing, asthma attacks, epileptic episodes, seizures, or any combination thereof).

[0031] Apnea-hypopnea index (AHI) is an index used to indicate the severity of sleep apnea during a sleep session. The AHI is calculated by dividing the number of apnea and / or hypopnea events experienced by a user during a sleep session by the total number of hours of sleep in that sleep session. The events may be, for example, pauses in breathing lasting at least 10 seconds. An AHI of less than 5 is considered normal. An AHI between 5 and 15 is considered to indicate mild sleep apnea. An AHI between 15 and 30 is considered to indicate moderate sleep apnea. An AHI of 30 or greater is considered to indicate severe sleep apnea. In children, an AHI greater than 1 is considered abnormal. Sleep apnea can be considered "controlled" when the AHI is normal or when the AHI is normal or mild. The AHI can also be used in combination with oxygen desaturation to indicate the severity of obstructive sleep apnea.

[0032] In general, the present disclosure describes systems and methods that utilize acoustic reflections to analyze a user interface connected to a respiratory treatment device via a conduit and classify and / or characterize the user interface. Classifying the user interface determines a category of the user interface directly or indirectly. As described in more detail below, an indirect user interface may be an indirect conduit or indirect frame user interface. Characterizing the user interface determines a particular type of user interface, e.g., a manufacturer and a particular model of the user interface. In one or more implementations, the user interface may be classified according to the disclosed methods without being characterized. Alternatively, in one or more implementations, the user interface may be characterized according to the disclosed methods without being directly classified. In that case, the user interface is indirectly classified through a characterization that implicitly identifies a category. Alternatively, in one or more implementations, the user interface may be (directly) classified and characterized according to the disclosed methods. In one or more implementations, classifying the user interface may, for example, validate the characterization of the user interface or create a subset of the user interfaces that characterize the user interface.

[0033] 1, a system 100 according to some implementations of the present disclosure is shown. The system 100 can be used to identify a user interface used by a user, among other users. The system 100 includes a control system 110, a storage device 114, an electronic interface 119, one or more sensors 130, and, optionally, one or more user devices 170. In some implementations, the system 100 further includes a respiratory treatment system 120 including a respiratory treatment device 122. As disclosed in further detail herein, the system 100 can be used to detect and / or identify a user interface 124.

[0034] The control system 110 includes one or more processors 112 (hereinafter processors 112). The control system 110 is generally used to control (e.g., operate) various components of the system 100 and / or analyze data acquired and / or generated by the components of the system 100. The processor 112 may be a general-purpose or special-purpose processor or microprocessor. Although one processor 112 is shown in FIG. 1, the control system 110 may include any suitable number of processors (e.g., one processor, two processors, five processors, ten processors, etc.) that may be in a single housing or may be located remotely from one another. The control system 110 (or any other control system) or a portion of the control system 110, such as the processor 112 (or any other processor or portion of any other control system), may be used to perform any one or more steps of the methods described and / or claimed herein. The control system 110 may be coupled to and / or located within, for example, the housing of the user device 170 and / or the housing of one or more of the sensors 130. Control system 110 may be centralized (within one such housing) or distributed (within two or more such housings that are physically distinct). In such implementations that include two or more housings to house control system 110, such housings may be located in close proximity to one another and / or remotely.

[0035] The storage device 114 stores machine-readable instructions executable by the processor 112 of the control system 110. The storage device 114 may be any suitable computer-readable storage device or medium, such as, for example, a random or serial access memory device, a hard drive, a solid state drive, a flash memory device, etc. Although one storage device 114 is shown in FIG. 1, the system 100 may include any suitable number of storage devices 114 (e.g., one storage device, two storage devices, five storage devices, ten storage devices, etc.). The storage device 114 may be coupled to and / or located within a housing of the respiratory treatment device 122 of the respiratory treatment system 120, a housing of the user device 170, a housing of one or more of the sensors 130, or any combination thereof. As with the control system 110, the storage device 114 may be centralized (within one such housing) or distributed (within two or more such housings that are physically distinct).

[0036] In some implementations, the storage device 114 stores a user profile associated with the user. The user profile may include, for example, demographic information associated with the user, biometric information associated with the user, medical information associated with the user, self-reported user feedback, sleep parameters associated with the user (e.g., sleep-related parameters recorded from one or more previous sleep sessions), or any combination thereof. The demographic information may include, for example, information indicative of the user's age, the user's gender, the user's race, family history (e.g., family history of insomnia or sleep apnea), the user's employment status, the user's education status, the user's socio-economic status, or any combination thereof. The medical information may include, for example, information indicative of one or more medical conditions associated with the user, medication usage by the user, or both. The medical information data may further include a fall risk assessment associated with the user (e.g., a Morse Fall Scale fall risk score), a Repeated Sleep Latency Test (MSLT) result, score, and / or a Pittsburgh Sleep Quality Index (PSQI) score or value. The self-reported user feedback may include information indicative of a self-reported subjective sleep score (poor, fair, good, etc.), a self-reported user's subjective stress level, a self-reported user's subjective level of fatigue, a self-reported user's subjective health status, recent life events experienced by the user, or any combination thereof.

[0037] The electronic interface 119 is configured to receive data (e.g., physiological data and / or acoustic data) from one or more sensors 130 so that the data can be stored in the storage device 114 and / or analyzed by the processor 112 of the control system 110. The electronic interface 119 can communicate with the one or more sensors 130 using a wired or wireless connection (e.g., RF communication protocol, WiFi communication protocol, Bluetooth communication protocol, IR communication protocol, cellular network, other optical communication protocol, etc.). The electronic interface 119 may include an antenna, a receiver (e.g., an RF receiver), a transmitter (e.g., an RF transmitter), a transceiver, or any combination thereof. The electronic interface 119 may also include another processor and / or another storage device the same as or similar to the processor 112 and storage device 114 described herein. In some implementations, the electronic interface 119 is coupled to or integrated with the user device 170. In other implementations, the electronic interface 119 is coupled to or integrated with (eg, within the housing) the control system 110 and / or the storage device 114 .

[0038] As noted above, in some implementations, the system 100 optionally includes a respiratory treatment system 120 (also referred to as a respiratory pressure treatment system). The respiratory treatment system 120 may include a respiratory treatment device 122 (also referred to as a respiratory pressure device), a user interface 124 (also referred to as a mask, a patient interface), a conduit 126 (also referred to as a tube or an air circuit), a display 128, a humidification tank 129, or a combination thereof. In some implementations, the control system 110, the memory device 114, the display 128, one or more of the sensors 130, and the humidification tank 129 are part of the respiratory treatment device 122. Respiratory pressure treatment refers to the application of an air supply to the entrance of a user's airway at a controlled target pressure that is nominally positive relative to the atmosphere throughout the user's respiratory cycle (unlike, for example, negative pressure therapies such as tank ventilators or cuirasses). Respiratory treatment system 120 is generally used to treat individuals suffering from one or more sleep-related breathing disorders (e.g., obstructive sleep apnea, central sleep apnea, mixed sleep apnea), other breathing disorders such as COPD, or other disorders that lead to respiratory insufficiency and that may be manifested during both sleep and wakefulness.

[0039] The respiratory treatment device 122 is generally used to generate pressurized air delivered to a user (e.g., using one or more motors driving one or more compressors). In some implementations, the respiratory treatment device 122 generates a continuous constant air pressure delivered to a user. In other implementations, the respiratory treatment device 122 generates two or more predetermined pressures (e.g., a first predetermined air pressure and a second predetermined air pressure). In still other implementations, the respiratory treatment device 122 is configured to generate a variety of different air pressures within a predetermined range. For example, the respiratory treatment device 122 can deliver at least about 6 cmH2O, at least about 10 cmH2O, at least about 20 cmH2O, between about 6 cmH2O and about 10 cmH2O, between about 7 cmH2O and about 12 cmH2O. The respiratory treatment device 122 can also deliver pressurized air at a predetermined flow rate, for example, between about 20 L / min and about 150 L / min, while maintaining a positive pressure (relative to ambient pressure). In some implementations, the control system 110 , the memory device 114 , the electronic interface 119 , or any combination thereof, may be coupled to and / or located within the housing of the respiratory treatment device 122 .

[0040] The user interface 124 engages a portion of the user's face and delivers pressurized air from the respiratory treatment device 122 to the user's airways to help prevent narrowing and / or obstruction of the airways while sleeping. This may also increase the user's oxygen intake while sleeping. Depending on the therapy being applied, the user interface 124 may form a seal with an area or portion of the user's face to facilitate delivery of gases at a pressure sufficiently different from ambient pressure, such as approximately 10 cmH2O positive pressure relative to ambient pressure, to make the therapy effective. In other forms of therapy, such as delivering oxygen, the user interface may not include a sufficient seal to facilitate delivery of gases to the airways at approximately 10 cmH2O positive pressure.

[0041] In some implementations, the user interface 124 is a face mask that covers the user's nose and mouth (see, e.g., FIG. 2). Alternatively, the user interface 124 is or includes a nasal mask that provides air to the user's nose, or a nasal pillow mask that delivers air directly to the user's nostrils. The user interface 124 can include a strap assembly having multiple straps (e.g., including hook-and-loop fasteners) for positioning and / or stabilizing the user interface 124 on a portion of the user's interface 124 in a desired position (e.g., on the face) and a conformal cushion (e.g., silicone, plastic, foam, etc.) that helps provide an airtight seal between the user interface 124 and the user. In some implementations, the user interface 124 can include a connector 127 and one or more vents 125. The one or more vents 125 can be used to allow carbon dioxide and other gases exhaled by the user to escape. In other implementations, the user interface 124 includes a mouthpiece (e.g., a night guard mouthpiece molded to fit the user's teeth, a jaw repositioning device, etc.). In some implementations, the connector 127 is separate from, but connectable to, the user interface 124 (and / or the conduit 126). The connector 127 is configured to connect and fluidly couple the user interface 124 to the conduit 126.

[0042] The conduit 126 allows air to flow between two components of the respiratory treatment system 120, such as the respiratory treatment device 122 and the user interface 124. In some implementations, there may be separate branches of this conduit for inhalation and exhalation. In other implementations, a single branch conduit is used for both inhalation and exhalation. Generally, the respiratory treatment system 120 forms an air pathway that extends between the motor of the respiratory treatment device 122 and the user and / or the user's airways. As such, the air pathway generally includes at least the motor of the respiratory treatment device 122, the user interface 124, and the conduit 126.

[0043] One or more of the respiratory treatment device 122, the user interface 124, the conduit 126, the display 128, and the humidification tank 129 may include one or more sensors (e.g., a pressure sensor, a flow sensor, or more generally, any of the other sensors 130 described herein). These one or more sensors may be used, for example, to measure the air pressure and / or flow rate of the pressurized air provided by the respiratory treatment device 122.

[0044] The display device 128 is generally used to display image(s), including still images, moving images, or both, and / or information related to the respiratory treatment device 122. For example, the display device 128 may display information regarding the status of the respiratory treatment device 122 (e.g., whether the respiratory treatment device 122 is on / off, the pressure of the air being delivered by the respiratory treatment device 122, the temperature of the air being delivered by the respiratory treatment device 122, etc.) and / or other information (e.g., a sleep score or a treatment score (such as those described in WO 2016 / 061629 and U.S. Patent Publication No. 2017 / 0311879, each of which is incorporated herein by reference in its entirety), TM The display device 128 may provide various information to the user, such as the user's progress, progress score, current date / time, user's personal information, user questionnaire, etc. In some implementations, the display device 128 functions as a human machine interface (HMI) that includes a graphical user interface (GUI) configured to display images as an input interface. The display device 128 may be an LED display, an OLED display, an LCD display, etc. The input interface may be, for example, a touch screen or touch sensitive board, a mouse, a keyboard, or any sensor system configured to sense inputs made by a human user interacting with the respiratory treatment device 122.

[0045] The humidification tank 129 is coupled or integrated with the respiratory treatment device 122 and includes a reservoir of water that can be used to humidify the pressurized air delivered from the respiratory treatment device 122. The respiratory treatment device 122 may include a heater that heats the water in the humidification tank 129 to humidify the pressurized air provided to the user. Additionally, in some implementations, the conduit 126 may also include a heating element (e.g., coupled and / or embedded in the conduit 126) that heats the pressurized air delivered to the user. The humidification tank 129 is fluidly coupled to a water vapor inlet of the air path and may deliver water vapor to the air path via the water vapor inlet or may be formed in-line with the air path as part of the air path itself. In other implementations, the respiratory treatment device 122 or the conduit 126 may include a waterless humidifier. The waterless humidifier may incorporate a sensor that interfaces with other sensors located elsewhere in the system 100.

[0046] Respiratory treatment system 120 may be, for example, a mechanical ventilator or a positive airway pressure (PAP) system, such as a continuous positive airway pressure (CPAP) system, an automatic positive airway pressure (APAP) system, a bilevel or variable positive airway pressure (BPAP or VPAP) system, or any combination thereof. A CPAP system delivers a predetermined air pressure (e.g., determined by a sleep physician) to a user. An APAP system automatically varies the air pressure delivered to a user, for example, based at least in part on respiratory data associated with the user. A BPAP or VPAP system is configured to deliver a first predetermined pressure (e.g., inspiratory positive airway pressure or IPAP) and a second predetermined pressure (e.g., expiratory positive airway pressure or EPAP) that is lower than the first predetermined pressure.

[0047] Referring to FIG. 2, a portion of the system 100 (FIG. 1) according to some implementations is shown. A user 210 and a bed partner 220 of the respiratory treatment system 120 are positioned in a bed 230 and lying on a mattress 232. A user interface 124 (e.g., a full face mask) can be worn by the user 210 during a sleep session. The user interface 124 is fluidly coupled and / or connected to a respiratory treatment device 122 via a conduit 126. The respiratory treatment device 122 then delivers pressurized air to the user 210 via the conduit 126 and the user interface 124 to increase air pressure in the throat of the user 210 to help prevent airway closure and / or narrowing during sleep. The respiratory treatment device 122 can include a display device 128 that allows the user to interact with the respiratory treatment device 122. The respiratory treatment device 122 can further include a humidification tank 129 that stores water used to humidify the pressurized air. 2, on a nightstand 240 directly adjacent to the bed 230, or more generally, on any surface or structure generally adjacent to the bed 230 and / or the user 210. The user may also wear a blood pressure device 180 and an activity tracker 190 while lying on the mattress 232 of the bed 230.

[0048] 1, the one or more sensors 130 of the system 100 may include a pressure sensor 132, a flow sensor 134, a temperature sensor 136, a motion sensor 138, a microphone 140, a speaker 142, a radio frequency (RF) receiver 146, an RF transmitter 148, a camera 150, an infrared (IR) sensor 152, a photoplethysmogram (PPG) sensor 154, an electrocardiogram (ECG) sensor 156, an electroencephalography (EEG) sensor 158, a capacitance sensor 160, a force sensor 162, a strain gauge sensor 164, an electromyogram (EMG) sensor 166, an oxygen sensor 168, an analyte sensor 174, a moisture sensor 176, a light detection and ranging (LiDAR) sensor 178, or any combination thereof. In general, each of the one or more sensors 130 is configured to output sensor data received and stored by the storage device 114 or one or more other storage devices. The sensor 130 may further include an electrooculogram (EOG) sensor, a peripheral oxygen saturation (SpO2) sensor, a galvanic skin response (GSR) sensor, a carbon dioxide (CO2) sensor, or any combination thereof.

[0049] Although the one or more sensors 130 are shown and described as including each of a pressure sensor 132, a flow sensor 134, a temperature sensor 136, a motion sensor 138, a microphone 140, a speaker 142, an RF receiver 146, an RF transmitter 148, a camera 150, an IR sensor 152, a PPG sensor 154, an ECG sensor 156, an EEG sensor 158, a capacitance sensor 160, a force sensor 162, a strain gauge sensor 164, an EMG sensor 166, an oxygen sensor 168, an analyte sensor 174, a moisture sensor 176, and a LiDAR sensor 178, more generally, the one or more sensors 130 may include any combination and any number of the sensors described and / or shown herein.

