Intelligent regulating system for temperature and humidity of breathing machine pipeline
By using multimodal adjustment model and image generation optimization algorithm in the ventilator, combined with expiratory physiological parameters, intelligent adjustment of the temperature and humidity of the ventilator pipeline is achieved, and the problem of humidified output in the existing technology does not meet the requirements , improves the accuracy and response speed of humidified output.
Patent Information
- Application Number
- CN202411058172.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing ventilator humidification output temperature and humidity adjustment methods rely on manual adjustment, and the operation is cumbersome and unable to respond to environmental changes in time, which may cause the humidification output to not meet the requirements and affect the health of users.
Adopt the ventilator pipeline temperature and humidity intelligent adjustment system to obtain user's facial image data and breath physiological parameters through camera equipment (such as mobile phones), use image generation optimization algorithms and machine learning algorithms to build a multimodal adjustment model, and adjust the pipeline in real time Temperature and humidity level to achieve automated control.
It reduces the cost of use, improves the accuracy and response speed of humidified output, ensures that humidified output meets the target temperature and humidity requirements, enhances the consideration of the user's facial condition, and improves the intelligent adjustment ability of the ventilator.
Smart Images

Figure CN118949204A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of intelligent medical devices, and in particular relates to an intelligent temperature and humidity adjustment system for ventilator pipelines. Background Art
[0002] The humidification output of a ventilator has temperature and humidity requirements. Humidification output that does not meet these requirements can cause harm to the user. A humidifier can ensure that the humidification output meets these requirements. Currently, most ventilators use manual humidification level settings. When the treatment environment, such as humidity, temperature, and real-time flow rate, changes, the humidification level needs to be readjusted. This is not only cumbersome, but if environmental changes are not detected in time, the humidification output may not meet the requirements for a long time, causing harm to the user.
[0003] In order to solve the above technical problems, the existing technical solution is: configure a heating circuit for the ventilator, and set a temperature sensor or a temperature and humidity sensor at the mask end (terminal) of the heating circuit. For heating circuits with temperature and humidity sensors, set fixed temperature and humidity target values. When environmental changes cause errors between the actual value and the target value, the sensor obtains the actual temperature and humidity values in real time, and automatically adjusts the humidification level of the humidifier and the heating level of the heating circuit according to the error, so that the humidification output is stable at the target temperature and humidity. For heating circuits with only temperature sensors, the relationship between the input variables, the temperature at the end of the heating circuit and the humidification output is fitted based on actual data. By controlling the temperature at the end of the heating circuit, indirect control of the humidification output is achieved.
[0004] The existing technical solution for automatic humidification and heating relies on a sensor at the end of the heating pipe. In addition, the existing technology intelligently adjusts the temperature and humidity of the ventilator based on the user's body temperature, respiratory monitoring conditions, and ambient temperature and humidity conditions. However, it does not take into account the actual facial condition of the user. Firstly, when using a ventilator, the cost of image acquisition equipment such as cameras is high. Secondly, the processing of facial images of users when using a ventilator is still underdeveloped. Therefore, the accuracy of the results obtained by combining the user's physiological parameters and image data is poor.
[0005] While accurate image recognition has been widely used in image recognition algorithms, the challenge remains to easily and cost-effectively capture these images for use in ventilator temperature and humidity control. Therefore, a system for intelligently regulating the temperature and humidity of ventilator circuits is urgently needed to address this issue. Summary of the Invention
[0006] In order to solve the above technical problems, the present invention proposes an intelligent temperature and humidity adjustment system for ventilator pipelines.
[0007] In a first aspect of the present invention, a method for intelligently adjusting the temperature and humidity of a ventilator pipeline is provided, wherein the method comprises:
[0008] First image data of the user when using the ventilator, and obtaining the user's first exhalation physiological parameter and the first ventilator pipeline temperature and humidity level in the ventilator working mode;
[0009] Performing feature processing on the first image data and the first exhalation physiological parameter to obtain a first adjustment feature;
[0010] constructing a ventilator pipeline multimodal regulation model using the first regulation feature and the first ventilator pipeline temperature and humidity level;
[0011] Acquire the second exhalation physiological parameter and second image data of the user to be adjusted in the ventilator working mode, process and generate a second adjustment feature, use the ventilator pipeline multimodal adjustment model to process the second adjustment feature, generate a second ventilator pipeline temperature and humidity level, and the ventilator controls the ventilator pipeline to adjust the temperature and humidity level according to the second ventilator pipeline temperature and humidity level.
