QCM humidity sensor based on locust bean gum / sodium bismuth titanate and its preparation method, application and respiratory monitoring system

Abstract

The present application discloses a QCM humidity sensor based on locust bean gum / sodium bismuth titanate and its preparation method, application and respiratory monitoring system, belonging to the field of QCM sensor technology. The QCM humidity sensor includes a humidity-sensitive coating and a substrate. The humidity-sensitive coating is a LBG / NBT composite film, which can alleviate the high-humidity swelling of the film, has good mechanical stability, good repeatability, short response / recovery time, high frequency shift response, high sensitivity, small humidity hysteresis, good long-term stability, excellent selectivity, and can quickly and accurately sense humidity changes on the surface of human skin, as well as quickly and accurately sense medium and low humidity changes. The respiratory monitoring system composed of a QCM humidity sensor, a QCM tester, an analysis module, a display module and a terminal can monitor and identify obstructive sleep respiratory status.

Description

Technical Field

[0001] The present application relates to the technical field of QCM sensors, and in particular to a QCM humidity sensor based on locust bean gum / sodium bismuth titanate and a preparation method, application and respiratory monitoring system thereof. Background Art

[0002] As a natural polymer produced from carob trees, locust bean gum (LBG) is not only biocompatible, non-toxic and biodegradable, but also has the additional advantages of wide availability and low cost, making it one of the important research materials in the field of biotechnology. From the perspective of molecular structure, the D-mannose main chain and galactose side chains of LBG are rich in surface functional groups, and the presence of a large number of hydrophilic hydroxyl groups enables LBG to lock water molecules through hydrogen bonds, resulting in strong water absorption.

[0003] However, LBG will swell after absorbing water, and its mechanical stability will decrease. Since the QCM sensor is based on the piezoelectric effect, after the sensing material absorbs water molecules, it indirectly reflects the humidity through the change of frequency, with nanogram-level detection accuracy, and high mechanical stability requirements for the sensing material.

[0004] Therefore, a composite thin film QCM sensor is urgently needed to solve the above problems. Summary of the invention

[0005] In view of this, the present application provides a QCM humidity sensor based on locust bean gum (LBG) / sodium bismuth titanate (NBT) and its preparation method, application and respiratory monitoring system, which can alleviate the high-humidity swelling of the film, has good mechanical stability, good repeatability, short response / recovery time, high frequency shift response, high sensitivity, small humidity hysteresis, good long-term stability, excellent selectivity, and can quickly and accurately sense humidity changes on the surface of human skin, as well as quickly and accurately sense medium and low humidity changes, thereby performing user skin humidity testing and obstructive sleep apnea monitoring.

[0006] Specifically, the following technical solutions are included:

[0007] In the first aspect, the present application provides a QCM humidity sensor based on locust bean gum / sodium bismuth titanate, wherein the QCM humidity sensor comprises a humidity sensitive coating and a substrate, wherein the humidity sensitive coating is a locust bean gum / sodium bismuth titanate (LBG / NBT) composite film, wherein the mass of the humidity sensitive coating is 8800 to 8850 ng, and the substrate is a QCM chip.

[0008] In some embodiments, in the LBG / NBT composite film, the composite films formed by LBG and NBT are overlapped and staggered, LBG is evenly distributed on the surface of NBT nanospheres, and the average diameter of the NBT nanospheres is 100 to 500 nm.

[0009] In some embodiments, the detection range of the QCM humidity sensor is 0-97% RH, the humidity hysteresis is as low as 2.75% RH@67% RH, the frequency shift response is as high as -5725.5 Hz, the sensitivity is as high as 58.99 Hz / % RH, and the water contact angle is 29.08-33.54°.

[0010] In a second aspect, the present application provides a method for preparing a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in the first aspect above, the preparation method comprising:

[0011] Step 1, preparing a mixed solution of NaOH solution, Bi(NO3)3 solution and Ti(OC4H9)4, the mixing volume ratio is (5-7):(4-8):(4-8), the concentration of the NaOH solution is 0.37-0.60 g / mL, and the concentration of the Bi(NO3)3 solution is 0.07-0.15 g / mL;

[0012] Step 2, calcining the mixed solution to obtain NBT, the calcination temperature is 180-200° C., and the calcination time is 24-36 hours;

[0013] Step 3, preparing a LBG / NBT mixed solution, wherein the mass ratio of LBG to NBT is (1-2):(1-2), and the mass fraction of the mixed solution is 0.15-0.25wt%;

[0014] Step 4: Apply the mixed solution to the electrodes on both sides of the QCM chip to obtain the QCM humidity sensor based on locust bean gum / sodium bismuth titanate.

[0015] In a third aspect, the present application provides an application of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in the first aspect above, wherein the application includes user skin humidity testing and obstructive sleep apnea monitoring.

[0016] In a fourth aspect, the present application provides a respiratory monitoring system, which includes a QCM sensor, a QCM tester, an analysis module, a display module and a terminal; the QCM sensor is a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in the first aspect above, which is used to obtain a respiratory humidity signal in real time and convert the respiratory humidity signal into a frequency shift signal; the QCM tester is connected to the QCM sensor, and is used to collect the frequency shift signal output by the QCM sensor in real time; the analysis module is loaded on the terminal, and is used to analyze and process the frequency shift signal obtained by the QCM tester to determine the apnea-hypopnea index and the degree of obstructive sleep apnea; the display module is loaded on the terminal, and is connected to the analysis module, and is used to display the apnea-hypopnea index, the degree of obstructive sleep apnea and to interact with the user; the terminal is connected to the QCM tester, the analysis module and the display module, and is used to provide an operating environment for the analysis module and the display module.

[0017] In some embodiments, the display module specifically includes a curve display unit, a user information display unit, a frequency shift signal loading unit, a key trigger unit, a apnea degree indication unit and a prompt alarm unit.

[0018] In some embodiments, the analysis module uses an embedded intelligent algorithm, which is a neural network recognition algorithm optimized by a genetic algorithm that has undergone macroscopic search and processing of the global optimum.

[0019] In some embodiments, the analysis module includes a population fitness unit, a selection calculation unit, a crossover operation unit, a mutation operation unit and a real number encoding unit, and has the functions of rising edge counting, apnea-hypopnea index AHI calculation, and respiratory condition classification.

