Lung function instrument protective sleeve with intelligent induction and self-prompting disinfection functions

Abstract

The invention discloses a pulmonary function instrument protection sleeve with intelligent induction and self-prompting disinfection functions, relates to the technical field of intelligent induction disinfection, aims to solve the technical problems that an existing protection sleeve is single in function and cannot guarantee timely disinfection of equipment, and is realized based on an intelligent induction and self-prompting disinfection system. The system comprises a spectrum monitoring module, an intelligent sensing and self-prompting module and a multi-modal fusion module, the spectrum monitoring module comprises an ultraviolet-visible light source, a micro spectrograph and a standard bacteria spectrum library, the ultraviolet-visible light source adopts a specific wavelength and is used for exciting biomolecules of nucleic acid and protein in bacteria to generate spectrum signals, and the micro spectrograph compares collected spectrum data with the standard bacteria spectrum library; identifying bacterial characteristics and calculating the concentration; and the multi-modal fusion module adopts an electronic sensor and performs data fusion with the spectrum monitoring module so as to improve the bacterial detection accuracy. The device has the advantage of guaranteeing timely disinfection of equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent sensing disinfection, and more particularly to a protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions. Background Art

[0002] In the field of medical device disinfection, spirometers, as key devices for monitoring human lung function, are crucial for disinfection. Currently, ultraviolet disinfection is a common method for disinfecting spirometers. Ultraviolet disinfection is highly efficient and rapid, effectively killing a wide range of pathogens and ensuring the sanitation and safety of the equipment. However, ultraviolet radiation can adversely affect the plastic casing of spirometers. Prolonged exposure to ultraviolet radiation can cause the plastic casing to become brittle and crack, compromising the device's seal and affecting the instrument's proper function and accuracy. To prevent this, the plastic casing of the spirometer must be shielded during ultraviolet disinfection.

[0003] Against this backdrop, protective covers for spirometers have emerged, designed to protect the plastic casing during UV disinfection. However, existing covers have significant drawbacks. Firstly, they only provide a simple physical shielding function and are unable to monitor bacterial contamination within the spirometer. This makes it difficult for medical staff to determine the degree of contamination during use and whether the equipment requires timely disinfection. This can lead to delays in disinfection when the equipment is severely contaminated, increasing the risk of cross-infection. Secondly, existing covers lack a reminder function. Even if the equipment is contaminated beyond the standard, they cannot alert medical staff to perform disinfection, significantly reducing the timeliness and effectiveness of disinfection efforts.

[0004] In view of this, we propose a pulmonary function meter protective cover with intelligent sensing and self-prompting disinfection functions. Summary of the Invention

[0005] The purpose of the present invention is to provide a pulmonary function instrument protective cover with intelligent sensing and self-prompting disinfection functions, so as to solve the technical problem that the existing protective cover has a single function and cannot ensure timely disinfection of the equipment.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions, which is implemented based on an intelligent sensing and self-prompting disinfection system, comprising:

[0007] Spectral monitoring module, intelligent sensing and self-prompt module, and multimodal fusion module;

[0008] The spectral monitoring module includes a UV-visible light source, a micro-spectrometer, and a standard bacterial spectral library. The UV-visible light source uses a specific wavelength to excite the biological molecules of nucleic acids and proteins in bacteria to produce spectral signals. The micro-spectrometer compares the collected spectral data with the standard bacterial spectral library to identify bacterial characteristics and calculate concentrations.

[0009] The multimodal fusion module uses electronic sensors and performs data fusion with the spectral monitoring module to improve the accuracy of bacteria detection and calculate the pollution index P;

[0010] The intelligent sensing and self-prompting module includes an audio-visual prompt device and a wireless communication module. When the pollution index P exceeds the set threshold, the audio-visual prompt device automatically emits an audio-visual prompt, and the wireless communication module remotely prompts medical staff to disinfect the pulmonary function meter.

