System for controlling sleep breathing disorder therapy device

Abstract

A system and method for controlling a sleep breathing disorder (SDB) therapy device. While a subject is undergoing therapy using an SDB therapy device, a signal is obtained and processed that is responsive to a change in sympathetic activity of the subject caused by respiration and baroreceptors of the subject to detect a change in sympathetic activity. For a change in sympathetic activity that exceeds a predetermined threshold, it is determined whether the change is chemoreceptor-induced or baroreceptor-induced. A control signal for controlling the SDB therapy device is generated depending on whether the change is chemoreceptor initiated or baroreceptor initiated.

Description

Technical Field

[0001] The present invention relates to the field of sleep breathing disorder therapy devices. Background Art

[0002] Sleep disordered breathing (SDB) diseases, such as obstructive sleep apnea (OSA) and central sleep apnea (CSA), are becoming more and more common. SDB diseases are usually treated with SDB therapy devices, such as positive airway pressure (PAP) devices, mandibular advancement devices (MADs), and positioning therapy devices.

[0003] PAP devices provide pressurized air to a subject to maintain the subject's airway open, thereby reducing SBD events. However, in subjects with cardiac comorbidities, the positive pressure of a PAP device may cause changes in intrathoracic pressure, which in turn can alter various cardiac parameters (e.g., heart rate, blood pressure, cardiac filling, ejection rate, stroke volume, and cardiac output). For example, in patients with chronic heart failure, changes in intrathoracic pressure cause changes in venous return, atrial filling, and stroke volume.

[0004] Changes in cardiac parameters activate baroreceptors, causing arousals. As a result, subjects with SDB and cardiac comorbidities often experience prolonged sleep onset latency and insomnia. The reduced sleep quality experienced by these subjects often results in reduced compliance with prescribed PAP therapy, as subjects often choose to reduce their use of the PAP device (e.g., use the PAP device for less time than recommended during the night) or even stop using the PAP device altogether if they believe that use of the PAP device will result in worse sleep.

[0005] Therefore, there is a need for improved SDB therapies, especially for subjects with SDB and cardiac comorbidities.

[0006] US patent application US7942824A1 discloses an integrated sleep diagnostic and therapeutic device, and more specifically, an integrated apnea diagnostic and therapeutic device. The sleep disorder treatment system may use a diagnostic device to perform various forms of analysis to determine or diagnose a subject's sleep disorder or a symptom of a subject's sleep disorder. Summary of the invention

[0007] The invention is defined by the claims.

[0008] According to an example of one aspect of the present invention, there is provided a processing system for controlling a sleep disordered breathing (SDB) therapy device that provides therapy to a subject during the subject's sleep period, the processing system being configured to: receive a first signal responsive to the subject's breathing during the sleep period; receive a second signal responsive to a change in sympathetic nerve activity triggered by the subject's baroreceptors; process the first signal and the second signal to detect a change in sympathetic nerve activity; in response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold: process the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and based on the determination of whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, generate a control signal for controlling the SDB therapy device.

[0009] By generating a control signal based on whether the change in the subject's sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, the SDB therapy device can be adjusted, which results in fewer arousals, thereby improving the subject's sleep quality.

[0010] The inventors have recognized that if the therapy provided by an SDB therapy device is adjusted to cause a slight increase in SBD events but a decrease in cardiac stress, then subjects who experience cardiac stress during typical use of the therapy device will experience improved sleep quality. In other words, the present invention is intended to provide a balance between reducing SDB events and limiting baroreceptor-induced arousals.

[0011] In some examples, the SDB therapy device is a positive airway pressure (PAP) device; wherein the control signal includes a first control signal, wherein the processing system is configured to: determine an regulated pressure to be provided by the PAP device based on a determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and generate the first control signal to control the PAP device to provide air to the subject at the regulated pressure.

[0012] The present invention is particularly beneficial for use with PAP devices because the pressure delivered by a PAP device may cause cardiac stress (and therefore baroreceptor-induced sympathetic activity and arousal). Baroreceptors are located in the ventricles and vascular structures. Receptor activation induces a sympathetic response. A strong sympathetic response can induce arousal.

[0013] By recognizing that the subject is experiencing baroreceptor-induced sympathetic activity, a regulated pressure can be determined that will reduce the baroreceptor-induced sympathetic activity so as to reduce the likelihood of baroreceptor-induced arousal. Similarly, if the subject is experiencing chemoreceptor-induced sympathetic activity, a regulated pressure can be determined that will reduce SDB events.

[0014] By automatically controlling the PAP device, the pressure of the PAP device can be adjusted in real time in an attempt to keep sympathetic nerve activity below a level that causes arousal. In this way, the pressure provided by the PAP device can be adjusted throughout the sleep period to reduce the likelihood of arousal during the sleep period, thereby improving sleep quality.

[0015] In some examples, in response to determining that the detected change in sympathetic nerve activity is chemoreceptor-induced, the adjusted pressure is an increased pressure; and in response to determining that the detected change in sympathetic nerve activity is baroreceptor-induced, the adjusted pressure is a decreased pressure.

[0016] The inventors have recognized that chemoreceptor-induced sympathetic nerve activity is the result of respiratory defects, indicating insufficient air pressure from the PAP device, while baroreceptor-induced activity is due to cardiac stress caused by excessive pressure. Chemoreceptors are located in the vascular structure of the lungs. Chemoreceptors are activated in response to insufficient oxygen intake or insufficient carbon dioxide release.

[0017] Thus, arousal can be minimized by increasing the pressure of the PAP device in response to chemoreceptor-induced sympathetic activity, and decreasing the pressure in response to baroreceptor-induced sympathetic activity.

[0018] In some examples, the processing system is configured to: generate a first control signal configured to control the PAP device to provide air to the subject at an adjusted pressure for a predetermined time period; and in response to the lapse of the predetermined time period, generate a second control signal configured to control the PAP device to provide air to the subject at a previous pressure.

[0019] The predetermined period of time may, for example, be measured in terms of absolute time, or may be a predetermined number of breathing cycles.The previous pressure may be the pressure at which the PAP device provided air to the subject immediately prior to adjustment of the pressure.

[0020] In some examples, the processing system is configured to: continue processing the first signal and the second signal to detect a change in sympathetic nerve activity; and in response to the change in sympathetic nerve activity failing to exceed a predetermined activity threshold, generate a second control signal configured to control the PAP device to provide air to the subject at a previous pressure.

[0021] In other words, once sympathetic nerve activity is restored to normal levels, the PAP device returns to its previous setting.

