Respiration monitoring system based on flexible strain sensor and monitoring analysis method

Abstract

The invention discloses a respiration monitoring system based on a flexible strain sensor and a monitoring analysis method, and belongs to the technical field of sensors and respiration monitoring equipment. The respiration monitoring system comprises a flexible strain sensor, a connecting line, a data acquisition card and an upper computer; the flexible strain sensor is a band-aid type flexible strain sensor and can be directly adhered to the skin surface of a human body, and the deformation of muscles related to respiration is monitored and converted into the change of a resistance signal. And the flexible strain sensor is connected with the data acquisition card through a connecting line. And the data acquisition card is connected with the upper computer and is used for transmitting the acquired resistance signals to the upper computer. The quantitative breathing mode judgment standard is provided through dynamic measurement of the respiratory muscle group, and the defect that judgment is mainly conducted through naked eyes of doctors for a long time can be overcome; the relation between the respiratory muscle group deformation and the lung function parameters is associated, and the problem that the actual respiration condition of part of patients does not conform to the detection condition in conventional lung function detection can be solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of sensors and respiratory monitoring equipment, and particularly relates to a respiratory monitoring system and a monitoring and analysis method based on a flexible strain sensor. Background Art

[0002] In recent years, flexible strain sensors have garnered widespread attention in the field of human physiological signal monitoring, particularly respiratory monitoring, due to their advantages such as high comfort, high sensitivity, and reusability. Compared with traditional rigid sensors, flexible strain sensors conform better to the human body surface, providing accurate and continuous respiratory signal monitoring data while reducing their cost. By collecting parameters such as respiratory rhythm and amplitude changes, they can assist in identifying various respiratory diseases. Combined with machine learning, they enable intelligent analysis, providing doctors with quantitative and reliable diagnostic evidence.

[0003] In clinical applications, flexible strain sensors can distinguish between chest, abdominal, and diaphragmatic breathing modes, and capture respiratory rate changes and abnormal rhythms in real time, indicating potential risks of respiratory function deterioration. For critically ill patients who rely on mechanical ventilation, such sensors can also be used to wean patients off mechanical ventilation.

[0004] The spontaneous breathing ability assessment before the patient is discharged, such as monitoring the rapid shallow breathing index (RSBI) and other indicators, can assist in determining the timing of ventilator removal and improve the scientificity and safety of clinical decision-making.

[0005] In addition, compared with traditional lung function testing methods (such as spirometry and plethysmography), flexible strain sensors provide a non-invasive solution that does not require special cooperation and can be used for a long time in daily environments. It is especially suitable for the dynamic monitoring needs of patients with severe diseases such as COPD and asthma. In the context of public health events (such as the XX epidemic), its feature of not requiring shared equipment also significantly reduces the risk of cross-infection. Therefore, the respiratory monitoring system based on flexible strain sensors provides a new technical path for clinical diagnosis and treatment, disease management and rehabilitation training. Respiratory monitoring is an important part of respiratory diseases and ICU management. Indirect signals can be measured by optical or radar sensors, such as 202411282732.4 respiratory monitoring method, equipment and readable storage medium, 202410840884.5 a respiratory monitoring method and device for radiotherapy, 202210068468.9 an abdominal respiratory monitoring system and method. Generally, IMU (inertial measurement unit), PPG (photoplethysmography) or frequency-modulated continuous wave radar signals are used for indirect signal measurement. There are problems such as long-term monitoring wearing discomfort, complex equipment, and susceptibility to interference from environmental reflections. In recent years, flexible Wearable respiratory monitoring sensors are gradually emerging, such as 202110343256.2 A wearable respiratory monitoring method, device and terminal equipment, which invented a wearable bracelet based on a three-axis acceleration sensor to achieve "remote, non-invasive, and continuous" monitoring, but has high requirements for algorithms and signal processing capabilities; in addition, most public reports or patents using flexible strain sensors simply measure parameters such as respiratory frequency or amplitude, providing a feasibility for respiratory monitoring, and do not use flexible sensors to conduct comprehensive and in-depth research on certain clinical pain points. Summary of the Invention