[0011] Preferably, in order to achieve accurate detection of bacteria by the spectral monitoring module, the spectral monitoring module establishes a standard bacterial spectral library and systematically collects characteristic absorption peaks, reflectance and fluorescence characteristic data of different bacteria;

[0012] During the data acquisition process, the spectrometer is used in the wavelength range λ∈[λ min ,λ max ] and obtain the spectral data S for each bacterial sample j. j (λ), in order to ensure the accuracy and reliability of the data, each bacterial sample is measured multiple times. Assuming the number of measurements is n, the final spectral data of the jth bacterial sample is obtained by averaging:

[0013]

[0014] Integrate the final spectral data of all bacterial samples to form a standard bacterial spectral library Where m is the number of bacterial species.

[0015] Preferably, in the actual operation of the spectrum monitoring module, in order to ensure that the acquired spectrum data is accurate and reliable, the collected spectrum data is first preprocessed, and a denoising algorithm based on wavelet transform is used to obtain the denoised spectrum data S denoised (λ);

[0016] After completing the spectral data preprocessing, the ultraviolet-visible light source is used to excite bacterial biomolecules to generate spectral signals. The excitation light with a specific wavelength λ0 is used to excite biomolecules such as nucleic acids and proteins in the bacteria to emit fluorescence. The micro-spectrometer compares the collected preprocessed spectral data with the standard bacterial spectral library. The comparison process is based on complex spectral feature analysis to obtain the spectral similarity Sim. When the Sim value exceeds the preset threshold Sim thWhen the corresponding type of bacterial contamination is detected, the bacterial concentration C is calculated by the concentration-spectral intensity relationship model. spec ;

[0017] Assume that the bacterial concentration C spec It satisfies a linear relationship with the spectral intensity I(λ0) at a specific wavelength λ0, and its mathematical model is:

[0018] C spec =a·I(λ0)+b;

[0019] Among them, a is the proportional coefficient, which reflects the influence of spectral intensity change on bacterial concentration, and b is the constant term.

[0020] Preferably, the electronic sensor used in the multimodal fusion module is an electrochemical sensor, which performs detection based on the electrochemical activity of bacterial metabolites;

[0021] In the signal processing of electrochemical sensors, let the original electrical signal output by the sensor be E raw , signal calibration and concentration conversion are performed using the following formula:

[0022] C met =g·(E raw -E offset );

[0023] Among them, C met is the concentration of bacterial metabolites, g is the sensor sensitivity coefficient, E offset is the zero offset voltage, in order to accurately calibrate g and E offset , multiple measurements were performed using bacterial metabolite solutions with standard concentrations;

[0024] Assume that the concentration of the standard solution is C std , the corresponding sensor output electrical signal is E std ,but:

[0025] C std =g·(E std -E offset );

[0026] Among them, C met is the concentration of bacterial metabolites, g is the sensor sensitivity coefficient, E offset is the zero offset voltage.

[0027] Preferably, in order to achieve data fusion between the multimodal fusion module and the spectral monitoring module, a support vector machine algorithm is used, and the multimodal fusion module uses a chemical sensor to obtain data from the perspective of bacterial metabolite detection, and performs fusion analysis with the spectral data collected by the spectral monitoring module;

[0028] In the data fusion process, the feature vector extracted by the spectrum monitoring module is set as X s =[x s1 ,x s2 ,…,x sn ], the feature vector extracted by the electrochemical sensor is X e =[x e1 ,x e2 ,…,x em ], merge the two into a new feature vector X = [X s ,X e ];

[0029] Introduced bacterial metabolite concentration C met and bacterial concentration C spec , define the pollution index P, and its calculation formula is:

[0030]

[0031] Where w1 and w2 are the bacterial concentrations C spec and bacterial metabolite concentration C met The weight coefficient, and satisfy w1+w2=1, α i is the Lagrange multiplier, y i is the sample label, K(X i ,X) is the kernel function and b1 is the bias term.

