Ventilator pressure regulation method, device and equipment based on body position recognition and household ventilator

Abstract

The application discloses a ventilator pressure regulation method and device based on body position recognition, equipment and a household ventilator. The method comprises the following steps: acquiring respiratory flow data and pipeline pressure data of a ventilator; calculating a body position indication feature according to the respiratory flow data and the pipeline pressure data; performing body position inference according to the body position indication feature to obtain a body position inference result; inputting the body position inference result into a body position-pressure mapping model to obtain a target treatment pressure corresponding to a sleep body position; and generating a pressure regulation instruction according to the target treatment pressure. The application does not depend on an additional acceleration sensor / body position sensor, but only analyzes the respiratory flow data and the pipeline pressure data to infer the sleep body position of a user, and performs differential pressure regulation according to the current body position, so that the accuracy and sensitivity of pressure regulation can be improved.

Description

Technical Field

[0001] This application relates to the field of ventilator control, and in particular to a ventilator pressure regulation method, device, equipment and home ventilator based on body position recognition. Background Technology

[0002] Sleep position has a significant impact on the severity of obstructive sleep apnea (OSA): approximately 60% of OSA patients present with "positional OSA"—the apnea-hypopnea index (AHI) is significantly higher in the supine position than in the lateral position (usually 2 to 4 times higher). This is because in the supine position, gravity causes the base of the tongue and soft palate to collapse backward, obstructing the airway.

[0003] Currently, home-use Automatic Positive Airway Pressure (APAP) ventilators passively adjust pressure based on the detection of airway events (pause, hypopnea, flow restriction, snoring), without considering body position. However, body position is one of the most significant known factors influencing airway obstruction. Existing APAP algorithms completely disregard positional information—after rolling from a side-lying position to a supine position, pressure is only initiated after an actual airway event occurs (potentially 10-30 seconds later), creating a treatment vacuum. Furthermore, after rolling from a supine to a side-lying position, the actual pressure required in the side-lying position is much lower than in the supine position, but the device still operates at the high pressure level established in the supine position. This necessitates a lengthy depressurization process (typically 5-20 minutes without events before gradually reducing pressure), resulting in prolonged high-pressure operation in the side-lying position, reducing comfort and increasing the risk of air leakage.

[0004] While some devices have accelerometers, these sensors are mounted on the main unit (placed on the bedside table) rather than worn by the patient, thus failing to accurately reflect the patient's position. Even if the sensor is placed at the tubing interface, changes in the tubing position when the patient turns over may not accurately reflect the trunk position. Furthermore, relying on additional accelerometers increases hardware costs, and position sensors require calibration when worn.

[0005] In summary, the pressure regulation of home ventilators in related technologies does not take into account body position factors, resulting in poor accuracy and sensitivity; adding an acceleration sensor / body position sensor would be costly and difficult to calibrate. Summary of the Invention

[0006] This application aims to propose a ventilator pressure regulation method, device, equipment, and home ventilator based on body position recognition, which can perform body position recognition without relying on speed sensors / body position sensors, and perform differentiated pressure regulation according to the current body position, thereby improving the accuracy and sensitivity of pressure regulation.

[0007] In a first aspect, embodiments of this application provide a ventilator pressure regulation method based on body position recognition, comprising the following steps: Acquire respiratory flow and tubing pressure data from the ventilator; Postural indication characteristics are calculated based on respiratory flow data and tubing pressure data. These postural indication characteristics include at least the flow limitation index, airway resistance index, and snoring energy ratio. Based on the posture indication features, posture inference is performed to obtain a posture inference result, which is used to indicate the user's sleeping posture. Input the body position inference results into the body position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; Pressure regulation instructions are generated based on the target therapeutic pressure.

[0008] According to some embodiments of this application, the position indication feature further includes a turning-over detection marker, which is used to indicate whether a turning-over event has occurred; the pressure regulation command includes a first pressure regulation command and a second pressure regulation command, and the generation of the pressure regulation command based on the target therapeutic pressure includes: When the turning detection sign indicates the presence of a turning event and the body position inference result indicates that the sleeping position is supine, a preventive pressure strategy including a safety factor is generated. A first treatment pressure is generated based on the target treatment pressure and the safety factor; A first pressure regulation command is generated based on the first treatment pressure, the target treatment pressure, and the preventive pressurization strategy; the first pressure regulation command is used to instruct the pipeline pressure to be adjusted to the first treatment pressure first, and then gradually increased to the target treatment pressure based on a preset pressurization rate; And / or, if the turning detection flag indicates the presence of a turning event and the position inference result indicates that the sleeping position is a lateral position, a pressure reduction strategy including a comfort factor is generated. A second treatment pressure is generated based on the target treatment pressure and the comfort factor; A second pressure regulation command is generated based on the second treatment pressure and the pressure reduction strategy; the second pressure regulation command is used to instruct the pipeline pressure to be adjusted to the second treatment pressure at a preset pressure reduction rate.