[0006] In view of the above problems, the present invention adopts an intelligent respiratory monitoring system constructed based on a flexible Band-Aid sensor, adheres the sensor to the skin surface of the main muscle groups and auxiliary muscle groups involved in human breathing, obtains breathing-related resistance signals such as frequency and amplitude, compares the strength of resistance signals of different muscle groups, establishes relevant quantitative standards, identifies the breathing pattern of the subjects, and solves the shortcomings of conventional judgment based on naked eye experience; through dynamic analysis of the stretching amount of human muscle groups, correlates respiratory function indicators such as FVC (forced vital capacity), FEV1 (one-second volume) and muscle fatigue index reflecting the strength of joint muscle groups, solves the problem of "results depending on blowing skills" in conventional insufflation lung function testing, solves the problem of "results relying on blowing skills" in clinical prediction of ventilator fatigue level, and realizes "interference-free and skill-free" detection of lung function and quantitative and accurate judgment of pre-ventilator withdrawal.

[0007] The technical solution adopted in the present invention is:

[0008] A respiratory monitoring system based on a flexible strain sensor, comprising a flexible strain sensor, a connecting line, a data acquisition card and a host computer;

[0009] The flexible strain sensor is a Band-Aid-style flexible strain sensor that can be directly adhered to the surface of human skin. It monitors the deformation of respiratory muscles and converts it into a change in resistance signal. The flexible strain sensor is connected to a data acquisition card via a connecting cable.

[0010] The data acquisition card is connected to the host computer and is used to transmit the acquired resistance signal to the host computer.

[0011] Furthermore, the flexible strain sensor includes an adhesive layer, a force-sensitive coating, a silver paste layer, a conductive base layer and an anti-interference layer from the outside to the inside.

[0012] Furthermore, the adhesive layer is made of medical waterproof tape material;

[0013] Furthermore, the force-sensitive coating is a conductive coating composed of a carbon-based material, and the carbon-based material in the force-sensitive coating is carbon black, carbon nanotubes or graphene. Preferably, the preparation method of the force-sensitive coating is: dispersing the carbon-based material in a solution-type polymer matrix to obtain a conductive paste, uniformly coating the dispersed conductive paste on the conductive base layer by blade coating or screen printing, and naturally drying to obtain the force-sensitive coating. Wherein, the solution-type polymer matrix is such as PDMS, TPU, Ecoflex or epoxy resin; the mass fraction of the carbon-based material in the solution-type polymer matrix is 2-10wt%; and the thickness of the force-sensitive coating is 10-30μm.

[0014] Furthermore, the silver paste layer is composed of a composite of silver powder and resin. It is applied to the conductive substrate by doctor blade coating or screen printing after the force-sensitive coating is applied. The silver paste layer is located at both ends of the force-sensitive coating, connecting the force-sensitive coating to the connecting wire. The silver powder accounts for 65–70% by weight. The resin is epoxy or acrylate. The silver paste layer has a thickness of 10–30 μm.

[0015] Furthermore, the conductive base layer is made of medical-grade polyethylene, polypropylene, polyvinyl alcohol, and thermoplastic polyurethane film with a thickness of 0.05-0.2 mm;

[0016] Furthermore, the anti-interference layer has a three-layer structure, which includes an insulating layer, a shielding layer and a protective layer from the inside to the outside. Preferably, the shielding layer material is PEDOT:PSS (100-300nm thick), 10-50nm thick copper foil or aluminum foil, 200-500nm thick graphene film, or 100-300nm thick carbon nanotube conductive film, which has excellent electromagnetic shielding effect. The insulating layer is arranged between the conductive base layer and the shielding layer for electrical isolation to prevent short circuits. The insulating layer material is polyimide or silicone with a thickness of 1-5μm. Preferably, the protective layer material is polyurethane (PU), TPU or Ecoflex, which is used to enhance the wear resistance and waterproof performance of flexible devices, with a thickness of 10-50μm.