[0020] In some embodiments, the population size defined by the embedded intelligent algorithm is 20, the genetic generation is 80, and the number of hidden nodes is set to 6.

[0021] The beneficial effects of the technical solution provided by the embodiments of the present application include at least:

[0022] The embodiment of the present application provides a QCM humidity sensor based on locust bean gum / sodium bismuth titanate and its preparation method and respiratory monitoring system. The QCM humidity sensor includes a humidity-sensitive coating and a substrate. The humidity-sensitive coating is a LBG / NBT composite film, which can alleviate the high-humidity swelling of the film, has good mechanical stability, good repeatability, short response / recovery time, high frequency shift response, high sensitivity, small humidity hysteresis, good long-term stability, excellent selectivity, and can quickly and accurately sense the humidity changes on the surface of human skin, as well as quickly and accurately sense the medium and low humidity changes, so as to perform user skin humidity testing and obstructive sleep breathing monitoring. The respiratory monitoring system composed of the QCM humidity sensor, QCM tester, analysis module, display module and terminal can monitor and identify the obstructive sleep breathing state. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] FIG. 1( a ) is a schematic diagram of the structure of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application; FIG. 1( b ) is a flow chart of the preparation process of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application; FIG. 1( c ) is a SEM image of a LBG / NBT composite film in a QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application;

[0025] Figure 2 XRD patterns of LBG / NBT composite film, pure LBG film and pure NBT film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiments of the present application;

[0026] Figure 3 FT-IR spectra of LBG / NBT composite film, pure LBG film and pure NBT film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiments of the present application;

[0027] Figure 4Schematic diagram of XPS results of LBG / NBT composite film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application: wherein, (a) is a wide range full spectrum scanning XPS spectrum of LBG / NBT composite film; (b) is a C1s spectrum of the XPS spectrum of LBG / NBT composite film; (c) is an O1s spectrum of the XPS spectrum of LBG / NBT composite film; (d) is a Na 1s spectrum of the XPS spectrum of LBG / NBT composite film; (e) is a Bi 4f spectrum of the XPS spectrum of LBG / NBT composite film; (f) is a Ti 2p spectrum of the XPS spectrum of LBG / NBT composite film;

[0028] Figure 5 The Na in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application 0.5 Bi 0.5 TEM image of TiO3 powder: (a) Na 0.5 Bi 0.5 Low magnification TEM image of TiO3 powder; (b) Na 0.5 Bi 0.5 High magnification TEM image of TiO3 powder; (c) Na 0.5 Bi 0.5 HRTEM image of TiO3 powder; (d) Na 0.5 Bi 0.5 Another HRTEM image of TiO3 powder;

[0029] Figure 6 Schematic diagram of the comparison of water contact angles of QCM humidity sensors based on locust bean gum / sodium bismuth titanate provided in the embodiments of the present application; wherein (a) is a schematic diagram of the water contact angle of a pure LBG film; (b) is a schematic diagram of the water contact angle of a pure NBT film; (c) is a schematic diagram of the water contact angle of a LBG / NBT composite film;

[0030] Figure 7 Schematic diagram of the process of water molecule adsorption by the LBG / NBT composite film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application;

[0031] Figure 8 Schematic diagram of dynamic frequency shift curves of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate, the pure LBG film QCM humidity sensor and the pure NBT film QCM humidity sensor under different RH environments provided in the embodiments of the present application;

[0032] Fig. 9Schematic diagram of frequency shift fitting curves of the locust bean gum / sodium bismuth titanate-based QCM humidity sensor, pure LBG film QCM humidity sensor, and pure NBT film QCM humidity sensor under different RH environments provided in the embodiments of the present application;

[0033] Fig.10 This is a rendering of the repeatability of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application;

[0034] Fig.11 Schematic diagram of response / recovery test curves of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate, the pure LBG film QCM humidity sensor, and the pure NBT film QCM humidity sensor provided in the embodiments of the present application;

[0035] Fig.12 A schematic diagram of a dynamic humidification / dehumidification frequency shift response curve of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application;

[0036] Fig.13 A schematic diagram of a humidity hysteresis curve of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application;

[0037] Fig.14 This is a rendering of the long-term stability of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application;

[0038] Fig.15 This is a selective effect diagram of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application;

[0039] Fig.16 A schematic diagram of the comparison of the conductivity spectra of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application; wherein, (a) is a schematic diagram of the conductivity spectrum of the pure LBG thin film QCM humidity sensor at different RH levels; (b) is a schematic diagram of the conductivity spectrum of the pure NBT thin film QCM humidity sensor at different RH levels; (c) is a schematic diagram of the conductivity spectrum of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application at different RH levels; (d) is a schematic diagram of the Q factor curves of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate, the pure NBT thin film QCM humidity sensor and the pure LBG thin film QCM humidity sensor at different RH levels provided in the embodiment of the present application;

[0040] Fig.17Skin humidity test result diagram of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application: wherein, (a) is a humidity test result diagram of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application when the finger moves up and down; (b) is a humidity test result diagram of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application when the finger moves left and right; (c) is a skin humidity test result diagram of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application when the user is resting and sweating; (d) is a humidity test result diagram of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application on the skin surface of different parts of the user;

[0041] Fig.18 A schematic diagram of the sleep breathing test results of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in an embodiment of the present application;

[0042] Fig.19 A system structure diagram of a respiratory monitoring system provided in an embodiment of the present application;

[0043] Fig. 20 A basic process of processing breathing curve samples in an analysis module of a breathing monitoring system provided in an embodiment of the present application and classifying and judging breathing segment waveforms by a classifier based on an embedded intelligent algorithm;

[0044] Fig.21 A schematic diagram of prediction results based on an embedded intelligent algorithm in a respiratory monitoring system provided in an embodiment of the present application: wherein (a) is a schematic diagram of a fitness change curve based on an embedded intelligent algorithm; (b) is an example diagram of a sleep respiratory signal processed by an embedded intelligent algorithm classifier; (c) is a comparison diagram of prediction results of a training set; (d) is a comparison diagram of prediction results of a test set;

[0045] Fig. 22 A confusion matrix of training data and a confusion matrix of test data of a classifier based on an embedded intelligent algorithm in a respiratory monitoring system provided in an embodiment of the present application: wherein (a) is a confusion matrix of training data of a classifier based on an embedded intelligent algorithm; (b) is a confusion matrix of test data of a classifier based on an embedded intelligent algorithm;

[0046] Fig.23 A schematic diagram of a display module in a respiratory monitoring system provided in an embodiment of the present application: wherein (a) is the display result of severe OSA; (b) is the display result of moderate OSA; (c) is the display result of mild OSA; and (d) is the display result of normal conditions.

