Intelligent high-end equipment remote operation and maintenance management system and method based on cloud computing
By using cloud computing and deep learning technologies to analyze patient care videos and vital sign data in the medical field, the problem that existing intelligent high-end equipment monitoring methods cannot detect abnormal behaviors in a timely manner is solved, and the ability to detect abnormal behaviors in a timely manner is realized. Improves medical care efficiency.
Patent Information
- Application Number
- CN202410552401.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-07
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the medical field, the existing intelligent high-end equipment monitoring methods have blind spots, and the abnormal behavior of patients cannot be detected in time, resulting in the missed optimal treatment opportunity.
Using artificial intelligence technology based on cloud computing and deep learning, the patient's care monitoring video and vital sign data are extracted and encoded, and a comprehensive analysis is used to determine whether the patient's behavior is abnormal.
It realizes comprehensive monitoring of patient behavior, promptly detect abnormal behaviors, avoid missing treatment opportunities, and improves the quality and efficiency of medical care.
Smart Images

Figure CN118919047A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent management, and more specifically to a cloud computing-based intelligent high-end equipment remote operation and maintenance management system and method. Background Art
[0002] With the development of the Internet of Things and computer cloud computing technology, intelligent equipment plays an important role in intelligent manufacturing, covering a variety of equipment types and widely used in various industrial scenarios. Intelligent high-end equipment usually has complex structures and highly integrated characteristics. Establishing an efficient and reliable remote operation and maintenance control system is a key means to ensure the normal operation of equipment and reduce management costs. However, in the medical field, intelligent high-end equipment for nursing is still uncommon. It mainly monitors the room through cameras. Nurses regularly check the situation in the room and use nursing staff and surveillance cameras to determine whether the patient has abnormal behavior. However, this method has some blind spots and may not detect the patient's abnormal behavior in time, resulting in missing the best time for treatment.
[0003] Therefore, we look forward to an intelligent high-end equipment remote operation and maintenance management system. Summary of the invention
[0004] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides a cloud computing-based intelligent high-end equipment remote operation and maintenance management system and method, which uses cloud computing technology and artificial intelligence technology based on the field of deep learning to extract and encode features of patient care monitoring videos within a predetermined time period and patient vital signs data at multiple predetermined time points within the predetermined time period, so as to obtain a classification result of whether the patient's behavior is abnormal. In this way, by comprehensively monitoring the patient's behavior, the patient's abnormal behavior can be discovered in time to avoid missing the best time for treatment.
[0005] According to one aspect of the present application, a remote operation and maintenance management method for intelligent high-end equipment based on cloud computing is provided, which includes:
[0006] Acquire patient care monitoring video within a predetermined time period and patient vital sign data at multiple predetermined time points within the predetermined time period;
[0007] Analyzing the patient care monitoring video within the predetermined time period to obtain a care-related feature vector;
[0008] Extracting and encoding the patient's vital sign data to obtain a patient behavior analysis feature vector;
[0009] The nursing-related feature vector and the patient's vital sign feature vector are comprehensively analyzed to obtain a classification result of whether the patient's behavior is abnormal.
[0010] According to another aspect of the present application, there is also provided a cloud computing-based intelligent high-end equipment remote operation and maintenance management system, which includes:
[0011] A patient behavior data acquisition module, used to acquire patient care monitoring videos within a predetermined time period and patient vital sign data at multiple predetermined time points within the predetermined time period;
[0012] A nursing monitoring feature extraction module is used to analyze the patient nursing monitoring video within the predetermined time period to obtain a nursing related feature vector;
[0013] A patient vital sign data analysis module, used for extracting and encoding features of the patient vital sign data to obtain a patient behavior analysis feature vector;
[0014] The patient behavior analysis module is used to comprehensively analyze the care-related feature vector and the patient's vital sign feature vector to obtain a classification result of whether the patient's behavior is abnormal.
[0015] Compared with the prior art, the cloud computing-based intelligent high-end equipment remote operation and maintenance management system and method provided by the present application uses cloud computing technology and artificial intelligence technology based on the field of deep learning to extract and encode features of patient care monitoring videos within a predetermined time period and patient vital sign data at multiple predetermined time points within the predetermined time period, so as to obtain a classification result of whether the patient's behavior is abnormal. In this way, by comprehensively monitoring the patient's behavior, the patient's abnormal behavior can be discovered in time to avoid missing the best time for treatment. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0017] Figure 1 The present invention is a flowchart of a remote operation and maintenance management method for intelligent high-end equipment based on cloud computing according to an embodiment of the present application.
[0018] Figure 2 The present invention provides a flowchart for analyzing the patient care monitoring video within the predetermined time period in the remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to an embodiment of the present application to obtain a care-related feature vector.