[0017] Furthermore, the connecting wire is a DuPont wire;

[0018] Furthermore, the host computer receives the resistance signal data collected by the data acquisition card through a data line, a WiFi module or a Bluetooth module for real-time processing.

[0019] Furthermore, the algorithms involved in the host computer are:

[0020] First, by collecting the resistance signal in the resting state at the initial stage of wearing, the baseline drift correction is automatically performed to ensure that the resistance signal accurately locates the zero point from the individual's normal breathing state and dynamically updates the baseline according to the signal fluctuation;

[0021] Then, during the monitoring process, the gain (sensitivity) is automatically adjusted according to the strength of the individual respiratory signal, normalizing the amplitude of the resistance signal of different subjects to ensure that the signal strength is within the optimal detection range;

[0022] At the same time, Kalman filtering is used to achieve real-time smoothing and noise suppression of resistance signals, improving data reliability and stability;

[0023] Finally, a temperature compensation function is added to correct the impact of temperature changes on the resistance signal, ensuring that the resistance signal remains accurate under different ambient temperatures.

[0024] Overall, the algorithm enables flexible sensors to achieve higher accuracy, individual adaptability and environmental robustness in respiratory monitoring through multiple processing such as adaptive adjustment, intelligent filtering and temperature compensation.

[0025] Furthermore, the resistance signal of the flexible strain sensor is related to the tidal volume (V T ) and the conventional lung function index forced vital capacity (FVC) are related as follows:

[0026]

[0027] Where ΔR is the resistance change of the flexible strain sensor caused by breathing, R0 is the initial resistance of the flexible strain sensor, ε is the strain (stretching ratio), and k is the strain sensitivity coefficient of the flexible strain sensor;

[0028] Assume that the radius change of the thorax in a certain direction is Δr, and the circumference change is ΔL(t)≈2πΔr(t). Divide by the original length to get the strain ε(t), then:

[0029] ε(t)∝Δr(t)

[0030] According to clinical research, chest circumference changes are related to tidal volume V T Linear relationship

[0031]

[0032] Combining the above two formulas, we can get

[0033] V T (t) = a·Δr(t)

[0034] Where a represents V T Proportional coefficient to Δr;

[0035] The linear approximation model finally established is:

[0036]

[0037] Where K is the comprehensive proportional coefficient, which is obtained through calibration experiments;

[0038] The subjects were given flexible strain sensors and the true tidal volume was monitored using a spirometer or ventilator. The ΔR / R0 and V of each breath were extracted. T Count the maximum and minimum resistance values in each breathing cycle and make linear / nonlinear regression model fitting formula:

[0039]

[0040] Wherein, b represents the constant of the fitting formula measured by experiment;

[0041] First, establish the relationship between the resistance signal and tidal volume, and then establish the relationship between the resistance signal and forced vital capacity (FVC), forced expiratory volume in the first second (FEV1), peak expiratory flow (PEF), etc. to complete the respiratory function evaluation.

[0042] Forced vital capacity (FVC) refers to the total amount of gas that can be exhaled by forced exhalation after maximal inspiration, and its relationship to tidal volume is:

[0043]

[0044] Where T is the duration of forced exhalation.

[0045] FEV1 refers to the volume of air exhaled in the first second of forced exhalation. T The changes are:

[0046]

[0047] PEF is the maximum flow rate during forced expiration. It is obtained by calculating the maximum rate of change of ΔR / R0:

[0048]

[0049] Where C is the calibration coefficient

[0050] Specific implementation steps:

[0051] (1) Quantitative determination of breathing pattern: Multiple flexible strain sensors are simultaneously attached to the subject's main respiratory muscle group and auxiliary respiratory muscle group, and connected to the data acquisition card and the host computer.

[0052] First, based on the flexible strain sensors attached to two of the main muscle groups, a period of resting resistance signals is collected to analyze the individual's basic breathing pattern and generate the subject's breathing baseline; it automatically determines whether there is drift and performs baseline correction.