[0012] Furthermore, the first image data or the second image data is acquired by a camera device, and the camera device is connected to the ventilator for data transmission, and the connection method is a wireless connection or a wired connection.
[0013] Furthermore, the camera device is a mobile phone, which is wirelessly connected to the ventilator. The ventilator has a ventilator pipeline data processing module, which performs feature processing on the user image data and exhalation physiological parameters collected by the mobile phone in the ventilator working mode.
[0014] Furthermore, before performing feature processing on the user image data and the exhalation physiological parameters, image optimization is performed on the user image data, and the image optimization adopts an image generation optimization algorithm.
[0015] Furthermore, the image generation optimization algorithm utilizes a generative adversarial network algorithm.
[0016] Furthermore, before using the ventilator pipeline multimodal adjustment model, the first image data or the second image data is segmented to obtain an input vector, and an image segmentation model is used to perform image segmentation, and the image obtained after image segmentation is converted into a vector value.
[0017] Furthermore, the first exhalation physiological parameter or the second exhalation physiological parameter includes the body temperature and blood oxygen saturation of the user during the exhalation period.
[0018] Furthermore, the ventilator circuit multimodal adjustment model adopts a machine learning algorithm, and the machine learning algorithm adopts a classifier based on the Fischer criterion.
[0019] A ventilator pipeline temperature and humidity intelligent adjustment system is also provided, comprising a data acquisition terminal and a ventilator pipeline multi-modal adjustment terminal, characterized in that:
[0020] The data acquisition terminal includes a ventilator image acquisition module, a ventilator parameter acquisition module and a ventilator pipeline data processing module, and the ventilator pipeline multimodal adjustment terminal includes a ventilator pipeline multimodal adjustment module and a ventilator pipeline control module, wherein:
[0021] A ventilator image acquisition module is configured to acquire first image data when a user uses the ventilator and second image data when a user to be adjusted uses the ventilator;
[0022] A ventilator parameter acquisition module is used to acquire a first expiratory physiological parameter of a user in a ventilator working mode and a second expiratory physiological parameter of a user to be adjusted in a ventilator working mode;
[0023] A ventilator circuit data processing module is configured to receive the first image data and the first expiratory physiological parameter, perform effective feature extraction to obtain a first adjustment feature, receive the second image data and the second expiratory physiological parameter, perform effective feature extraction to obtain a second adjustment feature, and receive the first ventilator circuit temperature and humidity level, and perform image optimization and image segmentation on the image data.
[0024] A ventilator pipeline multimodal adjustment module is connected to the ventilator pipeline data processing module, receives the first adjustment feature, and generates a ventilator pipeline multimodal adjustment model using the first ventilator pipeline temperature and humidity level. The ventilator pipeline multimodal adjustment model receives a second adjustment feature obtained by splicing the second image data processed by the ventilator pipeline data processing module and the second expiratory physiological parameter, and generates a second ventilator management temperature and humidity level using the second adjustment feature.
[0025] Ventilator pipeline control module: uses the second ventilator management temperature and humidity level to control the pipeline humidity, pipeline temperature and gas flow of the ventilator pipeline.
[0026] Furthermore, before performing feature processing on the user image data and the exhalation physiological parameters, image optimization is performed on the user image data, and the image optimization adopts an image generation optimization algorithm;
[0027] The image generation optimization algorithm utilizes a generative adversarial network algorithm.
[0028] The present invention takes into account the real facial condition of the user and uses a mobile phone, a low-cost camera device, to reduce the cost of use. It is highly developed for the facial image processing of the user when using a ventilator. The original image and non-original image captured by the mobile phone are used to generate a clearer facial image of the user using the ventilator by using an improved generative adversarial network method. The image segmentation method is used to obtain only facial images that can show the user's micro-expressions. Combined with the user's physiological parameters, more accurate adjustment of the ventilator pipeline control parameters can be obtained.
[0029] More embodiments and improved effects of the present invention will be further introduced in conjunction with the drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 This is a flow chart of a method for intelligently adjusting temperature and humidity in a ventilator pipeline according to the present invention;
[0031] Figure 2 This is a schematic structural diagram of a ventilator pipeline temperature and humidity intelligent adjustment system of the present invention;
[0032] Figure 3 is a schematic diagram of a facial image extracted when a user of the present invention uses a ventilator;
[0033] Figure 4 This is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0034] The invention is further described below with reference to the accompanying drawings and specific implementation methods.