[0019] Figure 3The present invention is a flowchart of linearly embedding the multiple patient regions of interest and the multiple equipment regions of interest in the remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to an embodiment of the present application to obtain multiple patient feature vectors of interest and multiple equipment feature vectors of interest.
[0020] Figure 4 It is a block diagram of a cloud computing-based intelligent high-end equipment remote operation and maintenance management system according to an embodiment of the present application.
[0021] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.
[0023] Figure 1 FIG. 1 is a flow chart of a remote operation and maintenance management method for intelligent high-end equipment based on cloud computing according to an embodiment of the present application. Figure 1 As shown, according to the cloud computing-based remote operation and maintenance management method for intelligent high-end equipment according to the embodiment of the present application, the method includes: S110, acquiring the patient care monitoring video within a predetermined time period and the patient's vital signs data at multiple predetermined time points within the predetermined time period; S120, analyzing the patient care monitoring video within the predetermined time period to obtain a care-related feature vector; S130, performing feature extraction and encoding on the patient's vital signs data to obtain a patient behavior analysis feature vector; and S140, comprehensively analyzing the care-related feature vector and the patient's vital signs feature vector to obtain a classification result of whether the patient's behavior is abnormal.
[0024] In step S110, the patient care monitoring video within the predetermined time period and the patient's vital signs data at multiple predetermined time points within the predetermined time period are obtained. As mentioned in the above background technology, in the medical field, intelligent high-end equipment for care is still uncommon. It is mainly to monitor the room through cameras, and the nursing staff regularly checks the situation in the room. The nursing staff and monitoring cameras are used to determine whether the patient has abnormal behavior. However, this method has some blind spots, and the patient's abnormal behavior may not be discovered in time, resulting in missing the best treatment opportunity. Therefore, a remote operation and maintenance management system for intelligent high-end equipment is expected.
[0025] In response to the above technical problems, a cloud computing-based intelligent high-end equipment remote operation and maintenance management system and method is proposed, which uses cloud computing technology and artificial intelligence technology based on deep learning to extract and encode features of patient care monitoring videos within a predetermined time period and patient vital signs data at multiple predetermined time points within the predetermined time period, so as to obtain a classification result of whether the patient's behavior is abnormal. In this way, by comprehensively monitoring the patient's behavior, the patient's abnormal behavior can be discovered in time to avoid missing the best time for treatment.
[0026] Specifically, in the medical field, a robot intelligent medical system can be built. The robot can patrol the ward and use sensors to monitor the patient's vital signs, including heart rate, respiration, blood pressure, and body temperature. By uploading this data to the cloud in real time, medical staff can remotely monitor the patient's health status.
[0027] More specifically, first, obtain the patient care monitoring video within a predetermined time period and the patient's vital signs data at multiple predetermined time points within the predetermined time period. Install a surveillance camera or monitoring equipment in the patient's room or ward, and transmit the video data to the cloud server through the network to obtain the patient care monitoring video. Alternatively, a special video acquisition device or system can be used to obtain the patient's monitoring video data regularly or in real time. At the same time, use vital signs monitoring equipment to monitor the patient's vital signs data in real time, such as a heart rate monitor, a blood pressure monitor, a respiratory monitor, etc. Connecting these monitoring devices to the cloud server can realize the automatic collection and transmission of vital signs data. By obtaining monitoring videos and vital signs data, combined with relevant data processing and analysis technologies, medical staff can fully understand the patient's condition, and can also promptly detect abnormal conditions and make corresponding treatments, thereby improving the quality and efficiency of medical care.
[0028] In step S120, the patient care monitoring video within the predetermined time period is analyzed to obtain a care-related feature vector.
[0029] Figure 2 The present invention is a flowchart of analyzing the patient care monitoring video within the predetermined time period to obtain the care-related feature vector in the remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to the embodiment of the present application. Figure 2As shown, analyzing the patient care monitoring video within the predetermined time period to obtain the care-related feature vector includes: S121, extracting multiple patient care monitoring key frames from the patient care monitoring video; S122, performing target detection on the multiple patient care monitoring key frames to obtain multiple patient regions of interest and multiple equipment regions of interest; S123, performing linear embedding coding on the multiple patient regions of interest and the multiple equipment regions of interest to obtain multiple patient interest feature vectors and multiple equipment interest feature vectors; S124, associating coding the multiple patient interest feature vectors and the multiple equipment interest feature vectors to obtain multiple collaborative care feature matrices; and S125, obtaining the care-related feature vector by using a care monitoring feature extraction module with a three-dimensional convolution kernel.