[0053] Then, during the actual detection process, the sensitivity is dynamically adjusted based on the amplitude of each user's resistance signal to ensure the appropriate response of multiple flexible strain sensors. If the resistance signal is too low, the gain is automatically increased; otherwise, the gain is reduced.

[0054] Then, the resistance signals of each flexible strain sensor are collected according to the subject's breathing baseline. The resistance signals are filtered and standardized, and the strengths of the resistance signals collected by different muscle groups are compared. If the resistance signals of the external intercostal muscles and pectoral muscles are strong, it is thoracic breathing. If the resistance signals of the diaphragm and external oblique muscles are stronger, it is abdominal breathing. If the cleidomastoid muscles and scalene muscles have strong resistance signals, there is a compensatory problem of the auxiliary respiratory muscles.

[0055] Specifically, let the chest respiratory muscle resistance be Rchest(t); the abdominal respiratory muscle resistance be Rabdomen(t); through peak detection, find the peak-to-valley difference (amplitude) of each respiratory cycle, that is, calculate the amplitude ΔR of the chest respiratory muscle resistance and the abdominal respiratory muscle resistance in each respiratory cycle chest and ΔR abdomen , then the breathing mode is determined by a proportional coefficient, defined as:

[0056]

[0057] When α>0.78: chest breathing; α<0.32: abdominal breathing; 0.32<α<0.78: mixed breathing. (The threshold range still needs to be adjusted based on the scope of large-sample trials).

[0058] Furthermore, the primary respiratory muscles include the external intercostal muscles, pectoral muscles, diaphragm, and external oblique muscles; the auxiliary respiratory muscles include the sternocleidomastoid muscles and scalene muscles;

[0059] (2) Non-inflated pulmonary function assessment: Flexible strain sensors are attached to the diaphragm and external oblique abdominal muscles (one centimeter below the xiphoid process, obliquely below the fifth rib) to measure respiratory rate and amplitude. The respiratory rate is determined by the number of respiratory waveform peaks in a visualization software over a period of time (generally times / minute), and the respiratory amplitude is obtained by subtracting the values of each respiratory waveform peak and trough. The resistance signal monitored by the flexible strain sensor is correlated with lung function indicators such as tidal volume, forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and peak expiratory flow (PEF) according to a formula.

[0060] (3) Quantitative judgment of pre-ventilator weaning: Attach the flexible strain sensor to the patient's main respiratory muscle group, and synchronously collect the resistance signals of multiple flexible strain sensors. Extract the respiratory rate, respiratory amplitude, and respiratory pattern from the pre-processed signals. If the patient's respiratory rate is significantly increased (such as greater than 5-10 times / minute) and the respiratory amplitude is reduced during the spontaneous breathing experiment, it indicates respiratory muscle fatigue and weaning should be carefully considered; if the respiratory rate is too fast (generally, a respiratory rate that is continuously higher than 28 times / minute at rest is considered too fast), it indicates excessive respiratory load and weaning should be carefully evaluated; if the chest and abdomen breathing patterns are repeatedly switched or the pattern is unstable, it indicates respiratory muscle dysfunction and weaning should be re-evaluated.

[0061] Beneficial effects of the present invention:

[0062] (1) It is easy and comfortable to wear, highly sensitive, stable, and portable, making it suitable for long-term respiratory monitoring. It is also convenient for subsequent establishment of a patient database, algorithm upgrades, and the construction of a remote smart medical system.

[0063] (2) Providing a quantitative standard for judging respiratory patterns through dynamic measurement of respiratory muscle groups, solving the long-standing problem of relying mainly on the naked eye of doctors;

[0064] (3) Correlate the relationship between respiratory muscle deformation and lung function parameters to solve the problem that some patients' actual breathing conditions do not match the test conditions in conventional lung function tests;

[0065] (4) When it is necessary to determine whether to pre-withdraw the ventilator, provide quantitative standards for the patient's fatigue level or respiratory disorder. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a schematic diagram of the composition of the intelligent respiratory monitoring system.