[0050] The one or more sensors 130 may be used to generate physiological data, acoustic data, or both associated with, for example, a user of the respiratory treatment system 120 (such as user 210 in FIG. 2), the respiratory treatment system, both the user and the respiratory treatment system 120, or other entities, objects, activities, etc. The physiological data generated by the one or more sensors 130 may be used by the control system 110 to determine a sleep-wake signal and one or more sleep-related parameters associated with the user during a sleep session. The sleep-wake signal may be indicative of one or more sleep stages (also referred to as sleep states), including distinct sleep stages such as sleep, wakefulness, relaxed wakefulness, micro-arousal, or rapid eye movement (REM) stages (which may include both typical and non-typical REM stages), a first non-REM stage (often referred to as “N1”), a second non-REM stage (often referred to as “N2”), a third non-REM stage (often referred to as “N3”), or any combination thereof. Methods for determining sleep stages based on physiological data generated by one or more sensors, such as sensor 130, are described, for example, in International Publication No. WO 2014 / 047310, U.S. Pat. No. 10,492,720, U.S. Pat. No. 10,660,563, U.S. Patent Application Publication No. 2020 / 0337634, WO 2017 / 132726, WO 2019 / 122413, U.S. Patent Application Publication No. 2021 / 0150873, WO 2019 / 122414, and U.S. Patent Application Publication No. 2020 / 0383580, each of which is incorporated by reference in its entirety.

[0051] The sleep-wake signal may also be time-stamped to indicate the time the user gets into bed, the time the user gets out of bed, the time the user attempts to fall asleep, etc. The sleep-wake signal may be measured by one or more sensors 130 during a sleep session at a predetermined sampling rate, for example, 1 sample / second, 1 sample / 30 seconds, 1 sample / minute, etc. Examples of one or more sleep-related parameters that may be determined for a user during a sleep session based at least in part on the sleep-wake signal include total time in bed, total time asleep, total wake time, sleep onset latency, wake-after parameter, sleep efficiency, fragmentation index, amount of time asleep, breathing rate consistency, time to fall asleep, time to wake up, percentage of sleep disturbances, movement count, or any combination thereof.

[0052] The physiological and / or acoustic data generated by the one or more sensors 130 may also be used to determine a respiratory signal associated with the user during a sleep session. The respiratory signal is generally indicative of the user's breathing during a sleep session. The respiratory signal may be indicative of, for example, respiratory rate, respiratory rate variability, inhalation amplitude, exhalation amplitude, inhalation / exhalation amplitude ratio, inhalation / exhalation duration ratio, number of events per hour, patterns of events, pressure settings of the respiratory treatment device 122, or any combination thereof. Events may include snoring, apnea, central apnea, obstructive apnea, mixed apnea, hypopnea, RERA, airflow limitation (e.g., an event resulting in no increase in airflow despite an increase in negative intrathoracic pressure indicating increased effort), mask leak (e.g., leaking from the user interface 124), restless legs, sleep disorders, choking, increased heart rate, heart rate variability, dyspnea, asthma attack, epileptic episodes, seizures, fever, coughing, sneezing, snoring, gasping, presence of illness such as a cold or flu, elevated stress levels, etc. Events may be detected by any means known in the art, such as those described in U.S. Pat. No. 5,245,995, U.S. Pat. No. 6,502,572, WO 2018 / 050913, WO 2020 / 104465, each of which is incorporated herein by reference in its entirety.

[0053] The pressure sensor 132 outputs pressure data that can be stored in the memory 114 and / or analyzed by the processor 112 of the control system 110. In some implementations, the pressure sensor 132 is an air pressure sensor (e.g., an air pressure sensor) that generates sensor data indicative of a user's breathing (e.g., inhalation and / or exhalation) and / or ambient pressure of the respiratory treatment system 120. In such implementations, the pressure sensor 132 can be coupled to or integrated with the respiratory treatment device 122. The pressure sensor 132 can be, for example, a capacitive sensor, an electromagnetic sensor, an inductive sensor, a resistive sensor, a piezoelectric sensor, a strain gauge sensor, an optical sensor, a potentiometric sensor, or any combination thereof. In one example, the pressure sensor 132 can be used to determine the user's blood pressure.

[0054] The flow sensor 134 outputs flow data that can be stored in the memory 114 and / or analyzed by the processor 112 of the control system 110. In some implementations, the flow sensor 134 is used to determine the airflow rate from the respiratory treatment device 122, the airflow rate through the conduit 126, the airflow rate through the user interface 124, or any combination thereof. In such implementations, the flow sensor 134 may be coupled to or integrated with the respiratory treatment device 122, the user interface 124, or the conduit 126. The flow sensor 134 may be a mass flow sensor, such as, for example, a rotary flow meter (e.g., a Hall effect flow meter), a turbine flow meter, an orifice flow meter, an ultrasonic flow meter, a hot wire sensor, a vortex sensor, a membrane sensor, or any combination thereof.

[0055] The temperature sensor 136 outputs temperature data that can be stored in the memory device 114 and / or analyzed by the processor 112 of the control system 110. In some implementations, the temperature sensor 136 generates temperature data indicative of the core body temperature of the user, the skin temperature of the user 210, the temperature of the air flowing from the respiratory treatment device 122 and / or through the conduit 126, the temperature within the user interface 124, the ambient temperature, or any combination thereof. The temperature sensor 136 may be, for example, a thermocouple sensor, a thermistor sensor, a silicon bandgap temperature sensor or semiconductor sensor, a resistance temperature detector, or any combination thereof.

[0056] The motion sensor 138 outputs motion data that can be stored in the memory 114 and / or analyzed by the processor 112 of the control system 110. The motion sensor 138 can be used to detect the movement of the user and / or any of the components of the respiratory treatment system 120, such as the respiratory treatment device 122, the user interface 124, or the conduit 126, during a sleep session. The motion sensor 138 can include one or more inertial sensors, such as an accelerometer, a gyroscope, and a magnetometer. The motion sensor 138 can be used to detect motion or acceleration associated with an arterial pulse, such as a pulse in or around the user's face and proximal to the user interface 124, and is configured to detect pulse shape, rate, amplitude, or volume characteristics. In some implementations, the motion sensor 138 alternatively or additionally generates one or more signals representative of the user's body movements, from which a signal representative of the user's sleep state can be obtained, for example via the user's respiratory movements.

[0057] The microphone 140 outputs acoustic data that can be stored in the storage device 114 and / or analyzed by the processor 112 of the control system 110. As described in more detail herein, the acoustic data generated by the microphone 140 can be played back as one or more sounds (e.g., sounds from the user) during a sleep session to determine one or more sleep-related parameters (e.g., using the control system 110). As described in more detail herein, the acoustic data from the microphone 140 can also be used to identify events experienced by the user during a sleep session (e.g., using the control system 110). In other implementations, the acoustic data from the microphone 140 represents noises associated with the respiratory treatment system 120. In some implementations, the system 100 includes multiple microphones (e.g., two or more microphones and / or an array of microphones using beamforming) such that audio data generated by each of the multiple microphones can be used to distinguish audio data generated by other microphones of the multiple microphones. The microphone 140 can be coupled or integrated with the respiratory treatment system 120 (or the system 100) in generally any configuration. For example, the microphone 140 may be located inside the respiratory treatment device 122, the user interface 124, the conduit 126, or other component. The microphone 140 may also be located adjacent to or coupled to the exterior of the respiratory treatment device 122, the exterior of the user interface 124, the exterior of the conduit 126, or the exterior of any other component. The microphone 140 may be a component of the user device 170 (e.g., the microphone 140 is a smartphone microphone). The microphone 140 may be integrated into the user interface 124, the conduit 126, the respiratory treatment device 122, or any combination thereof. In general, the microphone 140 may be located at any point within or adjacent to the air path of the respiratory treatment system 120, including at least the motor of the respiratory treatment device 122, the user interface 124, and the conduit 126.The air path is therefore also called the acoustic path.

[0058] The speaker 142 outputs sound waves that are audible to a user. In one or more implementations, the sound waves may be audible to a user of the system 100 or inaudible to a user of the system (e.g., ultrasound). The speaker 142 may be used, for example, as an alarm clock or to play alerts or messages to a user (e.g., in response to an event). In some implementations, the speaker 142 may be used to communicate acoustic data generated by the microphone 140 to a user. The speaker 142 may be coupled to or integrated with the respiratory treatment device 122, the user interface 124, the conduit 126, or the user device 170.

[0059] The microphone 140 and the speaker 142 can be used as separate devices. In some implementations, the microphone 140 and the speaker 142 can be combined into an acoustic sensor 141 (e.g., a SONAR sensor), for example, as described in International Publication Nos. WO 2018 / 050913 and WO 2020 / 104465, each of which is incorporated herein by reference in its entirety. In such implementations, the speaker 142 generates or emits sound waves at a predetermined interval and / or frequency, and the microphone 140 detects the reflection of the sound waves emitted from the speaker 142. The sound waves generated or emitted by the speaker 142 have a frequency that is inaudible to the human ear (e.g., below 20 Hz or above about 18 kHz) so as not to disturb the sleep of the user or the user's bed partner (e.g., bed partner 220 in FIG. 2). Based at least in part on data from the microphone 140 and / or the speaker 142, the control system 110 can determine the user's position and / or one or more sleep-related parameters described herein, such as a respiratory signal, a respiratory rate, an inhalation amplitude, an exhalation amplitude, an inhalation-to-exhalation ratio, a number of events per hour, a pattern of events, a sleep stage, a pressure setting of the respiratory treatment device 122, a mouth leak status, or any combination thereof. In this context, a SONAR sensor may be understood to involve active acoustic sensing, such as by generating / transmitting an ultrasonic or low frequency ultrasonic sensing signal (e.g., within a frequency range of about 17-23 kHz, 18-22 kHz, or 17-18 kHz) through the air. Such a system may be considered in connection with WO 2018 / 050913, WO 2020 / 104465, as discussed above. In some implementations, the speaker 142 is a bone conduction speaker. In some implementations, the one or more sensors 130 include (i) a first microphone that is the same as or similar to microphone 140 and integrated into acoustic sensor 141, and (ii) a second microphone that is the same as or similar to microphone 140 but separate and distinct from the first microphone that is integrated into acoustic sensor 141.

[0060] The RF transmitter 148 generates and / or emits radio waves having a predetermined frequency and / or a predetermined amplitude (e.g., in a high frequency band, in a low frequency band, long wave signals, short wave signals, etc.). The RF receiver 146 detects reflections of the radio waves emitted from the RF transmitter 148, and this data can be analyzed by the control system 110 to determine the user's location and / or one or more sleep-related parameters described herein. The RF receiver (either the RF receiver 146 and the RF transmitter 148, or another RF pair) can also be used for wireless communication between the control system 110, the respiratory treatment device 122, one or more sensors 130, the user device 170, or any combination thereof. Although the RF receiver 146 and the RF transmitter 148 are shown in FIG. 1 as separate and distinct elements, in some implementations the RF receiver 146 and the RF transmitter 148 are combined as part of the RF sensor 147 (e.g., a RADAR sensor). In some such implementations, the RF sensor 147 includes control circuitry. The specific form of RF communication may be Wi-Fi, Bluetooth (registered trademark), etc.

[0061] In some implementations, the RF sensor 147 is part of a mesh system. One example of a mesh system is a Wi-Fi mesh system, which may include mesh nodes, mesh router(s) and mesh gateway(s), each of which may be mobile / movable or fixed. In such implementations, the Wi-Fi mesh system includes a Wi-Fi router and / or a Wi-Fi controller, each of which includes an RF sensor identical or similar to the RF sensor 147, and one or more satellites (e.g., access points). The Wi-Fi router and satellites continuously communicate with each other using Wi-Fi signals. The Wi-Fi mesh system can be used to generate motion data based at least in part on changes in the Wi-Fi signal (e.g., differences in received signal strength) that occur between the router and satellite(s) due to the movement of an object or person partially obstructing the signal. This motion data can be indicative of movement, breathing, heart rate, gait, falls, behavior, etc., or any combination thereof.

[0062] The camera 150 outputs image data that can be played back as one or more images (e.g., still images, video images, thermal images, or a combination thereof) that can be stored in the storage device 114. The image data from the camera 150 can be used by the control system 110 to determine one or more sleep-related parameters described herein. For example, the image data from the camera 150 can be used to identify the user's location to determine the time the user enters the user's bed (such as bed 230 in FIG. 2) and the time the user leaves the bed 230. The camera 150 can also be used to track eye movement, pupil dilation (if one or both of the user's eyes are open), blink rate, or any changes during REM sleep. The camera 150 can also be used to track the user's location, which can affect the duration and / or severity of apnea episodes for a user with positional obstructive sleep apnea.

[0063] The IR sensor 152 outputs infrared image data that can be played back as one or more infrared images (e.g., still images, moving images, or both) that can be stored in the storage device 114. The infrared data from the IR sensor 152 can be used to determine one or more sleep-related parameters during a sleep session, including the user's temperature and / or the user's movement. The IR sensor 152 can also be used in combination with the camera 150 in measuring the user's presence, location, and / or movement. The IR sensor 152 can detect infrared light having a wavelength between about 700 nm and about 1 mm, for example, while the camera 150 can detect visible light having a wavelength between about 380 nm and about 740 nm.

[0064] The IR sensor 152 outputs infrared image data that can be played back as one or more infrared images (e.g., still images, moving images, or both) that can be stored in the storage device 114. The infrared data from the IR sensor 152 can be used to determine one or more sleep-related parameters during a sleep session, including the user's temperature and / or the user's movement. The IR sensor 152 can also be used in combination with the camera 150 in measuring the user's presence, location, and / or movement. The IR sensor 152 can detect infrared light having a wavelength between about 700 nm and about 1 mm, for example, while the camera 150 can detect visible light having a wavelength between about 380 nm and about 740 nm.

[0065] The PPG sensor 154 outputs physiological data associated with the user that can be used to determine one or more sleep-related parameters, such as, for example, heart rate, heart rate pattern, heart rate variability, cardiac cycle, respiration rate, inspiration amplitude, expiration amplitude, inspiration-to-expiration ratio, estimated blood pressure parameter(s), or any combination thereof. The PPG sensor 154 can be worn by the user, embedded in clothing and / or fabric worn by the user, embedded in and / or coupled to the user interface 124 and / or its associated headgear (e.g., straps, etc.), etc.

[0066] The ECG sensor 156 outputs physiological data related to the electrical activity of the user's heart. In some implementations, the ECG sensor 156 includes one or more electrodes positioned on or around a portion of the user during a sleep session. The physiological data from the ECG sensor 156 can be used, for example, to determine one or more sleep-related parameters described herein.

[0067] The EEG sensor 158 outputs physiological data related to the electrical activity of the user's brain. In some implementations, the EEG sensor 158 includes one or more electrodes placed on or around the user's scalp during a sleep session. The physiological data from the EEG sensor 158 can be used, for example, to determine the user's sleep state or sleep stage at any given time during a sleep session. In some implementations, the EEG sensor 158 can be integrated into the user interface 124 and / or its associated headgear (e.g., straps, etc.).

[0068] The capacitive sensor 160, the force sensor 162, and the strain gauge sensor 164 output data that can be stored in the memory device 114 and used by the control system 110 to determine one or more sleep-related parameters described herein. The EMG sensor 166 outputs physiological data related to electrical activity produced by one or more muscles. The oxygen sensor 168 outputs oxygen data indicative of the oxygen concentration of the gas (e.g., in the conduit 126 or in the user interface 124). The oxygen sensor 168 may be, for example, an ultrasonic oxygen sensor, an electrical oxygen sensor, a chemical oxygen sensor, an optical oxygen sensor, or any combination thereof. In some implementations, the one or more sensors 130 also include a galvanic skin response (GSR) sensor, a blood flow sensor, a respiration sensor, a pulse sensor, a blood pressure sensor, an oximetry sensor, or any combination thereof.

[0069] The analyte sensor 174 can be used to detect the presence of an analyte in the user's breath. Data output by the analyte sensor 174 can be stored in the storage device 114 and used by the control system 110 to determine the identity and concentration of any analytes in the user's breath. In some implementations, the analyte sensor 174 is placed near the user's mouth to detect the analyte in the user's breath. For example, if the user interface 124 is a face mask that covers the user's nose and mouth, the analyte sensor 174 can be placed in the face mask to monitor the user's mouth breathing. In other implementations, the analyte sensor 174 can be placed near the user's nose to detect the analyte in the user's breath, such as when the user interface 124 is a nasal mask or nasal pillows mask. In yet other implementations, the analyte sensor 174 can be placed near the user's mouth when the user interface 124 is a nasal mask or nasal pillows mask. In this implementation, the analyte sensor 174 can be used to detect whether air is inadvertently leaking from the user's mouth. In some implementations, the analyte sensor 174 is a volatile organic compound (VOC) sensor that can be used to detect carbon-based chemicals or compounds, such as carbon dioxide. In some implementations, the analyte sensor 174 can also detect whether the user is breathing through the nose or mouth. For example, if the presence of an analyte is detected by data output by the analyte sensor 174 located near the user's mouth or in a face mask (in implementations where the user interface 124 is a face mask), the control system 110 can use this data as an indication that the user is breathing through their mouth.