[0009] According to some embodiments of this application, in the step of calculating the postural indication characteristics based on respiratory flow data and tubing pressure data, the expression for the flow restriction index is: ; in, The traffic restriction index. This represents the median of the average mid-inspiratory flow rate during each cycle within the window. This represents the median peak inspiratory flow rate for each cycle within the window. And / or, the expression for the airway resistance index is: ; in, This is an indicator of airway resistance. This represents the average pipeline pressure during the intake phase. To set treatment pressure, This is the peak inspiratory flow rate; And / or, the expression for the snoring energy ratio is: ; in, The ratio of snoring energy. This refers to the energy of the respiratory flow signal in the 30~100Hz frequency band. This represents the total energy of the respiratory flow signal.

[0010] According to some embodiments of this application, the step of inferring body position based on the body position indication features to obtain a body position inference result includes: Obtain the first weighting coefficient corresponding to the flow restriction index, the second weighting coefficient corresponding to the airway resistance index, and the third weighting coefficient corresponding to the snoring energy ratio; The body position score is calculated based on the flow restriction index, the airway resistance index, the snoring energy ratio, the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient. Posture inference is performed based on the postural score and the preset postural judgment rules to obtain the postural inference result; The position determination rules include: if the position score is greater than a first threshold, the user's sleeping position is inferred to be supine; if the position score is less than a second threshold, the user's sleeping position is inferred to be lateral; the first threshold is greater than the second threshold.

[0011] According to some embodiments of this application, the step of generating a pressure regulation command based on the target therapeutic pressure includes: Obtain the upper and lower threshold values ​​for treatment pressure. The target treatment pressure is compared with the upper limit threshold and the lower limit threshold of the treatment pressure to generate a pressure adjustment command; Specifically, when the target treatment pressure is lower than the upper limit threshold of the treatment pressure but higher than the lower limit threshold of the treatment pressure, a pressure adjustment command is generated based on the target treatment pressure. If the target treatment pressure is higher than the upper limit threshold of the treatment pressure, a pressure adjustment command is generated based on the upper limit threshold of the treatment pressure. If the target treatment pressure is lower than the lower limit threshold of treatment pressure, a pressure adjustment command is generated based on the lower limit threshold of treatment pressure.

[0012] According to some embodiments of this application, after generating the pressure regulation command based on the target therapeutic pressure, the method further includes: Respiratory status is detected based on respiratory flow and tubing pressure, and respiratory status detection results are generated. If the respiratory status detection result indicates the presence of an airway event, the pressure of the ventilator is adjusted according to the preset safe treatment pressure.

[0013] According to some embodiments of this application, after generating the pressure regulation command based on the target therapeutic pressure, the method further includes: The total duration of supine position, average effective pressure in supine position, total duration of lateral position, average effective pressure in lateral position, number of times the user turned over, and number of apnea-hypopnea episodes in each position were recorded during each sleep session. The position-pressure mapping model is updated based on the total duration of supine position, the average effective pressure in supine position, the total duration of lateral decubitus position, the average effective pressure in lateral decubitus position, the number of times the patient turns over, and the number of apnea-hypopnea episodes in each position.

[0014] Secondly, embodiments of this application provide a ventilator pressure regulation device based on body position recognition, comprising: The data acquisition module is used to acquire respiratory flow data and tubing pressure data from the ventilator; The feature calculation module is used to calculate postural indication features based on respiratory flow data and tubing pressure data. The postural indication features include at least the flow limitation index, airway resistance index, and snoring energy ratio. The body position inference module is used to infer body position based on the body position indication features and obtain a body position inference result, which is used to indicate the user's sleeping position. The therapeutic pressure module is used to input the body position inference results into the body position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; The pressure regulation module is used to generate pressure regulation commands based on the target treatment pressure.

[0015] Thirdly, embodiments of this application provide an electronic device, the device comprising: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the ventilator pressure regulation method based on body position recognition as described in the first aspect.

[0016] Fourthly, embodiments of this application provide a home ventilator, including a control module, wherein the control module regulates ventilator pressure using a ventilator pressure regulation method based on body position recognition as described in the first aspect.

[0017] The ventilator pressure regulation method, apparatus, device, and home ventilator based on body position recognition of the embodiments of this application have at least the following beneficial effects: In this embodiment, respiratory flow data and tubing pressure data of the ventilator are first acquired; then, position indication features are calculated based on the respiratory flow data and tubing pressure data; position inference is performed based on the position indication features to obtain the position inference result; the position inference result is input into the position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; and a pressure adjustment command is generated based on the target therapeutic pressure. This application does not rely on additional accelerometers / position sensors, but infers the user's sleep position solely by analyzing respiratory flow data and tubing pressure data, and performs differentiated pressure adjustment based on the current position, which can improve the accuracy and sensitivity of pressure adjustment.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The present application will be further described below with reference to the accompanying drawings and embodiments, wherein: Figure 1 A schematic flowchart illustrating an embodiment of the ventilator pressure regulation method based on body position recognition provided in this application; Figure 2 A schematic diagram of the ventilator pressure regulation device based on body position recognition provided in this application; Figure 3 A schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation

[0020] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0021] In this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, without necessarily requiring or implying any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0022] To address the problems of the prior art, embodiments of this application provide a method, apparatus, device, and home ventilator based on body position recognition for ventilator pressure regulation. The following is a description of the body position recognition-based ventilator pressure regulation provided in this application embodiment.