[0030] More specifically, multiple patient care monitoring key frames are extracted from the patient care monitoring video. The patient care monitoring video is usually a continuous video stream. Extracting key frames can reduce the amount of data, save storage space and transmission costs. The key frame contains important information and key moments in the video. By extracting multiple key frames from the patient care monitoring video, the data processing process can be effectively optimized and the efficiency and accuracy of information extraction can be improved.
[0031] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, multiple patient care monitoring key frames are extracted from the patient care monitoring video, including: extracting multiple patient care monitoring key frames from the patient care monitoring video at a predetermined sampling frequency.
[0032] More specifically, target detection is performed on the multiple patient care monitoring key frames to obtain multiple patient regions of interest and multiple equipment regions of interest, that is, each of the multiple patient care monitoring key frames is passed through the patient target detection network and the room equipment target detection network to obtain the multiple patient regions of interest and the multiple equipment regions of interest. Through the patient target detection network, the position and dynamics of the patient in the monitoring screen can be accurately identified. Through the equipment target detection network, it is possible to timely detect whether there are abnormal conditions in the equipment in the room, such as equipment movement, missing or damage, which helps to ensure the safety of patients and the normal use of medical equipment. Identifying and extracting regions of interest for patients and equipment respectively helps to achieve refined management and monitoring of patients and equipment, and improve the quality and efficiency of medical care. Through the automated processing of the target detection network, the efficiency and accuracy of data processing can be improved.
[0033] Specifically, in the above-mentioned cloud computing-based remote operation and maintenance management method for intelligent high-end equipment, target detection is performed on the multiple patient care monitoring key frames to obtain multiple patient regions of interest and multiple equipment regions of interest, including: each patient care monitoring key frame in the multiple patient care monitoring key frames is passed through a patient target detection network and a room equipment target detection network to obtain the multiple patient regions of interest and the multiple equipment regions of interest.
[0034] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the patient target detection network and the room equipment target detection network are target detection networks based on anchor windows. In a specific example of the present application, the target detection network based on anchor windows is Fast R-CNN, Faster R-CNN or RetinaNet.
[0035] Specifically, in the above-mentioned remote operation and maintenance management method of intelligent high-end equipment based on cloud computing, each of the multiple patient care monitoring key frames is respectively passed through the patient target detection network and the room equipment target detection network to obtain the multiple patient interest regions and the multiple equipment interest regions, including: using the patient target detection network to process the patient care monitoring key frame with the following target detection network formula to obtain the region of interest portion corresponding to the patient; wherein the target detection network formula is:
[0036] ROI=(cls(ψ det ,B),Regr(ψ det ,B))
[0037] Among them, ψ det is the patient care monitoring key frame, B is the anchor frame, ROI is the region of interest corresponding to the patient, cls(ψ det ,B) indicates classification, Regr(ψ det ,B) indicates regression.
[0038] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the multiple patient regions of interest and the multiple equipment regions of interest are linearly embedded and encoded to obtain multiple patient feature vectors of interest and multiple equipment feature vectors of interest, including: S11, passing each of the multiple patient regions of interest through a patient linear embedding layer to obtain the multiple patient feature vectors of interest; and S12, passing the multiple equipment regions of interest through an equipment linear embedding layer to obtain the multiple equipment feature vectors of interest.
[0039] More specifically, multiple patient regions of interest are passed through the patient linear embedding layer to obtain patient feature vectors of interest. The patient linear embedding layer can help extract the key features of the patient region of interest and convert visual information into feature vectors with semantic information, which is helpful for subsequent feature representation and analysis. At the same time, the feature vector obtained by the patient linear embedding layer can reduce the original high-dimensional data to a lower-dimensional feature space, reduce the complexity of the data, and improve the efficiency of data processing.
[0040] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the patient linear embedding layer uses a learnable embedding matrix to fully connect encode each of the multiple patient regions of interest to obtain the multiple patient feature vectors of interest.
[0041] At the same time, multiple equipment regions of interest are passed through the equipment linear embedding layer to obtain equipment feature vectors of interest. The equipment linear embedding layer can help extract key features of equipment regions of interest, convert visual information into feature vectors with semantic information, and help to represent and analyze equipment features. The equipment feature vectors of interest obtained through the equipment linear embedding layer can help extract key features, reduce data complexity, and realize equipment feature expression and similarity comparison.
[0042] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the equipment linear embedding layer uses a learnable embedding matrix to fully connect encode each of the multiple equipment regions of interest to obtain the multiple equipment feature vectors of interest.