[0032] Preferably, the intelligent sensing and self-prompting module adopts a risk assessment model during the threshold setting process;

[0033] Assume that the hazard coefficient of different bacteria types l is h l , the frequency of use of the pulmonary function meter is f, then the bacterial concentration threshold C th It can be calculated by the following formula:

[0034]

[0035] Where p is the total number of bacterial types, w l is the weight coefficient of different bacteria types, and β is the safety factor.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention integrates intelligent sensing and self-prompting disinfection functions, which makes up for the defects of existing protective covers that cannot monitor pollution and prompt disinfection. The spectral monitoring module and the multimodal fusion module work together to accurately monitor the bacterial contamination of the pulmonary function meter in real time. Once the pollution index exceeds the threshold, the intelligent sensing and self-prompting module immediately emits sound and light and remote prompts to ensure that medical staff disinfect in time, effectively reduce the risk of cross infection, and solve the problem that the existing protective cover has a single function and cannot ensure timely disinfection of equipment.

[0038] 2. The present invention further improves the reliability of disinfection decisions by optimizing the detection process of the spectral monitoring module, establishes a standard bacterial spectral library, and provides an accurate reference for bacterial identification; it uses a wavelet transform denoising algorithm to process spectral data to reduce interference and make the detection results more accurate, which avoids excessive disinfection or untimely disinfection due to misjudgment of contamination. While reducing the risk of cross-infection, it also reduces unnecessary disinfection operations, extends the service life of the spirometer, and further improves the disinfection management of the equipment.

[0039] 3. By integrating a multimodal fusion module with intelligent sensing and self-prompting modules, this system dynamically adjusts the disinfection threshold based on factors such as the severity of different bacterial types and the frequency of spirometry use. This feature not only makes disinfection prompts more scientific and reasonable, but also improves device efficiency while ensuring effective disinfection. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0041] To facilitate those skilled in the art to understand the technical solution of the present invention, the technical solution of the present invention is further described with reference to the accompanying drawings.

[0042] Example 1, as Figure 1 As shown, a pulmonary function meter protective cover with intelligent sensing and self-prompting disinfection functions is implemented based on an intelligent sensing and self-prompting disinfection system, which includes:

[0043] Spectral monitoring module, intelligent sensing and self-prompt module, and multimodal fusion module;

[0044] The spectral monitoring module includes a UV-visible light source, a micro-spectrometer, and a standard bacterial spectral library. The UV-visible light source uses a specific wavelength to excite the biological molecules of nucleic acids and proteins in bacteria to produce spectral signals. The micro-spectrometer compares the collected spectral data with the standard bacterial spectral library to identify bacterial characteristics and calculate concentrations.

[0045] The multimodal fusion module uses electronic sensors and performs data fusion with the spectral monitoring module to improve the accuracy of bacteria detection and calculate the pollution index P;

[0046] The intelligent sensing and self-prompting module includes an audio-visual prompt device and a wireless communication module. When the pollution index P exceeds the set threshold, the audio-visual prompt device automatically emits an audio-visual prompt, and the wireless communication module remotely prompts medical staff to disinfect the pulmonary function meter.

[0047] In the embodiment of the present invention, in order to achieve accurate detection of bacteria by the spectral monitoring module, the spectral monitoring module establishes a standard bacterial spectral library and systematically collects characteristic absorption peaks, reflectance and fluorescence characteristic data of different bacteria;

[0048] During the data acquisition process, the spectrometer is used in the wavelength range λ∈[λ min ,λ max ] and obtain the spectral data S for each bacterial sample j. j (λ), in order to ensure the accuracy and reliability of the data, each bacterial sample is measured multiple times. Assuming the number of measurements is n, the final spectral data of the jth bacterial sample is obtained by averaging:

[0049]

[0050] Integrate the final spectral data of all bacterial samples to form a standard bacterial spectral library Here, m is the number of bacterial sample species. These data constitute the basic reference for subsequent bacterial identification and concentration calculation, so that the micro-spectrometer has a standard to follow when comparing spectral data.

[0051] In the embodiment of the present invention, in the actual operation of the spectrum monitoring module, in order to ensure that the acquired spectrum data is accurate and reliable, the collected spectrum data is first preprocessed by using a denoising algorithm based on wavelet transform;

[0052] Assume that the original spectrum data is S λ , decompose it into different scales through discrete wavelet transform, and obtain the wavelet coefficients W of different scales j and positions k j,k In the threshold processing stage, the soft threshold function is used to process the wavelet coefficients. The threshold is set to τ. The wavelet coefficients after processing are for:

[0053]

[0054] Then the denoised spectral data S is obtained by inverse wavelet transform denoised (λ), its mathematical expression is:

[0055]

[0056] in, is the wavelet coefficient after threshold processing, ψ j,k (λ) is the wavelet basis function. This preprocessing step improves the quality of spectral data and lays the foundation for the subsequent accurate identification of bacterial contamination.

