A ventilator control method based on event detection
By installing orifice plates and metal strain gauges in the ventilator pipeline, the respiratory waveform data are collected and processed, sleep apnea events and hypoventilation events are detected, and ventilation adjustment is carried out, the health hazards caused by respiratory stenosis in sleep apnea syndrome are solved. , precise control of the respiratory cycle and reduction of health risks.
Patent Information
- Application Number
- CN202210233601.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-10
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-03-10
AI Technical Summary
Sleep apnea syndrome causes respiratory tract stenosis or occlusion, resulting in insufficient respiratory flow or respiratory disorders, making it difficult to enter deep sleep, causing hypoxemia, hypercapnia and fluctuations in chest pressure, which is harmful to health and is difficult to be effective in the existing technology. Detect and regulate pause events and hypoventilation events during the respiratory cycle.
Using the ventilator control method based on event detection, by installing orifice plates and metal strain gauges in the ventilator pipeline, flow data is collected and converted into breathing waveforms, and apnea events and hypoventilation events are detected using wavelet transformation filtering and trigger detection algorithms. , and adjust the pipeline pressure through the ventilation adjustment algorithm for assisted breathing.
Effective detection and regulation of pause events and hypoventilation events in the respiratory cycle are achieved, the control accuracy of the ventilator is improved, the monitoring and treatment capabilities of sleep apnea are enhanced, and health risks are reduced.
Smart Images

Figure CN114618060B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of ventilator control, and in particular to a ventilator control method based on event detection. Background Art
[0002] Breathing is the process of exchanging gases between the human body and the external environment. When the respiratory system experiences airway collapse or obstruction, it directly leads to a decrease in the body's blood oxygen saturation, and carbon dioxide cannot be excreted from the body in time, causing respiratory diseases of varying degrees. Sleep apnea syndrome is the most common sleep-related respiratory disorder, and the consequences of undiagnosed sleep apnea can be very serious, ranging from increased blood pressure to heart failure.
[0003] The pathogenic mechanism of obstructive sleep apnea is the narrowing or occlusion of some parts of the respiratory tract, which leads to insufficient airflow or disordered breathing, making it difficult for the patient to enter deep sleep. At the same time, hypoxemia, hypercapnia, significant fluctuations in intrathoracic pressure and increased sympathetic nerve activity occur, which are very harmful to human health. In response to the above problems, this patent proposes a ventilator control method based on event detection, which realizes the detection of respiratory cycles through respiratory trigger detection, and detects apnea events or hypopnea events in each respiratory cycle, and controls and adjusts the ventilator according to the detection results. Summary of the invention
[0004] The present invention provides a ventilator control method based on event detection, the purpose of which is to (1) realize the detection of respiratory cycle based on respiratory trigger detection; (2) detect apnea events or hypopnea events in each respiratory cycle, and control and adjust the ventilator according to the detection results.
[0005] To achieve the above object, the present invention provides a ventilator control method based on event detection, comprising the following steps:
[0006] S1: collecting the ventilator pipeline flow, converting the collected ventilator pipeline flow into a respiratory waveform, and filtering the respiratory waveform using a wavelet transform method;
[0007] S2: Perform respiratory trigger detection on the respiratory waveform after filtering to obtain the exhalation trigger point and the inhalation trigger point of the respiratory waveform;
[0008] S3: According to the respiratory trigger detection result, event detection is performed on the respiratory waveform to detect whether there is an apnea event or a hypopnea event;
[0009] S4: If an apnea event or a hypopnea event is detected, the ventilation adjustment algorithm is used to control the pipeline pressure of the ventilator to perform assisted breathing, and an alarm is processed.