[0035] In order to solve the above technical problems, the present invention proposes an intelligent temperature and humidity adjustment system for ventilator pipelines.
[0036] In a first aspect of the present invention, a method for intelligently adjusting the temperature and humidity of a ventilator pipeline is provided, wherein the method comprises:
[0037] First image data of the user when using the ventilator, and obtaining the user's first exhalation physiological parameter and the first ventilator pipeline temperature and humidity level in the ventilator working mode;
[0038] Performing feature processing on the first image data and the first exhalation physiological parameter to obtain a first adjustment feature;
[0039] constructing a ventilator pipeline multimodal regulation model using the first regulation feature and the first ventilator pipeline temperature and humidity level;
[0040] Acquire the second exhalation physiological parameter and second image data of the user to be adjusted in the ventilator working mode, process and generate a second adjustment feature, use the ventilator pipeline multimodal adjustment model to process the second adjustment feature, generate a second ventilator pipeline temperature and humidity level, and the ventilator controls the ventilator pipeline to adjust the temperature and humidity level according to the second ventilator pipeline temperature and humidity level.
[0041] Furthermore, the first image data or the second image data is acquired by a camera device, and the camera device is connected to the ventilator for data transmission, and the connection method is a wireless connection or a wired connection.
[0042] Furthermore, the camera device is a mobile phone, which is wirelessly connected to the ventilator. The ventilator has a ventilator pipeline data processing module, which performs feature processing on the user image data and exhalation physiological parameters collected by the mobile phone in the ventilator working mode.
[0043] In this embodiment, the mobile phone can be mounted on the ventilator using a telescopic rod and pointed downward to take a picture of the user's face to obtain image data.
[0044] Furthermore, before performing feature processing on the user image data and the exhalation physiological parameters, image optimization is performed on the user image data, and the image optimization adopts an image generation optimization algorithm.
[0045] Furthermore, the image generation optimization algorithm uses a generative adversarial network algorithm, and its loss function is:
[0046]
[0047] Where K and G are the width and height of the image, respectively; x and y are the coordinates of the corresponding pixel positions; M is the original image taken by the mobile phone; S is the non-original image taken by the mobile phone; M ∈ (S) is the image generated by the generator, ∈ is the model parameter of the generator, is the weight and bias value of the adversarial network algorithm model, and the optimized image data is obtained using the image optimization algorithm based on the loss function.
[0048] Furthermore, before using the ventilator circuit multimodal adjustment model, the first image data or the second image data is segmented to obtain an input vector, and an image segmentation model is used to perform image segmentation, and the image obtained after the image segmentation is converted into a vector value, wherein the image segmentation model achieves image segmentation by minimizing the following energy functional F:
[0049]
[0050] Wherein, I(X) is the image of the first image data or the second image data to be segmented, C is the evolution curve, μ, v, λ1, and λ2 are all non-negative weight coefficients, c1 and c2 are the average grayscale values of image I inside and outside the contour, respectively, insi de (C) is the target area, outsi de (C) is the background area, length is the length constraint, and area is the area constraint. The specific expressions of the average grayscale values c1 and c2 of image I inside and outside the contour are as follows:
[0051]
[0052] Insi de (C) and outside (C) are the segmented region images. Insi de (C) and outside (C) are binarized again and converted into vector representations, where insi de (C) is the image data containing only the user's facial image.
[0053] In this embodiment, the segmented insi de (C) is as follows: Figure 3 Internal facial image wrapped with median lines.
[0054] Furthermore, the first exhalation physiological parameter or the second exhalation physiological parameter includes the body temperature and blood oxygen saturation of the user during the exhalation period.
[0055] Furthermore, the multimodal regulation model of the ventilator circuit adopts a machine learning algorithm, and the machine learning algorithm adopts a classifier based on the Fischer criterion, and the calculation formula is as follows:
[0056]
[0057] A is the regulation characteristic, W(S) is the temperature and humidity level of the output ventilator pipeline, W T is the normal vector perpendicular to the hyperplane, S I is the pixel value of the image data containing only the user’s facial image, S O The pixel value of image data other than the user's facial image.