[0047] The reference numerals in the figure represent: 1-humidity-sensitive coating, 2-substrate. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] In order to make the technical solutions and advantages of the present application more clear, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0050] In a first aspect of an embodiment of the present application, a QCM humidity sensor based on locust bean gum / sodium bismuth titanate is provided. As shown in FIG1( a), the QCM humidity sensor comprises a humidity-sensitive coating 1 and a substrate 2. The humidity-sensitive coating 1 is a LBG / NBT composite film. The mass of the humidity-sensitive coating 1 of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate is 8800 to 8850 ng, and the substrate 2 is a QCM chip.

[0051] In some embodiments, the fundamental frequency of the QCM chip may be 8 MHz, and the electrode material may be silver or gold.

[0052] In some embodiments, the quality of the humidity sensitive coating 1 is determined based on the fundamental frequency of the QCM chip, the frequency shift after the humidity sensitive film is coated, and the Sauerbrey equation.

[0053] In some embodiments, in the LBG / NBT composite film, the composite films formed by LBG and NBT are overlapped and staggered, LBG is evenly distributed on the surface of NBT nanospheres, and the average diameter of the NBT nanospheres is 100 to 500 nm.

[0054] In some embodiments, the detection range of the QCM humidity sensor is 0-97% RH, the humidity hysteresis is as low as 2.75% RH@67% RH, the frequency shift response is up to -5725.5 Hz, the sensitivity is up to 58.99 Hz / % RH, and the water contact angle is 29.08-33.54°.

[0055] The QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application can alleviate the high-humidity swelling of the film due to the presence of the LBG / NBT composite film. The QCM humidity sensor can quickly and accurately sense medium and low humidity changes, with a short response / recovery time, a high frequency shift response, and high sensitivity.

[0056] In a second aspect, the present application provides a method for preparing a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in the first aspect above, the preparation method comprising:

[0057] Step 1, preparing a mixed solution of NaOH solution, Bi(NO3)3 solution and Ti(OC4H9)4, the mixing volume ratio is (5-7):(4-8):(4-8), the concentration of NaOH solution is 0.37-0.60 g / mL, and the concentration of Bi(NO3)3 solution is 0.07-0.15 g / mL;

[0058] Step 2, calcining the mixed solution to obtain NBT, the calcination temperature is 180-200° C., and the calcination time is 24-36 hours;

[0059] Step 3, preparing a LBG / NBT mixed solution, wherein the mass ratio of LBG to NBT is (1-2):(1-2), and the mass fraction of the mixed solution is 0.15-0.25wt%;

[0060] Step 4: Apply the mixed solution to the electrodes on both sides of the QCM chip to obtain a QCM humidity sensor based on locust bean gum / sodium bismuth titanate.

[0061] In some embodiments, as shown in FIG. 1( a), the preparation method may be as follows: first, sodium bismuth titanate (Na 2 O 4 ) is prepared by a one-pot hydrothermal method. 0.5 Bi 0.5 TiO3, NBT) powder. Dissolve 2.92g Bi(NO3)3·5H2O in 30mL deionized water, stir and ultrasonically treat for 1h. Weigh 4.08mL Ti(OC4H9)4 and add it dropwise to the Bi(NO3)3 solution, continue stirring and ultrasonically treat for 1h. At the same time, dissolve 14.4g NaOH (granular) in 30mL deionized water, and slowly drop the NaOH solution into the ultrasonically treated mixed solution of Ti(OC4H9)4 and Bi(NO3)3 with a dropper while stirring, continue stirring for 2h and ultrasonically treat for 1h. Finally, transfer the above mixed emulsion solution to a 100mL polytetrafluoroethylene-lined autoclave and react at a high temperature of 180℃ for 24h. After washing the reactants with deionized water and anhydrous ethanol several times, dry them in a vacuum drying oven at 80℃ to obtain white NBT particles. Weigh equal amounts of LBG powder and prepared NBT powder, mix and grind them, and prepare a 0.2% w / w LBG / NBT solution (LBG:NBT=1:1) with deionized water. Then heat and stir the mixed solution in an 85°C water bath for 1 h. Use a spin coater to spin coat the LBG / NBT mixed solution on the electrodes on both sides of the QCM chip to obtain a QCM humidity sensor based on locust bean gum / sodium bismuth titanate.

[0062] In some embodiments, in order to evaluate the performance of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate (hereinafter referred to as LBG / NBT QCM humidity sensor), QCM humidity sensors coated with pure LBG and pure NBT films are prepared for performance comparison with the LBG / NBT QCM humidity sensor.

[0063] In some embodiments, the chemical materials to be used include locust bean gum powder (LBG) and bismuth nitrate pentahydrate (Bi(NO3)3·5H2O, AR, ≥99.0%) provided by Aladdin Chemical Co., Ltd. Tetrabutyl titanate (Ti(OC4H9)4, MW: 340.36, CP, ≥98.0%) and sodium hydroxide (NaOH, AR, ≥96.0%) purchased from Sinopharm Chemical Reagent Co., Ltd.

[0064] As shown in Figure 1(c), it can be seen that the composite film formed by LBG and NBT overlaps and interlaces, and it can be observed that LBG with good film-forming properties is evenly distributed on the surface of NBT nanospheres. Since NBT plays a supporting role, the entire LBG / NBT composite film becomes porous and rich in layers, which is conducive to providing water molecule adsorption sites. These characteristics improve the swelling phenomenon of LBG colloids with high viscosity due to water absorption, and improve the overall hydrophilicity of the composite film.

[0065] Figure 2 XRD patterns of LBG / NBT composite film, pure LBG film and pure NBT film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application. The interplanar spacing of LBG, NBT and LBG / NBT is characterized by an X-ray diffractometer (XRD) with CuKα radiation (λ=0.1541nm). Figure 2 The XRD patterns of LBG, NBT and LBG / NBT composite films are shown in Figure 1. The XRD pattern of LBG shows a broad peak near the diffraction angle of 2θ = 20°, proving that LBG plant gum as a polymer has no obvious crystalline morphology. No additional diffraction peaks were observed in the XRD pattern of NBT, and the diffraction peaks shown were consistent with the existing Na 0.5 Bi 0.5 The XRD pattern of the LBG / NBT composite film contains obvious characteristic peaks of LBG and NBT, which proves that LBG and NBT exist in the LBG / NBT composite film without other impurities.