[0010] As a further improvement method of the present invention:
[0011] The ventilator pipeline flow data collected in step S1 includes:
[0012] An orifice plate is installed in the ventilator pipeline. The surface of the orifice plate has a metal strain gauge. The opening diameter of the orifice plate is d. When the ventilator is started, a one-way outflow of flowing gas will be generated in the ventilator pipeline, causing the upper and lower surfaces of the orifice plate to deform, resulting in changes in the resistance of the metal strain gauges on the upper and lower surfaces of the orifice plate. The upper surface of the orifice plate is in the opposite direction of the outflow of the flowing gas. The metal strain gauges on the upper and lower surfaces of the orifice plate have the same structure. The pressure difference between the upper and lower surfaces of the orifice plate is expressed as:
[0013]
[0014] in:
[0015] R represents the resistance of the metal strain gauge;
[0016] ΔE represents the pressure difference between the upper and lower surfaces of the orifice plate;
[0017] L represents the length of the wire inside the metal strain gauge;
[0018] ΔL represents the change in the length of the wire inside the metal strain gauge on the upper surface of the orifice plate;
[0019] k represents the sensitivity of the resistance in the metal strain gauge, and the resistance sensitivity of the metal strain gauge selected in the present invention is 0.6;
[0020] Calculate the flow data in the ventilator pipeline:
[0021]
[0022] in:
[0023] q t represents the flow data in the ventilator pipeline at time t, and the flow data is a time series data sequence, then the sequence of flow data is represented by {q1, q2, ..., q t , ..., q n}, n represents the length of the time series for collecting ventilator pipeline flow data;
[0024] D represents the diameter of the ventilator tube;
[0025] ρ represents the density of the gas flowing in the ventilator pipeline;
[0026] ΔE t Represents the pressure difference between the upper and lower surfaces of the orifice plate in the ventilator pipeline at time t.
[0027] The step S1 converts the collected ventilator pipeline flow into a respiratory waveform, including:
[0028] The sequence of the ventilator pipeline flow data is represented as {q1, q2, ..., q t , ..., q n}, set the flow filter with resistance R1 and capacitance C1, then the cut-off flow of the flow filter is The sequence representation of the ventilator pipeline flow data is input into the flow filter, and the flow below the cutoff flow is The flow data is assigned a value of 0, and the digital sequence representation result is converted into an analog quantity using a digital-to-analog converter. The converted analog quantity is the respiratory waveform q(n).
[0029] In the step S1, the respiratory waveform converted from the ventilator pipeline flow is filtered using a wavelet transform method, including:
[0030] The respiratory waveform q(n) is filtered using the wavelet transform method, and the filtering process is as follows:
[0031] 1) The scale factor a is fixed, and the wavelet transform processing is performed on the respiratory waveform q(n) using the wavelet function ω(t). The wavelet transform processing formula is:
[0032]
[0033] in:
[0034] b represents the displacement factor of the wavelet function;
[0035] q a,b (t) represents the wavelet coefficient obtained by wavelet transform;
[0036] 2) Change the displacement factor of the wavelet function and repeat step 1) to obtain a set of wavelet coefficients under the scale factor a;
[0037] 3) Change the scale factor a, repeat steps 1) and 2) to obtain the wavelet coefficient q of the respiratory waveform q(n) under different scale factors and displacement factors a,b (n) until
[0038] 4) Determine the wavelet threshold as σ;
[0039] 5) Delete the wavelet coefficients less than the wavelet threshold σ, and retain the wavelet coefficients greater than the wavelet threshold σ. The retained wavelet coefficients are q′ a,b (n), the wavelet coefficients are reconstructed into a respiratory waveform using the inverse wavelet transform method. The formula of the inverse wavelet transform method is:
[0040]
[0041] in:
[0042] q′(n) is the reconstructed respiratory waveform, and the reconstructed respiratory waveform is the respiratory waveform after filtering.
[0043] In step S2, the respiratory waveform after filtering is subjected to respiratory trigger detection, including:
[0044] For the respiratory waveform after filtering, calculate the respiratory change rate Δq at adjacent moments i =q′(i)-q′(i-1), where q′(i) represents the analog value of the respiratory waveform after filtering at time i, Δq i represents the respiratory change rate at time i;
[0045] A gas detection sensor is set at the leakage hole of the ventilator mask, and the leakage value of the user in a complete breathing cycle is G1, the flow value of the user's exhaled in a complete breathing cycle is G2, and the orifice plate in the ventilator pipeline is used to count the flow value of the user flowing through the pipeline in a complete breathing cycle is G3. The calculated breathing detection threshold is:
[0046]
[0047]
[0048]
[0049] in:
[0050] Q h Indicates breath detection threshold;
[0051] Q s represents the inhalation detection threshold;
[0052] The process of the respiratory trigger detection is as follows:
[0053] For the analog quantity q′(i) at any time in the respiratory waveform, if q′(i)>Q s , q′(i-1)≤Q s , then the current moment i is the inhalation trigger point, where Indicates the flow change threshold, right q′(i)<Q h q′(i-1)>Q h , then the current moment i is the exhalation trigger point.