[0043] More specifically, the process of associating and encoding multiple patient feature vectors of interest and multiple equipment feature vectors of interest to obtain multiple collaborative care feature matrices can help achieve the association and collaboration between patients and equipment, and provide more comprehensive information and support for nursing work. By fusing the patient feature vector of interest and the equipment feature vector of interest, a comprehensive feature representation can be obtained, which contains the key information of patients and equipment. Through a certain encoding method (such as linear encoding, nonlinear encoding, etc.), the fused feature vector is converted into a new feature space to better capture the association information between patients and equipment. The encoded feature vector is organized into a matrix form, in which each row represents a patient's feature vector and each column represents a feature vector of an equipment, thereby forming a collaborative care feature matrix. By analyzing and processing this feature matrix, the collaborative features between patients and equipment can be extracted, and the association pattern and interaction between them can be revealed.
[0044] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the multiple patient interest feature vectors and the multiple equipment interest feature vectors are associated and encoded to obtain multiple collaborative care feature matrices, including: the multiple patient interest feature vectors and the multiple equipment interest feature vectors are associated and encoded using the following association coding formula to obtain multiple collaborative care feature matrices; wherein the association coding formula is:
[0045]
[0046] in, represents vector multiplication, M represents the classification feature matrix, F1 represents the plurality of patient feature vectors of interest, F2 represents the plurality of equipment feature vectors of interest, represents the transpose of the plurality of equipment feature-of-interest vectors.
[0047] More specifically, multiple collaborative care feature matrices are extracted by a care monitoring feature extraction module using a three-dimensional convolution kernel to obtain a care-related feature vector. The care monitoring feature extraction module is a convolutional neural network model using a three-dimensional convolution kernel. Convolutional neural networks (CNNs) perform well in processing images and sequence data. By using a three-dimensional convolution kernel, spatial and temporal information in the collaborative care feature matrix can be effectively extracted to help capture important features of care association. Convolutional neural networks can learn the spatial relationship between features. By applying a three-dimensional convolution kernel to the collaborative care feature matrix, the association information between patients and equipment can be effectively modeled, and the association features between patients and equipment can be extracted. At the same time, convolutional neural networks can learn features at different levels. Through multi-layer convolution and pooling operations, the information in multiple collaborative care feature matrices can be fused together to obtain a more representative care-related feature vector. Through the layer-by-layer abstraction of the convolutional neural network, the abstract features in the collaborative care feature matrix can be gradually extracted to help discover the potential association pattern between patients and equipment, and provide more powerful support for subsequent decision-making and analysis.
[0048] Specifically, in the above-mentioned remote operation and maintenance management method of intelligent high-end equipment based on cloud computing, the multiple collaborative care feature matrices are obtained by using a care monitoring feature extraction module using a three-dimensional convolution kernel to obtain a care-related feature vector, including: using each layer of the care monitoring feature extraction module using a three-dimensional convolution kernel to perform the following on the input data in the forward pass of the layer: convolution processing on the input data to obtain a convolution feature map; pooling the local feature matrix along the channel dimension of the convolution feature map to obtain a pooled feature map; nonlinear activation on the pooled feature map to obtain an activation feature map; wherein the output of the last layer of the care monitoring feature extraction module using a three-dimensional convolution kernel is the care-related feature vector, and the input of the first layer of the care monitoring feature extraction module using a three-dimensional convolution kernel is the multiple collaborative care feature matrices.
[0049] In step S130, feature extraction and encoding are performed on the patient's vital sign data to obtain a patient behavior analysis feature vector.
[0050] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the patient's vital signs data are feature extracted and encoded to obtain a patient behavior analysis feature vector, including: passing the patient's vital signs data through a patient vital signs timing encoder module to obtain a patient's vital signs feature vector.
[0051] More specifically, the patient's vital signs data is passed through the patient's vital signs time series encoder module to obtain the patient's vital signs feature vector. The patient's vital signs data usually contains time series information, such as heart rate, blood pressure and other data that change over time. Through the time series encoder module, these time series information can be effectively retained and utilized to help capture the dynamic changes and trends in the patient's vital signs data. The time series encoder module can effectively extract key features in the patient's vital signs data, such as periodic changes, emergencies, etc., so as to obtain a more representative patient vital signs feature vector, which helps medical staff better understand the patient's health status. Patient vital signs data are often multidimensional and contain multiple indicators and parameters. Through the time series encoder module, these multidimensional features can be effectively integrated and encoded to obtain a comprehensive patient vital signs feature vector, which more comprehensively reflects the patient's physiological state.