[0022] In some examples, the processing system is configured to repeatedly, at predetermined intervals: process the first signal and the second signal to detect a change in sympathetic nerve activity; in response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold: process the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; based on the determination of whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, determine an regulated pressure to be provided by the PAP device; and generate a first control signal configured to control the PAP device to provide air to the subject at the regulated pressure.

[0023] This allows the pressure of the PAP device to be adjusted to provide appropriate pressure throughout the sleep period, thereby keeping sympathetic nerve activity within normal levels. In this way, arousals can be reduced and the subject's sleep quality during the sleep period can be improved.

[0024] In some examples, the processing system is configured to: determine an initial adjusted pressure value based on a determination of whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; obtain a pressure threshold; compare the initial adjusted pressure value to the pressure threshold; in response to the initial adjusted pressure value exceeding the pressure threshold, set the pressure threshold to the adjusted pressure; and in response to the initial adjusted pressure value failing to exceed the pressure threshold, set the initial adjusted pressure value to the adjusted pressure.

[0025] In other words, a limit on the maximum pressure provided by the PAP device to the subject can be established.The pressure threshold can be a subject-specific value; for example, the subject-specific pressure threshold can be determined by a clinician during a titration period for the subject.

[0026] In some examples, the processing system is configured to obtain the stress threshold by: estimating a current sleep stage of the subject; and selecting the stress threshold from a set of stress thresholds based on the estimated current sleep stage.

[0027] The inventors have recognized that a subject can tolerate higher levels of pressure (i.e., no arousal occurs) when in deep sleep than in a light sleep stage. By selecting a pressure threshold based on the estimated current sleep stage, a higher pressure can be provided to the subject during deep sleep as needed, while the pressure provided during light sleep is limited, thereby reducing the likelihood of arousal.

[0028] The current sleep stage may be estimated by processing a first signal responsive to the subject's breathing, or, where used, a signal responsive to heart rate variability. Alternatively, the estimated sleep stage may be based on the current night time or the number of hours the PAP device has been used during the current sleep session.

[0029] In some examples, wherein, in response to the initial adjusted pressure value exceeding the pressure threshold, the processing system is further configured to generate a third control signal configured to control the PAP device to provide supplemental oxygen to the subject.

[0030] The provision of supplemental oxygen can reduce SDB events at lower pressure settings of the PAP device.

[0031] In some examples, the predetermined activity threshold is determined by: using a machine learning algorithm that is trained to predict the likelihood of arousal of the subject based on changes in sympathetic nervous activity; and wherein the predetermined activity threshold is set to a value of change in sympathetic nervous activity for which the likelihood of arousal predicted by the machine learning algorithm exceeds a predetermined likelihood of arousal.

[0032] In other words, the predetermined threshold is a subject-specific threshold, above which arousal may occur if no action is taken.

[0033] In some examples, the processing system is configured to determine whether the detected change in sympathetic nerve activity is induced by chemoreceptors or baroreceptors by: processing the first signal and the second signal to determine whether the detected change in sympathetic nerve activity corresponds to a sleep disordered breathing event; in response to a determination that the detected change in sympathetic nerve activity corresponds to a sleep disordered breathing event, determining that the detected change in sympathetic nerve activity is induced by chemoreceptors; and in response to a determination that the detected change in sympathetic nerve activity does not correspond to a sleep disordered breathing event, determining that the detected change in sympathetic nerve activity is induced by baroreceptors.

[0034] In other words, the presence of an SDB event immediately preceding a detected change in sympathetic nerve activity indicates that the change is chemoreceptor initiated, whereas the absence of an SDB event indicates that the change is baroreceptor initiated.

[0035] In some examples, the processing system is configured to generate a control signal by: processing a determination as to whether a detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced to determine control instructions for controlling the SDB therapy device; and generating a fourth control signal configured to control an output user interface to provide a user-perceivable output of the control instructions.

[0036] In other words, rather than directly controlling the SDB therapy device, the processing system may provide instructions to the user for controlling the SDB therapy device (eg, instructions to adjust one or more settings of the SDB therapy device).

[0037] A system for controlling sleep disordered breathing (SDB) is also provided, the therapy device providing therapy to a subject during a sleep period of the subject. The system includes: one or more first sensors, the one or more first sensors are configured to generate a first signal in response to the breathing of the subject during the sleep period; one or more second sensors, the one or more second sensors are configured to generate a second signal in response to changes in sympathetic nerve activity triggered by the subject's baroreceptors; and the above-mentioned processing system, the processing system is configured to receive the first signal and the second signal from the one or more first sensors and the one or more second sensors, respectively.

[0038] The one or more first sensors may include an airflow sensor. The one or more second sensors may include, for example, a heart sensor, a galvanic skin response sensor, a brain activity sensor, a muscle sensor, a pulse oximeter sensor, an exhaled gas sensor, a pressure sensor, and / or an ultrasonic sensor.

[0039] According to another aspect of the present invention, a computer-implemented method for controlling a sleep disordered breathing (SDB) therapy device is provided, the therapy device providing therapy to a subject during the subject's sleep period, the computer-implemented method comprising: receiving a first signal responsive to the subject's breathing during the sleep period; receiving a second signal responsive to a change in sympathetic nerve activity triggered by baroreceptors of the subject; processing the first signal and the second signal to detect a change in sympathetic nerve activity; in response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold: processing the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and based on the determination of whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, generating a control signal for controlling the SDB therapy device.

[0040] A computer program product is also provided, comprising computer program code means which, when executed on a computing device having a processing system, causes the processing system to carry out all of the steps of the above method.

[0041] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] For a better understanding of the invention, and to more clearly show how the same may be put into practice, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0043] Figure 1 A system for controlling a sleep disordered breathing (SDB) therapy device that provides therapy to a subject during a sleep period of the subject is illustrated in accordance with one embodiment of the present invention;

[0044] Figure 2 A system for controlling a SDB therapy device that provides therapy to a subject during a sleep period of the subject is illustrated according to another embodiment of the present invention; and

[0045] Figure 3 A computer-implemented method for controlling a SDB therapy device that provides therapy to a subject during the subject's sleep period is illustrated in accordance with one embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be described with reference to the accompanying drawings.

[0047] It should be understood that the detailed description and specific examples, although indicating exemplary embodiments of the apparatus, system and method, are for illustrative purposes only and are not intended to limit the scope of the invention. These and other features, aspects and advantages of the apparatus, system and method of the present invention will become better understood based on the following description, the appended claims and the accompanying drawings. It should be understood that the drawings are merely schematic and are not drawn to scale. It should also be understood that the same reference numerals are used to represent the same or similar parts in all drawings.