[0067] Figure 2 Schematic diagram of a breathing pattern recognition case.

[0068] Figure 3 Schematic diagram of the application process of the intelligent respiratory monitoring system.

[0069] Figure 4 Schematic diagram of the structure of the flexible strain sensor.

[0070] In the figure: 1 is the adhesive layer, 2 is the conductive base layer, 3 is the silver paste layer, 4 is the force-sensitive coating, and 5 is the anti-interference layer. DETAILED DESCRIPTION

[0071] The technical solution of the present invention is further described below with reference to embodiments.

[0072] Example 1

[0073] The flexible strain sensor includes an adhesive layer, a force-sensitive coating, a silver paste layer, a conductive base layer and an anti-interference layer from top to bottom. The force-sensitive coating and the silver paste layer are basically located in the same plane. The process sequence of the two is to first apply the force-sensitive coating to the conductive base layer, and then apply the silver paste layer to the conductive base layer. There is a certain overlapping area between the two ends of the force-sensitive coating.

[0074] The adhesive layer is made of medical waterproof tape material;

[0075] Force-sensitive coating: The carbon-based material graphene is added to the solution-based polymer matrix epoxy resin, with a typical mass fraction of 10wt%. To ensure uniform dispersion of the carbon material, an ultrasonic oscillator is used for pre-dispersion for 30 minutes, and mechanical stirring is performed at 5000rpm with a high-speed shearing machine for 15 minutes. To prevent agglomeration, 0.5wt% of a dispersant such as sodium dodecylbenzenesulfonate (SDBS) is added. The dispersed conductive paste is evenly coated on the conductive base layer by screen printing. The thickness is controlled, and the coating thickness is about 20μm. The pattern is designed by screen printing to achieve the desired rectangular morphology (40mm long and 10mm wide).

[0076] Silver paste layer: The silver paste layer is composed of a composite of silver powder and resin, with the silver powder accounting for 70% by weight and the resin being epoxy resin. Solvents and additives are used to control viscosity and dispersion stability. The solvent used is methyl ethyl ketone (MEK) at a mass fraction of 10%; the additive is a silicone leveling agent (such as BYK-306) at a mass fraction of 5% by weight. The silver paste is applied to the conductive substrate by screen printing, with a coating thickness of approximately 20μm. Drying process: Pre-dry at room temperature for 10 minutes, then dry at 90°C for 30 minutes, and then heat cure at a higher temperature of 110°C for a second time. After heat curing for 10 minutes, the resulting silver paste electrode has a square resistance of 0.6Ω / sq, meeting the sensor's resistance signal output requirements. Testing according to ASTM D3359 standards achieves adhesion grade 5B, demonstrating excellent mechanical adhesion, ensuring long-term and stable connection to the connection terminals for resistance signal acquisition.

[0077] Conductive base layer: Made of medical-grade thermoplastic polyurethane elastomer film (TPU), with a thickness of 0.2 mm;

[0078] The anti-interference layer has a three-layer structure, including an insulating layer, a shielding layer, and a protective layer. The insulating layer is placed between the conductive coating and the shielding layer to block direct current contact and prevent short circuits. It is made of polyimide or silicone material and has a thickness of 5μm. The shielding layer material is PEDOT:PSS (200nm thick), which has excellent flexibility and electromagnetic shielding properties. The protective layer material is Ecoflex, with a thickness of 20μm, which is used to improve the device's water resistance, wear resistance, and environmental stability, ensuring its long-term performance in multiple bending and outdoor environments.