[0052] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the patient's vital signs data is feature extracted and encoded to obtain a patient behavior analysis feature vector, including: arranging the patient's vital signs data into multiple sign input vectors according to the time dimension; normalizing the multiple sign input vectors based on the maximum value to obtain multiple normalized sign input vectors; using the fully connected layer of the patient's vital signs timing encoder module to fully connect the multiple normalized vital signs input vectors to extract high-dimensional implicit features of the eigenvalues at each position in the multiple normalized vital signs input vectors; and, using the one-dimensional convolutional layer of the patient's vital signs timing encoder module to one-dimensionally encode the multiple normalized vital signs input vectors to extract high-dimensional implicit correlation features of the correlation between the eigenvalues at each position in the multiple normalized vital signs input vectors.
[0053] In step S140, the nursing-related feature vector and the patient's vital sign feature vector are comprehensively analyzed to obtain a classification result of whether the patient's behavior is abnormal.
[0054] Specifically, in the above-mentioned cloud computing-based remote operation and maintenance management method for intelligent high-end equipment, the nursing-related feature vector and the patient's vital signs feature vector are comprehensively analyzed to obtain a classification result of whether the patient's behavior is abnormal, including: performing sparsity constraints on the nursing-related feature vector and the patient's vital signs feature vector based on implicit feature expression to obtain a patient behavior analysis feature vector; and, passing the patient behavior analysis feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the patient's behavior is abnormal.
[0055] In particular, in the technical solution of the present application, the care-related feature vector and the patient's vital signs feature vector come from different data sources and information types: the former is mainly extracted from surveillance videos based on visual data, while the latter is based on physiological data obtained by physiological monitoring equipment. There are fundamental differences between these two feature sources, and the information angles and contents they represent are different. In the technical solution of the present application, the care-related feature vector and the patient's vital signs feature vector are obtained by extracting key frames, target detection, linear embedding, etc. These steps may introduce repeated information or redundant features in the features. Even in the final fusion stage, if these repeated information are not well removed or processed, the final feature vector may contain a large number of similar or repeated features, which may make it difficult for the model to accurately learn effective feature representations, resulting in irrelevant or redundant features in the final patient behavior analysis feature vector. In order to solve this problem, in the technical solution of the present application, the care-related feature vector and the patient's vital signs feature vector are subjected to sparsity constraints based on implicit feature expression to obtain the patient behavior analysis feature vector.
[0056] Specifically, in the above-mentioned remote operation and maintenance management method of intelligent high-end equipment based on cloud computing, the nursing-related feature vector and the patient's vital sign feature vector are subjected to sparsity constraints based on implicit feature expression to obtain the patient behavior analysis feature vector, including: calculating the first autocorrelation sample covariance matrix of the nursing-related feature vector, wherein the eigenvalue of each position in the first autocorrelation sample covariance matrix is the variance between the eigenvalues of the corresponding two positions in the nursing-related feature vector; calculating the second autocorrelation sample covariance matrix of the patient's vital sign feature vector, wherein the eigenvalue of each position in the second autocorrelation sample covariance matrix is the variance between the eigenvalues of the corresponding two positions in the patient's vital sign feature vector; using principal component analysis to analyze the first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix. This covariance matrix is processed to obtain the feature implicit expression of the nursing-associated feature vector and the feature implicit expression of the patient's vital signs feature vector; L1 regularization is used to calculate the feature implicit constraint matrix between the feature implicit expression of the nursing-associated feature vector and the feature implicit expression of the patient's vital signs feature vector; the nursing-associated feature vector and the patient's vital signs feature vector are respectively multiplied by the feature implicit constraint matrix to map the nursing-associated feature vector and the patient's vital signs feature vector into the feature space of the feature implicit constraint matrix to obtain the mapped nursing-associated feature vector and the mapped patient's vital signs feature vector; and, the position-weighted sum between the mapped nursing-associated feature vector and the mapped patient's vital signs feature vector is calculated to obtain the patient behavior analysis feature vector.
[0057] Here, the care-related feature vector and the patient's vital sign feature vector are constrained by sparsity based on implicit feature expression, which helps to eliminate irrelevant or redundant features by retaining the most important features, thereby improving feature selectivity and reducing the complexity of the model and the risk of overfitting. In addition, the sparsity constraint based on implicit feature expression can make the fused feature training less sensitive to noise and outliers, thereby improving the robustness of the patient behavior analysis feature vector. At the same time, the patient behavior analysis feature vector after the sparsity constraint based on implicit feature expression contains only a small number of non-zero elements, making it easier to interpret, and also helps to understand the behavior of the model and identify the most important features. And because it only contains a small number of non-zero elements, it requires fewer resources in terms of storage and processing, which can significantly improve computational efficiency.