[0058] A ventilator pipeline temperature and humidity intelligent adjustment system is also provided, comprising a data acquisition terminal and a ventilator pipeline multi-modal adjustment terminal, characterized in that:
[0059] The data acquisition terminal includes a ventilator image acquisition module, a ventilator parameter acquisition module and a ventilator pipeline data processing module, and the ventilator pipeline multimodal adjustment terminal includes a ventilator pipeline multimodal adjustment module and a ventilator pipeline control module, wherein:
[0060] A ventilator image acquisition module is configured to acquire first image data when a user uses the ventilator and second image data when a user to be adjusted uses the ventilator;
[0061] A ventilator parameter acquisition module is used to acquire a first expiratory physiological parameter of a user in a ventilator working mode and a second expiratory physiological parameter of a user to be adjusted in a ventilator working mode;
[0062] A ventilator circuit data processing module is configured to receive the first image data and the first expiratory physiological parameter, perform effective feature extraction to obtain a first adjustment feature, receive the second image data and the second expiratory physiological parameter, perform effective feature extraction to obtain a second adjustment feature, and receive the first ventilator circuit temperature and humidity level, and perform image optimization and image segmentation on the image data.
[0063] A ventilator pipeline multimodal adjustment module is connected to the ventilator pipeline data processing module, receives the first adjustment feature, and generates a ventilator pipeline multimodal adjustment model using the first ventilator pipeline temperature and humidity level. The ventilator pipeline multimodal adjustment model receives a second adjustment feature obtained by splicing the second image data processed by the ventilator pipeline data processing module and the second expiratory physiological parameter, and generates a second ventilator management temperature and humidity level using the second adjustment feature.
[0064] Ventilator pipeline control module: uses the second ventilator management temperature and humidity level to control the pipeline humidity, pipeline temperature and gas flow of the ventilator pipeline.
[0065] Furthermore, before performing feature processing on the user image data and the exhalation physiological parameters, image optimization is performed on the user image data, and the image optimization adopts an image generation optimization algorithm;
[0066] The image generation optimization algorithm uses the generative adversarial network algorithm, and its loss function is:
[0067]
[0068] Where K and G are the width and height of the image, respectively; x and y are the coordinates of the corresponding pixel positions; M is the original image taken by the mobile phone; S is the non-original image taken by the mobile phone; M ∈ (S) is the image generated by the generator, ∈ is the model parameter of the generator, is the weight and bias value of the adversarial network algorithm model, and the optimized image data is obtained using the image optimization algorithm based on the loss function.
[0069] The present invention takes into account the real facial condition of the user and uses a mobile phone, a low-cost camera device, to reduce the cost of use. It is highly developed for the facial image processing of the user when using a ventilator. The original image and non-original image captured by the mobile phone are used to generate a clearer facial image of the user using the ventilator by using an improved generative adversarial network method. The image segmentation method is used to obtain only facial images that can show the user's micro-expressions. Combined with the user's physiological parameters, more accurate adjustment of the ventilator pipeline control parameters can be obtained.
[0070] Of course, it can be understood that each embodiment of the present invention can achieve one of the effects alone, and a combination of multiple embodiments of the present invention can achieve all of the above effects, but it is not required that every embodiment of the present invention achieve all of the above advantages and effects, because each embodiment of the present invention can constitute a separate technical solution and make one or more contributions to the existing technology.
[0071] For any module structure not specifically defined in this invention, the prior art shall prevail. The prior art mentioned in the aforementioned background and specific embodiments of this invention may be considered as part of this invention and used to understand the meaning of certain technical features or parameters. The scope of protection of this invention shall be based on the actual content of the claims.
Claims
1. A method for intelligently adjusting temperature and humidity of a ventilator pipeline, characterized in that: The method comprises: The first image data when the user uses the ventilator, and the first exhalation physiological parameter of the user and the first ventilator pipeline temperature and humidity level in the ventilator working mode are obtained; Performing feature processing on the first image data and the first exhalation physiological parameter to obtain a first adjustment feature; Constructing a multimodal regulation model of a ventilator pipeline using the first regulation feature and the first ventilator pipeline temperature and humidity level; The second exhalation physiological parameter and the second image data of the user to be adjusted in the ventilator working mode are obtained, and the second adjustment feature is generated by processing, and the second adjustment feature is processed by the ventilator pipeline multimodal adjustment model to generate the second ventilator pipeline temperature and humidity level. The ventilator controls the ventilator pipeline to adjust the temperature and humidity level according to the second ventilator pipeline temperature and humidity level.
2. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 1, characterized in that: The first image data or the second image data is acquired by a camera device, and the camera device is connected to the ventilator for data transmission, and the connection method is a wireless connection or a wired connection.
3. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 2, characterized in that: The camera device is a mobile phone, which is wirelessly connected to the ventilator. The ventilator is equipped with a ventilator pipeline data processing module, which performs feature processing on user image data and exhalation physiological parameters collected by the mobile phone in the ventilator working mode.
4. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 3, characterized in that: Before performing feature processing on the user image data and the exhalation physiological parameters, image optimization is performed on the user image data, and the image optimization adopts an image generation optimization algorithm.
5. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 4, characterized in that: The image generation optimization algorithm utilizes the generation adversarial network algorithm to generate weights and bias values of the adversarial network algorithm model, and utilizes the image optimization algorithm based on the loss function to obtain optimized image data.
6. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 2, characterized in that: Before using the ventilator pipeline multimodal adjustment model, the first image data or the second image data is segmented to obtain an input vector, and an image segmentation model is used to perform image segmentation, and the image obtained after image segmentation is converted into a vector value.
7. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 1, characterized in that: The first exhalation physiological parameter or the second exhalation physiological parameter includes the body temperature and blood oxygen saturation of the user during the exhalation period.
8. A method for intelligently adjusting temperature and humidity of a ventilator pipeline as claimed in claim 1, characterized in that: The ventilator circuit multimodal adjustment model adopts a machine learning algorithm, and the machine learning algorithm adopts a classifier based on the Fisher criterion.
9. A ventilator pipeline temperature and humidity intelligent adjustment system, comprising a data acquisition terminal and a ventilator pipeline multi-modal adjustment terminal, characterized in that: The data acquisition terminal includes a ventilator image acquisition module, a ventilator parameter acquisition module and a ventilator pipeline data processing module, and the ventilator pipeline multimodal adjustment terminal includes a ventilator pipeline multimodal adjustment module and a ventilator pipeline control module, wherein: A ventilator image acquisition module: acquiring first image data when a user uses a ventilator and second image data when a user to be adjusted uses a ventilator; A ventilator parameter acquisition module is used to acquire a first exhalation physiological parameter of a user in a ventilator working mode and a second exhalation physiological parameter of a user to be adjusted in a ventilator working mode; A ventilator pipeline data processing module: receiving the first image data and the first exhalation physiological parameter to extract effective features to obtain a first adjustment feature, receiving the second image data and the second exhalation physiological parameter to extract effective features to obtain a second adjustment feature, and receiving the first ventilator pipeline temperature and humidity level, and performing image optimization and image segmentation on the image data; A ventilator pipeline multimodal adjustment module: connected to the ventilator pipeline data processing module, receiving the first adjustment feature, and using the first ventilator pipeline temperature and humidity level to generate a ventilator pipeline multimodal adjustment model, the ventilator pipeline multimodal adjustment model receives the second image data processed by the ventilator pipeline data processing module and the second expiratory physiological parameter to obtain a second adjustment feature, and uses the second adjustment feature to generate a second ventilator management temperature and humidity level; Ventilator pipeline control module: utilizes the second ventilator to manage temperature and humidity levels to control the pipeline humidity, pipeline temperature and gas flow of the ventilator pipeline.
10. A ventilator pipeline temperature and humidity intelligent adjustment system as claimed in claim 9, characterized in that: Before performing feature processing on the user image data and the exhalation physiological parameters, performing image optimization on the user image data, wherein the image optimization adopts an image generation optimization algorithm; The image generation optimization algorithm utilizes a generative adversarial network algorithm.
Citation Information
Patent Citations
Image denoising method based on generative adversarial network
CN110473154A
FIB-SEM super-resolution algorithm based on generative adversarial network
CN114677281A
Intelligent breathing machine
CN115068759A
Intelligent adjusting method and system for temperature and humidity of breathing machine pipeline and medium
CN116570808A
Image amplification method, electronic equipment and storage medium
CN116824161A
Cited By
Trachea cannula temperature and humidity monitoring control system
CN119868745A