[0048] The present invention provides a system and method for controlling a sleep disordered breathing (SDB) therapy device. When a subject is undergoing therapy using the SDB therapy device, a signal of a change in the subject's sympathetic nerve activity responsive to the subject's breathing and baroreceptor-induced is obtained and processed to detect the change in sympathetic nerve activity. For a change in sympathetic nerve activity that exceeds a predetermined threshold, it is determined whether the change is chemoreceptor-induced or baroreceptor-induced. Based on whether the change is chemoreceptor-induced or baroreceptor-induced, a control signal for controlling the SDB therapy device is generated.

[0049] Embodiments are based at least in part on the recognition that therapies aimed at reducing sleep disordered breathing events can cause increases in baroreceptor-induced sympathetic activity, thereby causing increases in arousals. By controlling therapy based on whether a subject is experiencing changes in chemoreceptor- or baroreceptor-induced sympathetic activity, a compromise between reducing sleep disordered breathing events and reducing baroreceptor-induced arousals can be achieved, thereby improving the subject's sleep quality and increasing the subject's likelihood of continuing SDB therapy.

[0050] For example, the illustrative embodiments may be used in sleep disordered breathing therapy systems, including positive airway pressure systems (eg, APAP or BiPAP systems), mandibular advancement systems, and positional therapy systems.

[0051] Figure 1 A system 100 is illustrated for controlling a sleep disordered breathing (SDB) therapy device 110 that provides therapy to a subject 120 during a sleep period of the subject, according to one embodiment of the present invention. System 100 includes one or more first sensors 130, one or more second sensors 140, and a processing system 150. Processing system 150 itself is one embodiment of the present invention.

[0052] exist Figure 1 In the embodiment, the SDB therapy device 110 is a positive airway pressure (PAP) device; however, as explained in more detail below, in some embodiments, the system 100 can be used to control other types of SDB therapy devices, such as a mandibular advancement device (MAD) and / or a positioning therapy device. Figure 1 The PAP device shown includes a PAP machine that generates compressed air, a mask that fits over the subject's nose and mouth, and a tube that connects the mask to the PAP machine; however, any PAP device may be used (e.g., a CPAP machine, a BiPAP machine, or an APAP machine, a mask that fits only over the nose or mouth, etc.).

[0053] The subject 120 is preferably a subject having a sleep disordered breathing condition (eg, OSA or CSA) and a cardiac comorbidity (eg, hypertension, coronary artery disease, chronic heart failure, etc.).

[0054] exist Figure 1 In some embodiments, the processing system 150 is disposed within the PAP device 110; however, in some embodiments, the processing system 150 may be provided within a device separate from the SDB therapy device (e.g., the processing system may be configured to directly remotely control the SDB therapy device, and / or the processing system may control the SDB therapy device by providing control instructions to an output user interface, as described below).

[0055] One or more first sensors 130 are configured to generate a first signal 135 in response to the subject's breathing during a sleep period. Figure 1 In the present invention, one or more first sensors are disposed within the PAP device and include an airflow sensor. The inventors have advantageously recognized that airflow sensors are conventionally included in PAP devices, and therefore a signal responsive to the subject's breathing can be obtained for the subject using the PAP device without the need for an additional sensor.

[0056] In some examples, the one or more first sensors 130 may additionally or alternatively include one or more sensors external to the SDB therapy device. Suitable sensors for generating signals in response to the subject's breathing (e.g., by measuring chest and / or abdominal movement and / or by measuring breathing sounds) will be apparent to the skilled person. For example, the one or more sensors may include at least one of a microphone, an accelerometer, a gyroscope, an inertial measurement unit (IMU), an electrocardiogram (SCG) sensor, and / or a ballistocardiography (BCG) sensor. The one or more first sensors may be appropriately disposed in the SDB therapy device (e.g., in a PAP machine of a PAP device and / or in a mask of a PAP device), in a wearable device worn by the subject 120 (e.g., a smart watch, a chest strap, and / or an abdominal strap), and / or in a device positioned close to the subject during a sleep period (e.g., a microphone of a subject's smartphone).

[0057] One or more second sensors 140 are configured to generate second signals 145 in response to changes in baroreceptor-induced sympathetic nerve activity of subject 120 during a sleep period. Figure 1 In some embodiments, one or more second sensors are disposed within a smartwatch; however, one skilled in the art will readily appreciate that one or more second sensors may additionally or alternatively be disposed within other devices (e.g., other wearable devices such as a chest strap, an abdominal strap, a finger-mounted device, a wearable patch, etc.). In some embodiments, one or more second sensors (one or more) may be disposed within a SDB therapy device (e.g., within a mask of a PAP device).

[0058] One or more second sensors 140 can measure one or more of: cardiac activity (e.g., in response to changes in heart rate variability), brain activity (e.g., in response to activity associated with cortical arousal), muscle activity, sympathetic skin responses (e.g., in response to changes in skin temperature, sweating, and / or electrical skin impedance), blood oxygen levels, exhaled CO2 concentration, and / or changes in intrathoracic pressure.

[0059] For example, cardiac activity (particularly heart rate, heart rate variability, and / or blood pressure) can be measured using an electrocardiogram (ECG) sensor, a pulse oximetry sensor (e.g., a photoplethysmography (PPG) sensor), a ballistocardiography (BCG) sensor, a shockcardiography (SCG) sensor, and / or a blood pressure sensor. Brain activity can be measured using an electroencephalogram (EEG) sensor. Muscle activity can be measured using an electromyogram (EMG) sensor. Sympathetic skin response can be measured using a galvanic skin response sensor (also known as a galvanic skin response sensor). Blood oxygen level can be measured using a PPG sensor. Exhaled CO2 concentration can be measured using an exhaled gas sensor.

[0060] Various types of sensors can be used to (indirectly) measure changes in intrathoracic pressure. For example, one or more second sensors 140 can obtain alternative measurements of respiratory effort (e.g., PPG sensors can be used to measure regulation of peripheral blood flow and blood accumulation) or alternative measurements of neural respiratory drive (e.g., one or more surface EMG sensors can be used to measure electrical activity occurring in one or more muscles used for breathing). Other sensors that can be used to indirectly measure changes in intrathoracic pressure include respiratory belts, suprasternal notch pressure sensors, and / or ultrasound patches (e.g., for monitoring diaphragmatic activity and / or cardiac output).