[0057] After completing the spectral data preprocessing, the micro-spectrometer uses an ultraviolet-visible light source to excite bacterial biomolecules to produce spectral signals. This process has a precise wavelength control mechanism. Through the excitation light of a specific wavelength λ0, biomolecules such as nucleic acids and proteins in the bacteria are stimulated to fluoresce. The micro-spectrometer compares the collected preprocessed spectral data with a standard bacterial spectral library. This comparison process is based on complex spectral feature analysis.

[0058] In the process of spectral feature analysis, the spectral similarity calculation method is used, assuming that the spectral data in the standard bacterial spectral library is S std (λ), the unknown bacteria spectral data collected by the micro-spectrometer after preprocessing is S denoised,unk (λ), the calculation formula of spectral similarity Sim is:

[0059]

[0060] Where d represents the calculus symbol, and dλ represents the small change when performing the integral operation on λ;

[0061] When the Sim value exceeds the preset threshold Sim th When , it can be determined that the corresponding type of bacterial contamination is detected. At this time, the bacterial concentration is calculated through the concentration-spectral intensity relationship model;

[0062] Assume that the bacterial concentration C spec It satisfies a linear relationship with the spectral intensity I(λ0) at a specific wavelength λ0, and its mathematical model is:

[0063] C spec =a·I(λ0)+b;

[0064] Where a is the proportional coefficient, which reflects the degree of influence of spectral intensity changes on bacterial concentration; b is a constant term, which can be regarded as the basic concentration value under specific experimental conditions. The values of a and b are obtained by measuring the spectra of bacterial samples with known concentrations and fitting them using the least squares method.

[0065] The goal of the least squares method is to minimize the sum of squared errors E between the measured values and the fitted straight line;

[0066]

[0067] Calculate the partial derivatives of a and b respectively, and set the partial derivatives to 0, and get the equation system:

[0068]

[0069] Solving the system of equations can obtain the optimal values of a and b, thus ensuring the accuracy of the concentration calculation model.

[0070] In an embodiment of the present invention, the electronic sensor used in the multimodal fusion module is an electrochemical sensor. This sensor detects based on the electrochemical activity of bacterial metabolites. Its working principle is that specific substances produced during bacterial metabolism undergo redox reactions on the surface of the sensor electrode, generating electrical signals related to the concentration of the metabolites. At the same time, a Wi-Fi module or a Bluetooth communication module is used as a wireless communication module to achieve remote data transmission and device control.

[0071] In the signal processing of electrochemical sensors, let the original electrical signal output by the sensor be E raw , signal calibration and concentration conversion are performed using the following formula:

[0072] C met =g·(E raw -E offset );

[0073] Among them, C met is the concentration of bacterial metabolites, g is the sensor sensitivity coefficient, E offset is the zero offset voltage, in order to accurately calibrate g and E offset , multiple measurements were performed using bacterial metabolite solutions with standard concentrations;

[0074] Assume that the concentration of the standard solution is C std , the corresponding sensor output electrical signal is E std ,but:

[0075] C std =g·(E std -E offset );

[0076] By using at least two sets of measurement data of solutions with different standard concentrations, the simultaneous equations can be solved to obtain g and E. offset The value of can be accurately measured to achieve the accurate measurement of bacterial metabolite concentrations and provide reliable data support for the multimodal fusion module.