[0054] In step S3, event detection is performed on the respiratory waveform according to the respiratory trigger detection result, including:
[0055] According to the respiratory trigger detection results, a respiratory waveform within the adjacent <inhalation trigger point, exhalation trigger point> is used as a set of waveforms to be detected by the event g. i =(g i (1), g i (2), ..., g i (h)), where g i (1) represents the analog value of the inhalation trigger point of the event detection waveform, g i (h) represents the analog value of the exhalation trigger point of the waveform to be detected;
[0056] In each group of waveforms to be detected by events, the flow standard deviation of the waveforms in the previous 1 second is calculated every 10ms, and the statistical character count is set. The initial value of count is 0. If the calculated flow standard deviation is less than the preset apnea detection threshold, count++; if the statistical character of the waveform to be detected by events is greater than 10, it means that there is an apnea event at this moment; the flow standard deviation std of each group of waveforms to be detected by events is set to 0. i The calculation formula is:
[0057]
[0058] in:
[0059] g i,m represents the expectation of any i-th group of waveforms to be detected;
[0060] Calculate the hypopnea index V of any i-th group of event waveforms to be detected i :
[0061]
[0062] in:
[0063] G3(g i ) represents the flow value of any i-th group of event waveforms to be detected flowing through the ventilator pipeline;
[0064] G1(g i ) represents the leakage value of any i-th group of event waveforms to be detected;
[0065] If V i If the preset hypopnea threshold is less than 40%, it means that a hypopnea event occurs in the respiratory cycle.
[0066] If an apnea event or a hypopnea event is detected in step S4, a ventilation adjustment algorithm is used to control the pipeline pressure of the ventilator to perform assisted breathing, including:
[0067] The ventilation adjustment algorithm process is as follows:
[0068] If an apnea event is detected, the ventilator tube is extended by 1 cm within 10 seconds. 3 The water pressure is reduced by 0.5 cm until no respiratory events occur within 30 minutes. 3 If a hypopnea event is detected, the ventilator tube increases by 0.5 cm within 5 seconds. 3 The water pressure is reduced by 0.5 cm until no respiratory events occur within 30 minutes. 3 In a specific embodiment of the present invention, if an apnea event or a hypopnea event is detected, an alarm process is performed simultaneously.
[0069] Compared with the prior art, the present invention proposes a ventilator control method based on event detection, which has the following advantages:
[0070] First, this scheme proposes a scheme for collecting ventilator flow data. By installing an orifice plate in the ventilator pipeline, wherein the surface of the orifice plate has a metal strain gauge, and the opening diameter of the orifice plate is d, when the ventilator is started, a one-way outflow of flowing gas will be generated in the ventilator pipeline, causing the upper and lower surfaces of the orifice plate to deform, resulting in changes in the resistance of the metal strain gauges on the upper and lower surfaces of the orifice plate. The upper surface of the orifice plate is in the opposite direction of the outflow of the flowing gas, and the metal strain gauges on the upper and lower surfaces of the orifice plate have the same structure. Then the pressure difference between the upper and lower surfaces of the orifice plate is expressed as:
[0071]
[0072] Wherein: R represents the resistance of the metal strain gauge; ΔE represents the pressure difference between the upper and lower surfaces of the orifice plate; L represents the length of the wire in the metal strain gauge; ΔL represents the change in the length of the wire in the metal strain gauge on the upper surface of the orifice plate; k represents the sensitivity of the resistance in the metal strain gauge, and the resistance sensitivity of the metal strain gauge selected in the present invention is 0.6; the flow data in the ventilator pipeline is calculated:
[0073]
[0074] Where: q t represents the flow data in the ventilator pipeline at time t, and the flow data is a time series data sequence, then the sequence of flow data is represented by {q1, q2, ..., q t , ..., q n}, n represents the length of the time series of the collected ventilator pipeline flow data; D represents the diameter of the ventilator pipeline; ρ represents the density of the flowing gas in the ventilator pipeline; ΔE t Represents the pressure difference between the upper and lower surfaces of the orifice plate in the ventilator pipeline at time t.