[0061] The processing system 150 is configured to receive the first signal 135 from the one or more first sensors 130 and the second signal 145 from the one or more second sensors 140, and process the first signal and the second signal to detect a change in sympathetic nerve activity. If there is no change in the detected sympathetic nerve activity, the processing system may continue to receive and process the first signal and the second signal until a change in the sympathetic nerve activity is detected.

[0062] The detected change in sympathetic nerve activity may be a detected change in one or more parameters derivable from the first signal 135 and / or the second signal 145, which parameters change in response to the sympathetic nerve activity. For example, depending on the type(s) of sensors used to obtain the first signal and the second signal, the detected change in sympathetic nerve activity may include one or more of the following: a change in respiratory rate, a change in tidal volume, a change in airflow pattern, a change in heart rate, a change in heart rate variability, a change in blood pressure, a change in sympathetic skin response, a change in brain activity indicative of cortical arousal, a change in muscle activity indicative of cortical arousal, a change in blood oxygen level, and / or a change in exhaled CO2 concentration.

[0063] After detecting the change in sympathetic nerve activity, the processing system 150 determines whether the detected change in sympathetic nerve activity exceeds a predetermined activity threshold. The predetermined activity threshold can be a threshold that can distinguish between a change in sympathetic nerve activity that is too small to cause arousal and a change in sympathetic nerve activity that may cause arousal if no action is taken.

[0064] In some examples, the predetermined activity threshold may be a generic threshold that may be expected to distinguish between changes that may and may not cause arousal of any subject. The generic predetermined activity threshold may be a relatively low threshold to avoid excluding changes in sympathetic nervous activity that may cause arousal, particularly in light sleepers. A suitable value for the predetermined activity threshold will depend on the type of change in sympathetic nervous activity detected (i.e., the parameter(s) of the detected change may be derived from the first signal and / or the second signal).

[0065] In some examples, the predetermined activity threshold (for a given parameter) can be determined based on the phenotype of the subject. For example, the subjects can be grouped into phenotypes based on one or more of: age, body mass index, type of comorbidity, level of comorbidity, and / or severity of sleep-disordered breathing. A different predetermined activity threshold can be determined for each phenotype.

[0066] In other examples, the predetermined activity threshold (for a given parameter) can be a subject-specific threshold determined during a titration period for the subject by monitoring the physiological effects of changes in SDB therapy (e.g., changes in PAP settings) provided to the subject during the titration period.

[0067] In yet other examples, the predetermined activity threshold (for a given parameter) can be a subject-specific machine learning threshold. For example, the predetermined activity threshold can be determined by training a machine learning algorithm to predict the likelihood of arousal of a subject based on changes in sympathetic nerve activity. The predetermined activity threshold can then be set to a value of change in sympathetic nerve activity for which the likelihood of arousal predicted by the trained machine learning algorithm exceeds the predetermined likelihood of arousal.

[0068] A machine learning algorithm is any self-training algorithm that processes input data to generate or predict output data. Here, the input data includes changes in sympathetic nerve activity, and the output data includes the predicted likelihood of arousal if no action is taken.

[0069] Suitable machine learning algorithms for the present invention will be apparent to the skilled person. Examples of suitable machine learning algorithms include decision tree algorithms and artificial neural networks. Other machine learning algorithms, such as logistic regression, support vector machines, or naive Bayes models are suitable alternatives.

[0070] The structure of artificial neural networks (or simply, neural networks) is inspired by the human brain. Neural networks consist of layers, each layer includes multiple neurons. Each neuron contains a mathematical operation. Specifically, each neuron can include a different weighted combination of a single type of transformation (for example, the same type of transformation, sigmoid transformation, etc., but with different weights). In the process of processing input data, each neuron performs a mathematical operation on the input data to produce a numerical output. The output of each layer in the neural network is fed into the next layer in turn. The last layer provides the output.

[0071] Methods for training machine learning algorithms are well known. Typically, such methods include obtaining a training data set that includes training input data entries and corresponding training output data entries. An initialized machine learning algorithm is applied to each input data entry to generate a predicted output data entry. The error between the predicted output data entry and the corresponding training output data entry is used to modify the machine learning algorithm. The process can be repeated until the error converges and the predicted output data entry is sufficiently similar to the training output data entry (e.g., 1%). This is generally referred to as a supervised learning technique.

[0072] For example, in the case where the machine learning algorithm is formed by a neural network, the (weights) of the mathematical operations of each neuron can be modified until the error converges. Known methods of modifying neural networks include gradient descent, back propagation algorithms, and the like.

[0073] The training input data entries correspond to changes in the subject's sympathetic nerve activity. The training output data entries correspond to information about whether each change in sympathetic nerve activity causes arousal.

[0074] The training data may be obtained during one or more training sleep sessions. During one or more training sleep sessions, an initial predetermined activity threshold may be applied. Alternatively, no action to control the SDB therapy device may be taken during the training sleep session(s) (i.e., the training sleep session may be used purely to collect data).

[0075] In some examples, the predetermined activity threshold can be iteratively modified during one or more training sleep sessions until one or more predetermined conditions are met. The one or more predetermined conditions can be related to, for example, the number of hours of uninterrupted sleep or the number / frequency of awakenings during the sleep session. For example, the predetermined activity threshold can be iteratively modified by a machine learning algorithm, as described below, to control the SDB therapy device in response to changes in sympathetic nerve activity that exceed the current value of the predetermined activity threshold until four hours of uninterrupted sleep are achieved.

[0076] In response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold, processing system 150 is configured to process first signal 135 and second signal 145 to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced.

[0077] In some examples, a determination as to whether a detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced is based on whether the detected change in sympathetic nerve activity corresponds to a sleep disordered breathing (SDB) event (e.g., an obstructive apnea or hypopnea event). For example, in response to a determination that the detected change corresponds to an SDB event, the detected change in sympathetic nerve activity can be determined to be chemoreceptor-induced. Similarly, in response to a determination that the detected change does not correspond to an SDB event, the detected change in sympathetic nerve activity can be determined to be baroreceptor-induced.

[0078] To determine whether the detected change in sympathetic nerve activity corresponds to an SDB event, the processing system 150 may be configured to process the first signal 135 to detect an SDB event. For example, the processing system may determine values ​​of one or more respiratory parameters (e.g., inspiratory length, expiratory length, rise time of a respiratory cycle, fall time of a respiratory cycle, tidal volume, inspiratory flow dynamics, expiratory flow dynamics, etc.) based on the first signal, and process the value(s) to detect an SDB event.