[0023] Figure 1 A schematic flowchart of a ventilator pressure regulation method based on body position recognition provided in an embodiment of this application is shown. This method is applied to electronic devices, and includes the following steps: S101. Obtain respiratory flow data and tubing pressure data from the ventilator; S102. Calculate the postural indication characteristics based on respiratory flow data and tubing pressure data. The postural indication characteristics include at least the flow limitation index, airway resistance index, and snoring energy ratio. S103. Perform body position inference based on body position indication characteristics to obtain body position inference results, which are used to indicate the user's sleeping position. S104. Input the body position inference results into the body position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; S105. Generate pressure regulation instructions based on the target treatment pressure.

[0024] In this embodiment, respiratory flow data and tubing pressure data of the ventilator are first acquired; then, position indication features are calculated based on the respiratory flow data and tubing pressure data; position inference is performed based on the position indication features to obtain the position inference result; the position inference result is input into the position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; and a pressure adjustment command is generated based on the target therapeutic pressure. This application does not rely on additional accelerometers / position sensors, but infers the user's sleep position solely by analyzing respiratory flow data and tubing pressure data, and performs differentiated pressure adjustment based on the current position, which can improve the accuracy and sensitivity of pressure adjustment.

[0025] In step S101 above, respiratory flow data is acquired by collecting respiratory flow signals through a flow sensor, and tubing pressure data is acquired by collecting pressure signals through a pressure sensor.

[0026] It should be noted that flow sensors and pressure sensors are standard hardware components of ventilators. Currently, many home ventilators are equipped with flow sensors and pressure sensors for detection, so there is no need to add additional sensors to obtain respiratory flow data and tubing pressure data.

[0027] In step S102 above, calculating the body position indication characteristics based on respiratory flow data and tubing pressure data means calculating at least the flow restriction index, airway resistance index, and snoring energy ratio based on the respiratory flow data and tubing pressure data, and then inferring the sleep position based on the body position indication characteristics. The principle is that changes in body position alter respiratory signals and tubing pressure; the sleep position can be inferred based on the flow restriction index, airway resistance index, and snoring energy ratio. In addition to the flow restriction index, airway resistance index, and snoring energy ratio, other characteristics such as turning-over detection markers can be added to improve the reliability of the inference.

[0028] It should be noted that, taking the supine position, lateral position, and the process of turning over as examples, the physical mechanisms by which changes in body position affect respiratory mechanics are as follows: Supine position: Gravity causes the base of the tongue and soft palate to collapse backward, reducing the cross-section of the upper airway. This manifests as: (1) increased inspiratory flow restriction (more pronounced flattening); (2) increased snoring component in expiratory flow; (3) increased airway resistance; and (4) a higher flow limitation index (FLI) for the same pressure.

[0029] Lateral position: Gravity causes the base of the tongue and soft palate to shift laterally rather than collapse backward, thus maintaining a better airway cross-section. This manifests as: (1) a more sinusoidal inspiratory flow waveform (not flat-topped); (2) reduced snoring; (3) lower airway resistance; and (4) lower FLI.

[0030] Turning over: A large body movement lasting 3 to 15 seconds, with obvious disturbances in the respiratory flow signal (non-respiratory movement artifacts), followed by a systemic change in the breathing pattern.

[0031] Specifically, the expression for the traffic restriction index is: ; in, The traffic restriction index. This represents the median of the average mid-inspiratory flow rate during each cycle within the window. This represents the median peak inspiratory flow rate (FLI) for each cycle within the window. It should be noted that the FLI is typically 0.25–0.50 for the supine position and 0.05–0.20 for the lateral position.

[0032] Specifically, the expression for the airway resistance index is: ; in, This is an indicator of airway resistance. This represents the average pipeline pressure during the intake phase. To set treatment pressure, This is the peak inspiratory flow rate. It should be noted that... Reflects upper airway resistance, supine position Significantly higher than in the lateral decubitus position .

[0033] Specifically, the expression for the snoring energy ratio is: ; in, The ratio of snoring energy. This refers to the energy of the respiratory flow signal in the 30~100Hz frequency band. This represents the total energy of the respiratory flow signal. It should be noted that because there is more snoring in the supine position, the SER in the supine position is significantly higher than that in the lateral position.