[0070] The moisture sensor 176 outputs data that can be stored in the memory device 114 and used by the control system 110. The moisture sensor 176 can be used to detect moisture in various areas surrounding the user (e.g., inside the conduit 126 or the user interface 124, near the user's face, near the connection between the conduit 126 and the user interface 124, near the connection between the conduit 126 and the respiratory treatment device 122, etc.). Thus, in some implementations, the moisture sensor 176 can be coupled or integrated into the user interface 124 or the conduit 126 to monitor the humidity of the pressurized air from the respiratory treatment device 122. In other implementations, the moisture sensor 176 is located near any area where moisture levels need to be monitored. The moisture sensor 176 can also be used to monitor the humidity of the ambient environment surrounding the user, such as the air in the user's bedroom. The moisture sensor 176 can also be used to track the user's biological response to environmental changes.

[0071] One or more LiDAR sensors 178 can be used for depth sensing. This type of optical sensor (such as a laser sensor) can be used to detect objects and create a three-dimensional (3D) map of a surrounding environment, such as a living space. LiDAR typically uses a pulsed laser to measure time of flight. LiDAR is also referred to as 3D laser scanning. In one use case of such a sensor, a fixed or mobile device (such as a smartphone) with a LiDAR sensor 178 can measure and map an area more than 5 meters away from the sensor. LiDAR data can be fused with point cloud data estimated, for example, by an electromagnetic RADAR sensor. The LiDAR sensor 178 can also automatically create a geofence for a RADAR system by using artificial intelligence (AI) to detect and classify features in a space that may be problematic for the RADAR system, such as glass windows (which may be highly reflective to the RADAR). LiDAR can also be used to estimate a person's height, as well as changes in height that occur when a person sits down, falls, etc. LiDAR can be used to create a 3D mesh representation of the environment. In a further application, for solid surfaces through which radio waves pass (e.g., radio-transparent materials), LiDAR may reflect off such surfaces, thereby enabling classification of different types of obstacles.

[0072] 1, any combination of one or more sensors 130 may be integrated and / or coupled to any one or more of the components of the system 100, including the respiratory treatment device 122, the user interface 124, the conduit 126, the humidification tank 129, the control system 110, the user device 170, or any combination thereof. For example, the acoustic sensor 141 and / or the RF sensor 147 may be integrated and / or coupled to the user device 170. In such implementations, the user device 170 may be considered a secondary device that generates additional or secondary data used by the system 100 (e.g., the control system 110) according to some aspects of the disclosure. In some implementations, the pressure sensor 132 and / or the flow sensor 134 are integrated and / or coupled to the respiratory treatment device 122. In some implementations, at least one of the one or more sensors 130 is not coupled to the respiratory treatment device 122, the control system 110, or the user device 170, but is positioned generally adjacent to the user during a sleep session (e.g., placed on or in contact with a portion of the user, worn by the user, coupled or placed on a nightstand, coupled to a mattress, coupled to a ceiling, etc.). More generally, the one or more sensors 130 may be positioned in any suitable location relative to the user so as to generate physiological data related to the user and / or bed companions 220 during one or more sleep sessions.

[0073] Data from one or more sensors 130 may be analyzed to determine one or more sleep-related parameters, which may include a respiratory signal, a respiratory rate, a respiratory pattern, an inhalation amplitude, an exhalation amplitude, an inhalation-exhalation ratio, the occurrence of one or more events, the number of events per hour, a pattern of events, the average duration of an event, a range of event durations, a ratio of different event numbers, a sleep stage, an apnea-hypopnea index (AHI), or any combination thereof. The one or more events may include snoring, apnea, central apnea, obstructive apnea, mixed apnea, hypopnea, intentional user interface leak, unintentional user interface leak, mouth leak, coughing, restless legs, sleep disorder, choking, increased heart rate, dyspnea, asthma attack, epileptic episode, seizure, elevated blood pressure, hyperventilation, or any combination thereof. Many of these sleep-related parameters are physiological parameters, but some of the sleep-related parameters may be considered non-physiological parameters. Other types of physiological and non-physiological parameters may also be determined based on either the data from one or more sensors 130 or other types of data.

[0074] The user device 170 includes a display device 172. The user device 170 may be a mobile device, such as, for example, a smartphone, a tablet, a laptop, a game console, a smart watch, etc. Alternatively, the user device 170 may be an external sensing system, a television (e.g., a smart television), or another smart home device (e.g., a smart speaker(s), such as Google Home, Amazon Echo, Alexa, etc.). In some implementations, the user device 170 is a wearable device (e.g., a smart watch). The display device 172 is generally used to display an image(s), including still images, moving images, or both. In some implementations, the display device 172 functions as a human-machine interface (HMI), including a graphic user interface (GUI) and an input interface configured to display the image(s). The display device 172 may be an LED display, an OLED display, an LCD display, etc. The input interface may be, for example, a touch screen or touch-sensitive board, a mouse, a keyboard, or any sensor system configured to sense inputs made by a human user interacting with user device 170. In some implementations, one or more user devices 170 may be used by and / or included in system 100.

[0075] The blood pressure device 180 is generally used to assist in generating physiological data for determining one or more blood pressure measurements associated with a user. The blood pressure device 180 may include at least one of the one or more sensors 130 to measure, for example, a systolic blood pressure component and / or a diastolic blood pressure component.

[0076] In some implementations, the blood pressure device 180 is a blood pressure monitor that includes an inflatable cuff that can be worn by a user and a pressure sensor (e.g., pressure sensor 132 described herein). For example, as shown in the example of FIG. 2, the blood pressure device 180 can be worn on the upper arm of a user. In implementations in which the blood pressure device 180 is a blood pressure monitor, the blood pressure device 180 also includes a pump (e.g., a manually operated valve) that inflates the cuff. In some implementations, the blood pressure device 180 is coupled to a respiratory treatment device 122 of the respiratory treatment system 120, which provides pressurized air to inflate the cuff. More generally, the blood pressure device 180 can be communicatively coupled to and / or physically integrated (e.g., within a housing) with the control system 110, the memory device 114, the respiratory treatment system 120, the user device 170, and / or the activity tracker 190.

[0077] The activity meter 190 is generally used to assist in generating physiological data for determining activity metrics associated with a user. Activity metrics may include, for example, steps taken, distance traveled, steps climbed, duration of physical activity, type of physical activity, intensity of physical activity, time spent standing, respiration rate, average respiration rate, resting respiration rate, maximum respiration rate, respiration rate variability, heart rate, average heart rate, resting heart rate, maximum heart rate, heart rate variability, calorie expenditure, blood oxygen saturation, electrodermal activity (also known as skin conductance or galvanic skin response), or any combination thereof. The activity meter 190 may include one or more sensors 130 described herein, such as, for example, a motion sensor 138 (e.g., one or more accelerometers and / or gyroscope), a PPG sensor 154, and / or an ECG sensor 156.

[0078] In some implementations, the activity meter 190 is a wearable device that can be worn by a user, such as a smart watch, a wristband, a ring, or a patch. For example, referring to FIG. 2, the activity meter 190 is worn on the wrist of the user. The activity meter 190 can also be coupled to or integrated with clothing or apparel worn by the user. Further alternatively, the activity meter 190 can also be coupled to or integrated (e.g., within the same housing) with the user device 170. More generally, the activity meter 190 can be communicatively coupled to and / or physically integrated (e.g., within the housing) with the control system 110, the storage device 114, the respiratory treatment system 120, the user device 170, and / or the blood pressure monitor 180.

[0079] 1 as separate and distinct components of system 100, in some implementations control system 110 and / or storage 114 are integrated into user device 170 and / or respiratory treatment device 122. Alternatively, in some implementations control system 110 or portions thereof (e.g., processor 112) may be located in the cloud (e.g., integrated into a server, integrated into an Internet of Things (IoT) device, connected to the cloud, subject to edge cloud processing), located on one or more servers (e.g., remote servers, local servers, etc., or any combination thereof).

[0080] Although the system 100 is shown as including all of the above components, according to implementations of the present disclosure, a system for identifying a user interface used by a user can include more or fewer components. For example, a first alternative system includes the control system 110, the storage device 114, and at least one of the one or more sensors 130. As another example, a second alternative system includes the control system 110, the storage device 114, at least one of the one or more sensors 130, and a user device 170. As yet another example, a third alternative system includes the control system 110, the storage device 114, a respiratory treatment system 120, at least one of the one or more sensors 130, and a user device 170. As a further example, a fourth alternative system includes the control system 110, the storage device 114, a respiratory treatment system 120, at least one of the one or more sensors 130, a user device 170, and a blood pressure device 180 and / or an activity meter 190. Thus, various systems for changing pressure settings can be formed using any portion or portions of the components shown and described herein and / or in combination with one or more other components.

[0081] 2 , in some implementations, any or a combination of the control system 110, the memory device 114, and / or the one or more sensors 130 may be located on and / or within any surface and / or structure generally adjacent to the bed 230 and / or the user 210. For example, in some implementations, at least one of the one or more sensors 130 may be located at a first location on and / or within one or more components of the respiratory treatment system 120 adjacent to the bed 230 and / or the user 210. The one or more sensors 130 may be coupled to the respiratory treatment system 120, the user interface 124, the conduit 126, the display 128, the humidification tank 129, or a combination thereof.

[0082] Alternatively or additionally, at least one of the one or more sensors 130 can be positioned at a second location on and / or within the bed 230 (e.g., the one or more sensors 130 are coupled and / or integrated with the bed 230). Further, alternatively or additionally, at least one of the one or more sensors 130 can be positioned at a third location on and / or within the mattress 232 adjacent the bed 230 and / or the user 210 (e.g., the one or more sensors 130 are coupled and / or integrated with the mattress 232). Alternatively or additionally, at least one of the one or more sensors 130 can be positioned at a fourth location on and / or within a pillow generally adjacent the bed 230 and / or the user 210.

[0083] Alternatively or additionally, at least one of the one or more sensors 130 may be located at a fifth location on and / or within the bed 230 and / or nightstand 240 generally adjacent to the user 210. Alternatively or additionally, at least one of the one or more sensors 130 may be located at a sixth location such that it is coupled to and / or located on the user 210 (e.g., the one or more sensors 130 are embedded in or coupled to fabric, clothing, and / or a smart device worn by the user 210). More generally, at least one of the one or more sensors 130 may be located at any suitable location relative to the user 210 such that sensor data related to the user 210 can be generated.

[0084] In some implementations, a primary sensor, such as microphone 140, is configured to generate acoustic data associated with user 210 during a sleep session. For example, one or more microphones (the same as or similar to microphone 140 of FIG. 1 ) may be integrated into and / or coupled to (i) a circuit board of respiratory treatment device 122, (ii) conduit 126, (iii) connectors between components of respiratory treatment system 120, (iv) user interface 124, (v) headgear (e.g., straps) associated with the user interface, or (vi) combinations thereof.

[0085] In some implementations, in addition to the primary sensor, one or more secondary sensors can generate additional data. In some such implementations, the one or more secondary sensors include a microphone (e.g., microphone 140 of system 100), a flow sensor (e.g., flow sensor 134 of system 100), a pressure sensor (e.g., pressure sensor 132 of system 100), a temperature sensor (e.g., temperature sensor 136 of system 100), a camera (e.g., camera 150 of system 100), a vane-type air flow sensor (VAF), a hot air flow sensor (MAF), a cold air flow sensor, a laminar flow sensor, an ultrasonic sensor, an inertial sensor, or a combination thereof.

[0086] Alternatively or additionally, one or more microphones (the same as or similar to microphone 140 of FIG. 1 ) can be integrated and / or coupled to a collocated smart device such as user device 170, a TV, a watch (e.g., a mechanical watch or another smart device worn by the user), a pendant, mattress 232, bed 230, bedding placed on bed 230, a pillow, a speaker (e.g., speaker 142 of FIG. 1 ), a radio, a tablet-type device, an anhydrous humidifier, or a combination thereof. A collocated smart device may be any smart device that is within range to detect sounds emitted by the user, respiratory treatment system 120, and / or any part of system 100. In some implementations, a collocated smart device is a smart device that is in the same room as the user during a sleep session.

[0087] Alternatively or additionally, in some implementations, one or more microphones (identical or similar to microphone 140 of FIG. 1) may be remote from system 100 ( FIG. 1 ) and / or user 210 ( FIG. 2 ) provided there is an air passageway through which acoustic signals can be conducted thereto. For example, one or more microphones may be located in a room different from the room in which system 100 is housed. In some implementations, physical obstructions in the user's airway may be characterized and / or determined based at least in part on analysis of the acoustic data.

[0088] Here, a sleep session can be defined in several ways, for example, based at least in part on an initial start time and an end time. In some implementations, a sleep session is a duration during which a user is asleep, i.e., a sleep session has a start time and an end time, and the user does not wake up during the sleep session until the end time. That is, the time during which the user is awake is not included in the sleep session. According to this first definition of a sleep session, if a user wakes up and falls asleep multiple times throughout the night, each sleep segment separated by an awake segment is a sleep session.

[0089] Alternatively, in some implementations, a sleep session has a start time and an end time, and during a sleep session, the user can wake up without the sleep session ending as long as the cumulative duration that the user is awake is below the wake duration threshold. The wake duration threshold can be defined as a percentage of the sleep session. The wake duration threshold can be, for example, about 20% of the sleep session, about 15% of the sleep session duration, about 10% of the sleep session duration, about 5% of the sleep session duration, about 2% of the sleep session duration, etc., or any other threshold percentage. In some implementations, the wake duration threshold is defined as a certain time, for example, about 1 hour, about 30 minutes, about 15 minutes, about 10 minutes, about 5 minutes, about 2 minutes, etc., or any other time.

[0090] In some implementations, a sleep session is defined as the total time from when the user first goes to bed at night to when the user last gets out of bed the next morning. In other words, a sleep session can be defined as a period that begins at a first time (e.g., 10:00 PM) on a first date (e.g., Monday, January 6, 2020) that may be referred to as tonight when the user first goes to bed with the intention of going to bed (unless the user intends to watch TV or enjoy a smartphone first before going to bed), and ends at a second time (e.g., 7:00 AM) on a second date (e.g., Tuesday, January 7, 2020) that may be referred to as the next morning when the user first gets out of bed with the intention of not sleeping again the next morning.

[0091] In some implementations, a user can manually initiate a sleep session or manually initiate an end of a sleep session. For example, a user can select (e.g., click or tap) one or more user-selectable elements displayed on display 172 of user device 170 (FIG. 1) to manually initiate or end a sleep session.

[0092] 3, an example timeline 300 of a sleep session is shown. The timeline 300 begins with the time of going to bed (t bed ) and bedtime (t GTS ) and initial sleep time (t sleep ), the first micro-awakening MA1, the second micro-awakening MA2, the awakening A, and the wake-up time (t wake ) and wake-up time (t rise ) and.

[0093] bedtime t bed is associated with the time when the user first goes to bed (e.g., bed 230 in FIG. 2) before falling asleep (e.g., when the user lies or sits in bed). bedcan be determined based at least in part on a bedtime threshold duration to distinguish between a time when a user gets into bed to sleep and a time when a user gets into bed for other reasons (e.g., to watch television). For example, the bedtime threshold duration can be at least about 10 minutes, at least about 20 minutes, at least about 30 minutes, at least about 45 minutes, at least about 1 hour, at least about 2 hours, etc. As used herein, a bedtime t in relation to a bed is defined as a time when a user gets into bed to sleep and a time when a user gets into bed for other reasons (e.g., to watch television). bed More generally, the time of going to bed t bed can refer to the time when a user first sits down in any position to sleep (e.g., on a couch, in a chair, in a sleeping bag, etc.).

[0094] GTS is the time when the user goes to bed (t bed ) is associated with the time that the user first attempts to fall asleep after getting into bed. For example, after getting into bed, the user may engage in one or more activities (e.g., reading, watching television, listening to music, using user device 170, etc.) to relax before attempting to fall asleep. sleep ) is the time when the user first falls asleep. For example, the initial sleep time (t sleep ) may be the time when the user first entered the first non-REM stage.

[0095] wake up time t wake is the time associated with the time the user wakes up without going back to sleep (e.g., different from the time the user wakes up in the middle of the night and goes back to sleep). After the user initially falls asleep, they may experience one of many involuntary micro-arousals (e.g., micro-arousals MA1 and MA2) having short durations (e.g., 5 seconds, 10 seconds, 30 seconds, 1 minute, etc.). wake In contrast, the user goes back to sleep again after each micro-awakening MA1 or MA2. Similarly, the user may have one or more conscious awakenings (e.g., awakening A1) after initially falling asleep (e.g., waking up to go to the bathroom, caring for children or pets, sleepwalking, etc.). However, the user goes back to sleep again after awakening A. Thus, at wake-up time t wakecan be defined, for example, based at least in part on the awakening threshold duration (eg, 15 minutes or more, 20 minutes or more, 30 minutes or more, 1 hour or more, etc.).