[0079] Methods for detecting SDB events based on signals responsive to a subject's breathing are known: see, e.g., Nakano et al. (2007), "Automatic Detection of Sleep Disordered Breathing from Single-Channel Airflow Recordings," European Respiratory Journal, 29:728-736; Haidar et al. (2017), "Detection of Sleep Apnea Events from Nasal Airflow Using Convolutional Neural Networks," ICONIP 2017: Neural Information Processing, pp. 819-827; Yang et al. (2019), "Network-Based Detection of Sleep Apnea and Hypopnea Events Based on Airflow Signals," 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 2576-2579; Koley and Dey (2013), "Real-Time Adaptive Apnea and Hypopnea Event Detection Method for Portable Sleep Apnea Monitoring Devices," IEEE Transactions on Biomedical Engineering, pp. 2576-2579. Biomedical Engineering)》, 60(12):3354-3363.

[0080] Additional physiological parameters such as cardiac parameters and / or blood oxygen levels (e.g., derivable from the second signal 145) can be used in combination with one or more respiratory parameters to detect SDB events. Methods for detecting SDB events using additional physiological parameters in combination with one or more respiratory parameters are known: for example, see John et al. (2021), "SomnNET: SpO2-based deep learning network for sleep apnea detection in smartwatches," 2021 "43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society", pp. 1961-1964; and Saha et al. (2020), "Portable diagnosis of sleep apnea and validation of individual event detection," Sleep Medicine, 69:51-57.

[0081] In response to failing to detect an SDB event, processing system 150 may determine that the detected change in sympathetic nerve activity does not correspond to an SDB event, and therefore determine that the change in sympathetic nerve activity is baroreceptor-induced.

[0082] In response to the detected SDB event, the processing system 150 may determine a time between the detected SDB event and the detected change in sympathetic nerve activity, and determine whether the change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced based on whether the time between the SDB event and the change in sympathetic nerve activity exceeds a predetermined time interval. In response to the time between the SDB event and the change in sympathetic nerve activity exceeding the predetermined time interval, the change in sympathetic nerve activity may be determined to be baroreceptor-induced, and in response to the time between the SDB event and the change in sympathetic nerve activity failing to exceed the predetermined time interval, it may be determined to be chemoreceptor-induced.

[0083] Changes in sympathetic nerve activity caused by baroreceptors tend to occur within one respiratory cycle of an SDB event, while changes in sympathetic nerve activity caused by chemoreceptors generally occur within several respiratory cycles after an SDB event (e.g., after about 5 respiratory cycles). Therefore, the predetermined time interval can be the time between 2 and 5 respiratory cycles (e.g., 3 respiratory cycles). The predetermined time interval can be determined based on a population average of respiratory cycle lengths (e.g., based on an average respiratory rate of 12-15 breaths per minute). Alternatively, the average respiratory cycle of the subject 120 can be determined based on the first signal 135 and is used to determine the predetermined time interval.

[0084] In some examples, processing system 150 may additionally or alternatively determine whether a detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced based on whether the detected change includes a change in oxygen saturation and / or a change in carbon dioxide concentration in the subject's exhaled air, where this information is available (i.e., where one or more sensors include a PPG sensor and / or an exhaled gas sensor). For example, in response to a change in oxygen saturation detected in conjunction with an SDB event, the detected change in sympathetic nerve activity may be determined to be chemoreceptor-induced. Similarly, in response to a change in exhaled CO2 concentration detected in conjunction with an SDB event, the detected change in sympathetic nerve activity may be determined to be chemoreceptor-induced.

[0085] The processing system 150 is then configured to process the determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced to generate a control signal for controlling the SDB therapy device 110. The generated control signal may directly control the SDB therapy device and / or control an output user interface ( Figure 1) to output control instructions for controlling the SDB therapy device.

[0086] In embodiments where SDB therapy device 110 is a PAP device, the control signal may be configured to control the pressure at which the PAP device provides air to subject 120 based on whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced. In other words, processing system 150 may be configured to process a determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced to determine a regulated pressure of the PAP device, and to generate control signal 155a configured to control the PAP device to provide air to the subject at the regulated pressure (e.g., by controlling the speed of motor 160 of the PAP device).

[0087] For example, in response to determining that the detected change in sympathetic nerve activity is induced by chemoreceptors, processing system 150 may determine that the pressure of air provided to subject 120 should be increased, and determine a new increased pressure at which to provide air to the subject. Processing system may then control PAP device 110 to provide air to the subject at the new, increased pressure.

[0088] The changes in chemoreceptor-induced sympathetic nerve activity are changes in sympathetic nerve activity due to respiratory defects. The decrease in oxygen uptake and carbon dioxide release caused by the respiratory defects triggers the activation of chemoreceptors. Therefore, the changes in chemoreceptor-induced sympathetic nerve activity indicate that the current pressure provided by PAP is insufficient to treat the subject's sleep-disordered breathing and that a higher PAP pressure is needed to reduce the likelihood of SDB events and chemoreceptor-induced arousals, thereby improving sleep quality.

[0089] Similarly, in response to determining that the detected change in sympathetic nerve activity is baroreceptor-induced, processing system 150 can determine that the pressure of air provided to subject 120 should be reduced, and determine a new reduced pressure at which to provide air to the subject. Processing system can then control PAP device 110 to provide air to the subject at the new reduced pressure.

[0090] Baroreceptor-induced changes in sympathetic nerve activity are changes in sympathetic nerve activity caused by changes in intrathoracic pressure, which can be caused by excessive current pressures provided by a PAP device. Changes in cardiac parameters (including venous return, atrial filling, and stroke volume) caused by changes in intrathoracic pressure trigger activation of baroreceptors. Therefore, changes in baroreceptor-induced sympathetic nerve activity can indicate the need for lower PAP pressures to reduce the likelihood of baroreceptor-induced arousals and thus improve sleep quality.

[0091] In some examples, the pressure may be adjusted incrementally until a desired effect is achieved. For example, processing system 150 may continue to receive and process first signal 135 and second signal 145 while controlling PAP device 110 to adjust the pressure in incremental steps until processing system determines that the sympathetic nervous response has stabilized.

[0092] In some examples, having determined an appropriate adjustment to pressure for a particular change in sympathetic nerve activity in this manner, the processing system can record the determined adjusted pressure. The processing system can then control the PAP device to provide air at that adjusted pressure (i.e., without requiring incremental steps) in response to the specific change in detected sympathetic nerve activity. In other words, the processing system can learn subject-specific pressure adjustments based on incrementally adjusting pressure.