[0079] Example 2

[0080] Respiratory pattern recognition for a 26-year-old male. First, connect multiple flexible strain sensors to a multi-channel data acquisition card via DuPont cables. Connect the data acquisition card to a computer (host computer) via a data cable. Peel back the protective layer from the flexible strain sensors. Attach the first flexible strain sensor vertically one centimeter below the subject's xiphoid process. Attach the second flexible strain sensor diagonally along the subject's fifth rib. Ensure full skin contact between the flexible strain sensors. Open the host computer program. The subject first performed resting breathing (i.e., without holding his breath). After a period of normal breathing, the real-time respiratory rate was 13.37 Hz, the peak respiratory amplitude of the chest wave was 29.849 kΩ, the trough respiratory amplitude was 29.802 kΩ, and ΔRchest = 0.047 kΩ; the peak respiratory amplitude of the abdominal wave was 33.362 kΩ, the trough respiratory amplitude was 32.875 kΩ, and ΔRabdomen = 0.487 kΩ; α = 0.047 / 0.047 + 0.487 = 0.088, and α = 0.088 was obtained, indicating that the subject was breathing abdominally, which was consistent with the doctor's experience.

[0081] Example 3

[0082] Lung function assessment, taking a 30-year-old male as an example. Flexible strain sensors are attached to the diaphragm and external oblique abdominal muscles (one centimeter below the xiphoid process, obliquely below the fifth rib) to measure respiratory frequency and amplitude, and a spirometer is used to measure the true tidal volume. Through linear regression analysis, K≈500mL is obtained; when △R / R0=0.02 in the entire respiratory cycle, the constant b=2240 and T=1.907 measured by the fitting formula are substituted into the tidal volume estimation model, and other lung function indicators can be further derived. FVC refers to the total amount of gas that can be exhaled by forced exhalation after maximum inhalation. By calculating the cumulative V in the maximum expiratory cycle T Perform calculations,

[0083]

[0084] FVC=(10+b)(T+1)=(2250*2.907)=6540.75ml

[0085] Where T is the duration of forced expiration. The subject's FVC is rounded to 6541 mL. FEV1 is the volume of air exhaled in the first second of forced expiration. It is calculated by calculating the change in VT during the first 1 second.

[0086]

[0087] FVC=(10+b)(1+1)=(2250*2)=4500ml

[0088] The subject's FEV1 = 4500 mL, PEF refers to the maximum flow rate during forced expiration. It is estimated by calculating the maximum value of the ΔR / R0 change rate.

[0089]

[0090] Where C is the calibration coefficient. Select a segment of experimental data and use the waveform visualization software to find the portion with the highest rate of change. △R(t) = R0 + at, R0 = 30 kΩ, a = 22.5. Taking a one-step derivative, PEF = a / R0 * C. In the experiment, C was measured to be 1000 L / min. The subject's PEF was 750 L / min.

[0091] Example 4

[0092] The quantitative assessment of pre-ventilator weaning was conducted using a 75-year-old patient with asthma and emphysema as an example. The subject was relatively old, so the respiratory rate threshold was adjusted to 24 breaths / minute. During the spontaneous breathing test (SBT), the following changes were monitored: Respiratory rate (f): At baseline, the patient's respiratory rate was 18 breaths / minute. Ten minutes after the SBT, the respiratory rate increased to 28 breaths / minute, exceeding the resting warning value of 24 breaths / minute, indicating excessive respiratory load. Respiratory amplitude (ΔR): At baseline, ΔR = 4.167 kΩ. Ten minutes after the SBT, ΔR decreased to 2.696 kΩ, a 36% decrease, indicating respiratory muscle fatigue. At the same time, repeated switching of chest and abdominal breathing patterns and unstable respiratory rhythm were detected, indicating respiratory muscle dysfunction. Based on a comprehensive assessment, it was recommended to delay weaning, continue respiratory muscle function training, and regularly assess the patient's spontaneous breathing ability.

[0093] The above embodiments are only used to illustrate the present invention. Any equivalent transformations and improvements based on the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. A respiratory monitoring system based on a flexible strain sensor, characterized in that: The respiratory monitoring system includes a flexible strain sensor, a connecting line, a data acquisition card and a host computer; The flexible strain sensor is a band-aid-type flexible strain sensor that can be directly adhered to the surface of human skin. It monitors the deformation of muscles related to breathing and converts it into changes in resistance signals. The flexible strain sensor is connected to the data acquisition card via a connecting line. The data acquisition card is connected to the host computer and is used to transmit the acquired resistance signal to the host computer.