[0077] In an embodiment of the present invention, in order to achieve data fusion between the multimodal fusion module and the spectral monitoring module, a support vector machine algorithm is used. The multimodal fusion module uses a chemical sensor to obtain data from the perspective of bacterial metabolite detection, and performs fusion analysis with the spectral data collected by the spectral monitoring module;

[0078] In the data fusion process, the feature vector extracted by the spectrum monitoring module is set as Xs =[x s1 ,x s2 ,…,x sn ], the feature vector extracted by the electrochemical sensor is X e =[x e1 ,x e2 ,…,x em ], merge the two into a new feature vector X = [X s ,X e ];

[0079] Introduced bacterial metabolite concentration C met and bacterial concentration C spec , define the pollution index P, and its calculation formula is:

[0080]

[0081] Where w1 and w2 are the bacterial concentrations C spec and bacterial metabolite concentration C met The weight coefficient, and satisfy w1+w2=1, α i is the Lagrange multiplier, y i is the sample label, K(X i ,X) is the kernel function, b1 is the bias term;

[0082] Kernel function K(X i ,X) is very important. Common kernel functions include linear kernel functions Polynomial kernel function and radial basis kernel function K(X i ,X)=exp(-γ||X i -X|| 2 ), and select the optimal kernel function and its parameters through cross-validation method.

[0083] In the embodiments of the present invention, the structure and performance of the protective cover itself are the basis for the various functional modules to function. Therefore, the protective cover is made of materials with protective performance and compatibility, and reasonable space is reserved for integrating the spectral monitoring module, the intelligent sensing and self-prompt module, and the multimodal fusion module. At the same time, the structural design of the protective cover takes into account the heat dissipation requirements of each module, and adopts a combination of heat dissipation holes and thermal conductive materials to ensure that each functional module maintains stable working performance during long-term operation.

[0084] In an embodiment of the present invention, the intelligent sensing and self-prompting module uses a risk assessment model during the threshold setting process;

[0085] Assume that the hazard coefficient of different bacteria types l is h l , the frequency of use of the pulmonary function meter is f, then the bacterial concentration threshold Cth It can be calculated by the following formula:

[0086]

[0087] Where p is the total number of bacterial types, w l is the weight coefficient of different bacteria types, and β is the safety factor, which can be adjusted according to actual needs;

[0088] The multimodal fusion module uses machine learning and data fusion technology to jointly analyze data and continuously optimize algorithm parameters and model structure to improve detection accuracy;

[0089] In the process of optimizing the detection accuracy, a method combining cross-validation and grid search is used. Let the data set be D and divide it into K subsets D1, D2, ..., D K , for each candidate model parameter combination θ, calculate its average accuracy ACC(θ) under K subset fold cross validation:

[0090]

[0091] By traversing the candidate parameter space, the parameter combination that maximizes Acc(θ) is selected as the optimal model parameter, thereby improving the detection accuracy of the system and ensuring that the entire protective cover can reliably realize intelligent sensing and self-prompting disinfection functions.

[0092] The embodiments disclosed in the present invention are preferred embodiments, but are not limited to them. Ordinary technicians in this field can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. As long as they do not deviate from the spirit of the present invention, they are all within the scope of protection of the present invention.

Claims

1. A protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions, characterized in that: The protective cover is implemented based on an intelligent sensing and self-prompting disinfection system, which includes: Spectral monitoring module, intelligent sensing and self-prompt module, and multimodal fusion module; The spectral monitoring module includes a UV-visible light source, a micro-spectrometer, and a standard bacterial spectral library. The UV-visible light source uses a specific wavelength to excite the biological molecules of nucleic acids and proteins in bacteria to produce spectral signals. The micro-spectrometer compares the collected spectral data with the standard bacterial spectral library to identify bacterial characteristics and calculate concentrations. The multimodal fusion module uses electronic sensors and performs data fusion with the spectral monitoring module to improve the accuracy of bacteria detection and calculate the pollution index P; The intelligent sensing and self-prompting module includes an audio-visual prompt device and a wireless communication module. When the pollution index P exceeds the set threshold, the audio-visual prompt device automatically emits an audio-visual prompt, and the wireless communication module remotely prompts medical staff to disinfect the pulmonary function meter.