[0096] Similarly, the wake-up time t rise is associated with the time that the user gets out of bed and remains out of bed (as opposed to, for example, getting up during the night to go to the bathroom, to care for children or pets, sleepwalking, etc.) for the purpose of ending a sleep session. In other words, the wake-up time t rise is the time when the user last left the bed without returning to the bed until the next sleep session (e.g., the next night). Therefore, the wake-up time t rise For example, t can be defined at least in part based on a wake threshold duration (e.g., when the user has been out of bed for 15 minutes or more, 20 minutes or more, 30 minutes or more, 1 hour or more, etc.). bed may also be defined at least in part based on a wake threshold duration (eg, when the user has been out of bed for 4 hours or more, 6 hours or more, 8 hours or more, 12 hours or more, etc.).

[0097] As mentioned above, the user first bed From the last t rise In some implementations, the user may wake up once more during the night and get out of bed at the latest wake-up time t wake and / or last wake-up time t rise is determined or determined based at least in part on a predetermined threshold duration of time following an event (e.g., falling asleep or leaving bed). Such threshold duration may be customized for a user. For a standard user who goes to bed at night and wakes up and gets out of bed in the morning, the user may wake up (t wake ) or wake up (t rise ) to bedtime (t bed ) and went to bed (t GTS ) or fall asleep (t sleep) can be used from about 12 hours to about 18 hours. For users who spend longer periods in bed, shorter threshold periods (e.g., from about 8 hours to about 14 hours) may be used. The threshold period may be initially selected and / or later adjusted based at least in part on the system monitoring the user's sleep behavior.

[0098] Total time in bed (TIB) is the time from the time you get into bed to the time you sleep. bed From wake-up time t rise t . Total sleep time (TST) is the duration from the initial sleep time t to the wake time, excluding conscious or unconscious awakenings and / or microarousals. Typically, total sleep time (TST) will be less than total time in bed (TIB) (e.g., 1 minute less, 10 minutes less, 1 hour less, etc.). For example, referring to timeline 300 of FIG. 3, total sleep time (TST) is the duration from the initial sleep time t sleep and the alarm time t wake , but excluding the duration of the first micro-awakening MA1, the second micro-awakening MA2, and awakening A. As shown, in this example, the total sleep time (TST) is less than the total time in bed (TIB).

[0099] In some implementations, total sleep time (TST) can be defined as total continuous sleep time (PTST). In such implementations, the total continuous sleep time excludes a predetermined initial portion or period of the first non-REM stage (e.g., light sleep stage). For example, the predetermined initial portion can be about 30 seconds to about 20 minutes, about 1 minute to about 10 minutes, about 3 minutes to about 5 minutes, etc. The total continuous sleep time is an indicator of continuous sleep and smoothes the sleep / wake hypnogram. For example, when a user first falls asleep, the user may enter the first non-REM stage for a very short time (e.g., about 30 seconds), then return to the wake stage for a short time (e.g., 1 minute), and then return to the first non-REM stage. In this example, the total continuous sleep time excludes the first instance of the first non-REM stage (e.g., about 30 seconds).

[0100] In some implementations, a sleep session is started by bedtime (t bed ) and wake-up time (t rise ), i.e., a sleep session is defined as the total time in bed (TIB). In some implementations, a sleep session is defined as the time from the initial sleep time (t sleep ) and wakes up at (t wake ) In some implementations, a sleep session is defined as a total sleep time (TST). In some implementations, a sleep session is defined as a time from bedtime (t GTS ) and wakes up at (t wake ) In some implementations, a sleep session is defined as ending at bedtime (t GTS ) and wake-up time (t rise ) In some implementations, a sleep session is defined as ending at bedtime (t bed ) and wakes up at (t wake In some implementations, a sleep session is defined as ending at an initial sleep time (t sleep ) and wake-up time (t rise ) is defined as ending in

[0101] 4, an example hypnogram 350 is shown corresponding to the timeline 300 (FIG. 3) according to some implementations. As shown, the hypnogram 350 includes a sleep-wake signal 351, a wake stage axis 360, a REM stage axis 370, a light sleep stage axis 380, and a deep sleep stage axis 390. The intersection of the sleep-wake signal 351 with one of the axes 360-390 indicates the sleep stage at any given time during the sleep session.

[0102] The sleep-wake signal 351 may be generated based at least in part on physiological data associated with the user (e.g., generated by one or more of the sensors 130 described herein). The sleep-wake signal may indicate one or more sleep stages including an awake state, a relaxed awake state, a micro-arousal, a REM stage, a first non-REM stage, a second non-REM stage, a third non-REM stage, or any combination thereof. In some implementations, one or more of the first non-REM stage, the second non-REM stage, and the third non-REM stage may be grouped and classified as a light sleep stage or a deep sleep stage. For example, the light sleep stage may include a first non-REM stage, and the deep sleep stage may include a second non-REM stage and a third non-REM stage. 4, the hypnogram 350 includes a light sleep stage axis 380 and a deep sleep stage axis 390, although in some implementations the hypnogram 350 may include axes for each of a first non-REM stage, a second non-REM stage, and a third non-REM stage. In other implementations, the sleep-wake signal may indicate a respiration signal, a respiration rate, an inhalation amplitude, an exhalation amplitude, an inhalation / exhalation amplitude ratio, an inhalation / exhalation duration ratio, a number of events per hour, a pattern of events, or any combination thereof. Information describing the sleep-wake signal may be stored in the memory device 114.

[0103] The hypnogram 350 may be used to determine one or more sleep-related parameters, such as, for example, sleep onset latency (SOL), wake after sleep onset (WASO), sleep efficiency (SE), sleep fragmentation index, sleep blocks, or any combination thereof.

[0104] Sleep onset latency (SOL) is the time to bed (t GTS ) to the initial sleep time (t sleep) in order to fall asleep. In other words, sleep onset latency indicates the time it takes for a user to actually fall asleep after first attempting to fall asleep. In some implementations, sleep onset latency is defined as a sustained sleep onset latency (PSOL). A sustained sleep onset latency differs from a sleep onset latency in that it is defined as the duration from bedtime to a predetermined amount of sustained sleep. In some implementations, the predetermined amount of sustained sleep can include at least 10 minutes of sleep, for example, in the second non-REM stage, the third non-REM stage, and / or a REM stage including 2 minutes or less of wakefulness, the first non-REM stage, and / or transitions therebetween. In other words, a sustained sleep onset latency requires, for example, up to 8 minutes of sustained sleep in the second non-REM stage, the third non-REM stage, and / or a REM stage. In other implementations, the predetermined amount of continuous sleep may include at least 10 minutes of sleep in the first non-REM stage, the second non-REM stage, the third non-REM stage, and / or the REM stage after the initial sleep time. In such implementations, the predetermined amount of continuous sleep may exclude any microarousals (e.g., a 10-second microarousal does not start the 10 minutes over).

[0105] A wake-on-sleep onset (WASO) is associated with the total time that the user is awake between the initial sleep time and the wake-up time. Thus, a wake-on-sleep onset includes short-term microarousals (e.g., microarousals MA1 and MA2 shown in FIG. 4) during a sleep session, whether conscious or unconscious. In some implementations, a wake-on-sleep onset (WASO) is defined as a persistent wake-on-sleep onset (PWASO) that includes only total times of waking with a predetermined length (e.g., 10 seconds or more, 30 seconds or more, 60 seconds or more, about 5 minutes or more, about 10 minutes or more, etc.).

[0106] The sleep efficiency (SE) is determined as a ratio between the total time in bed (TIB) and the total sleep time (TST). For example, if the total time in bed is 8 hours and the total sleep time is 7.5 hours, the sleep efficiency for the sleep session is 93.75%. The sleep efficiency indicates the sleep hygiene of the user. For example, if the user goes to bed and spends time on other activities (e.g., watching TV) before sleep, the sleep efficiency decreases (e.g., the user is penalized). In some implementations, the sleep efficiency (SE) can be calculated at least in part based on the total time in bed (TIB) and the total time the user attempts to fall asleep. In such implementations, the total time the user attempts to fall asleep is defined as the duration from the Go to Bed (GTS) time to the wake-up time as described herein. For example, if the total sleep time is 8 hours (e.g., from 11:00 PM to 7:00 AM), the bedtime is 10:45 PM, and the wake-up time is 7:15 AM, the sleep efficiency parameter is calculated as approximately 94%.

[0107] The fragmentation index is determined based at least in part on the number of awakenings during the sleep session. For example, if a user had two micro-awakenings (e.g., micro-awakenings MA1 and MA2 shown in FIG. 4), the fragmentation index can be expressed as 2. In some implementations, the fragmentation index is scaled between a predetermined range of integers (e.g., between 0 and 10).

[0108] A sleep block is associated with a transition between any sleep stage (e.g., a first non-REM stage, a second non-REM stage, a third non-REM stage, and / or REM) and a wake stage. A sleep block can be calculated, for example, with a resolution of 30 seconds.

[0109] In some implementations, the systems and methods described herein generate or analyze a hypnogram including a sleep-wake signal and determine a time to bed (t bed ), bedtime (t GTS ), initial sleep time (t sleep ), one or more first microarousals (e.g., MA1 and MA2), and a wake-up time (t wake), wake-up time (t rise ), or any combination thereof.

[0110] In another implementation, one or more sensors 130 are used to measure the time of initiation of bed (t bed ), bedtime (t GTS ), initial sleep time (t sleep ), one or more first microarousals (e.g., MA1 and MA2), and a wake-up time (t wake ), wake-up time (t rise ), or any combination thereof. bed may be determined based at least in part on data generated by, for example, motion sensor 138, microphone 140, camera 150, or any combination thereof. Bedtime may be determined based at least in part on data from, for example, motion sensor 138 (e.g., data indicative of no motion by the user), camera 150 (e.g., data indicative of no motion by the user and / or data indicative of the user turning off the lights), microphone 140 (e.g., data indicative of the television being turned off), user device 170 (e.g., data indicative of the user no longer using user device 170), pressure sensor 132 and / or flow sensor 134 (e.g., data indicative of the user turning on respiratory treatment device 122, data indicative of the user putting on user interface 124, etc.), or any combination thereof.

[0111] 5 is a diagram illustrating the generation of acoustic data responsive to acoustic reflections indicative of one or more features of a user interface, according to an embodiment of the present disclosure. The generation of acoustic data includes an acoustic sensor 141 including a microphone 140 (or any type of audio transducer) and a speaker 142. However, as noted above, the speaker 142 may instead be replaced by another device capable of generating an acoustic signal, such as a motor of the respiratory treatment device 122. The microphone 140 and speaker 142 are shown in particular locations relative to a conduit 126 that is connected to the respiratory treatment device 122, not shown. However, the locations of the microphone 140 and speaker 142 may differ from those shown, as noted above.

[0112] The speaker 142 emits an acoustic signal 302 within the conduit 126. The acoustic signal 302 is in the form of sound. The sound may be one or more of a standard sound (e.g., an original unmodified sound from a commercial sound application), a custom sound, a non-audible frequency, a white noise sound, a broadband impulse, a continuous sine wave, a square wave, a sawtooth wave, a frequency modulated sine wave (e.g., a chirp). According to some other implementations, the acoustic signal 302 may be in the form of one or more of an audible sound or an ultrasonic sound. According to some other implementations, the acoustic signal 302 is in the form of a non-audible sound, where the sound is not audible based on one or both of the frequency of the sound (e.g., a frequency that is outside the frequency range of human hearing) or the amplitude of the sound (e.g., an amplitude that is low enough that the sound is not loud enough for human perception).

[0113] In one or more implementations, the acoustic signal 302 is emitted at a particular time, such as when a user first puts on the user interface, after a user removes the user interface, after detecting that an apnea or hypopnea event is occurring (e.g., after detecting that an apnea or hypopnea event is occurring using a respiratory treatment device), etc. The particular monitoring time is selected, for example, to be at least 4 seconds in duration at 0.1 second intervals.

[0114] The acoustic signal 302 travels down the length L of the conduit 126 until it contacts a feature 304 of the user interface 124, such as at or near a junction 306 between the user interface 124 and the conduit 126. The feature 304 may include an expansion (or constriction) of a path 308 formed through the conduit 126 and the user interface 124. The expansion at the feature 304 causes a change in the acoustic impedance of the acoustic signal 302 and an acoustic reflection 310. The acoustic reflection 310 travels back down the length L of the conduit 126 until it reaches the microphone 140. The microphone 140 detects the acoustic reflection 310 and generates acoustic data in response to the acoustic reflection 310. As described below with reference to Figures 6 and 14A, the generated acoustic data is analyzed to classify and / or characterize the user interface.

[0115] The acoustic signal 302 can continue beyond feature 304 to the user interface 124. Although not shown, the user interface 124 may include one or more additional features that change the acoustic impedance of the acoustic signal 302 further producing an acoustic reflection 310. Thus, although only feature 304 is shown and described as causing the acoustic reflection 310, there may be multiple features of the user interface 124 that can all contribute to the acoustic reflection 310.

[0116] 6 is a flow diagram of a method 400 for characterizing a user interface according to an embodiment of the present disclosure. For convenience, the following description is directed to method 400 being performed by a respiratory treatment system, such as respiratory treatment system 120. However, method 400 may be performed by one or more other devices, such as one or more user devices 170. Such user devices 170 may be in communication with respiratory treatment system 120, for example, to analyze and characterize data as described below.

[0117] In step 402, the respiratory treatment system 120 generates acoustic data associated with the acoustic reflection of the acoustic signal. The acoustic data may be generated based on an acoustic transducer that detects the acoustic reflection. For example, the acoustic transducer may be a microphone that converts pressure waves of the acoustic reflection into an electrical signal representing the data.

[0118] The acoustic reflection is at least partially indicative of one or more features of a user interface coupled to a respiratory treatment device via a conduit. As the acoustic signal travels down the conduit, it interacts with one or more features of the conduit, one or more features of the connection between the conduit and the user interface, one or more features of the user interface, etc. This interaction may cause the acoustic signal to be reflected to produce an acoustic reflection. As will be described in more detail below, in Figures 15-20, the acoustic reflection is caused, at least in part, by one or more features of the user interface and therefore one or more features of the user interface.

[0119] In one or more implementations, the acoustic signal may be transmitted to a conduit connected to the user interface via an acoustic transducer, such as a speaker. Alternatively, or in addition, the acoustic signal may be transmitted into the conduit via a motor of a respiratory treatment device connected to the conduit. For example, the motor of the respiratory treatment device may emit a sound as it forces air through the conduit. The emitted sound may be the acoustic signal, or may be at least a portion of the acoustic signal.

[0120] In one or more implementations, the acoustic signal may be a sound that is audible to a human (e.g., an average human with average human hearing). A sound is audible to a human if it has an amplitude large enough for human detection and has a frequency within the frequency range of human hearing, typically from about 20 Hz to about 20,000 Hz.

[0121] In one or more implementations, the acoustic signal may be a sound that is inaudible to humans (e.g., an average human with average human hearing). A sound is inaudible if the amplitude of the sound is less than the lowest amplitude perceptible by a human. Alternatively, or in addition, a sound is inaudible to a human if it has a frequency outside the frequency range of human hearing, about 20 Hz to about 20,000 Hz as discussed above. For example, in one or more implementations, the acoustic signal may be an ultrasonic sound having a frequency higher than the highest frequency perceptible by human hearing.

[0122] In one or more implementations, the method 400 may be performed when the user interface is connected to a user (e.g., worn by the user or engaged with the user's face), i.e., emitting an acoustic signal and generating acoustic data based on acoustic reflection by the user interface connected or coupled to the user. The user interface may also be connected or coupled to the user while the respiratory treatment system is providing treatment to the user, before providing treatment to the user, or after providing treatment to the user. For example, a flow of pressurized air may be provided to the user interface via a conduit at the time of generating the acoustic data. The delivery of treatment may generate the acoustic signal, for example, by a motor of the respiratory treatment device that emits the acoustic signal. Alternatively, the acoustic signal may be emitted in the conduit without pressurized air flowing through the conduit to the user interface.

[0123] Alternatively, the method may be performed when the user interface is not connected or coupled to a user. In particular, acoustic reflections indicative of one or more features of the user interface do not require the presence of a user. Thus, the method 400 may be performed before a user wears the user interface or after a user removes the user interface.