2. A respiratory monitoring system based on a flexible strain sensor according to claim 1, characterized in that: The flexible strain sensor includes an adhesive layer, a force-sensitive coating, a silver paste layer, a conductive base layer, and an anti-interference layer from the outside to the inside. The adhesive layer is made of medical waterproof tape material; The force-sensitive coating is a conductive coating composed of a carbon-based material. The carbon-based material in the force-sensitive coating is carbon black, carbon nanotubes or graphene; the thickness is 10-30 μm; The silver paste layer is composed of a composite of silver powder and resin, and is prepared on the conductive base layer by blade coating or screen printing after the force-sensitive coating is applied. The silver paste layer is located at both ends of the force-sensitive coating and is used to connect the force-sensitive coating to the connecting wire. The silver powder accounts for 65-70wt% by weight. The resin is epoxy resin or acrylate. The thickness of the silver paste layer is 10-30μm. The conductive base layer is made of medical-grade polyethylene, polypropylene, polyvinyl alcohol, and thermoplastic polyurethane film with a thickness of 0.05-0.2mm; The anti-interference layer has a three-layer structure, including an insulation layer, a shielding layer and a protective layer from the inside to the outside.

3. A respiratory monitoring system based on a flexible strain sensor according to claim 2, characterized in that: The preparation method of the force-sensitive coating is as follows: dispersing the carbon-based material in a solution-type polymer matrix to obtain a conductive paste, uniformly coating the dispersed conductive paste on a conductive base layer by blade coating or screen printing, and naturally drying to obtain a force-sensitive coating; wherein, the solution-type polymer matrix is such as PDMS, TPU, Ecoflex or epoxy resin; the mass fraction of the carbon-based material in the solution-type polymer matrix is 2-10wt%.

4. A respiratory monitoring system based on a flexible strain sensor according to claim 2, characterized in that: The shielding layer material is 100-300nm thick PEDOT:PSS, 10-50nm thick copper foil or aluminum foil, 200-500nm thick graphene film, or 100-300nm thick carbon nanotube conductive film, which has excellent electromagnetic shielding effect; the insulating layer is arranged between the conductive base layer and the shielding layer for electrical isolation to prevent short circuit. The insulating layer material is polyimide or silicone with a thickness of 1-5μm; the protective layer material is polyurethane, TPU or Ecoflex, which is used to enhance the wear resistance and waterproof performance of the flexible device and has a thickness of 10-50μm.

5. A respiratory monitoring system based on a flexible strain sensor according to claim 1, characterized in that: The connecting line is a DuPont line; the host computer receives the resistance signal data collected by the data acquisition card through a data line, WiFi module or Bluetooth module for real-time processing.