[0066] The Fourier transform infrared spectroscopy (FT-IR) spectra of LBG / NBT composite film, pure LBG film and pure NBT film are shown in Figure 2. Figure 3 As shown, for pure LBG, 3423 cm -1The band at 2910 cm represents the -OH stretching vibration. -1 The characteristic peak at 1670 cm corresponds to the CH stretching vibration, and the cyclic stretching and vibration peaks of galactose and mannose appear at 1670 cm -1 This is attributed to the C-C bond between the C atoms of mannose and galactose. -1 The band at 1144 cm corresponds to the bending or deformation of the CH bonds in -CH and -CH2. -1 The characteristic peak at 1029 cm -1 The characteristic peak observed at 830 cm is attributed to COH stretching vibration. -1 and 661cm -1 The characteristic peak at 830cm is attributed to the vibration of Ti-O octahedron, confirming the formation of titanate structure. -1 The characteristic peak at 661cm is formed by the stretching vibration of Ti-O. -1 The characteristic peak at is formed by the bending vibration of bismuth oxide. It can be clearly observed that the FT-IR spectrum of the LBG / NBT composite film has no additional peaks compared with the pure LBG film and the pure NBT film, proving that LBG and NBT have been successfully composited.

[0067] Figure 4 (a) to Figure 4 (f) shows the XPS spectra of the LBG / NBT composite film, pure LBG film and pure NBT film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application, such as Figure 4 As shown in (a), the wide-range full spectrum scan only contains Bi, C, Ti, O and Na elements in the sample, proving that there are no other impurities in the LBG / NBT composite film. Figure 4 (b) shows the high-resolution XPS spectrum of C1s. The XPS spectrum of C 1s can be fitted into two peaks. The peaks at 284.08 eV and 286.83 eV correspond to the CC bond and CO bond, respectively, which are attributed to the galactomannan component in LBG. The high-resolution XPS spectrum of O 1s is deconvoluted into three characteristic peaks, such as Figure 4 (c) The main peak at 532.53 eV is related to the OH bond in the hydrophilic plant gum LBG, and the fitting peaks at 530.43 eV and 535.53 eV are related to the metal-oxygen bond and OO bond in the NBT perovskite film. Figure 4 The Na 1s peak in (d) is located at 1072.88 eV, confirming that the oxidation state of the Na element in the NBT humidity sensitive film is +1. The XPS spectrum data of Bi 4f is shown in Figure 4 (e) is shown. 4f5 / 2 and Bi 4f 7 / 2 The characteristic peaks of Bi 3+ The data is consistent. Figure 4 In (f), the two main peaks at 464.13 eV and 458.18 eV correspond to Ti 2p 1 / 2 and Ti 2p 3 / 2 , indicating that Ti is in the +4 valence state.

[0068] In order to further study the microstructure and nanostructure of the prepared film, the Na 0.5 Bi 0.5 TiO3 powder was characterized by TEM. Figure 5 (a) and Figure 5 (b) It can be clearly observed that Na 0.5 Bi 0.5 The diameter of TiO3 nanospheres is 100 to 500 nm and they are assembled from nanosheets. Figure 5 (c) and Figure 5 (d) is Na 0.5 Bi 0.5 HRTEM image of TiO3 nanospheres. The lattice spacing of 0.392nm and 0.275nm correspond to the (110) and (101) planes of NBT, respectively. The clear lattice spacing is a good proof of the crystalline morphology of NBT nanospheres.

[0069] The water contact angle of pure LBG film, pure NBT film and LBG / NBT composite film was measured by water contact angle analyzer (HARKE-SPCAXIS) to confirm the hydrophilicity of the humidity sensitive film. The volume of deionized water used for immersion in each experiment was set to 3 μL. The water contact angle results are shown in Figure 6 (a) to Figure 6 (c) as shown. Figure 6 (a) shows that the water contact angle of pure LBG film is 64.41o. Figure 6 (b) shows that the water contact angle of the pure NBT film is 54.96°, and the surface contact angles of the two films are less than 90°, which indicates that both LBG and NBT films have hydrophilic properties. Figure 6 As can be seen in (c), the water contact angle of the LBG / NBT composite film is 33.54o, which is smaller than that of the pure LBG film and the pure NBT film. Among the three films, the water contact angle of the LBG / NBT composite film is the smallest, indicating that it has the strongest hydrophilicity, which proves that the blending of LBG and NBT improves the overall hydrophilicity of the composite film.

[0070] Figure 7Schematic diagram of the process of water molecule adsorption by the LBG / NBT composite film in the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application, see Figure 7 LBG is rich in hydrophilic functional groups and has a strong adsorption capacity for water molecules. The addition of the auxiliary hydrophilic material NBT not only provides more water molecule adsorption sites for the composite film, but also the NBT in the form of nanospheres plays the role of supporting the LBG hydrophilic film, effectively increasing the porosity of the hydrophilic composite film, and avoiding the swelling effect caused by the agglomeration of the strongly hydrophilic plant gum LBG after absorbing a large amount of water. The synergistic effect of the two materials improves the humidity sensitivity of the composite film. The adsorption of water molecules on the surface of the LBG / NBT composite film can be divided into two stages: chemical adsorption and physical adsorption. Under low humidity conditions, hydrogen bonds are formed between the hydroxyl groups on the mannose main chain and the galactose side chains of LBG and water molecules, and the water molecules are chemically adsorbed on the hydroxyl groups. At the same time, a chemical adsorption layer is also formed between the sodium ions and oxygen ions on the surface of NBT and the water molecules, providing more binding sites for the formation of hydrogen bonds between the hydrophilic film and the water molecules ( Figure 7 , left). Then the load on the QCM electrode surface changes, and it starts to output a frequency shift response to humidity. Under high humidity conditions, the subsequently adsorbed water molecules form a physical adsorption layer on top of the first chemical adsorption layer through double hydrogen bonds ( Figure 7 , right). As humidity increases, the composite film swells after absorbing a large amount of water, increasing the energy consumption of the QCM while reducing the mechanical stability of the sensor. The presence of NBT adds more voids to the composite film, improves the dispersion of the LBG hydrophilic material and increases the quality factor of the QCM.