[0034] Specifically, the calculation method for the energy of the respiratory flow signal in the 30~100Hz frequency band is as follows: For the flow signal, take an N-point sampling window (N=256, sampling rate fs=100Hz, frequency resolution Δf=fs / N≈0.39Hz), and perform a Fast Fourier Transform (FFT) on the signal x(n) within the window to obtain the spectrum X(k), where k=0,1,...,N / 2-1 corresponds to the frequency f(k)=k×Δf. The energy of a specific frequency band (e.g., 30~100Hz) is calculated as the sum of the squares of the amplitudes within the corresponding frequency index range of that frequency band: E_band = ∑|X(k)|², where k∈[k1, k2], k1= f_low / Δf k2= f_high / Δf Total energy E_total = ∑|X(k)|², k = 0 to N / 2-1. For the 30~100Hz frequency band: k1 = 30 / 0.39 =77, k2= 100 / 0.39 =256 (limited by the 50Hz Nyquist frequency, the actual value is k2=N / 2-1=127, corresponding to 50Hz). Therefore, SER = E_{30-50Hz} / E_total (limited by the sampling rate, the actual usable frequency band is 30~50Hz).

[0035] In step S103 above, inferring body position based on body position indication features means inferring the user's current sleeping position based on the features calculated above.

[0036] For example, in this step, the positional inference can be achieved by substituting the flow restriction index, airway resistance index, and snoring energy ratio into a weighted formula for calculating a positional score, thus obtaining a positional score. Then, the user's sleeping position is inferred based on the position score. For example, a PS greater than 0.6 indicates a supine position, and a PS less than 0.4 indicates a lateral position.

[0037] For example, the body position inference in this step can also use a Hidden Markov Model (HMM) instead of a weighted formula, modeling the body position as a state transition chain of "supine → roll over → side-lying → roll over → supine", thereby inferring the user's sleeping position.

[0038] It should be noted that, in addition to supine and lateral positions, prone and semi-lateral positions can also be added to further refine the classification of body positions.

[0039] In step S104 above, inputting the position inference result into the position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position means inputting the position inference result, such as supine or lateral position, into the pre-built position-pressure mapping model to obtain the target therapeutic pressure under the corresponding position. The target therapeutic pressure refers to the optimal therapeutic pressure under that position.

[0040] For example, suppose Targeted therapeutic pressure in the supine position. Target therapeutic pressure for the lateral decubitus position. First, set initial values ​​for the target therapeutic pressure corresponding to the same position, constructing an initial position-pressure mapping model. The initial values ​​for the target therapeutic pressure in the supine and lateral decubitus positions can be set as follows: ; ; Among them, P prescriptionPrescription pressure, also known as physician-prescribed pressure, is the baseline therapeutic pressure determined by a sleep specialist through pressure titration during polysomnography (PSG) or based on clinical experience. It serves as the initial reference pressure and benchmark for pressure adjustment in the APAP algorithm. Prescription pressure is used as the initial therapeutic pressure for the supine position, while the initial therapeutic pressure for the lateral position is lower than that for the supine position. .

[0041] After each position change, the actual required pressure for the current position is determined using the existing APAP event-driven response of the ventilator, and the mapping model is updated as follows: ; in, For the updated target treatment pressure, To address the pressure of the target treatment before the update, This represents the effective pressure for stable APAP operation in the current body position, such as the 95th percentile pressure during event-free periods. The learning rate can be set to 0.1 by default. It should be noted that... Usually more Increased by 2-5 cmH2O.

[0042] In step S105 above, generating a pressure adjustment command based on the target treatment pressure means adjusting the ventilator's pressure according to the optimal treatment pressure corresponding to the current body position, and adjusting the ventilator's tubing pressure to the target treatment pressure.

[0043] In some embodiments, the position indication feature further includes a turning detection marker, which is used to indicate whether a turning event has occurred; the pressure regulation command includes a first pressure regulation command and a second pressure regulation command, and the pressure regulation command generated based on the target therapeutic pressure may include: When the turning detection sign indicates that a turning event has occurred, and the position inference result indicates that the sleeping position is supine, a preventive pressure strategy including a safety factor is generated. The first treatment pressure is generated based on the target treatment pressure and safety factor; A first pressure regulation command is generated based on the first treatment pressure, the target treatment pressure, and the preventive pressurization strategy; the first pressure regulation command is used to instruct the tubing pressure to be adjusted to the first treatment pressure first, and then gradually increased to the target treatment pressure based on a preset pressurization rate; And / or, when the turning detection flag indicates the presence of a turning event and the position inference result indicates that the sleeping position is a lateral position, a pressure reduction strategy including a comfort factor is generated; A second therapeutic pressure is generated based on the target therapeutic pressure and comfort level. A second pressure regulation command is generated based on the second treatment pressure and the pressure reduction strategy; the second pressure regulation command is used to instruct the pipeline pressure to be adjusted to the second treatment pressure at a preset pressure reduction rate.