[0124] While step 402 has been described as being based on acoustic signals and acoustic reflections, in one or more implementations, acoustic data may be generated that is generated from multiple acoustic reflections from multiple acoustic signals. The acoustic signals may be the same acoustic signal transmitted repeatedly, or may be different acoustic signals transmitted together, or may be a single acoustic signal transmitted separately. In the case of multiple acoustic signals with multiple acoustic reflections, the generated acoustic data may be an average of multiple acoustic reflections from the multiple acoustic signals.

[0125] In step 404, the respiratory treatment system 120 analyzes the generated acoustic data. The analysis includes windowing the generated acoustic data based at least in part on at least one characteristic of the respiratory treatment system, such as at least one characteristic of the user interface, at least one characteristic of the conduit, etc. The at least one characteristic of the user interface may be, for example, an extension from the connection of the conduit to the chamber, which is formed by the cushion of the full face mask against the user's face. The at least one characteristic of the conduit may be, for example, a length of the conduit. More specifically, the length may be a distance from the acoustic sensor to where the conduit connects to the user interface. In one or more implementations, the windowing may be based at least in part on at least one characteristic of the one or more characteristics of the user interface. Windowing is the selection of a portion of the generated acoustic data to be used for further analysis and / or characterization of the user interface, as described below. Windowing focuses on generated acoustic data that can distinguish and identify various user interfaces present in the respiratory treatment system. Thus, the windowing may focus on generated acoustic data that is indicative of a user interface. Windowing can advantageously ignore regions of the signal that may confound the estimation of the user interface, such as different respiratory treatment devices, varying lengths of conduits, differences in respiratory treatment systems, etc. Thus, windowing the generated acoustic data selects only regions of the acoustic signal and corresponding data associated with the user interface. As a result, windowing can ignore regions of the acoustic signal and corresponding data that may confound the estimation of the user interface, such as the factors listed above.

[0126] At least one feature has a corresponding point in the generated acoustic data that serves as a reference point. The reference point provides a basis for analyzing the generated acoustic data for different types of user interfaces, conduits, respiratory treatment devices, etc. Thus, the reference point serves as a reference for further analyzing the generated acoustic data and provides consistency between different acoustic data generated under different circumstances.

[0127] In one or more implementations, the reference point may correspond to a location of one or more features along a path formed at least in part by the conduit and the user interface. For example, the reference point may correspond to a feature of the conduit or a feature of the connection between the conduit and the user interface. In one or more implementations, the reference point may correspond to a feature of the user interface itself. In one or more implementations, the reference point is the same for a given user interface. In one or more implementations, the reference point is the same for all potentially different user interfaces. The feature may be common to different types of user interfaces such that the feature, or at least a variation of the feature, is present in all respiratory treatment systems that include a user interface connected to a conduit. For example, the feature may be an expansion of the acoustic path that occurs at the connection between the conduit and the user interface. Such an expansion may not be the same for each combination of conduit and user interface, but each combination of conduit and user interface will include such an expansion. Thus, the expansion at the connection of the conduit and the user interface may be a feature that corresponds to a reference point user in the windowing of the generated acoustic data.

[0128] Broadly speaking, the fiducial may be any feature that causes a change in acoustic impedance and is therefore present in the acoustic data generated from the acoustic reflection. In one or more implementations, the change in acoustic impedance may be based at least in part on a narrowing of the pathway formed by the conduit and the user interface at (i) the user interface, (ii) the connection between the conduit and the user interface, or (iii) a combination thereof. Alternatively, the change in acoustic impedance may be based at least in part on a widening of the pathway at (i) the user interface, (ii) the connection between the conduit and the user interface, or (iii) a combination thereof.

[0129] In one or more implementations described further below, the analysis of the generated acoustic data includes performing a deconvolution of the generated acoustic data, such as calculating a cepstrum of the generated acoustic data. In that case, the reference point is a minimum point within a predetermined interval of the deconvolution of the generated acoustic data. Alternatively, the reference point is a maximum point within a predetermined interval of the deconvolution of the generated acoustic data. As described above, the reference point may be based on a connection between the user interface and the conduit, and the predetermined interval may be based on a length of the conduit. More specifically, the predetermined interval may be based on a distance between the connection between the conduit and the user interface and the acoustic sensor.

[0130] 7, there are shown computed cepstrums 500a and 500b resulting from generated acoustic data for two different acoustic reflections. The two cepstrums 500a and 500b are shown merely for convenience of explanation and are not intended to be limiting. For example, one cepstrum can be calculated for generated acoustic data characterizing a user interface. The two cepstrums 500a and 500b are two cepstrums computed from different generated acoustic data for the same user interface.

[0131] These plots show quefrency on the x-axis and amplitude on the y-axis. Quefrency is similar to time or distance, but the x-axis units are simply samples or number of samples of acoustic reflections. The y-axis units may be units in which amplitude is measured. Values ​​near the 250 value on the x-axis are large minima 502a and 502b that can be used as reference points to analyze the generated acoustic data. This corresponds to an extension of the pathway within the conduit, such as the connection between the conduit and the user interface.

[0132] Referring again to step 404 of FIG. 6, in one or more implementations, the windowing can include a first windowing of the generated acoustic data and a second windowing of the generated acoustic data. In one or more implementations, the first windowing can identify or locate a feature of the user interface that corresponds to the reference point. The second windowing can identify or locate an additional feature of the user interface. The first windowing of the generated acoustic data and the second windowing of the generated acoustic data can vary depending on an amount of the generated acoustic data selected before the reference point of the generated acoustic data, an amount of the generated acoustic data selected after the reference point of the generated acoustic data, or a combination thereof. In terms of the length of the path through the conduit and the user interface, the first windowing can capture only about 10 cm before and after the reference point, for example, about 10 cm before and after the connection point between the conduit and the user interface. In contrast, the second windowing can capture about 35 cm before the reference point and about 50 cm after the reference point.

[0133] The amount of data captured by the windowing process may vary based on the amount of generated acoustic data required for the particular analysis associated with the windowing process. For example, the first windowing process is simply to identify a fiducial point. As a result, the size of the resulting window of the windowing process may be small since it is purely focused on determining the fiducial point. Once the fiducial point is determined, a second windowing process may occur, which may result in a larger second window. The larger second window may capture the generated acoustic data that is sent to the characterization step described below. Since the second window generated in the second windowing process may be larger than the first window generated in the first windowing process, the characterization step may benefit from the additional data using a larger window that captures more features that characterize the user interface compared to the first window. Alternatively, the size of the windows generated during the first and second windowing processes may be inverted, such that the first window is larger than the second window. The larger first window may aid in determining the correct fiducial point for the analysis.

[0134] Referring to Figure 8, windowed cepstrums 600a and 600b, corresponding to cepstrums 500a and 500b, respectively, are shown after windowing has been performed as described above. The windowing removed the generated acoustic data after about the 350 value on the x-axis of Figure 7 and removed the generated acoustic data before about the 200 value on the x-axis of Figure 7. The windowing is performed to remove generated acoustic data that is not used in characterizing the user interface and that potentially interferes with further analysis and characterization of the acoustic data to determine the user interface. The remaining generated acoustic data is approximately 80 on either side of reference points 502a and 502b.

[0135] 9, a further result of windowing is alignment with respect to reference points 502a and 502b, as shown by windowed cepstrum 600a relative to windowed cepstrum 600b, which allows for further consistency between additional analysis of the generated acoustic data and the features of the analyzed acoustic data, as described below.

[0136] As mentioned above, in one or more implementations, the respiratory treatment system 120 analyzing the generated acoustic data may include deconvolution of the generated acoustic data prior to windowing the generated acoustic data. In one or more implementations, the deconvolution of the generated acoustic data may include computing a cepstrum of the generated acoustic data. The cepstrum identifies distances indicative of the location of one or more features of the user interface associated with the acoustic reflection. More specifically, the cepstrum indicates the location of each of one or more physical features along the acoustic path of the acoustic signal by identifying one or more distances associated with each of the corresponding one or more acoustic reflections. Calculation of the cepstrum may be accomplished by processing the acoustic reflection in a series of stages: computing a frequency spectrum of the acoustic reflection via a fast Fourier transform, taking the natural logarithm of the absolute amplitude of this frequency spectrum, and performing an inverse Fourier transform to compute the real part of the signal to generate the cepstrum. This process may be repeated for multiple time intervals to estimate an average cepstrum. The resulting cepstrum signal transforms the acoustic reflection into the quefrency domain, which is similar to the reflection decomposition in time and / or distance. In one or more implementations, the cepstrum may be calculated using a Fast Fourier Transform of 4096 samples of the acoustic reflections, optionally with 50% sample overlap, where these 4096 samples represent acoustic data generated in approximately 0.2 seconds.

[0137] In one or more implementations, the analysis of the generated acoustic data may include calculating a cepstral derivative to determine a rate of change of the cepstral signal. As described below with respect to step 406, the cepstral derivative may provide additional information used to characterize the user interface. The cepstral derivative may also allow the network to find alternative features from the same input.

[0138] In one or more implementations, the respiratory treatment system 120 analyzing the generated acoustic data may include normalizing the generated acoustic data. Normalizing the generated acoustic data may account for confounding conditions, such as microphone gain, respiration amplitude, treatment pressure, changes in conduit length, curled or stretched conduits, different treatment pressures, ambient temperature, or combinations thereof. In one or more implementations, normalizing the generated acoustic data may include subtracting a mean value from the generated acoustic data, dividing by a standard deviation of the generated acoustic data, or combinations thereof. The purpose of the normalization is to input the cepstrum of all analyzed acoustic data of similarity and amplitude into the characterization in a natural range. In one or more implementations, the purpose of the normalization is to make the mean value of the cepstrum always zero.

[0139] Referring to Figure 10, there are shown normalized cepstrums 800a and 800b, which correspond to cepstrums 500a and 500b, respectively. As shown by the Y-axis values ​​in Figure 10 relative to Figure 9, the amplitudes have been adjusted so that the cepstrum has similar amplitude and is within the range of natural characterization, as described below.

[0140] 6, in step 406, the respiratory treatment system 120 characterizes a user interface based at least in part on the analyzed acoustic data. Characterizing the user interface may include inputting the analyzed acoustic data into a supervised machine learning model (e.g., a (deep) neural network such as a convolutional neural network) to determine the user interface. In one or more implementations, determining the user interface may include determining the form factor or type of the user interface (e.g., full face, nasal, pillow, etc.), the model of the user interface (e.g., the AirFit manufactured by ResMed), the type of ... TM F20 mask), determining the size of one or more elements of the user interface, or a combination thereof.

[0141] In one or more implementations, the convolutional neural network includes one or more convolutional layers and one or more max pooling layers. In one or more implementations, the CNN includes one or more convolutional layers. In one or more implementations, the CNN may include N features. Each of the N features may be L samples in width. The CNN may further enable max pooling of the M samples. Each feature of the N features is a unique plot or wave in the analyzed acoustic data, such as in a cepstrum. In one or more implementations, since N is equal to 64, the CNN may account for 64 unique plots or features in the analyzed acoustic data. Each sample of the L samples in width represents a data point generated based on an acoustic reflection. Thus, the L samples represent L data points generated in response to an acoustic reflection. The L samples approximately match an expected size of a feature in an acoustic reflection. In one or more implementations, the L samples may approximately correspond to a length of about 25 cm. The M samples can be, for example, 20 samples, which allows for natural time invariance in the model to account for shifts between different mask reflections. In one or more implementations, the ratio of N to M can be 1:1 to 4:1. In one or more implementations, the output of the convolutional layer can be fed to one or more fully connected dense layers before classification takes place.

[0142] Although the disclosed embodiments have been shown and described with respect to one or more implementations, equivalent changes and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. It should be noted that while a particular feature of the invention may be disclosed with respect to only one of several implementations, such feature may be combined with one or more other features in other implementations as may be desirable and advantageous for any given or particular application.

[0143] In one or more implementations, the CNN is trained based on an additional training set by adding 20 samples of random circular shifts to the features used to train the CNN. Specifically, a transformation can be applied to the features in the dataset used to train the CNN, which makes them more robust to time portions of acoustic signal shifts, shifts in acoustic data within a window, addition of noise, and other types of shifts for a more robust system.

[0144] Although the focus of this application is to identify user interfaces, in one or more implementations, the systems and methods disclosed herein can be used to identify conduits based on the same methods disclosed for identifying user interfaces. For example, different conduits (e.g., different diameters, lengths, materials, etc.) can be used within a respiratory treatment system. The systems and methods can be used to determine the different conduits.

[0145] In one or more implementations, the conduit 126 can be identified instead of the user interface. Further details regarding the identification of the conduit 126 can be found in U.S. Provisional Patent Application No. 63 / 107,763, filed October 30, 2020, which is incorporated by reference in its entirety. In one or more implementations, the conduit is identified along with the identification of the user interface. If both are identified, the order of identification includes both being identified simultaneously or one being identified first (e.g., the conduit or the user interface) and then the other.

[0146] In one or more implementations, identifying either the conduit or the user interface can narrow the possible choices for identifying the other of the conduit or the user interface. For example, a particular conduit may be compatible with only a subset of all user interfaces of a respiratory treatment system. Thus, identifying the conduit can limit subsequent identification of the user interface from only those user interfaces in the subset. With similar narrowing, simplification of characterization can be achieved by first identifying the user interface and then identifying the conduit.

[0147] The characterization of user interfaces above generally focuses at the highest level on the type of user interface, which may be refined to specific models, aesthetic styles, and the like. However, user interfaces such as user interface 124 in FIGS. 1 and 2 are differentiated in different categories, which may be considered as a distinction between user interfaces broader than, or at least as broad as, the types (e.g., full face, nasal, nasal pillows, etc.). In accordance with some implementations, the categories are defined, at least in part, as directly connected user interfaces (referred to herein as “direct category” user interfaces) or indirectly connected user interfaces (referred to herein as “indirect category” user interfaces). In one or more implementations, the distinction between direct and indirect connections is based on the element of the user interface that connects to the conduit of the respiratory treatment system. That is, the description of direct and indirectly connected user interfaces relates to how pressurized air is delivered from the end of the conduit of the respiratory treatment system through the user interface to the user. As described in more detail below, a directly connected user interface generally directly connects the conduit of the respiratory treatment system to the cushion and / or frame of the user interface, optionally via a connector of the user interface. Thus, pressurized air is delivered from the conduits of the respiratory treatment system (optionally via a connector) directly to a chamber formed by the cushion (or cushion and frame) of the user interface against the user's face.

[0148] 11A and 11B show perspective and exploded views, respectively, of one implementation of a directly connected user interface (a "direct category" user interface) according to aspects of the disclosure. The direct category of user interface 900 generally includes a cushion 930 and a frame 950 that define a volume of space around a user's and / or nose. During use, the volume of space receives pressurized air for passage to the user's airway. In some embodiments, the cushion 930 and the frame 950 of the user interface 900 form an integral component of the user interface. The user interface 900 assembly may further be considered to include headgear 910, which in the case of the user interface 900 is generally a strap assembly, and optionally a connector 970. The headgear 910 is configured to be generally disposed around at least a portion of the user's head when the user wears the user interface 900. The headgear 910 may be coupled to the frame 950 and disposed on the user's head such that the user's head is disposed between the headgear 910 and the frame 950. The cushion 930 is disposed between the user's face and the frame 950 to form a seal with the user's face. An optional connector 970 is coupled at one end to the frame 950 and / or the cushion 930 and is configured to couple to a conduit of a respiratory treatment device (not shown). Pressurized air can flow directly from a conduit of the respiratory treatment system through the connector 970 to the volume of space defined by the cushion 930 (or the cushion 930 and the frame 950) of the user interface 900. From the user interface 900, the pressurized air reaches the user's airway through the user's mouth, nose, or both. Alternatively, if the user interface 900 does not include the connector 970, the conduit of the respiratory treatment system can be connected directly to the cushion 930 and / or the frame 950.

[0149] In the case of an indirectly connected user interface (an "indirect category" user interface), as described in more detail below, the conduit of the respiratory treatment system indirectly connects with the cushion and / or frame of the user interface. Another element of the user interface other than the connector is between the conduit of the respiratory treatment system and the cushion and / or frame. This additional element routes the pressurized air from the conduit of the respiratory treatment system to a volume of space formed between the cushion (or cushion and frame) of the user interface and the user's face. Thus, the pressurized air is indirectly routed from the conduit of the respiratory treatment system to a volume of space defined by the cushion (or cushion and frame) of the user interface relative to the user's face. Furthermore, according to some implementations, the indirectly connected user interface category can be described as at least two different categories: "indirect headgear" and "indirect conduit." In the indirect headgear category, the conduit of the respiratory treatment system is connected to a headgear conduit, optionally via a connector, which in turn is connected to the cushion (or cushion and frame). The headgear is thus configured to deliver pressurized air from the conduits of the respiratory treatment system to the cushion (or cushion and frame) of the user interface, and the headgear conduits in the headgear of the user interface are thus configured to deliver pressurized air from the conduits of the respiratory treatment system to the cushion of the user interface.