[0075] At the same time, this scheme proposes a respiratory trigger detection method. For the respiratory waveform after filtering, the respiratory change rate Δq at adjacent moments is calculated. i =q′(i)-q′(i-1), where q′(i) represents the analog value of the respiratory waveform after filtering at time i, Δq i represents the breathing change rate at time i; a gas detection sensor is set at the leakage hole of the ventilator ventilation mask, and the leakage value of the user in a complete breathing cycle is G1, the flow value of the user's inhalation in a complete breathing cycle is G2, and the orifice plate in the ventilator pipeline is used to count the flow value flowing through the pipeline in a complete breathing cycle of the user is G3, then the calculated breathing detection threshold is:
[0076]
[0077]
[0078]
[0079] Where: Q h represents the exhalation detection threshold; Qs represents the inhalation detection threshold. The process of the respiratory trigger detection is: for the analog quantity q′(i) at any time in the respiratory waveform, if q′(i)>Qs, q′(i-1)≤Q s , then the current moment i is the inhalation trigger point, where represents the flow change threshold, if q′(i)<Q h , q′(i-1)>Q h , then the current moment i is the exhalation trigger point; compared with the traditional algorithm, this scheme proposes a respiratory trigger point detection scheme. When the user inhales, the flow in the ventilator pipeline will increase significantly, and the respiratory change rate will suddenly increase. Therefore, the respiratory waveform is greater than the inhalation detection threshold and the respiratory change rate is greater than the flow change threshold. When the user exhales, the opposite is true, and the detection of the respiratory trigger point is realized. When the leakage of the ventilator mask increases, the exhalation Q h The detection threshold will be greatly reduced, while the flow rate change rate remains unchanged, ensuring that respiratory trigger detection can be successfully achieved at any time. According to the respiratory trigger detection results, this scheme uses a respiratory waveform within the adjacent <inhalation trigger point, exhalation trigger point> as a set of waveforms to be detected by the event g i =(g i (1), g i (2), ..., g i (h)), where g i (1) represents the analog value of the inhalation trigger point of the event detection waveform,
[0080] gi(h) represents the analog value of the exhalation trigger point of the waveform to be detected, thereby realizing the detection of apnea events and hypopnea events. If an apnea event or a hypopnea event is detected, the pipeline pressure of the ventilator is adjusted to perform assisted breathing. BRIEF DESCRIPTION OF THE DRAWINGS
[0081] Figure 1 A schematic flow chart of a ventilator control method based on event detection provided by one embodiment of the present invention;
[0082] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0083] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0084] S1: Collect the ventilator pipeline flow, convert the collected ventilator pipeline flow into a respiratory waveform, and use the wavelet transform method to filter the respiratory waveform.
[0085] The ventilator pipeline flow data collected in step S1 includes:
[0086] An orifice plate is installed in the ventilator pipeline. The surface of the orifice plate has a metal strain gauge. The opening diameter of the orifice plate is d. When the ventilator is started, a one-way outflow of flowing gas will be generated in the ventilator pipeline, causing the upper and lower surfaces of the orifice plate to deform, resulting in changes in the resistance of the metal strain gauges on the upper and lower surfaces of the orifice plate. The upper surface of the orifice plate is in the opposite direction of the outflow of the flowing gas. The metal strain gauges on the upper and lower surfaces of the orifice plate have the same structure. The pressure difference between the upper and lower surfaces of the orifice plate is expressed as:
[0087]
[0088] in:
[0089] R represents the resistance of the metal strain gauge;
[0090] ΔE represents the pressure difference between the upper and lower surfaces of the orifice plate;
[0091] L represents the length of the wire inside the metal strain gauge;
[0092] ΔL represents the change in the length of the wire inside the metal strain gauge on the upper surface of the orifice plate;
[0093] k represents the sensitivity of the resistance in the metal strain gauge, and the resistance sensitivity of the metal strain gauge selected in the present invention is 0.6;
[0094] Calculate the flow data in the ventilator pipeline:
[0095]
[0096] in:
[0097] q t represents the flow data in the ventilator pipeline at time t, and the flow data is a time series data sequence, then the sequence of flow data is represented by {q1, q2, ..., q t , ..., q n}, n represents the length of the time series for collecting ventilator pipeline flow data;
[0098] D represents the diameter of the ventilator tube;
[0099] ρ represents the density of the gas flowing in the ventilator pipeline;
[0100] ΔE t Represents the pressure difference between the upper and lower surfaces of the orifice plate in the ventilator pipeline at time t.