[0058] Specifically, in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management method, the first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix are processed using the principal component analysis method to obtain the characteristic implicit expression of the nursing-related feature vector and the characteristic implicit expression of the patient's vital sign feature vector, including: creating a Spring Web application; creating a data model class, the data model class is used to represent the first autocorrelation sample covariance matrix, the second autocorrelation sample covariance matrix, the characteristic implicit expression of the nursing-related feature vector and the characteristic implicit expression of the patient's vital sign feature vector; writing the principal component analysis method in the Spring Web application; creating a REST API endpoint, the REST API endpoint is used to allow the client to send the first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix; and deploying the Spring Web application on a Web server.
[0059] In this way, by using the Spring Web mechanism to deploy the sparsity constraint feature fusion algorithm based on implicit feature expression, the following can be achieved: 1. RESTful API: Spring Web provides a convenient framework to create RESTful APIs, allowing clients to interact with applications through HTTP requests. This makes it easy to deploy algorithms and clients can easily access it from various platforms and devices. 2. Lightweight and extensible: Spring Web is a lightweight framework that does not have a significant impact on the performance of the application. It is also highly extensible and can handle applications from small to large. 3. Security: Spring Web provides built-in security features such as authentication and authorization to protect your application from unauthorized access. 4. Easy to deploy: Spring Web applications can be easily deployed to various web servers such as Apache Tomcat, Nginx, and Cloud Foundry. This makes it simple to deploy algorithms to production environments. 5. Community support: Spring Web has a large and active community that provides documentation, tutorials, and examples. This makes it easy to learn and use the framework. In addition, the use of Spring Web mechanisms allows algorithms to be integrated with other Spring components, such as data access, business logic, and messaging. This can simplify the development and maintenance of applications. In summary, Spring Web mechanism provides a powerful and easy-to-use platform to deploy sparsity-constrained feature fusion algorithms based on implicit feature expression. It provides advantages such as RESTful API, lightweight and scalability, security, easy deployment, and community support.
[0060] More specifically, the code portion for processing the first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix using the principal component analysis method to obtain the characteristic implicit expression of the nursing association feature vector and the characteristic implicit expression of the patient's vital sign feature vector is as follows:
[0061]
[0062]
[0063] Furthermore, the patient behavior analysis feature vector is passed through a classifier to obtain a classification result, which is then used to indicate whether the patient's behavior is abnormal. By classifying the patient behavior analysis feature vector through a classifier, the difference between normal behavior and abnormal behavior can be identified. This helps medical staff to promptly detect abnormal behaviors or health problems that may exist in patients, and improves the accuracy and efficiency of abnormality detection.
[0064] Specifically, in the above-mentioned remote operation and maintenance management method of intelligent high-end equipment based on cloud computing, the patient behavior analysis feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the patient's behavior is abnormal, including: using the classifier to process the patient behavior analysis feature vector with the following patient behavior analysis formula to obtain the classification result; wherein the patient behavior analysis formula is:
[0065] O=softmax{(W n ,B n ):…:(W1,B1)|X}
[0066] Among them, W1 to W n is the weight matrix, B1 to B n is a bias vector, X is a patient behavior analysis feature vector, softmax represents a softmax function, and O represents the classification result.
[0067] In summary, according to the embodiment of the present application, the remote operation and maintenance management method of intelligent high-end equipment based on cloud computing has been explained, which uses cloud computing technology and artificial intelligence technology based on the field of deep learning to extract and encode features of patient care monitoring videos within a predetermined time period and patient vital signs data at multiple predetermined time points within the predetermined time period to obtain a classification result of whether the patient's behavior is abnormal.
[0068] In this way, by comprehensively monitoring the patient's behavior, abnormal behavior of the patient can be discovered in time to avoid missing the best time for treatment.
[0069] Based on the same inventive concept, the embodiments of the present application also provide a cloud computing-based intelligent high-end equipment remote operation and maintenance management system, which can be used to implement the methods described in the above embodiments, as described in the following embodiments. Since the principle of solving the problem by the cloud computing-based intelligent high-end equipment remote operation and maintenance management system is similar to the above method, the implementation of the cloud computing-based intelligent high-end equipment remote operation and maintenance management system can refer to the implementation of the above method, and the repeated parts will not be repeated. As used below, the terms "unit" or "module" can be a combination of software and / or hardware that implements predetermined functions. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.
[0070] Figure 4 FIG. 1 is a block diagram of a cloud computing-based intelligent high-end equipment remote operation and maintenance management system according to an embodiment of the present application. Figure 4 As shown, according to the embodiment of the present application, the cloud computing-based intelligent high-end equipment remote operation and maintenance management system 100 includes: a patient behavior data acquisition module 110, which is used to acquire the patient care monitoring video within a predetermined time period and the patient's vital signs data at multiple predetermined time points within the predetermined time period; a care monitoring feature extraction module 120, which is used to analyze the patient care monitoring video within the predetermined time period to obtain a care-related feature vector; a patient vital signs data analysis module 130, which is used to extract and encode the patient's vital signs data to obtain a patient behavior analysis feature vector; and a patient behavior analysis module 140, which is used to comprehensively analyze the care-related feature vector and the patient's vital signs feature vector to obtain a classification result of whether the patient's behavior is abnormal.