[0150] 12A and 12B show perspective and exploded views, respectively, of one implementation of an indirect conduit user interface 1000 according to aspects of the disclosure. The indirect conduit user interface 1000 includes a cushion 1030 and a frame 1050. In some embodiments, the cushion 1030 and the frame 1050 form an integral component of the user interface 1000. The indirect conduit user interface 1000 may further be considered to include head gear 1010, such as a strap assembly, optionally a connector 1070, and a user interface conduit 1090 (often referred to in the art as a "mini-tube" or "flexi-tube"). In general, the user interface conduit (i) is more flexible than the conduit 126 of the respiratory treatment system, (ii) has a diameter smaller than the diameter of the conduit 126 of the respiratory treatment system, or both (i) and (ii). The user interface conduit may have a shorter length than the conduit. Similar to the user interface 900, the headgear 1010 is configured to be generally disposed around at least a portion of the user's head when the user wears the user interface 1000. The headgear 1010 is coupled to the frame 1050 and positioned on the user's head such that the user's head is disposed between the headgear 1010 and the frame 1050. The cushion 1030 is positioned between the user's face and the frame 1050 to form a seal with the user's face. The optional connector 1070 is configured to couple to the frame 1050 and / or the cushion 1030 at one end and to a conduit 1090 of the user interface 1000 at the other end. In other implementations, the conduit 1090 may be directly connected to the frame 1050 and / or the cushion 1030. The conduit 1090 is configured to connect to a conduit 126 ( FIG. 12A ) of a respiratory treatment system (not shown) at an opposite end relative to the frame 1050 and the cushion 1030. Pressurized air can flow from the respiratory treatment system conduit 126 (FIG. 12A), through the user interface conduit 1090, and optionally the connector 1070, into the spatial volume defined by the cushion 1030 (or the cushion 1030 and frame 1050) of the user interface 1000 against the user's face.From the volume of space, the pressurized air reaches the user's airway through the user's mouth, nose, or both.

[0151] Given the above configuration, the user interface 1000 is an indirectly connected user interface because the pressurized air is not delivered directly from the conduit 126 (FIG. 12A) of the respiratory treatment system (not shown) but rather from the conduit 126 (FIG. 12A) of the respiratory treatment system (not shown) to the cushion 1030 (or the cushion 1030 and frame 1050) via the user interface conduit 1090.

[0152] 13A and 13B show perspective and exploded views, respectively, of one implementation of an indirect headgear user interface 1100 according to an embodiment of the present disclosure. The indirect headgear user interface 1100 includes a cushion 1130. The indirect headgear user interface 1100 may further be considered to include a headgear 1110 (including a strap 1110a and a headgear conduit 1110b), and optionally a connector 1170. Similar to the user interfaces 900 and 1000, the headgear 1110 is configured to be generally positioned around at least a portion of a user's head when the user wears the user interface 1100. The headgear 1110 includes a strap 1110a coupled to the headgear conduit 1110b and positioned on the user's head such that the user's head is positioned between the strap 1110a and the headgear conduit 1110b. The cushion 1130 is disposed between the user's face and the headgear conduit 1110b to form a seal with the user's face. The connector 1170 is configured to couple to the headgear 1110 at one end and to a conduit of a respiratory treatment system at the other end. In other implementations, the connector 1170 may be optional or the headgear 1110 may be directly connected to a conduit of a respiratory treatment system. The headgear conduit 1110b may be configured to deliver pressurized air from the conduit of the respiratory treatment system to the cushion 1130, or specifically to a volume of space enclosed by the user's cushion around the user's mouth and / or nose. Thus, the headgear conduit 1110b is hollow and provides a passageway for the pressurized air. Both sides of the headgear conduit 1110b may be hollow to provide two passageways for the pressurized air. Alternatively, only one side of the headgear conduit 1110b may be hollow to provide a single passageway. 13A and 13B, the headgear conduit 1110b includes two passageways that are positioned on either side of the user's head / face in use, or only one passageway of the headgear conduit 1110b can be hollow to provide a single passageway.Pressurized air can flow from the conduits of the respiratory treatment system, through connector 1170 (if present) and headgear conduit 1110b, into the volume of space between the cushion 1130 and the user's face. From there, the pressurized air can reach the user's airway through the user's mouth, nose, or both.

[0153] Given the above configuration, the user interface 1100 is an indirect headgear user interface because pressurized air is not delivered directly from a conduit of the respiratory treatment system to the volume of space between the cushion 1130 and the user's face, but rather from a conduit of the respiratory treatment system via a headgear conduit 1110b to the volume of space between the cushion 1130 and the user's face.

[0154] In one or more implementations, the difference between the direct and indirect categories is defined in terms of the distance the pressurized air travels after exiting the conduit of the respiratory treatment device before reaching the volume of space defined by the cushion of the user interface that forms a seal with the user's face, excluding the connector of the user interface that connects to the conduit. This distance will be less, such as less than 1 centimeter (cm), less than 2 cm, less than 3 cm, less than 4 cm, or less than 5 cm, for a direct category user interface than for an indirect category user interface. This is because the pressurized air travels through additional elements between the conduits of the respiratory treatment system, for example, the user interface conduit 1090 or the headgear conduit 1110b, before reaching the volume of space defined by the cushion (or cushion and frame) of the user interface that forms a seal with the user's face for an indirect category user interface.

[0155] 14A is a flow diagram of a method 1200 for classifying a user interface, according to an embodiment of the disclosure. For convenience, the following description is of method 1200 performed by a respiratory treatment system, such as respiratory treatment system 120. However, one or more other devices, such as one or more user devices 170, may perform method 1200. Such user devices 170 may, for example, be in communication with respiratory treatment system 120, which analyzes and classifies data as described below.

[0156] In step 1202, the respiratory treatment system 120 generates acoustic data related to an acoustic reflection of the acoustic signal. Similar to step 402 above, the acoustic data may be generated based on an acoustic transducer that detects the acoustic reflection. For example, the acoustic transducer may be a microphone that converts pressure waves of the acoustic reflection into an electrical signal representative of the data. The acoustic reflection is at least partially indicative of one or more features of a user interface coupled to the respiratory treatment device via a conduit. As the acoustic signal travels down the conduit, it interacts with one or more features of the conduit, one or more features of the connection between the conduit and the user interface, one or more features of the user interface, etc. This interaction may cause the acoustic signal to be reflected to generate an acoustic reflection. The acoustic reflection is at least partially caused by one or more features of the user interface and is therefore indicative of one or more features of the user interface.

[0157] In one or more implementations, the acoustic signal may be transmitted to a conduit connected to the user interface via an acoustic transducer, such as a speaker. Alternatively, or in addition, the acoustic signal may be transmitted into the conduit via a motor of a respiratory treatment device connected to the conduit. For example, the motor of the respiratory treatment device may emit a sound as it forces air through the conduit. The emitted sound may be the acoustic signal, or may be at least a portion of the acoustic signal.

[0158] In one or more implementations, the acoustic signal may be a sound that is audible to a human (e.g., an average human with average human hearing). A sound is audible to a human if it has an amplitude large enough for human detection and has a frequency within the frequency range of human hearing, typically from about 20 Hz to about 20,000 Hz.

[0159] In one or more implementations, the acoustic signal may be a sound that is inaudible to humans (e.g., an average human with average human hearing). A sound is inaudible if the amplitude of the sound is less than the lowest amplitude perceptible by a human. Alternatively, or in addition, a sound is inaudible to a human if it has a frequency outside the frequency range of human hearing, about 20 Hz to about 20,000 Hz as discussed above. For example, in one or more implementations, the acoustic signal may be an ultrasonic sound having a frequency higher than the highest frequency perceptible by human hearing.

[0160] In one or more implementations, the method 1200 is performed while the user interface is connected or worn by the user, i.e., it may emit acoustic signals and generate acoustic data based on acoustic reflections by the user interface connected or coupled to the user. The user interface may also be connected or coupled to the user while the respiratory treatment system is providing treatment to the user, before providing treatment to the user, or after providing treatment to the user. For example, a flow of pressurized air may be provided to the user interface via a conduit at the time of generating the acoustic data. The delivery of treatment may generate the acoustic signal, for example, by a motor of the respiratory treatment device that emits the acoustic signal. Alternatively, the acoustic signal may be emitted in the conduit without pressurized air flowing through the conduit to the user interface.

[0161] Alternatively, the method may be performed when the user interface is not connected or coupled to a user. In particular, acoustic reflections indicative of one or more features of the user interface do not require the presence of a user. Thus, the method 1200 may be performed before a user wears the user interface or after a user removes the user interface.

[0162] While step 1202 has been described as being based on acoustic signals and acoustic reflections, in one or more implementations, acoustic data may be generated from multiple acoustic reflections from multiple acoustic signals. The acoustic signals may be the same acoustic signal transmitted repeatedly, or may be different acoustic signals transmitted together, or may be a single acoustic signal transmitted separately. In the case of multiple acoustic signals with multiple acoustic reflections, the generated acoustic data may be an average of multiple acoustic reflections from the multiple acoustic signals.

[0163] In one or more implementations, the generated acoustic data may include a first reflection within the acoustic reflections of the acoustic signal. Additionally or alternatively, the generated acoustic data may include a second reflection within the acoustic reflections of the acoustic signal. Additionally or alternatively, the generated acoustic data may include a third reflection within the acoustic reflections of the acoustic signal. The analysis performed in one or more of the following steps may be from the first reflections, the second reflections, the third reflections, or a combination thereof, for example a combination of the first reflections and the second reflections.

[0164] Although step 402 of method 400 of FIG. 6 and step 1202 of method 1200 of FIG. 14A are described in different parts of this specification and shown as different steps in different flow diagrams, in one or more implementations, step 402 and step 1202 may be the same step.

[0165] In step 1204, the respiratory treatment system 120 analyzes the generated acoustic data to associate one or more features of the user interface with one or more signatures in the generated acoustic data. The one or more signatures may include, for example, a maximum amplitude of the transform, a minimum amplitude of the transform, a standard deviation of the transform, a skewness of the transform, a kurtosis of the transform, a median absolute value of the transform, a sum of absolute values ​​of the transform, a sum of positive regions of the transform, a sum of negative regions of the transform, a fundamental frequency, an energy corresponding to a fundamental frequency, an average energy, at least one resonant frequency of the conduit and user interface combination, a change in at least one resonant frequency, a number of peaks in a range, a peak prominence, a distance between peaks, or combinations thereof. In general, a signature may be comprised of one or more acoustic features in the acoustic data. In one or more implementations, one or more of the signatures may vary over time in one or more of the above aspects. Because the change in the aspect over time may be specific to a category, the change in the aspect over time may be considered one or more signatures specific to a particular category. Changes may occur during an iteration of method 1200, such as consecutive iterations over several minutes, or during multiple iterations of method 1200 over the course of a sleep session or night, or over multiple iterations of method 1200 over several nights, weeks, months, years, etc.

[0166] In one or more implementations, analyzing the generated acoustic data includes generating a spectrum of the generated acoustic data, such as computing a transform (e.g., a discrete Fourier transform DFT or a discrete cosine transform DCT) of the generated acoustic data. Then, analyzing the generated acoustic data may further include computing a logarithm of the spectrum. Then, analyzing the generated acoustic data may include computing a cepstrum of the log-spectrum, such as computing an inverse transform (e.g., an iDFT or an iDCT) of the log-spectrum. The computed cepstrum may then be analyzed to obtain one or more signatures that can be used to classify the user interface. Non-linear operations may also be used instead of logarithmic calculations.

[0167] In one or more implementations, the analysis of the generated acoustic data does not require analyzing all of the generated acoustic data. For example, in one or more implementations, analyzing the generated acoustic data may include selecting a segment of the spectrum as described above. After selecting the segment, the analysis may further include computing a direct transform (e.g., DFT or DCT) of the segment of the spectrum. From the direct transform, one or more features of the user interface may be associated with one or more signatures in a Fourier transform of the segment to identify a category of the user interface.

[0168] Alternatively, analyzing the generated acoustic data may include selecting a segment of a log spectrum of said generated acoustic data. From the log spectrum, the analysis may include computing a transform of the log spectrum, such as a Fourier transform of the log spectrum. Alternatively, the analysis may include computing an inverse transform of the log spectrum, or an inverse transform of the transform of the log spectrum.

[0169] In one or more particular implementations, Figure 14B illustrates a flow diagram of a particular method for analyzing acoustic data generated from step 1202, according to some implementations of the present disclosure. The method of Figure 14B is a sub-method within step 1204 of the method of Figure 14A.

[0170] In step 1204-1, a spectrum is generated from the acoustic data. For example, a transform such as a DFT may be generated from the acoustic data. In step 1204-2, a log is taken from the spectrum generated in step 1204-1. In step 1204-3, a cepstrum may then be generated from the log spectrum of step 1204-2. In step 1204-4, one or more signatures may be identified from the cepstrum generated in step 1204-3.

[0171] Instead of returning to step 1204-1 and proceeding to step 1204-2, in step 1204-5, a subset of the spectra may be extracted, as described further below with respect to Figures 22 and 23. In step 1204-6, a direct transform, such as a DFT, may then be computed from the subset spectra. In step 1204-7, one or more signatures may be identified from the direct transform of the subset spectra generated in step 1204-6.

[0172] Instead of returning to step 1204-2 and proceeding to step 1204-3, in step 1204-8 a subset of the log-spectrum may be extracted, as described further below with respect to Figures 22 and 23. Then, in step 1204-9 a direct transform, such as a DFT, may be computed from the subset log-spectrum. Alternatively, in step 1204-10 an inverse transform, such as an iDFT, may be computed from the subset log-spectrum. After steps 1204-9 and 1204-10, in step 1204-7 one or more signatures may be identified from either the direct transform or the inverse transform from the subset log-spectrum.

[0173] Referring back to FIG. 14A, a segment of the generated acoustic signal can be selected by trimming the generated acoustic data based on the general location of one or more signatures indicative of a category of user interfaces in the generated acoustic data. For example, FIG. 22 shows a spectrum of an exemplary generated acoustic data. A hashed segment 2001 of the plot of the generated acoustic data can be a segment of the generated acoustic data corresponding to acoustic data associated with a user interface feature that can be used to classify the user interface. Thus, the generated acoustic data before and after the segment 2001 can be removed, and only the generated acoustic data in the segment 2001 can be analyzed (or further analyzed) to classify the user interface. FIG. 23 shows a further transformation of the segment 2001 of FIG. 22, for example, an inverse Fourier transform of the log spectrum of the generated acoustic data plotted in FIG. 22. According to the plot of FIG. 23, the signatures in the plot can be used to classify the user interface.

[0174] In particular, in step 1206, the respiratory treatment system 120 classifies the user interface based at least in part on the one or more signatures. As described above, the category of the user interface may be, for example, direct frame, indirect frame, or indirect conduit. The one or more signatures may be fed, for example, into a machine learning model (e.g., linear regression, logistic regression, nearest neighbor, decision tree, PCA, naive Bayes classifier, K-means, random forest, etc.) and returned a category (e.g., direct conduit, indirect conduit, indirect frame). In one or more implementations, the user interface may be classified at least in part based on one or more signatures that match one or more known signatures of one or more user interfaces of a known category. The user interface may be determined based on a category of known user interfaces having matching signatures.

[0175] In one or more implementations, the signatures used to classify the user interface can be associated with a first reflection in the generated acoustic data. Alternatively, the signatures used to classify the user interface can be associated with reflections other than the first reflection in the generated acoustic data, e.g., a second reflection, a third reflection, etc. Alternatively, the signatures used to classify the user interface may include a combination of reflections, e.g., a combination of a first reflection and a second reflection, a combination of a first reflection and a third reflection, a combination of a first reflection, a second reflection and a third reflection, a combination of a second reflection and a third reflection, etc.