[0101] The step S1 converts the collected ventilator pipeline flow into a respiratory waveform, including:
[0102] The sequence of the ventilator pipeline flow data is represented as {q1, q2, ..., q t , ..., q n}, set the flow filter with resistance R1 and capacitance C1, then the cut-off flow of the flow filter is The sequence representation of the ventilator pipeline flow data is input into the flow filter, and the flow below the cutoff flow is The flow data is assigned a value of 0, and the digital sequence representation result is converted into an analog quantity using a digital-to-analog converter. The converted analog quantity is the respiratory waveform q(n).
[0103] In the step S1, the respiratory waveform converted from the ventilator pipeline flow is filtered using a wavelet transform method, including:
[0104] The respiratory waveform q(n) is filtered using the wavelet transform method, and the filtering process is as follows:
[0105] 1) The scale factor a is fixed, and the wavelet transform processing is performed on the respiratory waveform q(n) using the wavelet function ω(t). The wavelet transform processing formula is:
[0106]
[0107] in:
[0108] b represents the displacement factor of the wavelet function;
[0109] q a,b (t) represents the wavelet coefficient obtained by wavelet transform;
[0110] 2) Change the displacement factor of the wavelet function and repeat step 1) to obtain a set of wavelet coefficients under the scale factor a;
[0111] 3) Change the scale factor a, repeat steps 1) and 2) to obtain the wavelet coefficient q of the respiratory waveform q(n) under different scale factors and displacement factors a,b (n) until
[0112] 4) Determine the wavelet threshold as σ;
[0113] 5) Delete the wavelet coefficients less than the wavelet threshold σ, and retain the wavelet coefficients greater than the wavelet threshold σ. The retained wavelet coefficients are q′ a,b (n), the wavelet coefficients are reconstructed into a respiratory waveform using the inverse wavelet transform method. The formula of the inverse wavelet transform method is:
[0114]
[0115] in:
[0116] q′(n) is the reconstructed respiratory waveform, and the reconstructed respiratory waveform is the respiratory waveform after filtering.
[0117] S2: Perform respiratory trigger detection on the filtered respiratory waveform to obtain an exhalation trigger point and an inhalation trigger point of the respiratory waveform.
[0118] In step S2, the respiratory waveform after filtering is subjected to respiratory trigger detection, including:
[0119] For the respiratory waveform after filtering, calculate the respiratory change rate Δq at adjacent moments i =q′(i)-q′(i-1), where q′(i) represents the analog value of the respiratory waveform after filtering at time i, Δq i represents the respiratory change rate at time i;
[0120] A gas detection sensor is set at the leakage hole of the ventilator mask, and the leakage value of the user in a complete breathing cycle is G1, the flow value of the user's exhaled in a complete breathing cycle is G2, and the orifice plate in the ventilator pipeline is used to count the flow value of the user flowing through the pipeline in a complete breathing cycle is G3. The calculated breathing detection threshold is:
[0121]
[0122]
[0123]
[0124] in:
[0125] Q h Indicates breath detection threshold;
[0126] Q s represents the inhalation detection threshold;
[0127] The process of the respiratory trigger detection is as follows:
[0128] For the analog quantity q′(i) at any time in the respiratory waveform, if q′(i)>Q s , q′(i-1)≤Q s , then the current moment i is the inhalation trigger point, where represents the flow change threshold, if q′(i)<Q h , q′(i-1)>Q h , then the current moment i is the exhalation trigger point.
[0129] S3: According to the respiratory trigger detection result, event detection is performed on the respiratory waveform to detect whether there is an apnea event or a hypopnea event.