[0071] The performances of pure LBG thin film QCM sensor, pure NBT thin film QCM sensor and LBG / NBT composite thin film QCM sensor were tested one by one. Figure 8 The dynamic response curves of pure LBG film, pure NBT film and LBG / NBT composite film QCM sensors are shown. All sensors were tested in a wide humidity range from dry conditions without humidification to high humidity conditions of 97% RH. As the relative humidity in the test environment increases, the amount of water molecules adsorbed by the three films increases, and the frequency response of the three sensors also increases. In addition, compared with the pure LBG film QCM sensor and the pure NBT film QCM sensor, the response of the LBG / NBT composite film QCM sensor is significantly higher. Compared with the pure LBG film QCM sensor (38.01Hz / %RH) and the pure NBT film QCM sensor (13.22Hz / %RH), the QCM humidity sensor based on the LBG / NBT composite film has a higher sensitivity (58.99Hz / %RH), and its performance improvement is attributed to the porous structure and abundant hydrophilic groups of the LBG / NBT composite film.

[0072] Fig. 9 The fitted exponential function relationship between RH and frequency shift (Δf) of the three tested sensors is given. The exponential function of the pure LBG film QCM humidity sensor is |Δf| = -8.52e x / 16.22 -2223.83, the fitting function of the pure NBT film QCM humidity sensor is |Δf|=-30.44e x / 25.23 +41.74, the fitting function of the LBG / NBT composite film QCM sensor is |Δf|=-27.92e x / 18.4 -246.82. It can be clearly found that the exponential function of the LBG / NBT composite film QCM sensor has a greater goodness of fit. The LBG / NBT composite film QCM sensor was exposed to three different RH levels for three cycles of repeatability testing. The test results are shown in Fig.10 The response of the sensor in each test cycle is almost the same, indicating that the LBG / NBT composite film QCM sensor has good repeatability. Fig.11 The response time and recovery time of the three QCM sensors are shown. The response times of the pure LBG thin film QCM sensor, the pure NBT thin film QCM sensor and the LBG / NBT composite thin film QCM sensor are approximately 65s, 21s and 23s, respectively, and the recovery times are approximately 3s, 4s and 3s, respectively.

[0073] Fig.12 The dynamic response curves of the LBG / NBT composite thin film QCM sensor from dry conditions without humidification to high humidity (@97%RH) and then dehumidification to dryness are shown. It can be observed that the response curve has good symmetry and a narrow hysteresis loop. Fig.13 The humidity hysteresis curve of the LBG / NBT composite thin film QCM sensor is shown, where the humidity hysteresis can be determined to be 2.75%RH (@67%RH). The above results prove that the LBG / NBT composite thin film QCM sensor has excellent reversibility and low humidity hysteresis. The pure LBG thin film QCM sensor, the pure NBT thin film QCM sensor and the LBG / NBT composite thin film QCM sensor were tested once a week under humidity conditions of 11%, 33%, 52%, 75% and 85%RH to test the long-term stability of the sensor for a total of 35 days. The test results are shown in Fig.14 As shown in the figure, the frequency response of the QCM sensor has almost no significant change during the entire test process, proving its good stability. The LBG / NBT composite film QCM sensor was tested in different gas environments (3% RH, 85% RH, ethanol 400ppm, NH3400ppm, acetone 400ppm and methane 400ppm). The specific results are shown in the figure. Fig.15Obviously, the response of the LBG / NBT composite film QCM sensor to relative humidity is much higher than that of the interfering gas in the control group, which proves that the sensor has excellent selectivity.

[0074] Based on the above test results, the LBG / NBT composite thin film QCM sensor was studied in depth. The conductivity spectrum analysis method was used to effectively analyze the resonance characteristics of the LBG / NBT composite thin film QCM sensor. Fig.16 (a) to Fig.16 (d) shows the conductivity test results of pure LBG film QCM sensor, pure NBT film QCM sensor and LBG / NBT composite film QCM sensor at different RH. It can be clearly observed that with the increase of humidity, the peak value of the conductivity decreases and the center frequency of the conductivity moves from higher frequency to lower frequency. In addition, from low relative humidity to high relative humidity, the half bandwidth of the conductivity spectrum gradually widens, which is due to the increase in the viscosity of the humidity-sensitive coating 1 after the QCM sensor adsorbs water molecules at higher relative humidity. Fig.16 (d) shows the Q factors of the three QCM sensors at the resonance frequency at different RH levels. The Q factors of the pure LBG film QCM sensor, the pure NBT film QCM sensor, and the LBG / NBT composite film QCM sensor decrease with the increase of RH, which is due to the increase of film viscosity and the energy consumed by the QCM sensor due to the adsorption of water molecules. The LBG / NBT composite film QCM sensor has the largest Q factor among the three QCM sensors, which is much higher than the pure LBG film QCM sensor and the pure NBT film QCM sensor. This is because the introduction of NBT can effectively support the LBG film, improve the quality factor and reduce energy dissipation. Therefore, the overall performance of the LBG / NBT composite film QCM sensor is better than that of a single material.

[0075] Table 1 shows the performance parameters of QCM humidity sensors attached with pure LBG, pure NBT, LBG / NBT composite film and other humidity sensitive films. The results show that the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application has a larger frequency shift response (-5725.2Hz), higher sensitivity (58.99Hz / %RH) and shorter response / recovery time (23s / 3s). These indicators are better than other QCM humidity sensors listed in the table.

[0076] Table 1 QCM humidity sensor performance parameter comparison table

[0077]

[0078]

[0079] In summary, the preparation method of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application, the prepared QCM humidity sensor can alleviate the high-humidity swelling of the film, can quickly and accurately sense medium and low humidity changes, has good mechanical stability, good repeatability, short response / recovery time, high frequency shift response, high sensitivity, small humidity hysteresis, good long-term stability, and excellent selectivity.

[0080] In a third aspect, the present application provides an application of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in the first aspect above, and the applications include user skin humidity testing and obstructive sleep apnea monitoring.