[0044] In this embodiment, when the turning-over detection indicator indicates a turning-over event and the position inference result indicates a supine sleeping position, a preventative pressure-increasing strategy including a safety factor is generated. A first therapeutic pressure is generated based on the target therapeutic pressure and the safety factor, and a first pressure regulation command is generated based on the first therapeutic pressure, the target therapeutic pressure, and the preventative pressure-increasing strategy. Furthermore, when the turning-over detection indicator indicates a turning-over event and the position inference result indicates a lateral sleeping position, a pressure-reducing strategy including a comfort factor is generated; a second therapeutic pressure is generated based on the target therapeutic pressure and the comfort factor; and a second pressure regulation command is generated based on the second therapeutic pressure and the pressure-reducing strategy. This embodiment achieves proactive pressure regulation based on position inference results and turning-over events, performing preventative pressure increase when the user turns to a supine position and rapid pressure reduction when turning to a lateral position, replacing the traditional event-driven passive response, and can greatly improve sensitivity.

[0045] It should be noted that the postural indication feature also includes a turning-over detection flag. This refers to detecting turning-over by setting a turning-over detection flag in the postural indication feature. When a large non-respiratory disturbance (amplitude > 3 times normal breathing, irregular frequency) lasting 3-15 seconds appears in the flow signal, it is marked as a turning-over event. The expression for the turning-over detection flag TURN is as follows: ; Specifically, when a rolling over event is detected and the Postural Score (PS) turns to the supine position, a first pressure regulation instruction is generated through a preventative pressure strategy, expressed as follows: ; in, The primary treatment pressure; For safety, the default value is 0.9, which means that the pressure is first increased to 90% of the optimal pressure in the supine position, and then fine-tuned to 100%. The pressure increase rate can be 2 cmH2O / min. According to tests, the preventive pressure increase strategy adopted in this implementation method is 15 to 60 seconds faster than the event response of traditional APAP.

[0046] Specifically, when a turning event is detected and the Postural Score (PS) shifts to the lateral decubitus position, a second pressure regulation command is generated using a pressure reduction strategy, expressed as follows: ; in, P target2 This is the second treatment pressure; For comfort factor, the default value is 1.05, which means reducing the pressure to 105% of the optimal pressure in the lateral decubitus position, with a 5% margin; the decompression rate is 1 cmH2O / min. After testing, the decompression strategy adopted in this implementation method is 3 to 5 times faster than the decompression rate of traditional APAP events.

[0047] It should be noted that when the body position is uncertain, a conservative strategy can be added. For example, a third pressure regulation instruction can be generated through the conservative strategy. The third pressure regulation instruction indicates that pressure regulation is performed through a third therapeutic pressure. The expression for the conservative strategy when the body position is uncertain is as follows: ; in, The third therapeutic pressure is the midpoint between the optimal pressures of the two body positions, which is then fine-tuned in conjunction with the event response of normal APAP.

[0048] In some implementations, the expression for the flow restriction index in the step of calculating the postural indication characteristic based on respiratory flow data and tubing pressure data is: ; in, The traffic restriction index. This represents the median of the average mid-inspiratory flow rate during each cycle within the window. This represents the median peak inspiratory flow rate for each cycle within the window. And / or, the expression for the airway resistance index is: ; in, This is an indicator of airway resistance. This represents the average pipeline pressure during the intake phase. To set treatment pressure, This is the peak inspiratory flow rate; And / or, the expression for the snoring energy ratio is: ; in, The ratio of snoring energy. This refers to the energy of the respiratory flow signal in the 30~100Hz frequency band. This represents the total energy of the respiratory flow signal.

[0049] In this embodiment, the flow restriction index is calculated using the above expressions. airway resistance index Compared to snoring energy This can improve the accuracy of body position inference.

[0050] In some implementations, inferring body position based on postural indication features to obtain a postural inference result may include: Obtain the first weighting coefficient of the corresponding flow restriction index, the second weighting coefficient of the corresponding airway resistance index, and the third weighting coefficient of the corresponding snoring energy ratio; The body position score is calculated based on the flow restriction index, airway resistance index, snoring energy ratio, first weighting coefficient, second weighting coefficient, and third weighting coefficient. Posture inference is performed based on the postural score and the preset postural judgment rules to obtain the postural inference result; The position determination rules include: if the position score is greater than the first threshold, the user's sleeping position is inferred to be supine; if the position score is less than the second threshold, the user's sleeping position is inferred to be lateral; and the first threshold is greater than the second threshold.

[0051] In this embodiment, firstly, the first weighting coefficient corresponding to the flow restriction index, the second weighting coefficient corresponding to the airway resistance index, and the third weighting coefficient corresponding to the snoring energy ratio are obtained; then, a body position score is calculated based on the flow restriction index, airway resistance index, snoring energy ratio, first weighting coefficient, second weighting coefficient, and third weighting coefficient; next, body position inference is performed based on the body position score and preset body position determination rules to obtain the body position inference result. This allows for accurate and rapid inference of the current body position.