[0071] Here, those skilled in the art can understand that the specific functions and operations of each unit and module in the above-mentioned cloud computing-based intelligent high-end equipment remote operation and maintenance management system 100 have been referred to above. Figures 1 to 3 The invention has been introduced in detail in the description of the remote operation and maintenance management method of intelligent high-end equipment based on cloud computing, and therefore, its repeated description will be omitted.
[0072] In summary, the cloud computing-based intelligent high-end equipment remote operation and maintenance management system according to the embodiment of the present application has been explained, which uses cloud computing technology and artificial intelligence technology based on the field of deep learning to extract and encode features of patient care monitoring videos within a predetermined time period and patient vital sign data at multiple predetermined time points within the predetermined time period to obtain a classification result of whether the patient's behavior is abnormal. In this way, by comprehensively monitoring the patient's behavior, the patient's abnormal behavior can be discovered in time to avoid missing the best time for treatment.
[0073] The devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is an electronic device, and the electronic device may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
[0074] In a typical example, the electronic device specifically includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the above data processing steps when executing the program.
[0075] Figure 5 A block diagram of an electronic device according to an embodiment of the present application. Figure 5 As shown, based on the above data processing method, this specification also provides an electronic device, at least one processor; a memory connected to the at least one processor in communication; wherein the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute any of the above data processing methods. The processor can take the form of a computer-readable medium, logic gate, switch, application specific integrated circuit (Application Specific Integrated Circuit, ASIC), programmable logic controller and embedded microcontroller, etc., such as a microprocessor or processor and a computer-readable program code (such as software or firmware) that can be executed by the (micro) processor. The memory includes but is not limited to random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), cache (Cache), hard disk (Hard Disk Drive, HDD) or memory card (Memory Card). The electronic device can be a server of the method described in the above embodiment, etc. Of course, the electronic device can also include module structures such as input devices and transmission modules. In this embodiment, the functions and effects specifically implemented by the electronic device can be explained in comparison with other embodiments, and will not be repeated here.
[0076] In addition, the embodiments of this specification also provide a computer-readable storage medium on which computer instructions are stored, and when the instructions are executed, the steps of the method described in any one or more of the above embodiments are implemented. The storage medium may include a physical device for storing information, which is usually to digitize the information and then store it in a medium using electrical, magnetic or optical methods. The storage medium may include: devices that store information using electrical energy, such as various types of memories, such as RAM, ROM, etc.; devices that store information using magnetic energy, such as hard disks, floppy disks, magnetic tapes, magnetic core memories, bubble memories, and U disks; devices that store information optically, such as CDs or DVDs. Of course, there are other types of readable storage media, such as quantum memories, graphene memories, and so on.
[0077] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0078] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, devices or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0079] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.
[0080] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0081] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims
1. A remote operation and maintenance management method for intelligent high-end equipment based on cloud computing, characterized in that: include: Acquire patient care monitoring video within a predetermined time period and patient vital sign data at multiple predetermined time points within the predetermined time period; Analyzing the patient care monitoring video within the predetermined time period to obtain a care-related feature vector; Extracting and encoding the patient's vital sign data to obtain a patient behavior analysis feature vector; The nursing-related feature vector and the patient's vital sign feature vector are comprehensively analyzed to obtain a classification result of whether the patient's behavior is abnormal.
2. The remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to claim 1 is characterized in that: Analyze the patient care monitoring video within the predetermined time period to obtain a care-related feature vector, including: extracting a plurality of patient care monitoring key frames from the patient care monitoring video; Performing target detection on the plurality of patient care monitoring key frames to obtain a plurality of patient regions of interest and a plurality of equipment regions of interest; Performing linear embedding coding on the multiple patient regions of interest and the multiple equipment regions of interest to obtain multiple patient feature vectors of interest and multiple equipment feature vectors of interest; Associatively encoding the plurality of patient feature vectors of interest and the plurality of equipment feature vectors of interest to obtain a plurality of collaborative care feature matrices; The multiple collaborative care feature matrices are passed through a care monitoring feature extraction module using a three-dimensional convolution kernel to obtain a care-related feature vector.
3. The remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to claim 2 is characterized in that: Extracting a plurality of patient care monitoring key frames from the patient care monitoring video includes: A plurality of patient care monitoring key frames are extracted from the patient care monitoring video at a predetermined sampling frequency.