[0176] According to some implementations, the direct category may include one or more signatures whose maximum amplitude of the cepstrum is greater than a threshold. Figures 15 and 16 are plots of the calculated cepstrum of a direct user interface according to some implementations of the present disclosure. Referring to Figure 15, the plot shows the cepstrum of a direct category user interface (full face mask, specifically AirFit F10™) as shown in Figures 11A and 11B. The maximum amplitude of the cepstrum meets (e.g., is greater than) a threshold, which at least partially indicates that the associated user interface is a direct user interface. For example, the threshold may be a value of 0.2 on the y-axis. Because the maximum peak 1301 is greater than the threshold 0.2, the cepstrum indicates that the user interface belongs to the direct category. Similarly, in Figure 16, the plot shows the calculated cepstrum of another direct category user interface (nasal mask, specifically AirFit N20 Classic). The maximum amplitude of the cepstrum meets (e.g., is greater than) a threshold, such as the same threshold of 0.2 in Figure 15. Specifically, the maximum peak 1401 is greater than the same threshold of 0.2. Thus, the cepstrum indicates that the user interface belongs to the direct category.

[0177] According to some implementations, the indirect frame category includes one or more signatures with a mean cepstral value less than a threshold. Figures 17 and 18 are cepstrum plots of example user interfaces of the indirect frame category according to some implementations of the present disclosure. Referring to Figure 17, the plot shows the cepstrum of a user interface of the indirect frame category (full face mask, specifically AirFit F30i™) as shown in Figures 12A and 12B. The maximum amplitude of the cepstrum meets (e.g., is less than) a threshold, which indicates that the associated user interface is an indirect frame user interface. For example, the threshold may be a y-axis value of 0.2. Since there are no peaks greater than the threshold of 0.2, the cepstrum meets the threshold, which at least partially indicates that the user interface belongs to the indirect frame category. Furthermore, the cepstrum plot of Figure 17 shows a distributed cepstral signature, which also at least partially indicates that the user interface belongs to the indirect frame category. Referring to Figure 18, the plot shows the cepstrum of another indirect frame category user interface (nose mask, specifically AirFitN30i). The maximum amplitude of the cepstrum meets a threshold of less than 0.2. The cepstrum plot in Figure 18 shows a distributed cepstral signature that at least partially indicates that the user interface belongs to the indirect frame category.

[0178] According to some implementations, the indirect conduit category may include one or more signatures that meet a threshold number of peaks. This may be the result of, for example, a connection between a conduit of the respiratory treatment system and a conduit of the user interface, and a connection between a conduit of the user interface and a connector of the user interface. FIGS. 19 and 20 are cepstrum plots of an example indirect frame conduit user interface according to some implementations of the present disclosure. Referring to FIG. 19, the plot shows the cepstrum of a user interface in the indirect conduit category (a nasal mask, specifically the AirFitN20™), as shown in FIGS. 13A and 13B. The cepstrum includes two prominent peaks 1701 and 1703. The prominent peak 1701 may represent, for example, a connection between a conduit of the respiratory treatment system and a conduit of the user interface. The prominent peak 1703 may represent, for example, a connection between a conduit of the user interface and a connector of the user interface. Therefore, based on the presence of these two prominent peaks 1701 and 1703, the cepstrum indicates that the user interface category is indirect conduit.

[0179] 20, similar to FIG 19, prominent peak 1801 may represent, for example, a connection between a conduit of a respiratory treatment system and a conduit of a user interface. Prominent peak 1803 may represent, for example, a connection between a conduit of a user interface and a connector of the user interface. Thus, based on the presence of these two prominent peaks 1801 and 1803, the cepstrum indicates that the category of the user interface is an indirect conduit.

[0180] Based on the above analysis, a signature in the generated acoustic data can indicate a category of the user interface. In one or more implementations, a single signature, such as a prominent peak that meets a large threshold, can be used to classify the user interface. However, in one or more implementations, multiple signatures can be used to classify the user interface. In one or more implementations, the classification can be similar to characterization. For example, the classification can be performed based on inputting the analyzed acoustic data into a supervised machine learning model (e.g., a (deep) neural network such as a convolutional neural network) to determine the user interface.

[0181] Referring to FIG. 21, as described above, the signature used to determine the category of the user interface can be related to the first and second order reflections in the generated acoustic data. Box 1901 in FIG. 21 shows a portion of the cepstrum corresponding to the first order reflection. Box 1903 in FIG. 21 shows a portion of the cepstrum corresponding to the second order reflection. The amplitude of the cepstrum corresponding to the first order reflection may be greater than the amplitude of the cepstrum corresponding to the second order reflection. Furthermore, the shapes of the plots of the first and second order reflections are similar. In one or more implementations, the signature in the second order reflection can be used to confirm the signature in the first order reflection with respect to the category of the user interface. In one or more implementations, the presence or absence of additional reflections (e.g., second order reflections and / or third order reflections) can be used as a signature for classifying the user interface. In one or more implementations, the ratio of the second order reflections (and / or third order reflections) to the first order reflections can be used as a signature for classifying the user interface.

[0182] As described with respect to FIG. 14A, categorization of a user interface may be performed as an independent process. Alternatively, in one or more implementations, categorization of a user interface may be performed as part of a method for characterizing a user interface to determine a particular make, type, model, etc. of the user interface. Categorization of a user interface may be performed prior to characterizing the user interface. Once a category of a user interface is determined, the category may be used to define a subset of user interfaces for characterizing the user interface. This may improve the characterization of the user interface by removing user interfaces of different categories that may adversely affect the characterization.

[0183] Alternatively, the user interface may be characterized first. The user interface may then be classified. The category of the user interface may be used to verify that the make, type, model, etc. of the determined user interface matches the determined category of the user interface. If the category does not match, the user interface may be re-characterized with or without considering the category in the characterization process.

[0184] In one or more implementations, during the characterization of a user interface, a confidence score may be determined. The confidence score may be determined, for example, based on how well the user interface is characterized. Depending on the confidence score, the characterized user interface may be classified to verify that the category of the user interface matches the type, model, manufacturer, etc. of the user interface. For example, the user interface may be classified if the confidence score is less than a threshold or greater than a threshold. The category may be used to verify the characterization of the user interface. Thus, although method 400 of FIG. 6 and method 1200 of FIG. 14A are described in different parts of this specification and shown in different flow diagrams, in one or more implementations, elements of methods 400 and 1200 may overlap such that the same steps of the same method achieve parallel results in the described methods 400 and 1200.

[0185] In some implementations of the present disclosure, analyzing the acoustic data may include estimating a length of the conduit 126. The estimated length of the conduit 126 can be used to define an analysis window of the acoustic data (e.g., a window of acoustic data that is used to characterize and / or classify a user interface after windowing the acoustic data).

[0186] 5, 6, 14A and 14B, the microphone 140 can detect reflections 310 from the features 304 and various other reflections of the acoustic signal 302. In some implementations, the reflections 310 can be used to determine the length of the conduit. The acoustic data representing the acoustic reflections 310 can be used to generate a time domain intensity signal. The time domain intensity signal is a measurement of the intensity and amplitude of the reflections 310 over a period of time as measured by the microphone 140. The time domain intensity signal can be converted to a frequency domain intensity signal, which can then be used to determine the length of the conduit 126. The length of the conduit 126 can be used to assist in windowing the acoustic data and / or identifying an acoustic signature of the user interface 124.

[0187] In some implementations, because the acoustic data represents multiple reflections from multiple acoustic signals, the time-domain intensity signal may represent more than just the acoustic signal 302. For example, reflections from each of the multiple acoustic signals may be averaged together and the time-domain intensity signal generated from the averaging.

[0188] FIG. 24A shows a plot of a frequency domain intensity signal 2400 that can be obtained from the time domain intensity signal. In some implementations, the frequency domain intensity signal 2400 can be obtained from the time domain intensity signal by Fourier transforming the time domain intensity signal. The frequency domain intensity signal 2400 represents the intensity of various different frequencies of the acoustic reflection 310 across a frequency band. Thus, the frequency domain intensity signal 2400 shows the frequency spectrum of the reflection 310 of the acoustic signal 302. In FIG. 24A, the Y-axis represents the absolute value of the intensity of the reflection 310, shown as |x|. The intensity is plotted on a logarithmic scale using arbitrary units. The X-axis represents the frequency of the reflection 310 measured in Hertz (Hz). In some implementations, the intensity is plotted on an absolute scale rather than a logarithmic scale.

[0189] 24A represents various frequencies of reflections 310 from an acoustic signal 302. The acoustic signal 302 and the reflections 310 form standing waves in the conduit 126 and / or the user interface 124. The conduit 126 can support standing waves at various frequencies therein. Generally, the lowest frequency of standing waves supported by the conduit 126 is referred to as the resonant frequency.

[0190] The wavelength of the standing wave at the resonant frequency can be obtained from the resonant frequency and is generally given by 2L C where L C is the length of conduit 126. The strength of standing waves propagating in conduit 126 at the resonant frequency is generally greater than the strength of standing waves propagating in conduit 126 at frequencies other than the resonant frequency.

[0191] The frequency domain intensity signal 2400 is indicative of the intensity of standing waves having various frequencies propagating within the conduit 126. The peaks in the frequency domain intensity signal 2400 correspond to different standing waves supported within the conduit 126. However, because the conduit 126 is coupled to both the respiratory apparatus 122 and the user interface 124, the frequency domain intensity signal 2400 is also representative of standing waves propagating partially within the respiratory apparatus 122, the user interface 124, or both. Thus, some of the peaks in the frequency domain intensity signal 2400 correspond to standing waves propagating partially within the respiratory apparatus 122, the user interface 124, or both.

[0192] For example, the frequency domain intensity signal 2400 includes a peak 2402 at a low frequency, which is larger than the remaining peaks of the frequency domain intensity signal. The peak 2402 corresponds to a standing wave propagating primarily within the respiratory device 122. The frequency domain intensity signal 2400 also includes peaks 2404, 2406, and 2408, which represent standing waves propagating at least partially within the user interface 124. Finally, the frequency domain intensity signal 2400 includes a series of peaks 2410 within the portion 2401 of the frequency domain intensity signal 2400. As shown in FIG. 24A, the peaks 2410 are arranged as a generally periodic series of peaks of decreasing intensity. The series of peaks 2410 corresponds primarily to a standing wave propagating at a resonant frequency within the conduit 126 and higher harmonics at that resonant frequency, but may also correspond to a standing wave propagating at a resonant frequency within the user interface 124 and higher harmonics thereof. These higher harmonics are 2L C / n, where n is a positive integer greater than or equal to 2.

[0193] 24B shows a plot of a wave periodic intensity signal 2420 that can be obtained from the frequency domain intensity signal 2400. In some implementations, the wave periodic intensity signal 2420 is obtained by Fourier transforming a portion 2401 of the frequency domain intensity signal 2400 that includes a periodic series of peaks 2410. The wave periodic intensity signal 2420 plots the intensity of the reflection 310 in the portion 2401 of the frequency domain intensity signal against the period of the standing wave. The Y-axis of the plot represents the intensity of the reflection 310, shown as |x|. The intensity is plotted on an absolute scale. The X-axis represents the period (e.g., the inverse of the frequency) of the reflection 310 in the portion 2401 of the frequency domain intensity signal 2400.

[0194] As shown, the wave periodic intensity signal 2420 includes a peak 2422, which is significantly larger than the other peaks of the wave periodic intensity signal 2420. Because the peak 2422 is larger than the other peaks, the peak 2422 corresponds to a standing wave propagating at a resonant frequency in the conduit 126. The location of the peak 2422 along the x-axis represents the period of the standing wave propagating at the resonant frequency. By taking the inverse of the period of this standing wave, the resonant frequency can be determined. The wavelength λ of this standing wave is then generally equal to v / f, where v is the speed of sound in the conduit 126. In some implementations, the speed of sound v is based at least in part on the temperature and / or humidity of the air in the conduit 126. Once the wavelength of the standing wave propagating at the resonant frequency is known, the length of the conduit 126 can be determined by the inverse of that frequency, λ=2L. C Therefore, the length of the conduit 126 is proportional to the wavelength of the standing wave propagating at the resonant frequency and inversely proportional to the resonant frequency.

[0195] In some implementations, a peak other than peak 2422 may be selected as the peak representing the resonant frequency of the conduit 126. For example, due to the presence of the respiratory apparatus 122 and / or the user interface 124, the intensity of a standing wave propagating at a frequency other than the resonant frequency may appear to be greater than the peak that actually represents the resonant frequency of the standing wave. In these cases, the lowest frequency peak of the wave period intensity signal 2420 may be selected as the peak representing the resonant frequency of the conduit 126.

[0196] By Fourier transforming the initial time domain intensity signal and frequency domain intensity signal 2400, along with any pre-processing steps, noise in the data can generally be minimized to more easily identify the resonant frequency of the conduit. However, in some implementations, the resonant frequency can be identified from the frequency domain intensity signal 2400 rather than the wave period intensity signal. For example, after identifying the series of peaks 2410 in the frequency domain intensity signal 2400, the resonant frequency may be selected directly from these series of peaks 2410. The peak with the lowest frequency in the series of peaks 2410 may be selected as the peak representing the resonant frequency of the conduit 126.

[0197] In some implementations, the frequency domain intensity signal 2400 itself may be windowed prior to obtaining the wave periodic intensity signal 2420. In these implementations, a portion of the frequency domain intensity signal that is expected to contain the resonant frequency of the conduit 126 is selected, and the wave periodic intensity signal 2420 is obtained from only this selected portion. By identifying the portion of the frequency domain intensity signal that is expected to contain the resonant frequency, additional portions of the frequency domain intensity signal are removed to reduce the complexity in analyzing the acoustic data. The portion of the frequency domain intensity signal that contains the resonant frequency may be selected based on a variety of factors. In some implementations, a predetermined estimate of the length of the conduit 126 is used to provide a frequency range in which the resonant frequency is expected to fall. For example, if it is known that the conduit 126 is typically between 1.5 meters and 2.0 meters, these length estimates may be used to estimate the portion of the frequency domain intensity signal in which the resonant frequency is expected to lie. Additionally or alternatively, the estimate can be based on the speed of sound within the conduit 126, the sampling rate of the acoustic data, known conditions both inside the conduit 126 and outside the conduit 126 (such as temperature or humidity that may affect the speed of sound within the conduit 126), and one skilled in the art can use other factors to estimate the portion of the frequency domain intensity signal in which the resonant frequency resides.

[0198] In some implementations, instead of applying a Fourier transform to the frequency domain intensity signal to obtain a wave periodic intensity signal, other transforms and / or analyses can be applied. For example, an inverse Fourier transform can be applied to the frequency domain intensity signal 2400 to obtain a new time domain intensity signal, and the resonant frequency can be identified from the new time domain intensity signal. In another example, a cross-correlation analysis can be applied to the initial time domain intensity signal and the frequency domain intensity signal 2400 to determine the resonant frequency of the conduit 126.

[0199] In some implementations, various different pre-processing steps can be applied to the frequency domain intensity signal or the wave period intensity signal. In one implementation, the pre-processing includes detrending, which can remove variations in the frequency domain intensity signal that are not related to the length of the conduit 126. For example, the acoustic data can include artifacts (peaks, dips, etc.) due to physical characteristics of components other than the conduit 126, external noise and / or disturbances, data noise, etc. In additional or alternative implementations, the pre-processing includes spectral windowing. Spectral windowing can be used to improve the quality of the frequency domain intensity signal, which is a Fourier transform of the time domain intensity signal. In some cases, applying a fast Fourier transform algorithm to the time domain intensity signal can introduce various artifacts and other errors into the frequency domain intensity signal. For example, if the endpoints of the time domain intensity signal are discontinuous, the frequency domain intensity signal can be slightly altered from the actual frequency domain version of the time domain intensity signal. Spectral windowing can be used to correct for discontinuities in the endpoints of the time domain intensity signal, so that the frequency domain intensity signal can more accurately represent the actual frequency spectrum of the reflections of the acoustic signal.

[0200] Thus, when windowing the generated acoustic data in step 404 of method 400, the windowing of the generated acoustic data can be based at least in part on the length of the conduit 126 determined according to FIGS. 24A and 24B. By determining the length of the conduit 126, an approximate location on the x-axis of the cepstrum where the conduit 126 ends and the user interface 124 begins can be identified. This location can then be used to define an analysis window of the acoustic data (e.g., can be used to window the acoustic data). The windowed acoustic data can then be used to characterize the user interface 124 as described herein. In some implementations, these techniques can be used to first estimate the length of the conduit 126 to provide an estimate of the location of the user interface 124. One or more fiducial points can then be identified based on the estimated location of the user interface 124, and further analysis of the user interface 124 and / or the conduit 126 can be performed. The fiducial points can be identified and the user interface 124 and / or the conduit 126 can be analyzed as described herein. Additionally, the determined length of the conduit 126 may also be used as part of any acoustic signature that aids in characterizing and / or classifying the user interface 124. In further implementations, the length of the conduit 126 may be estimated using techniques other than those described herein with respect to Figures 24A and 24B.