[0052] Specifically, postural scoring The expression for (Posture Score) is: ; in, As the first weighting coefficient, This is the second weighting coefficient. This is the third weighting coefficient. For example, in this embodiment... The first weighting coefficient for FLI has the highest value because it is the most sensitive indicator for distinguishing between supine and lateral recumbent positions. These are the flow restriction index, airway resistance index, and snoring energy ratio normalized to [0,1], respectively.

[0053] One rule for determining body position in this embodiment is as follows: ; In addition to evaluating body position solely based on the score range of the first and second thresholds, it can also be determined using a turning-over detection marker. Specifically, another body position determination rule in this embodiment is as follows: ; It should be noted that the two judgment rules can also be combined, for example: ; Since the change in body position after rolling over is certain, the judgment threshold is lowered within 3 minutes after the rolling over detection is triggered, that is, the threshold of 0.5 is used instead of the threshold of 0.6 / 0.4.

[0054] It should be noted that this application may also add a postural inference confidence linkage: when the postural score PS is in the uncertainty range (0.4~0.6), no large-scale pressure adjustment is performed, only a fine adjustment is performed, so as to improve safety.

[0055] In some implementations, generating a pressure regulation command based on the target therapeutic pressure may include: Obtain the upper and lower threshold values ​​for treatment pressure. The target treatment pressure is compared with the upper and lower thresholds of the treatment pressure to generate a pressure adjustment command. Specifically, when the target treatment pressure is lower than the upper limit threshold of treatment pressure but higher than the lower limit threshold of treatment pressure, a pressure adjustment command is generated based on the target treatment pressure. When the target treatment pressure is higher than the upper limit threshold of treatment pressure, a pressure regulation command is generated based on the upper limit threshold of treatment pressure. When the target treatment pressure is lower than the lower limit threshold of treatment pressure, a pressure adjustment command is generated based on the lower limit threshold of treatment pressure.

[0056] In this embodiment, the upper and lower limits of treatment pressure are first obtained. Then, the target treatment pressure is compared with these thresholds, and a pressure adjustment command is generated based on the comparison result. This achieves lower and upper limit protection for treatment pressure adjustment, improving safety.

[0057] It should be noted that the upper limit of treatment pressure refers to the maximum pressure prescribed by the doctor, while the lower limit of treatment pressure refers to the minimum pressure prescribed by the doctor.

[0058] Specifically, if the target treatment pressure does not exceed the upper or lower threshold of the treatment pressure, the target treatment pressure will be used as the actual adjusted pressure value; if it exceeds the lower or upper limit, the upper or lower threshold will be used as the actual adjusted pressure value, as follows: Lower limit protection: The pressure under any body position shall not be lower than The expression is as follows: ; in, This is the lower limit threshold for treatment pressure.

[0059] Upper limit protection: Pressure should not exceed [a certain value] in any position. The expression is as follows: ; in, This is the upper limit threshold for treatment pressure.

[0060] In some implementations, after generating the pressure regulation command based on the target therapeutic pressure, the method may further include: Respiratory status is detected based on respiratory flow and tubing pressure, and respiratory status detection results are generated. If the respiratory status test results indicate the presence of an airway event, the ventilator pressure is adjusted according to the preset safe treatment pressure.

[0061] In this embodiment, respiratory status is first detected based on respiratory flow and tubing pressure to generate a respiratory status detection result. Then, if the respiratory status detection result indicates the presence of an airway event, the ventilator pressure is adjusted according to a preset safe treatment pressure. This event fallback mechanism enhances safety.

[0062] Specifically, respiratory status detection based on respiratory flow and tubing pressure is used to determine whether an airway event has occurred after treatment pressure adjustment. If an airway event still occurs after pressure adjustment, it indicates that the position inference may be incorrect, and the ventilator pressure should be adjusted immediately according to the preset safe treatment pressure, such as reverting to the event-driven pressurization of traditional APAP, to ensure treatment safety.

[0063] In some implementations, after generating the pressure regulation command based on the target therapeutic pressure, the method may further include: The total duration of supine position, average effective pressure in supine position, total duration of lateral position, average effective pressure in lateral position, number of times the user turned over, and number of apnea-hypopnea episodes in each position were recorded during each sleep session. The position-pressure mapping model was updated based on the total duration of supine position, the average effective pressure in supine position, the total duration of lateral position, the average effective pressure in lateral position, the number of times the patient turned over, and the number of apnea-hypopnea episodes in each position.

[0064] In this implementation, the total duration of the user's supine position, average effective pressure in the supine position, total duration of the user's lateral position, average effective pressure in the lateral position, number of times the user turned over, and number of apnea-hypopnea episodes in each position are first statistically analyzed during each sleep session. Then, the position-pressure mapping model is updated based on these parameters. By establishing a long-term learned position-pressure mapping model, the optimal therapeutic pressure for each individual in each position can be continuously updated, further improving the therapeutic effect.

[0065] It should be noted that each sleep cycle refers to the entire duration of each sleep session, such as each night. For users who sleep at other times, each sleep cycle can also be a different time period.