6. A monitoring and analysis method for a respiratory monitoring system based on a flexible strain sensor according to any one of claims 1 to 5, characterized in that: The specific steps are as follows: (1) Quantitative determination of breathing pattern: Multiple flexible strain sensors are simultaneously attached to the subject's main respiratory muscles and auxiliary respiratory muscles, and connected to a data acquisition card and a host computer; First, the flexible strain sensors attached to two of the major muscle groups collect resting resistance signals for a period of time to analyze the individual's basic breathing pattern and generate the subject's breathing baseline. The system automatically determines whether there is drift and performs baseline correction. Then, during the actual detection process, the sensitivity is dynamically adjusted according to the amplitude of each user's resistance signal to ensure that the multiple flexible strain sensors have a moderate response; If the resistance signal is too small, the gain will be automatically increased; otherwise, the gain will be reduced; Then, based on the subject's breathing baseline, the resistance signals of each flexible strain sensor are collected. After filtering and standardization, the resistance signals collected by different muscle groups are compared. If the resistance signals of the external intercostal muscles and pectoral muscles are strong, it is thoracic breathing. If the resistance signals of the diaphragm and external oblique muscles are stronger, it is abdominal breathing. If the resistance signals of the cleidomastoid muscles and scalene muscles are strong, there is a compensatory problem of the auxiliary respiratory muscles. Specifically, let the chest respiratory muscle resistance be Rchest(t); the abdominal respiratory muscle resistance be Rabdomen(t); through peak detection, find the peak-to-valley difference of each respiratory cycle, that is, calculate the amplitude ΔR of the chest respiratory muscle resistance and the abdominal respiratory muscle resistance in each respiratory cycle chest and ΔR abdomen , then the breathing mode is determined by a proportional coefficient, defined as: When α>0.78: chest breathing; α<0.32: abdominal breathing; 0.32<α<0.78: mixed breathing; The main respiratory muscles include the external intercostal muscles, pectoral muscles, diaphragm, and external oblique muscles; the auxiliary respiratory muscles include the sternocleidomastoid muscles and scalene muscles; (2) Non-inflated lung function assessment: flexible strain sensors are attached to the diaphragm and external oblique muscles to measure respiratory rate and amplitude. The respiratory rate is determined by the number of respiratory waveform peaks in the visualization software over a period of time, and the respiratory amplitude is obtained by subtracting the values of each respiratory waveform peak and trough. The resistance signal monitored by the flexible strain sensor is correlated with the lung function indicators tidal volume, forced vital capacity (FVC), forced expiratory volume in the first second (FEV1), and peak expiratory flow (PEF) according to the formula; (3) Quantitative judgment of pre-ventilator weaning: Attach the flexible strain sensor to the patient's main respiratory muscle group, and synchronously collect the resistance signals of multiple flexible strain sensors. Extract the respiratory rate, respiratory amplitude, and respiratory pattern from the pre-processed signals. If the patient's respiratory rate increases significantly and the respiratory amplitude decreases during the spontaneous breathing experiment, it indicates respiratory muscle fatigue and weaning should be considered carefully. If the respiratory rate is too fast, it indicates that the respiratory load is too heavy and weaning should be evaluated carefully. If the chest and abdomen breathing patterns switch repeatedly or the pattern is unstable, it indicates respiratory muscle dysfunction and weaning should be re-evaluated.

7. A monitoring and analysis method for a respiratory monitoring system based on a flexible strain sensor according to any one of claims 1 to 5, characterized in that: Flexible strain sensor resistance signal and tidal volume V in lung function index T The correlation process with the conventional lung function indicator forced vital capacity FVC is as follows: Where ΔR is the resistance change of the flexible strain sensor caused by breathing, R0 is the initial resistance of the flexible strain sensor, ε is the strain, and k is the strain sensitivity coefficient of the flexible strain sensor; Assume that the radius change of the thorax in a certain direction is Δr, and the circumference change is ΔL(t)≈2πΔr(t). Divide by the original length to get the strain ε(t), then: ε(t)∝Δr(t) According to clinical research, chest circumference changes are related to tidal volume V T Linear relationship Combining the above two formulas, we can get IN T (t)=a·Δr(t) Where a represents V T Proportional coefficient to Δr; The linear approximation model finally established is: Where K is the comprehensive proportional coefficient, which is obtained through calibration experiments; The subjects were given flexible strain sensors and the true tidal volume was monitored using a spirometer or ventilator. The ΔR / R0 and V of each breath were extracted. T Count the maximum and minimum resistance values in each breathing cycle and make linear / nonlinear regression model fitting formula: Wherein, b represents the constant of the fitting formula measured by experiment; First, the relationship between the resistance signal and tidal volume is established, and then the relationship between the resistance signal and forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and peak expiratory flow (PEF) is established to complete the respiratory function assessment. Forced vital capacity (FVC) refers to the total amount of gas that can be exhaled by forced exhalation after maximal inspiration, and its relationship to tidal volume is: Where T is the duration of forced exhalation; FEV1 refers to the volume of air exhaled in the first second of forced exhalation; it is calculated by V T The changes are: PEF is the maximum flow rate during forced expiration; it is obtained by calculating the maximum rate of change of ΔR / R0: Where C is the calibration coefficient.

Images