2. A protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions according to claim 1, characterized in that: In order to achieve accurate detection of bacteria by the spectral monitoring module, the spectral monitoring module establishes a standard bacterial spectrum library and systematically collects characteristic absorption peaks, reflectance and fluorescence characteristic data of different bacteria; During the data acquisition process, the spectrometer is used in the wavelength range λ∈[λ min ,λ max ] and obtain the spectral data S for each bacterial sample j. j (λ), in order to ensure the accuracy and reliability of the data, each bacterial sample is measured multiple times. Assuming the number of measurements is n, the final spectral data of the jth bacterial sample is obtained by averaging: Integrate the final spectral data of all bacterial samples to form a standard bacterial spectral library Where m is the number of bacterial species.

3. A protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions according to claim 2, characterized in that: In the actual operation of the spectrum monitoring module, in order to ensure the accuracy and reliability of the acquired spectrum data, the collected spectrum data is first preprocessed and a denoising algorithm based on wavelet transform is used to obtain the denoised spectrum data S denoised (λ); After completing the spectral data preprocessing, the ultraviolet-visible light source is used to excite bacterial biomolecules to generate spectral signals. The excitation light with a specific wavelength λ0 is used to excite biomolecules such as nucleic acids and proteins in the bacteria to emit fluorescence. The micro-spectrometer compares the collected preprocessed spectral data with the standard bacterial spectral library. The comparison process is based on complex spectral feature analysis to obtain the spectral similarity Sim. When the Sim value exceeds the preset threshold Sim th When the corresponding type of bacterial contamination is detected, the bacterial concentration C is calculated by the concentration-spectral intensity relationship model. spec ; Assume that the bacterial concentration C spec It satisfies a linear relationship with the spectral intensity I(λ0) at a specific wavelength λ0, and its mathematical model is: C spec =a·I(λ0)+b; Among them, a is the proportional coefficient, which reflects the influence of spectral intensity change on bacterial concentration, and b is the constant term.

4. A protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions according to claim 3, characterized in that: The electronic sensor used in the multimodal fusion module is an electrochemical sensor, which detects based on the electrochemical activity of bacterial metabolites; In the signal processing of electrochemical sensors, let the original electrical signal output by the sensor be E raw , signal calibration and concentration conversion are performed using the following formula: C met =g·(E raw -E offset ); Among them, C met is the concentration of bacterial metabolites, g is the sensor sensitivity coefficient, E offset is the zero offset voltage, in order to accurately calibrate g and E offset , multiple measurements were performed using bacterial metabolite solutions with standard concentrations; Assume that the concentration of the standard solution is C std , the corresponding sensor output electrical signal is E std ,but: C std =g·(E std -E offset ); Among them, C met is the concentration of bacterial metabolites, g is the sensor sensitivity coefficient, E offset is the zero offset voltage.

5. A protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions according to claim 4, characterized in that: In order to achieve data fusion between the multimodal fusion module and the spectral monitoring module, the support vector machine algorithm is adopted. The multimodal fusion module uses chemical sensors to obtain data from the perspective of bacterial metabolite detection, and performs fusion analysis with the spectral data collected by the spectral monitoring module; In the data fusion process, the feature vector extracted by the spectrum monitoring module is set as X s =[x s1 ,x s2 ,…,x sn ], the feature vector extracted by the electrochemical sensor is X e =[x e1 ,x e2 ,…,x em ], merge the two into a new feature vector X = [X s ,X e ]; Introduced bacterial metabolite concentration C met and bacterial concentration C spec , define the pollution index P, and its calculation formula is: Where w1 and w2 are the bacterial concentrations C spec and bacterial metabolite concentration C met The weight coefficient, and satisfy w1+w2=1, α i is the Lagrange multiplier, y i is the sample label, K(X i ,X) is the kernel function and b1 is the bias term.

6. A protective cover for a pulmonary function meter with intelligent sensing and self-prompting disinfection functions according to claim 5, characterized in that: The intelligent sensing and self-prompting module uses a risk assessment model during the threshold setting process; Assume that the hazard coefficient of different bacteria types l is h l , the frequency of use of the pulmonary function meter is f, then the bacterial concentration threshold C th It can be calculated by the following formula: Where p is the total number of bacterial types, w l is the weight coefficient of different bacteria types, and β is the safety factor.

Images