[0066] For example, after each night's treatment, the following statistics were recorded: total duration and average effective pressure in supine position, total duration and average effective pressure in lateral position, number of turns, and AHI in each position.

[0067] Long-term learning and updating and The individual optimal value was determined, and the stress difference index was calculated. The details are as follows: ; in, Typically 2~5 cmH2O; if The patient has "non-positional OSA," meaning that positional differences have little value, indicating that the benefit of stress regulation in this application is relatively small; if The patient is a typical case of "postural OSA", indicating that the stress regulation benefit of this application is the greatest.

[0068] It should be noted that the body position-pressure mapping model in this application can use Gaussian process regression instead of linear learning to establish a more flexible nonlinear pressure prediction model.

[0069] In summary, this application infers body position based on respiratory signal characteristics, eliminating the need for additional accelerometers and resulting in zero hardware cost increase. Utilizing three-dimensional features of FLI, airway resistance, and snoring energy, the accuracy of body position recognition reaches over 85%. Preventative pressurization when turning to the supine position reaches the target pressure 15-60 seconds earlier than traditional event-driven approaches, reducing the incidence of airway events after positional changes by approximately 40%-60%. Rapid depressurization when turning to the lateral position reduces the average treatment pressure during the lateral position period by 2-4 cmH2O, significantly improving comfort and reducing air leakage. Online learning of the body position-pressure mapping model allows the system to gradually and accurately master the optimal pressure for each patient in different positions. After 5-10 nights of use, the model converges, and treatment accuracy continues to improve. A safety-assured event fallback mechanism ensures that even if the body position inference is occasionally incorrect, it does not affect treatment safety.

[0070] Based on the ventilator pressure regulation method based on body position recognition provided in the above embodiments, this application also provides a specific implementation of a ventilator pressure regulation device based on body position recognition.

[0071] like Figure 2 As shown, the ventilator pressure regulating device 200 based on body position recognition provided in this application embodiment may include: The data acquisition module 201 is used to acquire respiratory flow data and tubing pressure data of the ventilator; The feature calculation module 202 is used to calculate the postural indication features based on respiratory flow data and tubing pressure data. The postural indication features include at least the flow limitation index, airway resistance index and snoring energy ratio. The body position inference module 203 is used to infer body position based on body position indication features and obtain body position inference results, which are used to indicate the user's sleeping position. The therapeutic pressure module 204 is used to input the body position inference results into the body position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position. Pressure regulation module 205 is used to generate pressure regulation commands based on the target treatment pressure.

[0072] The ventilator pressure regulation device 200 based on body position recognition in this application embodiment is used to execute the ventilator pressure regulation method based on body position recognition in the above embodiment. Its specific processing procedure is the same as that of the ventilator pressure regulation method based on body position recognition in the above embodiment, and will not be described in detail here.

[0073] Figure 3 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0074] The electronic device may include a processor 301 and a memory 302 storing computer program instructions.

[0075] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0076] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.

[0077] In some embodiments, memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0078] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any of the ventilator pressure regulation methods based on body position recognition in the above embodiments.

[0079] In one example, the electronic device may also include a communication interface 303 and a bus 310. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.

[0080] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0081] Bus 310 may include hardware, software, or both. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0082] Furthermore, in conjunction with the ventilator pressure regulation method based on body position recognition in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the ventilator pressure regulation methods based on body position recognition in the above embodiments.

[0083] This application also relates to a home ventilator, including a control module, which regulates ventilator pressure using the ventilator pressure regulation method based on body position recognition described in the above embodiments.

[0084] Specifically, the control module uses an MCU to run position inference and differentiated pressure control algorithms. The core hardware of a home ventilator also includes a flow sensor, pressure sensor, fan, and Flash memory.

[0085] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0086] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0087] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0088] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0089] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A ventilator pressure regulation method based on body position recognition, characterized in that, Includes the following steps: Acquire respiratory flow and tubing pressure data from the ventilator; Postural indication characteristics are calculated based on respiratory flow data and tubing pressure data. These postural indication characteristics include at least the flow limitation index, airway resistance index, and snoring energy ratio. Based on the posture indication features, posture inference is performed to obtain a posture inference result, which is used to indicate the user's sleeping posture. Input the body position inference results into the body position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; Pressure regulation instructions are generated based on the target therapeutic pressure.