[0081] Skin humidity monitoring is one of the popular directions of humidity monitoring. The relative humidity of the human skin surface varies with different parts. For example, the relative humidity of the back skin, which is always wrapped in clothes, is in the range of 60-100% RH, while the relative humidity of the ventral skin of the fingers and the facial skin, which are often exposed to the air, is in the range of 40-80% RH and 45-60% RH, respectively. In general, it is scientifically ideal for the relative humidity of the human skin surface to be maintained at around 50% RH. Therefore, the humidity sensor developed for monitoring human skin humidity needs to have high sensitivity in the medium and low humidity stages and a faster response / recovery speed. In this way, when measuring the humidity on the surface of human skin, the humidity sensor can not only distinguish between air humidity and human body humidity, but also enable the measured response value to recover quickly from the air humidity.

[0082] The skin moisture of a 25-year-old healthy female was tested using the LBG / NBT composite thin film QCM sensor provided in the embodiments of the present application. Fig.17 (a) and Fig.17 (b) is the experimental result of the LBG / NBT composite thin film QCM sensor in testing finger humidity. At the beginning of the test, the index finger was placed about 3 mm away from the humidity sensitive coating 1, and then the finger was moved vertically within a distance of about 3 to 5 mm from the surface of the LBG / NBT composite thin film QCM sensor. The test results are shown in Figure 2. Fig.17 As shown in (a), the response value is the largest when the finger is 3 mm away from the surface of the LBG / NBT composite thin film QCM sensor, and the response value is the smallest when the finger is 5 mm away from the surface of the LBG / NBT composite thin film QCM sensor. Fig.17 (b) shows the response of the sensor when the finger moves left and right above the LBG / NBT composite thin film QCM sensor at a speed of about 2 cm / s. It can be observed that the response of the LBG / NBT composite thin film QCM sensor changes periodically, and the closer the finger is to the surface of the LBG / NBT composite thin film QCM sensor, the larger the response value is, and the response value decreases when the finger moves away. Fig.17(c) shows the skin humidity of the human body in a calm state and in a moving state. It can be seen that the skin surface humidity of the human body is lower in a calm state. Since the human body dissipates heat through sweating during exercise, the skin humidity of the human body after exercise will gradually increase, and the response value of the LBG / NBT composite film QCM sensor will increase accordingly. Fig.17 (d) is the test result of skin surface humidity of different parts of human body. The skin humidity of cheek, forehead, elbow, back of hand, palm and finger surface was tested respectively. The relative humidity of these parts were 27.9%RH, 29.4%RH, 34.1%RH, 42.5%RH, 53%RH and 55.6%RH respectively, which is in line with the biological law of human body.

[0083] Obstructive sleep apnea (OSA) is characterized by recurrent pharyngeal collapse during sleep, resulting in intermittent hypoxemia and sleep disruption. OSA occurs in approximately 13% to 33% of men and approximately 16% to 19% of women and is becoming increasingly common in adults. If not monitored, the irregular sleep and intermittent hypoxia caused by OSA, in addition to increased sympathetic nerve activity, can lead to reduced quality of life. The severity of OSA is usually quantified by the apnea-hypopnea index (AHI), which is the sum of respiratory events (including apneas and hypopneas) per hour of sleep. Apnea is defined as the absence of a respiratory signal for more than 10 seconds and a decrease in airflow amplitude of more than 90%. Hypopnea is defined as a decrease in airflow amplitude of more than 30% and lasting for more than 10 seconds. The severity of OSA is classified according to the size of the AHI parameter as shown in Table 2.

[0084] Table 2 Classification of OSA severity corresponding to different AHI parameters

[0085]

[0086] Most monitoring methods are costly, inaccessible, and have a poor tolerance process due to interference with the user's sleep, which is a burden for the user. Therefore, the present application provides an application of a QCM humidity sensor based on locust bean gum / sodium bismuth titanate to perform sleep respiration monitoring through a LBG / NBT composite thin film QCM sensor.

[0087] Through the excellent test results of the LBG / NBT composite film QCM sensor on skin humidity, it can be found that the sensor has a good response sensitivity to medium and low humidity, which meets the needs of detecting abnormal breathing data in the sleep breathing signal in OSA characteristics. Fig.18An 18-year-old female volunteer simulated normal breathing and OSA-related abnormal breathing. The sampling interval of the QCM tester was set to 0.25s. The LBG / NBT composite film QCM sensor was fixed under the volunteer's nose about 3cm away from the nose wing through a breathing mask, and the relative humidity after inhalation was used as the benchmark. Fig.18 The respiratory waveform monitored by the LBG / NBT composite film QCM sensor includes normal breathing and OSA-related respiratory events (apnea and hypopnea). Fig.18 The dark block marks the respiratory signal during apnea, and the light block marks the respiratory curve during hypopnea. It can be observed that the respiratory waveforms related to hypopnea and apnea collected by the LBG / NBT composite film QCM sensor have obvious characteristics, which can be significantly distinguished from the respiratory waveform during normal breathing, and are suitable for the collection of sleep breathing data and the calculation of auxiliary AHI parameters.

[0088] In summary, the application of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate provided in the embodiment of the present application can perform user skin humidity testing and obstructive sleep apnea monitoring, and the detection effect is excellent. The application of portable sensors can minimize the use of bulky monitors, thereby minimizing interference with sleep quality, pain and discomfort, and can provide a more objective non-contact assessment of OSA severity, which is very suitable for home-based pre-self-screening of obstructive sleep apnea.

[0089] Fourthly, Fig.19 As shown, the present application provides a respiratory monitoring system, which includes a QCM sensor, a QCM tester, an analysis module, a display module and a terminal; the QCM sensor is a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in the first aspect above, which is used to obtain a respiratory humidity signal in real time and convert the respiratory humidity signal into a frequency shift signal; the QCM tester is connected to the QCM sensor, and is used to collect the frequency shift signal output by the QCM sensor in real time; the analysis module is loaded on the terminal, and is used to analyze and process the frequency shift signal obtained by the QCM tester, and determine the apnea-hypopnea index and the degree of obstructive sleep apnea; the display module is loaded on the terminal, and is connected to the analysis module, and is used to display the apnea-hypopnea index, the degree of obstructive sleep apnea and to interact with the user; the terminal is connected to the QCM tester, the analysis module and the display module, and is used to provide an operating environment for the analysis module and the display module.

[0090] In some embodiments, the display module specifically includes a curve display unit, a user information display unit, a frequency shift signal loading unit, a key trigger unit, a apnea degree indication unit and a prompt alarm unit.