[0130] In step S3, event detection is performed on the respiratory waveform according to the respiratory trigger detection result, including:
[0131] According to the respiratory trigger detection results, a respiratory waveform within the adjacent <inhalation trigger point, exhalation trigger point> is used as a set of waveforms to be detected by the event g. i =(g i (1), g i (2), ..., g i (h)), where g i (1) represents the analog value of the inhalation trigger point of the event detection waveform, g i (h) represents the analog value of the exhalation trigger point of the waveform to be detected;
[0132] In each group of waveforms to be detected by events, the flow standard deviation of the waveforms in the previous 1 second is calculated every 10ms, and the statistical character count is set. The initial value of count is 0. If the calculated flow standard deviation is less than the preset apnea detection threshold, count++; if the statistical character of the waveform to be detected by events is greater than 10, it means that there is an apnea event at this moment; the flow standard deviation std of each group of waveforms to be detected by events is set to 0. i The calculation formula is:
[0133]
[0134] in:
[0135] g i,m represents the expectation of any i-th group of waveforms to be detected;
[0136] Calculate the hypopnea index V of any i-th group of event waveforms to be detected i :
[0137]
[0138] in:
[0139] G3(g i ) represents the flow value of any i-th group of event waveforms to be detected flowing through the ventilator pipeline;
[0140] G1(g i ) represents the leakage value of any i-th group of event waveforms to be detected;
[0141] If V i If the preset hypopnea threshold is less than 40%, it means that a hypopnea event occurs in the respiratory cycle.
[0142] S4: If an apnea event or a hypopnea event is detected, the ventilation adjustment algorithm is used to control the pipeline pressure of the ventilator to perform assisted breathing, and an alarm is processed.
[0143] If an apnea event or a hypopnea event is detected in step S4, a ventilation adjustment algorithm is used to control the pipeline pressure of the ventilator to perform assisted breathing, including:
[0144] The ventilation adjustment algorithm process is as follows:
[0145] If an apnea event is detected, the ventilator tube is extended by 1 cm within 10 seconds. 3 The water pressure is reduced by 0.5 cm until no respiratory events occur within 30 minutes. 3 If a hypopnea event is detected, the ventilator tube increases by 0.5 cm within 5 seconds. 3 The water pressure is reduced by 0.5 cm until no respiratory events occur within 30 minutes. 3 In a specific embodiment of the present invention, if an apnea event or a hypopnea event is detected, an alarm process is performed simultaneously.
[0146] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.
[0147] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.
[0148] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A ventilator control device based on event detection, characterized in that: The method involved in the device includes: S1: collecting the ventilator pipeline flow, converting the collected ventilator pipeline flow into a respiratory waveform, and filtering the respiratory waveform using a wavelet transform method; S2: Perform respiratory trigger detection on the respiratory waveform after filtering to obtain the exhalation trigger point and the inhalation trigger point of the respiratory waveform; Perform respiratory trigger detection on the filtered respiratory waveform, including: For the respiratory waveform after filtering, calculate the respiratory change rate at adjacent moments ,in It represents the analog value of the respiratory waveform after filtering at time i, represents the respiratory change rate at time i; A gas detection sensor is installed at the leakage hole of the ventilator mask to count the leakage value of the user in a complete breathing cycle. The flow rate of the user's breath in a complete breathing cycle is , using the orifice plate in the ventilator pipeline to calculate the flow rate through the pipeline during a complete breathing cycle of the user , then the calculated breathing detection threshold is: ; ; ; in: Indicates breath detection threshold; represents the inhalation detection threshold; The process of the respiratory trigger detection is as follows: For the analog quantity at any time in the respiratory waveform ,like , , , then the current moment i is the inhalation trigger point, where Indicates the flow change threshold. If , , , then the current moment i is the exhalation trigger point; S3: According to the respiratory trigger detection result, event detection is performed on the respiratory waveform to detect whether there is an apnea event or a hypopnea event; According to the respiratory trigger detection results, event detection is performed on the respiratory waveform, including: According to the respiratory trigger detection results, a section of the respiratory waveform within the adjacent <inhalation trigger point, exhalation trigger point> is used as a set of waveforms to be detected by the event ,in The analog value representing the inhalation trigger point of the waveform to be detected. An analog value representing the exhalation trigger point of the waveform to be detected by the event; In each group of waveforms to be detected by an event, the flow standard deviation of the waveform in the previous 1 second is calculated every 10ms, and the statistical character count is set. The initial value of count is 0. If the calculated flow standard deviation is less than the preset apnea detection threshold, count++; if the statistical character of the waveform to be detected by an event is greater than 10, it means that an apnea event exists at this moment; the flow standard deviation of each group of waveforms to be detected by an event is The calculation formula is: ; in: represents the expectation of any i-th group of waveforms to be detected; Calculate the hypopnea index of any i-th group of event waveforms to be detected : ; in: represents the flow value of any i-th group of event waveforms to be detected flowing through the ventilator pipeline; represents the leakage value of any i-th group of event waveforms to be detected; like If the pre-set hypopnea threshold is less than 40%, it means that there is a hypopnea event in the respiratory cycle; S4: If an apnea event or a hypopnea event is detected, the ventilation adjustment algorithm is used to control the pipeline pressure of the ventilator and an alarm is processed; The ventilation adjustment algorithm process is as follows: If an apnea event is detected, the ventilator pipeline increases within 10 seconds. water pressure until no respiratory events occur within 30 minutes, then reduce If a hypoventilation event is detected, the ventilator tubing increases within 5 seconds. water pressure until no respiratory events occur within 30 minutes, then reduce water pressure.