[0093] In some examples, pressure adjustments made in response to specific changes in the subject's sympathetic nervous response can be determined during a titration period of the subject.

[0094] In some examples, when determining the adjusted pressure, a pressure threshold may be considered (e.g., so that the adjusted pressure does not exceed the pressure threshold). For example, as described above, the initial adjusted pressure value may be determined based on a determination of whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced. The initial adjusted pressure value may then be compared to the pressure threshold: in response to the initial adjusted pressure value exceeding the pressure threshold, the pressure threshold is set to the adjusted pressure, and in response to the initial adjusted pressure value failing to exceed the pressure threshold, the initial adjusted pressure is set to the adjusted pressure.

[0095] In other words, the pressure threshold represents the maximum allowable pressure of air that can be provided by PAP device 110 to subject 120, e.g., a pressure corresponding to a possible number of baroreceptor-induced arousals exceeding the threshold. For example, the pressure threshold may be a subject-specific value determined by a clinician during a PAP titration period.

[0096] In some examples, the pressure threshold may depend on the current sleep stage of subject 120. For example, processing system 150 may estimate the current sleep stage of the subject and select the pressure threshold from a set of pressure thresholds based on the estimated current sleep stage. This recognizes that the subject may tolerate higher pressure PAP therapy during deep sleep stages (i.e., without experiencing baroreceptor-induced arousals).

[0097] A time-based approach may be used to estimate the subject's current sleep stage, for example, based on the current time of night or the number of hours PAP device 110 has been used during the current sleep session (either of which provides an estimate of the current number of sleep hours).

[0098] Alternatively, the current sleep stage of the subject can be estimated by processing the first signal 135 and / or the second signal 145 using known methods for estimating sleep stages based on physiological signals. See, for example, Bakker et al. (2021), "Estimating sleep stages using cardiopulmonary signals: Validation of a new algorithm across a wide range of sleep-disordered breathing severity," J Clin Sleep Med., 17(7):1343-1354; Long et al. (2014), "Analysis of respiratory effort amplitude for automatic sleep stage classification," Biomedical Signal Processing and Control, 14:197-205; Radha et al. (2019), "Sleep stage classification from heart rate variability using long short-term memory neural networks," Scientific Reports, 9:14149; Beattie et al. (2017), "Estimating sleep stages in healthy adult populations from optical plethysmography and accelerometer signals," Physiological Measurement. Measurement, 38(11):1968-1979; Aboalayon et al. (2016), “Sleep stage classification using EEG signal analysis: a comprehensive survey and new investigations,” Entropy, 18(9):272; and IET Wireless Sensor Systems, 12(5-6):123-133; Imtiaz (2021), “A systematic review of wearable sleep stage sensing technologies,” Sensors, 21(5):1562.

[0099] Having estimated the current sleep stage, processing system 150 may select a pressure threshold corresponding to the estimated sleep stage from a set of pressure thresholds including a pressure threshold for each sleep stage. The pressure threshold for each sleep stage may be a subject-specific value determined by a clinician during a PAP titration session.

[0100] In the event that the initial regulated pressure exceeds the pressure threshold and the pressure threshold is set to the regulated pressure, the pressure provided by PAP device 110 to subject 120 may be insufficient to keep the number / frequency of SDB events below an acceptable (e.g., clinically defined) level. To further reduce the number / frequency of SDB events without increasing the PAP pressure above the pressure threshold, processing system 150 may be configured to additionally control PAP device 110 to provide supplemental oxygen to the subject in response to the initial regulated pressure value exceeding the pressure threshold. The supplemental oxygen improves the oxygen saturation of the subject, thereby reducing SDB events without increasing the pressure provided by the PAP device.

[0101] In some examples, processing system 150 can be configured to control PAP device 110 to return to a previous pressure in response to satisfaction of a predetermined condition. In the context of the present disclosure, "previous pressure" is the pressure of air provided by PAP device to subject 120 before pressure adjustment. For example, the previous pressure can be the pressure prescribed to the subject by a clinician.

[0102] The predetermined condition may be, for example, time-dependent. For example, processing system 150 may control PAP device 110 to provide air to subject 120 at an adjusted pressure for a predetermined period of time, and in response to the lapse of the predetermined period of time, control the PAP device to provide air to the subject at a previous pressure. In some examples, processing system 150 may control the PAP device to return to the previous pressure after a first adjustment of pressure during a sleep period, and then control the PAP device to maintain the adjusted pressure in response to subsequent detection of changes in sympathetic nerve activity and adjustment of pressure.

[0103] In other examples, the predetermined condition may relate to sympathetic nerve activity of subject 120. For example, processing system 150 may continue to process first signal 135 and second signal 145 to detect a change in sympathetic nerve activity while PAP device 110 provides air to the subject at an adjusted pressure, and in response to the detected change in sympathetic nerve activity failing to exceed a predetermined activity threshold, control the PAP device to provide air at a previous pressure. Alternatively, the processing system may control the PAP device to provide air at a previous pressure in response to a determination that a physiological parameter of the detected change in sympathetic nerve activity has returned (within a predetermined range) to a value of the physiological parameter before the change occurred. For example, if the change in sympathetic nerve activity is an increase in heart rate variability, the PAP device may be controlled to return to its previous pressure in response to the heart rate variability returning to or falling below a value that the heart rate variability had before the increase.

[0104] In some examples, processing system 150 can be configured to repeat the steps of processing first signal 135 and second signal 145 at predetermined intervals to detect a change in sympathetic nerve activity, and in response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold, process the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, determine an regulated pressure based on whether the change is chemoreceptor-induced or baroreceptor-induced, and generate a control signal to control PAP device 110 to provide air to subject 120 at the regulated pressure.

[0105] Processing system 150 may repeat the above steps in addition to or instead of returning to a previous pressure (ie, to monitor additional changes in sympathetic nerve pressure after the PAP device has returned to a previous pressure) (ie, operating as a closed-loop feedback system).

[0106] By operating as a closed-loop feedback system, processing system 150 can control PAP device 110 in a manner that takes into account the subject's response to the adjusted pressure. Due to the complexity of cardiopulmonary interactions, changes in sympathetic activity may not always be correctly classified as chemoreceptor-induced or baroreceptor-induced, as respiratory defects can induce cardiac responses and vice versa. In such cases, the pressure provided by the PAP device may be incorrectly adjusted (i.e., such that the adjusted pressure causes a negative subject response). By continuing to monitor changes in the subject's sympathetic activity and adjusting the pressure provided by the PAP device accordingly, the subject's negative response can be corrected by readjusting the pressure.