[0201] 25A and 25B, the determined length of the conduit 126 can also be used to adjust the characterization of the user interface 124 based on the acoustic data. As described herein, the length of the conduit 126 is assumed when analyzing the user interface 124, identifying features of the user interface 124, and characterizing and / or classifying the user interface 124. However, the conduit 126 may have a length that differs from expected due to, for example, manufacturing tolerances, different designs, stretching, damage, etc. Additionally, the apparent effective length of the conduit 126 may vary depending on the amount of time it takes sound to travel through the conduit 126. This amount of time may vary based on temperature, humidity, position / orientation of the conduit 126, etc. If the conduit 126 has a length (actual or apparent) that differs from expected, the acoustic signature identified using the techniques described herein may be unexpectedly distorted and the identified acoustic signature may not be associated with the user interface 124 that was tested / used by the user. Thus, an unexpected length of the conduit 126 may reduce the accuracy of the characterization and / or classification of the user interface 124, and various aspects related to the characterization and / or classification of the user interface 124 may need to be adjusted based on the length of the conduit 126.

[0202] 25A shows an upper plot of cepstrum 2500A and a lower plot of cepstrum 2500B. Cepstrum 2500A is determined from acoustic data generated by a user interface 124 and a first type of conduit 126. Cepstrum 2500B is determined from acoustic data generated by the same user interface 124 from a second type of conduit 126. For example, the first type of conduit 126 can be a ResMed® SlimLine® conduit and the second type of conduit 126 can be a ResMed® ClimateLine® conduit. The first and second types of conduit 126 have different lengths.

[0203] Cepstrums 2500A and 2500B are shown with quefrency on the X-axis and amplitude on the Y-axis. In these implementations, quefrency corresponds to a measure of distance along the path that an acoustic signal propagates. Thus, cepstrums 2500A and 2500B are used to indicate distances along this path associated with reflections of the acoustic signal. These reflections are in turn associated with various features of the user interface. Thus, cepstrums 2500A and 2500B can be used to characterize and / or classify the respective user interfaces 124. Although the illustrated implementation shows the use of a cepstrum to characterize and / or classify the user interfaces 124, analysis of acoustic data can generally include performing a deconvolution of the acoustic data. The deconvolution of the acoustic data can then be used to characterize and / or classify the user interfaces 124. Cepstrum is a particular type of deconvolution that can be utilized.

[0204] Since the two cepstrums 2500A and 2500B are determined from acoustic data generated using the same user interface 124, the two cepstrums 2500A and 2500B should generally have the same pattern. However, different lengths of the conduits 126 result in distortion of the cepstrum 2500B. As shown in FIG. 25A, the cepstrum 2500B is stretched or smeared relative to the cepstrum 2500A. For example, the cepstrum 2500B includes a peak 2502B that is located anteriorly relative to the corresponding peak 2502A of the cepstrum 2500A due to the stretching or smearing of the cepstrum 2500B. The distortion of the cepstrum 2500B due to different lengths of the conduits may also affect the location of the peaks along the Y-axis. If the cepstrum 2500B is elongated relative to what would be expected from the expected length of the conduit 126 used to obtain the cepstrum 2500B, the peaks may be shorter along the Y axis than they would otherwise be. Conversely, if the length of the conduit 126 is shorter than expected, the cepstrum 2500B may be more compressed than normal and the peaks may be higher along the Y axis than they would otherwise be. Thus, the characteristics of the user interface 124 obtained from the cepstrum 2500B will differ from the characteristics of the user interface 124 obtained from the cepstrum 2500A, even if both cepstra were generated using the same user interface 124. Thus, as shown in FIG. 25A, the length of the conduit 126 may adversely affect the characterization of the user interface 124.

[0205] FIG. 25B shows an upper plot of a corrected cepstrum 2504A and a lower plot of a corrected cepstrum 2504B. Cepstrum 2504A is a modified version of cepstrum 2500A and includes peak 2506A. Cepstrum 2504B is a modified version of cepstrum 2500B and includes peak 2506B. Cepstrums 2504A and 2504B have been corrected by scaling cepstrums 2500A and 2500B by a factor equal to the ratio of (i) the measured length of the respective conduit 126 to (ii) the assumed length of the conduit 126. For example, if a particular conduit is assumed to be 2 meters long, and analysis of the conduit reveals that the conduit is actually 1.8 meters long, then the new location along the x-axis of each data point in the cepstrum is its old location along the x-axis multiplied by 1.8 / 2. As shown in FIG. 25B, peak 2506A of corrected cepstrum 2504A aligns with peak 2506B of corrected cepstrum 2504B, which is expected if the actual length of conduit 126 were the same.

[0206] In the illustrated implementation, both cepstrums 2500A and 2500B are scaled to the same length to remove the stretching effect caused by the unknown length of the conduit. When two conduits 126 and their respective user interfaces 124 are being compared, scaling to the same assumed length allows for easier comparison. However, cepstrums generated from acoustic data can generally be scaled to any length. For example, in some implementations, two conduits 126 assumed to have two different lengths can each be scaled to their respective assumed lengths even though their lengths are different. In general, the cepstrum can be scaled to the assumed length to remove distortion caused by the actual length of the conduit 126. As described herein, this distortion can include the cepstrum being stretched and the cepstrum being compressed.

[0207] Thus, the acoustic data used to characterize and / or classify the conduit 126 may be stretched and / or compressed to compensate for differences in the effective lengths of the conduit 126 so that the length of the conduit 126 does not distort the characterization and / or classification of the user interface 124. This scaling (e.g., stretching and / or compressing) of the acoustic data may be performed in the time domain (time signal), frequency domain (spectrum), quefrency domain (cepstrum), other domains, or any combination of domains. The scaling may be performed on the spectrum and / or cepstrum, logarithm of the spectrum and / or cepstrum, direct transform of the spectrum and / or cepstrum, direct transform of the logarithm of the spectrum and / or cepstrum, inverse transform of the spectrum and / or cepstrum, inverse transform of the logarithm of the spectrum and / or cepstrum, or other data set. The scaling may also be performed on the acoustic data itself before any manipulations or transformations are performed on the acoustic data. Scaling can be performed using a variety of signal processing methods, such as interpolation, upsampling, phase vocoding techniques (which allow for control of the frequency shift applied to the signal), etc. Thus, scaling can be used to help characterize user interface 124, as described herein with reference to method 400 and Figure 6, and can also be used to help classify user interface 124, as described herein with reference to method 1200 and Figures 14A and 14B.

[0208] Additional details regarding determining the length of conduit 126 and / or identifying conduit 126 can be found in U.S. Provisional Application No. 63 / 107,763, filed October 30, 2020, the entire contents of which are incorporated by reference herein.

[0209] While various embodiments of the present invention have been described above, it should be understood that they are presented for illustrative purposes only, not for limitation. Many modifications to the disclosed embodiments can be made in accordance with the disclosures herein without departing from the spirit or scope of the present invention. Thus, the scope of the present invention should not be limited by any of the above-described embodiments. Rather, the scope of the present invention should be defined according to the claims and their equivalents.

[0210] One or more elements or aspects or steps or portions thereof from any one or more of claims 1 to 126 below may be combined with one or more elements or aspects or steps or portions thereof from any one or more of the other claims 1 to 126 or combinations thereof to form one or more further implementations and / or claims of the disclosure.

[0211] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Furthermore, to the extent that the terms "including," "includes," "having," "has," "with," or variations thereof are used in either the detailed description and / or claims, such terms are inclusive in a similar manner as the term "comprising."

[0212] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art. Furthermore, terms as defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense unless expressly defined.

[0213] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of and priority to U.S. Provisional Patent Application No. 63 / 108,161, filed October 30, 2020, and U.S. Provisional Patent Application No. 63 / 036,303, filed June 8, 2020, each of which is incorporated by reference in its entirety.

Claims

1. A step of generating acoustic data associated with an acoustic reflection of an acoustic signal emitted by a sound emitting means and traveling through a conduit and a user interface, the acoustic reflection being caused by the acoustic signal contacting one or more features of the user interface and detected by an acoustic sensor, the acoustic reflection at least partially indicating the one or more features of the user interface coupled to a respiratory treatment device via the conduit; Analyzing the generated acoustic data, The analyzing step includes computing a deconvolution of the generated acoustic data; computing a deconvolution of the generated acoustic data includes computing a cepstrum of the generated acoustic data; The analyzing step further includes calculating a derivative of the cepstrum to determine a rate of change of the cepstral signal; said analyzing including windowing the generated acoustic data based at least in part on at least one feature of the one or more features of the user interface; characterizing the user interface based at least in part on the analyzed acoustic data.

2. The method of claim 1 , wherein said windowing the generated acoustic data comprises determining reference points within the generated acoustic data.

3. The method of claim 2 , wherein the reference point is a minimum or maximum point within a predetermined interval of deconvolution of the generated acoustic data on a path formed at least in part by the conduit and a user interface.

4. The method of claim 3 , wherein the reference points correspond to positions of the one or more features along the path formed at least in part by the conduit and the user interface that correspond to the predetermined interval of deconvolution of the generated acoustic data.

5. the one or more features cause a change in acoustic impedance; the change in acoustic impedance is due, at least in part, to a narrowing or widening of the pathway at (i) the user interface, (ii) a junction of the conduit and the user interface, or (iii) a combination thereof; The method according to claim 4.

6. the windowing process includes a first windowing process of the generated acoustic data obtained from a first range defined by a position before a reference point and a position after the reference point located on a path formed at least in part by the conduit and a user interface, and a second windowing process of the generated acoustic data obtained from a second range defined by a different position before the reference point and a different position after the reference point; the first windowing process and the second windowing process vary depending on an amount of the generated acoustic data selected from before the reference point in the generated acoustic data, after the reference point in the generated acoustic data, or a combination thereof.

6. The method according to any one of claims 1 to 5.

7. the cepstrum identifies a distance associated with an acoustic reflection between a position of the one or more features of the user interface and a position of the sound emitting means; the distance being indicative of a location of one or more features of the user interface; The method of claim 1.

8. analyzing the generated acoustic data includes normalizing the generated acoustic data by subtracting a mean value from the generated acoustic data, dividing by a standard deviation of the generated acoustic data, or a combination thereof.

8. The method according to any one of claims 1 to 7.

9. 9. The method of claim 1, wherein characterizing the user interface comprises inputting the analyzed acoustic data into a convolutional neural network to determine a form factor of the user interface, a model of the user interface, a size of one or more elements of the user interface, or a combination thereof.

10. 10. The method of claim 9, wherein the convolutional neural network includes N features with max pooling of M samples of the N features, where the ratio of N to M is between 1:1 and 4:

1.

11. 11. The method of claim 1, further comprising emitting the acoustic signal to the conduit connected to the user interface via an acoustic transducer or via a motor of the respiratory treatment device connected to the conduit.

12. The method of claim 11 , wherein the acoustic signal is an ultrasonic sound.

13. The method of claim 1 , wherein the user interface is not connected to a user during the generation of the acoustic data.

14. The method of claim 1 , further comprising providing a flow of pressurized air through the conduit and into the user interface during the generation of the acoustic data.

15. the generated acoustic data is generated from a plurality of acoustic reflections from a plurality of acoustic signals; the generated acoustic data is an average of the acoustic reflections from the acoustic signals.

15. The method according to any one of claims 1 to 14.

16. analyzing the generated acoustic data includes identifying one or more signatures that correlate to the one or more features of the user interface; the one or more signatures each comprise one or more acoustic features within the generated acoustic data; The method of claim 1 , further comprising classifying the user interface based at least in part on the one or more signatures.

17. The user interface is characterized from a subset of user interfaces; The method of claim 16 , wherein the subset of user interfaces is defined based at least in part on a category of the user interfaces.

18. 18. The method of claim 16 or claim 17, further comprising validating the characterized user interfaces based at least in part on the characterized user interfaces satisfying the user interface category.

19. determining a confidence score that the user interface is correctly characterized; The method of claim 18 , wherein validating the characterized user interface occurs when the confidence score meets a confidence threshold.

20. A memory storing machine-readable instructions; generating acoustic data associated with an acoustic reflection of an acoustic signal emitted by a sound emitting means and traveling through a conduit and a user interface, the acoustic reflection being caused by the acoustic signal contacting one or more features of the user interface and detected by an acoustic sensor and at least partially indicative of the one or more features of the user interface coupled to a respiratory treatment device via the conduit; Analyzing the generated acoustic data, The analyzing step includes computing a deconvolution of the generated acoustic data; computing the deconvolution of the generated acoustic data includes computing a cepstrum of the generated acoustic data; The analyzing step further includes calculating a derivative of the cepstrum to determine a rate of change of the cepstral signal; said analyzing including windowing the generated acoustic data based at least in part on at least one feature of the one or more features of the user interface; characterizing the user interface based at least in part on the analyzed acoustic data; a control system including one or more processors configured to execute the machine-readable instructions to perform the Including, the system.

21. 21. The system of claim 20, wherein said windowing the generated acoustic data comprises determining reference points within the generated acoustic data.

22. 22. The system of claim 21, wherein the reference point is a minimum or maximum point within a predetermined interval of deconvolution of the generated acoustic data on a path formed at least in part by the conduit and a user interface.

23. 23. The system of claim 22, wherein the reference points correspond to positions of the one or more features along the path formed at least in part by the conduit and the user interface that correspond to the predetermined interval of deconvolution of the generated acoustic data.

24. the one or more features cause a change in acoustic impedance; the change in acoustic impedance is due, at least in part, to a narrowing or widening of the pathway at (i) the user interface, (ii) a junction of the conduit and the user interface, or (iii) a combination thereof; 24. The system of claim 23.

25. the windowing process includes a first windowing process of the generated acoustic data obtained from a first range defined by a position before a reference point and a position after the reference point located on a path formed at least in part by the conduit and a user interface, and a second windowing process of the generated acoustic data obtained from a second range defined by a different position before the reference point and a different position after the reference point; the first windowing process and the second windowing process vary depending on an amount of the generated acoustic data selected from before the reference point in the generated acoustic data, after the reference point in the generated acoustic data, or a combination thereof. A system according to any one of claims 20 to 24.

26. the cepstrum identifies a distance associated with an acoustic reflection between a position of the one or more features of the user interface and a position of the sound emitting means; the distance being indicative of a location of one or more features of the user interface; 21. The system of claim 20.

27. 27. The system of claim 20, wherein analyzing the generated acoustic data comprises normalizing the generated acoustic data by subtracting a mean value from the generated acoustic data, dividing by a standard deviation of the generated acoustic data, or a combination thereof.

28. 28. The system of claim 20, wherein characterizing the user interface comprises inputting the analyzed acoustic data into a convolutional neural network to determine a form factor of the user interface, a model of the user interface, a size of one or more elements of the user interface, or a combination thereof.

29. 30. The system of claim 28, wherein the convolutional neural network includes N features with max pooling of M samples of the N features, where the ratio of N to M is between 1:1 and 4:

1.

30. 30. The system of any one of claims 20 to 29, wherein the one or more processors are further configured to execute the machine-readable instructions to cause the conduit connected to the user interface to emit the acoustic signal via an acoustic transducer or via a motor of the respiratory treatment device connected to the conduit.

31. 31. The system of claim 30, wherein the acoustic signal is an ultrasonic sound.

32. 32. The system of claim 20, wherein the user interface is not connected to a user during the generation of the acoustic data.

33. 33. The system of any one of claims 20 to 32, wherein the one or more processors are further configured to execute the machine-readable instructions to generate the acoustic data while providing a flow of pressurized air through the conduit and into the user interface.

34. the generated acoustic data is generated from a plurality of acoustic reflections from a plurality of acoustic signals; the generated acoustic data is an average of the acoustic reflections from the acoustic signals. A system according to any one of claims 20 to 33.

35. analyzing the generated acoustic data includes identifying one or more signatures that correlate to the one or more features of the user interface; The system further includes classifying the user interface based at least in part on the one or more signatures; the one or more signatures each being comprised of one or more acoustic features within the generated acoustic data. A system according to any one of claims 20 to 34.

36. The user interface is characterized from a subset of user interfaces; the subset of user interfaces is defined based at least in part on a category of the user interfaces.

36. The system of claim 35.

37. 37. The system of claim 35 or claim 36, wherein the system further comprises validating the characterized user interfaces based at least in part on the characterized user interfaces satisfying the user interface category.

38. The system further includes determining a confidence score that the user interface is correctly characterized; 40. The system of claim 37, wherein validating the characterized user interface occurs when the confidence score meets a confidence threshold.

Images