[0091] It should be noted that the calculation of the AHI parameter is an important step in classifying the severity of OSA. In order to obtain a highly accurate AHI value, it is necessary to quickly and accurately distinguish between normal breathing and respiratory events such as hypopnea and apnea in the sleep breathing signal. To this end, a machine learning algorithm is used to classify the collected sleep breathing signals.

[0092] In some embodiments, the analysis module includes using an embedded intelligent algorithm, which is a neural network recognition algorithm optimized by a genetic algorithm that performs macroscopic search and global optimization. The embedded intelligent algorithm uses a genetic algorithm GA that is good at macroscopic search and global optimization to optimize the BP neural network. By complementing each other between the genetic algorithm and the BP neural network, the GA algorithm can be used to find the optimal range globally, and then the BP neural network can be used to further obtain the optimal solution, thereby improving the classification accuracy and operation speed of the algorithm.

[0093] In some embodiments, the number of hidden nodes set by the embedded intelligent algorithm is 6. In order to imitate the biological evolution mechanism, the defined population size is 20 and the genetic generation is 80.

[0094] In some embodiments, the analysis module includes a population fitness unit, a selection calculation unit, a crossover operation unit, a mutation operation unit and a real number encoding unit, and has the functions of rising edge counting, apnea-hypopnea index AHI calculation, and respiratory condition classification.

[0095] In some embodiments, the formula for calculating the population fitness value F in the population fitness unit is:

[0096]

[0097] In the formula, f i It represents the expected output value, f(X i ) represents the actual output value.

[0098] In some embodiments, in the selection calculation unit, the probability of selecting excellent sample data individuals from the original population for reproduction to generate the next generation of sample data is calculated by the formula:

[0099]

[0100] In the formula, p i represents the individual's selection probability, F i represents the population fitness value of this individual (i), and N represents the population size.

[0101] In some embodiments, in the crossover operation unit, a new individual with strong adaptability is generated by exchanging and combining two random samples in the population, and the formula of the crossover operation is:

[0102] a kj =a kj (1-b)+a lj

[0103] a lj =a lj (1-b)+a ij

[0104] In the formula, a k 、a l represents two randomly selected sample individuals in the crossover operation at j, b represents the parameter of the crossover function, which is defined as 0.4, and a i represents the individual ready to mutate, a ij Represents the original individual a i The new individual formed after mutation at position j.

[0105] In some embodiments, in the mutation operation unit, a mutation operation is performed on a portion of a selected individual in the original population to generate better individuals to maintain the diversity of the population. The mutation operation formula is:

[0106]

[0107] In the formula, a i represents the individual ready to mutate, a max 、a min Represents the variant individual a ij The boundary conditions, r, r2 represent random numbers, g represents the number of iterations, G max , G min Indicates the maximum and minimum evolution times, a ij Represents the original individual a i The new individual formed after mutation at position j.

[0108] In some embodiments, the weight and threshold calculation formula of the BP neural network optimized by the GA algorithm is as follows:

[0109]

[0110]

[0111] In the formula, w ij represents the weight between the i-th neuron and the j-th neuron in the hidden layer, w jk represents the connection weight between the neurons in the hidden layer and the output layer, k represents the kth neuron in the output layer, and α j represents the bias from the input layer to the hidden layer, β k represents the bias from the input layer to the hidden layer, η represents the learning rate, which is set to 0.008, and H jrepresents the output of the hidden layer. m represents the number of nodes in the output layer, x i represents the input variable and e represents the prediction error.

[0112] In some embodiments, in the real number coding unit, a method of hybrid coding of the number of hidden layer nodes and network weights is adopted, and the coding form is as follows:

[0113]

[0114] Where W i represents the weight associated with the i-th hidden layer node, and the code string length L = (n i +n0)n h +1, where: n i is the number of nodes in the input layer, n0 is the number of nodes in the output layer, and n h is the number of hidden layer nodes.

[0115] Fig. 20 The present application provides a basic process for the respiratory curve sample processing of the analysis module in a respiratory monitoring system and the classification and judgment of the respiratory segment waveform by a classifier based on an embedded intelligent algorithm, and the frequency shift signal output by the QCM sensor collected by the sleep respiratory curve. In some embodiments, the basic information of the subject to be tested and the sleep respiratory waveform to be tested must first be input, and the input respiratory waveform must be divided into respiratory data segments at a period of 5 seconds for subsequent algorithm training. The embedded intelligent algorithm program (i.e., a neural network recognition algorithm optimized by a genetic algorithm that has undergone macroscopic search and processing of the global optimum) is written in the MATLAB environment to train and classify normal breathing and respiratory events including apnea and hypopnea in sleep respiratory waveforms. During the experiment, four adult testers simulated a total of 240 sets of data of normal breathing and respiratory events (apnea and hypopnea) in the sleep state, from which 110 sets of normal breathing segments and respiratory event segments were intercepted and selected for training the classifier based on the embedded intelligent algorithm. The classified respiratory segment data are sorted and used to further calculate the AHI parameters, and finally the OSA severity classification results of the subject are displayed according to the calculated AHI value. Fig.21 (a) shows the fitness change curve based on the embedded intelligent algorithm, with the error value set as the fitness value. It can be observed that as the number of iterations increases, the error becomes smaller and smaller, and when the number of iterations exceeds 32, the error value remains stable. The results show that the constructed classifier based on the embedded intelligent algorithm has a small error, which meets the expected requirements. Fig.21(b) shows an example of a sleep breathing curve after being processed by an embedded intelligent algorithm classifier. The trained embedded intelligent algorithm classifier classifies the processed breathing segments. A total of two classifications are set: normal breathing and respiratory events. The respiratory signals of hypopnea and apnea are all classified as respiratory events. The classified respiratory segments are sorted in chronological order and drawn, with normal breathing segments displayed as 1 and respiratory event segments displayed as 2. When a rising edge is detected and the classification result of the second breathing segment after the rising edge is also displayed as 2, it means that a respiratory event occurred at that time point. Therefore, the total number of respiratory events in this section of sleep breathing signal can be calculated by counting the number of rising edges, and the AHI parameter of the subject can be further calculated in combination with the total duration of the sleep breathing segment. Fig.21 (c) and Fig.21 (d) Comparison of the prediction results of the training set and the test set of the embedded intelligent algorithm classifier for respiratory segment classification. It can be observed that in the training set, the prediction results of the embedded intelligent algorithm classifier are consistent with the actual results. In the test set, the accuracy of the classifier is 98.3%, which has a high precision.