[0107] The length of the predetermined interval may depend on whether the PAP device has returned to a previous pressure, or whether the system is operating as a closed-loop feedback system, in which case the length of the predetermined interval may be determined based on the expected time it takes the subject to respond to the most recent adjustment to the pressure provided by the PAP device. The length of the predetermined time interval may depend on the type of most recent pressure adjustment. For example, a change in pressure in response to a detected chemoreceptor-induced change may take longer to produce an observable effect on the subject (e.g., about 1 minute, i.e., the time it takes for the subject's blood volume to be pumped through the lungs) than a change in response to a detected baroreceptor-induced change (which is typically observable in a cardiac parameter within a few heartbeats).

[0108] although Figure 1 The control signal 155a in is shown in the context of controlling the pressure provided by a PAP device, but the processing system 150 may additionally or alternatively be configured to directly control therapy provided by other SDB therapy devices.

[0109] For example, where the SDB therapy device is a positional therapy device, the processing system 150 can generate control signals that control the positional therapy device to change the subject's head and / or body position (e.g., by vibrating until a desired position is achieved). Where the SDB therapy device is a mandibular advancement device having a motorized device for adjusting jawbone protrusion, the processing system can generate control signals that control the motorized device to adjust jawbone protrusion (e.g., to increase jawbone protrusion in response to changes in sympathetic nerve activity triggered by chemoreceptors, and to decrease jawbone protrusion in response to changes triggered by baroreceptors).

[0110] As previously described, the control signals generated by the processing system 150 may control the output user interface. Figure 2 is shown in Figure 2 Illustrated is a system 200 for controlling a SDB therapy device that provides therapy to a subject during the subject's sleep period, according to another embodiment of the present invention.

[0111] System 200 and Figure 1 The system 200 is the same as the system 100 , except that the system 200 further includes an output user interface 170 , and the control signal is a control signal 155 b for controlling the output user interface 170 .

[0112] The processing system 150 can control the output user interface 170 by processing the determination of whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced to determine control instructions for controlling the SDB therapy device, and generate a control signal 155b configured to control the output user interface to provide a user-perceivable output of the control instructions.

[0113] The control instructions determined by the processing system 150 will depend on the type of SDB therapy device to be controlled. For example, in the case of a PAP device, the control instructions may be instructions to adjust the pressure provided by the PAP device to a new adjusted pressure determined by the processing system as described above. In the case of a mandibular advancement device, the control instructions may be instructions to adjust the jawbone protrusion provided by the mandibular advancement device.

[0114] Suitable user-perceivable outputs for providing control instructions to the user will be apparent to a skilled person and may depend on the control instructions provided. For example, the control instructions may be provided to the user in the form of a text display and / or a sound (e.g., a verbal instruction). In some examples, the control instructions may be provided to subject 120, while in other examples, the control instructions may additionally or alternatively be provided to a clinician. Figure 2 In , output user interface 170 is a computer monitor; however, skilled artisans will readily appreciate that any output user interface (eg, a mobile device or a speaker) capable of providing a suitable user-perceivable output may be used.

[0115] Although the figures illustrate the generation of control signal 155a and control signal 155b as separate embodiments, in some examples, the processing system 150 can be configured to generate control signals for directly controlling the SDB therapy device 110 and control signals for controlling an output user interface to provide a user-perceivable output of control instructions.

[0116] In some examples, the processing system 150 can determine whether to generate a control signal for directly controlling the SDB therapy device 110 or to generate a control signal for controlling the output user interface 170 based on one or more predetermined rules. For example, the processing system 150 can be configured to control the PAP device to provide air at an adjusted pressure as long as the adjusted pressure falls within a predetermined pressure range, and to control the output user interface in response to the adjusted pressure not falling within the predetermined pressure range. This reduces clinical intervention in modifying the PAP therapy provided to the subject, thereby improving clinical efficiency, while allowing clinician input in situations where the recommended change in adjusted pressure is significantly different from the pressure prescribed for the subject.

[0117] Figure 3 Illustrated is a computer-implemented method 300 for controlling a SDB therapy device that provides therapy to a subject during the subject's sleep period, according to one embodiment of the present invention.

[0118] Method 300 begins at step 310 where a first signal responsive to breathing of a subject during a sleep session is received.

[0119] At step 320, a second signal is received in response to a change in baroreceptor-induced sympathetic nerve activity of the subject.

[0120] At step 330, the first signal and the second signal are processed to detect changes in sympathetic nerve activity.

[0121] At step 340, it is determined whether the detected change in sympathetic nerve activity exceeds a predetermined activity threshold.

[0122] In response to determining that the detected change in sympathetic nerve activity exceeds the predetermined activity threshold, the method proceeds to step 350 where the first signal and the second signal are processed to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced.

[0123] At step 360, the determination as to whether the change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced is processed to generate a control signal for controlling the SDB therapy device.

[0124] It should be understood that the disclosed methods are computer-implemented methods. Thus, the concept of a computer program is also proposed, which includes code means for implementing any of the described methods when said program is run on a processing system.

[0125] Those skilled in the art will be able to easily develop a processing system for implementing any of the methods described herein. Therefore, each step of the flowchart may represent a different action performed by the processing system and may be performed by a corresponding module of the processing system.

[0126] As described above, the system utilizes a processing system to perform data processing. The processing system may be implemented in a variety of ways using software and / or hardware to perform the desired functions. The processing system typically employs one or more microprocessors that may be programmed using software (e.g., microcode) to perform the desired functions. The processing system may be implemented as a combination of dedicated hardware to perform some functions and one or more programmed microprocessors and associated circuitry to perform other functions.

[0127] Examples of circuits that may be used in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field programmable gate arrays (FPGAs).Thus, the processing system may be embodied as a digital and / or analog processing system.

[0128] In various embodiments, the processing system may be associated with one or more storage media, such as volatile and non-volatile computer memory (such as RAM, PROM, EPROM, and EEPROM). The storage media may be encoded with one or more programs that, when executed on one or more processing systems and / or controllers, perform the desired functions. The various storage media may be fixed within the processing system, or the controller may be removable so that one or more programs stored thereon may be loaded into the processing system.

[0129] Based on a study of the drawings, the disclosure and the appended claims, those skilled in the art can understand and implement variations of the disclosed embodiments when practicing the claimed invention. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0130] The functionality implemented by a processor may be implemented by a single processor or by multiple separate processing units, which may be collectively considered to constitute a “processor.” Such processing units may in some cases be remote from one another and communicate with one another in a wired or wireless manner.