2. The ventilator pressure regulation method based on body position recognition according to claim 1, characterized in that, The postural indication feature also includes a turning-over detection marker, which is used to indicate whether a turning-over event has occurred; The pressure regulation command includes a first pressure regulation command and a second pressure regulation command. Generating the pressure regulation command based on the target treatment pressure includes: When the turning detection sign indicates the presence of a turning event and the body position inference result indicates that the sleeping position is supine, a preventive pressure strategy including a safety factor is generated. A first treatment pressure is generated based on the target treatment pressure and the safety factor; A first pressure regulation command is generated based on the first treatment pressure, the target treatment pressure, and the preventive pressurization strategy; the first pressure regulation command is used to instruct the pipeline pressure to be adjusted to the first treatment pressure first, and then gradually increased to the target treatment pressure based on a preset pressurization rate; And / or, if the turning detection flag indicates the presence of a turning event and the position inference result indicates that the sleeping position is a lateral position, a pressure reduction strategy including a comfort factor is generated. A second treatment pressure is generated based on the target treatment pressure and the comfort factor; A second pressure regulation command is generated based on the second treatment pressure and the pressure reduction strategy; the second pressure regulation command is used to instruct the pipeline pressure to be adjusted to the second treatment pressure at a preset pressure reduction rate.

3. The ventilator pressure regulation method based on body position recognition according to claim 1, characterized in that, In the step of calculating the postural indication characteristics based on respiratory flow data and tubing pressure data, the expression for the flow restriction index is: ； in, The traffic restriction index. This represents the median of the average mid-inspiratory flow rate during each cycle within the window. This represents the median peak inspiratory flow rate for each cycle within the window. And / or, the expression for the airway resistance index is: ； in, This is an indicator of airway resistance. This represents the average pipeline pressure during the intake phase. To set treatment pressure, This is the peak inspiratory flow rate; And / or, the expression for the snoring energy ratio is: ； in, The ratio of snoring energy. This refers to the energy of the respiratory flow signal in the 30~100Hz frequency band. This represents the total energy of the respiratory flow signal.

4. The ventilator pressure regulation method based on body position recognition according to claim 1, characterized in that, The step of inferring body position based on the body position indication features to obtain the body position inference result includes: Obtain the first weighting coefficient corresponding to the flow restriction index, the second weighting coefficient corresponding to the airway resistance index, and the third weighting coefficient corresponding to the snoring energy ratio; The body position score is calculated based on the flow restriction index, the airway resistance index, the snoring energy ratio, the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient. Posture inference is performed based on the postural score and the preset postural judgment rules to obtain the postural inference result; The position determination rules include: if the position score is greater than a first threshold, the user's sleeping position is inferred to be supine; if the position score is less than a second threshold, the user's sleeping position is inferred to be lateral; the first threshold is greater than the second threshold.

5. The ventilator pressure regulation method based on body position recognition according to claim 1, characterized in that, The step of generating a pressure regulation command based on the target therapeutic pressure includes: Obtain the upper and lower threshold values ​​for treatment pressure. The target treatment pressure is compared with the upper limit threshold and the lower limit threshold of the treatment pressure to generate a pressure adjustment command; Specifically, when the target treatment pressure is lower than the upper limit threshold of the treatment pressure but higher than the lower limit threshold of the treatment pressure, a pressure adjustment command is generated based on the target treatment pressure. If the target treatment pressure is higher than the upper limit threshold of the treatment pressure, a pressure adjustment command is generated based on the upper limit threshold of the treatment pressure. If the target treatment pressure is lower than the lower limit threshold of treatment pressure, a pressure adjustment command is generated based on the lower limit threshold of treatment pressure.

6. The ventilator pressure regulation method based on body position recognition according to claim 1, characterized in that, After generating the pressure regulation command based on the target treatment pressure, the method further includes: Respiratory status is detected based on respiratory flow and tubing pressure, and respiratory status detection results are generated. If the respiratory status detection result indicates the presence of an airway event, the pressure of the ventilator is adjusted according to the preset safe treatment pressure.

7. The ventilator pressure regulation method based on body position recognition according to claim 1, characterized in that, After generating the pressure regulation command based on the target treatment pressure, the method further includes: The total duration of supine position, average effective pressure in supine position, total duration of lateral position, average effective pressure in lateral position, number of times the user turned over, and number of apnea-hypopnea episodes in each position were recorded during each sleep session. The position-pressure mapping model is updated based on the total duration of supine position, the average effective pressure in supine position, the total duration of lateral position, the average effective pressure in supine position, the number of times the patient turns over, and the number of apnea-hypopnea episodes in each position.

8. A ventilator pressure regulation device based on body position recognition, characterized in that, include: The data acquisition module is used to acquire respiratory flow data and tubing pressure data from the ventilator; The feature calculation module is used to calculate postural indication features based on respiratory flow data and tubing pressure data. The postural indication features include at least the flow limitation index, airway resistance index, and snoring energy ratio. The body position inference module is used to infer body position based on the body position indication features and obtain a body position inference result, which is used to indicate the user's sleeping position. The therapeutic pressure module is used to input the body position inference results into the body position-pressure mapping model to obtain the target therapeutic pressure corresponding to the sleep position; The pressure regulation module is used to generate pressure regulation commands based on the target treatment pressure.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the ventilator pressure regulation method based on body position recognition as described in any one of claims 1-7.

10. A home-use ventilator, characterized in that, The system includes a control module, which regulates ventilator pressure using the ventilator pressure regulation method based on body position recognition as described in any one of claims 1-7.

Images