[0116] Fig. 22 The confusion matrix of the training data and the confusion matrix of the test data of the classifier based on the embedded intelligent algorithm are shown. The confusion matrix reflects that the embedded intelligent algorithm classifier has good classification results for the processed sleep breathing segment signals, which shows that the embedded intelligent algorithm classifier has achieved good training results and can be used for the subsequent calculation of AHI parameters and the judgment of OSA severity. The above results show that the built embedded intelligent algorithm classifier can achieve the expected training results. Fig.23 A schematic diagram of a display module in a respiratory monitoring system provided in an embodiment of the present application shows an interface for AHI parameter calculation and OSA severity classification. This program interface for OSA severity classification has user interaction functions such as curve display, user information display, frequency shift signal loading, button triggering, apnea degree indication, prompt alarm and result integration. There is a code for calculating AHI parameters and judging the corresponding OSA severity in the button callback. The OSA severity judgment process is controlled through human-computer interaction, and the OSA severity judgment result can be intuitively displayed, accompanied by indicator lights of different colors. The OSA severity judgment program can display the individual basic information of the subject and the sleep breathing waveform to be detected, and give the AHI calculation result and OSA severity judgment in combination with the trained embedded intelligent algorithm classifier. The text judgment result is generated in the display box, and the indicator light will also give a warning according to the judgment result. Then the system will organize the breathing judgment result and related suggestions to generate an editable electronic version of the sleep report. Fig.23 (a) to Fig.23(d) shows the judgment results of OSA severity as severe, moderate, mild and normal, and these four judgment results correspond to red, orange, yellow and green indicator lights respectively. Due to the introduction of an embedded intelligent algorithm classifier with high training completion to construct the respiratory signal classification link and AHI parameter calculation program, the OSA severity judgment interface has the advantages of fast AHI parameter calculation speed and high OSA severity judgment results for the input sleep breathing data.

[0117] In summary, the respiratory monitoring system provided in the embodiment of the present application has high monitoring accuracy and the result of judging the degree of sleep apnea is relatively accurate.

Claims

1. QCM humidity sensor based on locust bean gum / sodium bismuth titanate, characterized in that: The QCM humidity sensor comprises a humidity sensitive coating (1) and a substrate (2), wherein the humidity sensitive coating (1) is a locust bean gum / sodium bismuth titanate (LBG / NBT) composite film, the mass of the humidity sensitive coating (1) is 8800-8850 ng, the substrate (2) is a QCM chip, and in the LBG / BNT composite film, the composite films formed by LBG and BNT overlap and interlace, and LBG is evenly distributed on the surface of BNT nanospheres, and the average diameter of the BNT nanospheres is 100-500 nm.

2. The QCM humidity sensor based on locust bean gum / sodium bismuth titanate according to claim 1, characterized in that: The detection range of the QCM humidity sensor is 0~97%RH, the humidity hysteresis is as low as 2.75%RH at 67%RH, the frequency shift response is as high as -5725.5Hz, the sensitivity is as high as 58.99Hz / %RH, and the water contact angle is 29.08~33.54°.

3. The method for preparing the QCM humidity sensor based on locust bean gum / sodium bismuth titanate according to any one of claims 1-2, characterized in that: The preparation method comprises: Step 1, preparing a mixed solution of NaOH solution, Bi(NO3)3 solution and Ti(OC4H9)4, the mixing volume ratio of which is (5-7):(4-8):(4-8), the concentration of the NaOH solution is 0.37-0.60 g / mL, and the concentration of the Bi(NO3)3 solution is 0.07-0.15 g / mL; Step 2, calcining the mixed solution to obtain NBT, the calcination temperature is 180~200 o C, calcination time is 24~36h; Step 3, preparing a LBG / NBT mixed solution, wherein the mass ratio of LBG to NBT is (1-2): (1-2), and the mass fraction of the mixed solution is 0.15-0.25wt%; Step 4: Apply the mixed solution to the electrodes on both sides of the QCM chip to obtain the QCM humidity sensor based on locust bean gum / sodium bismuth titanate.

4. The use of the QCM humidity sensor based on locust bean gum / sodium bismuth titanate according to any one of claims 1-2, characterized in that: The applications include user skin moisture testing and obstructive sleep apnea monitoring.

5. A respiratory monitoring system, characterized in that The respiratory monitoring system includes a QCM sensor, a QCM tester, an analysis module, a display module and a terminal; The QCM sensor is a QCM humidity sensor based on locust bean gum / sodium bismuth titanate as described in any one of claims 1-2, which is used to obtain a respiratory humidity signal in real time and convert the respiratory humidity signal into a frequency shift signal; The QCM tester is connected to the QCM sensor and is used to collect the frequency shift signal output by the QCM sensor in real time; The analysis module is loaded on the terminal and is used to analyze and process the frequency shift signal acquired by the QCM tester to determine the apnea-hypopnea index and the degree of obstructive sleep apnea; The display module is mounted on the terminal and connected to the analysis module, and is used to display the apnea-hypopnea index, the obstructive sleep apnea degree, and to perform user interaction; The terminal is connected to the QCM tester, the analysis module and the display module, and is used to provide an operating environment for the analysis module and the display module.

6. The respiratory monitoring system according to claim 5, characterized in that: The display module specifically includes a curve display unit, a user information display unit, a frequency shift signal loading unit, a key trigger unit, a apnea degree indicating unit and a prompt alarm unit.

7. The respiratory monitoring system according to claim 5, characterized in that: The analysis module uses an embedded intelligent algorithm, which is a neural network recognition algorithm optimized by a genetic algorithm that has undergone macroscopic search and processing of the global optimum.

8. The respiratory monitoring system according to claim 5, characterized in that: The analysis module comprises a population fitness unit, a selection calculation unit, a crossover operation unit, a variation operation unit and a real number encoding unit, and has the functions of rising edge counting, apnea hypopnea index AHI calculation and respiratory status classification.

9. The respiratory monitoring system according to claim 7, characterized in that: The population size defined by the embedded intelligent algorithm is 20, the genetic generation is 80, and the number of hidden nodes is set to 6.

Images