[0131] The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0132] The computer program may be stored / distributed on suitable media such as optical storage media or solid-state media provided with or as part of other hardware, but may also be distributed in other forms such as via the Internet or other wired or wireless telecommunications systems.

[0133] If the term "suitable for" is used in the claims or the specification, it should be noted that the term "suitable for" is intended to be equivalent to the term "configured to". If the term "device" is used in the claims or the specification, it should be noted that the term "device" is intended to be equivalent to the term "system", and vice versa.

[0134] Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. A processing system (150) for controlling a sleep disordered breathing (SDB) therapy device (110) that provides therapy to a subject (120) during a sleep period of the subject, the processing system being configured to: receiving a first signal (135) responsive to the subject's breathing during the sleep period; receiving a second signal (145) responsive to a baroreceptor-induced change in sympathetic nerve activity of the subject; processing the first signal and the second signal to detect changes in sympathetic nerve activity; In response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold: processing the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and Based on the determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, a control signal (155a, 155b) is generated for controlling the SDB therapy device.

2. The processing system (150) of claim 1, wherein: The SDB therapy device (110) is a positive airway pressure (PAP) device; wherein the control signal comprises a first control signal (155a), and The processing system is configured to: determining a regulated pressure to be provided by the PAP device based on a determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and The first control signal (155a) is generated to control the PAP device to provide air to the subject (120) at a regulated pressure.

3. The processing system (150) of claim 2, wherein: In response to said determination that the detected change in sympathetic nerve activity is chemoreceptor-induced, the adjusted pressure is an increased pressure; and In response to the determination that the detected change in sympathetic nerve activity is baroreceptor-induced, the adjusted pressure is a reduced pressure.

4. The processing system (150) according to claim 2 or 3, wherein the processing system is configured to: generating the first control signal (155a) configured to control the PAP device (110) to provide air to the subject (120) at the adjusted pressure for a predetermined period of time; and In response to the lapse of the predetermined period of time, a second control signal is generated, the second control signal configured to control the PAP device to provide air to the subject at a previous pressure.

5. The processing system (150) according to claim 2 or 3, wherein the processing system is configured to: continuing to process the first signal (135) and the second signal (145) to detect changes in sympathetic nerve activity; and In response to the change in sympathetic nerve activity failing to exceed the predetermined activity threshold, a second control signal is generated, the second control signal being configured to control the PAP device (110) to provide air to the subject (120) at a previous pressure.

6. The processing system (150) according to any one of claims 2 to 5, wherein the processing system is configured to repeatedly at predetermined intervals: processing the first signal (135) and the second signal (145) to detect changes in sympathetic nerve activity; In response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold: processing the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; determining a regulated pressure to be provided by the PAP device based on the determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and A first control signal (155a) is generated, the first control signal being configured to control the PAP device (110) to provide air to the subject (120) at the regulated pressure.

7. The processing system (150) according to any one of claims 2 to 6, the processing system being configured to: determining an initially adjusted pressure value based on the determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; Obtaining stress threshold; comparing the initially adjusted pressure value with the pressure threshold; In response to the initially adjusted pressure value exceeding the pressure threshold, setting the pressure threshold to the adjusted pressure; and In response to the initial adjusted pressure value failing to exceed the pressure threshold, the initial adjusted pressure value is set as the adjusted pressure.

8. The processing system (150) according to claim 7, wherein the processing system is configured to obtain the pressure threshold by: estimating a current sleep stage of the subject (120); and The pressure threshold is selected from a set of pressure thresholds according to the estimated current sleep stage.

9. The processing system (150) of claim 7 or 8, wherein in response to the initially adjusted pressure value exceeding the pressure threshold, the processing system is further configured to generate a third control signal, wherein the third control signal is configured to control the PAP device (110) to provide supplemental oxygen to the subject (120).

10. The processing system (150) according to any one of claims 1 to 9, wherein the predetermined activity threshold is determined by: using a machine learning algorithm trained to predict the likelihood of arousal of the subject (120) based on changes in sympathetic nerve activity; and The predetermined activity threshold is set to a value of change in sympathetic nerve activity for which the likelihood of arousal predicted by the machine learning algorithm exceeds a predetermined likelihood of arousal.

11. The processing system (150) according to any one of claims 1 to 10, the processing system being configured to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced by: processing the first signal (135) and the second signal (145) to determine whether the detected change in sympathetic nerve activity corresponds to a sleep disordered breathing event; responsive to said determination that the detected change in sympathetic nerve activity corresponds to a sleep disordered breathing event, determining that the detected change in sympathetic nerve activity is chemoreceptor-induced; In response to the determination that the detected change in sympathetic nerve activity does not correspond to a sleep disordered breathing event, it is determined that the detected change in sympathetic nerve activity is baroreceptor-induced.

12. The processing system (150) according to any one of claims 1 to 11, the processing system being configured to generate the control signal (155a, 155b) by: processing the determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced to determine control instructions for controlling the SDB therapy device (110); and A fourth control signal (155b) is generated, the fourth control signal being configured to control an output user interface (170) to provide a user-perceivable output of the control instruction.

13. A system (100) for controlling a sleep disordered breathing (SDB) therapy device (110) that provides therapy to a subject (120) during a sleep period of the subject, the system comprising: one or more first sensors (130) configured to generate a first signal (135) in response to the subject's breathing during the sleep period; one or more second sensors (140) configured to generate a second signal (145) in response to a change in baroreceptor-induced sympathetic nerve activity of the subject; and The processing system (155) according to any one of claims 1 to 12, the processing system being configured to receive the first signal and the second signal from the one or more first sensors and the one or more second sensors, respectively.

14. A computer-implemented method (300) for controlling a sleep disordered breathing (SDB) therapy device (110) that provides therapy to a subject (120) during a sleep period of the subject, the computer-implemented method comprising: receiving a first signal (135) responsive to the subject's breathing during the sleep period; receiving a second signal (145) in response to a change in sympathetic nerve activity initiated by a baroreceptor of the subject; processing the first signal and the second signal to detect changes in sympathetic nerve activity; In response to the detected change in sympathetic nerve activity exceeding a predetermined activity threshold: processing the first signal and the second signal to determine whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced; and Based on the determination as to whether the detected change in sympathetic nerve activity is chemoreceptor-induced or baroreceptor-induced, a control signal (155a, 155b) is generated for controlling the SDB therapy device.

15. A computer program product comprising computer program code means which, when executed on a computing device having a processing system, causes the processing system to carry out all of the steps of the method (300) according to claim 14.

Images