2. A ventilator control device based on event detection as claimed in claim 1, characterized in that: The ventilator pipeline flow data collected in step S1 includes: An orifice plate is installed in the ventilator pipeline. The surface of the orifice plate has a metal strain gauge. The opening diameter of the orifice plate is When the ventilator is started, a one-way outflow of flowing gas will be generated in the ventilator pipeline, causing the upper and lower surfaces of the orifice plate to deform, resulting in changes in the resistance of the metal strain gauges on the upper and lower surfaces of the orifice plate. The upper surface of the orifice plate is in the opposite direction of the outflow of the flowing gas. The metal strain gauges on the upper and lower surfaces of the orifice plate have the same structure. The pressure difference between the upper and lower surfaces of the orifice plate is expressed as: ; in: R represents the resistance of the metal strain gauge; Indicates the pressure difference between the upper and lower surfaces of the orifice plate; L represents the length of the wire inside the metal strain gauge; Indicates the change in the length of the wire inside the metal strain gauge on the upper surface of the orifice plate; k represents the sensitivity of the resistance inside the metal strain gauge; Calculate the flow data in the ventilator pipeline: ; in: represents the flow data in the ventilator pipeline at time t. The flow data is a time series data sequence. The sequence of flow data is expressed as , n represents the length of the time series for collecting ventilator pipeline flow data; D represents the diameter of the ventilator tube; Indicates the density of the gas flowing in the ventilator pipeline; Represents the pressure difference between the upper and lower surfaces of the orifice plate in the ventilator pipeline at time t.
3. A ventilator control device based on event detection as claimed in claim 2, characterized in that: The step S1 converts the collected ventilator pipeline flow into a respiratory waveform, including: The sequence of the ventilator pipeline flow data is expressed as , set the resistance to , the capacitance is The flow filter has a cut-off flow rate of ; Input the sequence representation of the ventilator pipeline flow data into the flow filter, and filter the flow below the cutoff flow The flow data is assigned a value of 0, and the digital sequence representation result is converted into an analog quantity using a digital-to-analog converter. The converted analog quantity is the respiratory waveform. .
4. A ventilator control device based on event detection as described in any one of claims 2-3, characterized in that: In the step S1, the respiratory waveform converted from the ventilator pipeline flow is filtered using a wavelet transform method, including: Using wavelet transform method to analyze respiratory waveform Perform filtering processing, the filtering processing process is as follows: 1) Fixed scale factor , using the wavelet function Respiratory waveform Perform wavelet transform processing, the wavelet transform processing formula is: ; in: b represents the displacement factor of the wavelet function; Represents the wavelet coefficients obtained by wavelet transform processing; 2) Change the displacement factor of the wavelet function and repeat step 1) to obtain the scale factor The set of wavelet coefficients under ; 3) Change the scale factor , repeat steps 1) and 2) to get the breathing waveform Wavelet coefficients under different scale factors and displacement factors ,until ; 4) Determine the wavelet threshold ; 5) Delete the values less than the wavelet threshold The wavelet coefficients greater than the wavelet threshold are retained The wavelet coefficients retained are , the wavelet coefficients are reconstructed into a respiratory waveform using the inverse wavelet transform method. The formula of the inverse wavelet transform method is: ; in: The reconstructed respiratory waveform is the respiratory waveform after filtering.