4. The remote operation and maintenance management method for intelligent high-end equipment based on cloud computing according to claim 3 is characterized in that: Performing target detection on the multiple patient care monitoring key frames to obtain multiple patient regions of interest and multiple equipment regions of interest, including: Each of the plurality of patient care monitoring key frames is passed through a patient target detection network and a room equipment target detection network to obtain the plurality of patient interest regions and the plurality of equipment interest regions.
5. The remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to claim 4 is characterized in that: Performing linear embedding coding on the multiple patient regions of interest and the multiple equipment regions of interest to obtain multiple patient feature vectors of interest and multiple equipment feature vectors of interest, including: Pass each of the plurality of patient regions of interest through a patient linear embedding layer to obtain the plurality of patient feature vectors of interest; The plurality of equipment regions of interest are passed through an equipment linear embedding layer to obtain the plurality of equipment feature vectors of interest.
6. The remote operation and maintenance management method for intelligent high-end equipment based on cloud computing according to claim 5 is characterized in that: Extracting and encoding the patient's vital sign data to obtain a patient behavior analysis feature vector includes: The patient's vital sign data is passed through a patient's vital sign time series encoder module to obtain a patient's vital sign feature vector.
7. The remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to claim 6 is characterized in that: Comprehensively analyzing the nursing-related feature vector and the patient's vital sign feature vector to obtain a classification result of whether the patient's behavior is abnormal, including: The care-related feature vector and the patient's vital sign feature vector are constrained by sparsity based on implicit feature expression to obtain the patient behavior analysis feature vector; The patient behavior analysis feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the patient's behavior is abnormal.
8. The remote operation and maintenance management method for intelligent high-end equipment based on cloud computing according to claim 7 is characterized in that: The care-related feature vector and the patient's vital signs feature vector are constrained by sparsity based on implicit feature expression to obtain the patient behavior analysis feature vector, including: Calculating a first autocorrelation sample covariance matrix of the nursing-associated feature vector, wherein an eigenvalue at each position in the first autocorrelation sample covariance matrix is a variance between eigenvalues at two corresponding positions in the nursing-associated feature vector; Calculating a second autocorrelation sample covariance matrix of the patient's vital sign feature vector, wherein the eigenvalue of each position in the second autocorrelation sample covariance matrix is the variance between the eigenvalues of two corresponding positions in the patient's vital sign feature vector; Using principal component analysis to process the first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix to obtain a characteristic implicit expression of a nursing-related characteristic vector and a characteristic implicit expression of a patient's vital sign characteristic vector; Using L1 regularization to calculate an implicit inter-feature constraint matrix between the implicit feature expression of the care-related feature vector and the implicit feature expression of the patient's vital sign feature vector; Multiplying the care-related feature vector and the patient's vital sign feature vector by the feature implicit constraint matrix respectively to map the care-related feature vector and the patient's vital sign feature vector into the feature space of the feature implicit constraint matrix to obtain a mapped care-related feature vector and a mapped patient's vital sign feature vector; A position-weighted sum between the mapped care association feature vector and the mapped patient vital sign feature vector is calculated to obtain the patient behavior analysis feature vector.
9. The remote operation and maintenance management method of intelligent high-end equipment based on cloud computing according to claim 8 is characterized in that: The first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix are processed by using a principal component analysis method to obtain a characteristic implicit expression of a nursing-related characteristic vector and a characteristic implicit expression of a patient's vital sign characteristic vector, including: Create a Spring web application; Creating a data model class, wherein the data model class is used to represent the first autocorrelation sample covariance matrix, the second autocorrelation sample covariance matrix, the feature implicit expression of the care-related feature vector, and the feature implicit expression of the patient's vital sign feature vector; Writing the principal component analysis method in the Spring Web application; Creating a REST API endpoint, wherein the REST API endpoint is used to allow a client to send the first autocorrelation sample covariance matrix and the second autocorrelation sample covariance matrix; Deploy the Spring Web application on a Web server.
10. A cloud computing-based intelligent high-end equipment remote operation and maintenance management system, characterized in that: include: A patient behavior data acquisition module, used to acquire patient care monitoring videos within a predetermined time period and patient vital sign data at multiple predetermined time points within the predetermined time period; A nursing monitoring feature extraction module is used to analyze the patient nursing monitoring video within the predetermined time period to obtain a nursing related feature vector; A patient vital sign data analysis module, used for extracting and encoding features of the patient vital sign data to obtain a patient behavior analysis feature vector; The patient behavior analysis module is used to comprehensively analyze the care-related feature vector and the patient's vital sign feature vector to obtain a classification result of whether the patient's behavior is abnormal.
