Nursing system and method for elderly hip arthroplasty patient

Abstract

The invention relates to the field of intelligent detection, and provides a nursing system and method for elderly hip replacement patients. According to the method, electrocardiogram data, time queue data of non-invasive blood pressure, time queue data of pulse blood oxygen saturation, time queue data of end-expiratory carbon dioxide and time queue data of body temperature of a patient in the operation process of old-age hip replacement are analyzed in real time, and then whether the physiological state of the patient in the operation is stable or not is judged. Therefore, a more objective evaluation result can be obtained, the risk of subjective judgment is reduced, potential unstable risks can be found based on real-time analysis, and then a timely intervention basis is provided for medical staff.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection, and more specifically, to a nursing system and method for elderly patients undergoing hip replacement surgery. Background Art

[0002] Hip fracture is a common trauma in the elderly, usually occurring in the elderly with osteoporosis, with falls being the main cause. Hip fracture has a huge impact on the health of the elderly, and is known as the "last fracture in life" due to its high disability and mortality rates. Hip replacement is one of the main treatments for hip fractures in the elderly.

[0003] During hip replacement surgery, the patient's physiological state may change due to a variety of factors such as surgical operation, anesthesia effect, blood loss, etc. In other words, the patient's physiological state may fluctuate at any time. If the potential risk of instability is not discovered in time, the best time for intervention may be delayed. In the past, medical staff relied on their own experience to judge whether the patient's physiological condition was stable during surgery, which was highly subjective and only focused on individual indicators while ignoring the overall situation.

[0004] Therefore, an optimized care program for elderly patients undergoing hip replacement is needed. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present application provides a system and method for caring for elderly patients undergoing hip replacement.

[0006] A method for nursing elderly patients undergoing hip replacement, comprising:

[0007] Obtain the electrocardiogram data, time-cohort data of non-invasive blood pressure, time-cohort data of pulse oxygen saturation, time-cohort data of end-tidal carbon dioxide, and time-cohort data of body temperature of patients during hip replacement surgery in the elderly;

[0008] Performing data analysis on the electrocardiogram data of the patient to obtain the patient's electrocardiogram semantic feature vector;

[0009] Performing data analysis on the time queue data of the patient's non-invasive blood pressure, the time queue data of the pulse oxygen saturation, the time queue data of the end-tidal carbon dioxide, and the time queue data of the body temperature to obtain a characteristic vector of the patient's physiological monitoring parameters;

[0010] Performing feature fine attribute alignment and combination based on scale standardization adjustment on the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector to obtain a patient's intraoperative monitoring index feature vector;

[0011] Based on the patient's intraoperative monitoring index feature vector, an evaluation result of the patient's intraoperative physiological state is obtained.

[0012] A nursing system for elderly patients undergoing hip replacement, comprising:

[0013] The patient intraoperative monitoring data acquisition module is used to obtain the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data and body temperature time queue data during the elderly hip replacement surgery;

[0014] The patient's intraoperative electrocardiogram data analysis module is used to analyze the patient's electrocardiogram data to obtain the patient's electrocardiogram semantic feature vector;

[0015] A patient intraoperative physiological monitoring data analysis module is used to analyze the time queue data of the patient's non-invasive blood pressure, the time queue data of pulse oxygen saturation, the time queue data of end-tidal carbon dioxide, and the time queue data of body temperature to obtain a patient physiological monitoring parameter feature vector;

[0016] A patient intraoperative monitoring indicator feature fusion module is used to align and combine the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector based on scale standardization adjustment to obtain a patient intraoperative monitoring indicator feature vector;

[0017] The patient's intraoperative physiological state assessment module is used to obtain the assessment result of the patient's intraoperative physiological state based on the patient's intraoperative monitoring index feature vector.

[0018] This application has significant technical effects due to the adoption of the above technical solutions:

[0019] The present application provides a nursing system and method for elderly hip replacement patients, which can determine whether the patient's intraoperative physiological state is stable by real-time analysis of the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data, and body temperature time queue data during the elderly hip replacement surgery. In this way, a more objective evaluation result can be obtained, reducing the risk of subjective judgment, and based on real-time analysis, potential unstable risks can be discovered, thereby providing timely intervention basis for medical staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] 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.

[0021] Figure 1 The present invention is a flowchart of a method for caring for elderly patients undergoing hip replacement according to an embodiment of the present application.

[0022] Figure 2 This is a flowchart of step S120 in the nursing method for elderly patients undergoing hip replacement according to an embodiment of the present application.

[0023] Figure 3 This is a flowchart of step S130 in the nursing method for elderly patients undergoing hip replacement according to an embodiment of the present application.

[0024] Figure 4 This is a system block diagram of a care system for elderly hip replacement patients according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] 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.

[0026] Hip fractures are common in the elderly, especially those with osteoporosis, with falls being the main cause. Such fractures pose a serious threat to the health of the elderly, as they often lead to high disability and mortality rates, and are therefore known as "the last fracture in life." Hip replacement is a major treatment option for elderly hip fractures.

[0027] During hip replacement surgery, the patient's physiological condition may fluctuate due to the combined effects of surgical manipulation, anesthesia, blood loss, and other factors. In other words, the patient's physiological stability may be challenged at any time. Failure to identify and respond to these potential unstable factors in a timely manner may miss the critical opportunity for intervention. In the past, medical staff mainly relied on personal experience to assess the patient's physiological stability during surgery. This approach is highly subjective and may ignore the patient's overall condition due to excessive focus on individual indicators, resulting in a one-sided assessment. Therefore, an optimized care plan for elderly patients undergoing hip replacement is needed.

[0028] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also shown a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks has provided new solutions and solutions for the care of elderly patients undergoing hip replacement.

[0029] Figure 1Flow chart of the nursing method for elderly patients undergoing hip replacement according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the nursing method for elderly hip replacement patients includes: S110, obtaining the patient's electrocardiogram data, time queue data of non-invasive blood pressure, time queue data of pulse blood oxygen saturation, time queue data of end-tidal carbon dioxide and time queue data of body temperature during the elderly hip replacement operation; S120, performing data analysis on the patient's electrocardiogram data to obtain the patient's electrocardiogram semantic feature vector; S130, performing data analysis on the patient's time queue data of non-invasive blood pressure, time queue data of pulse blood oxygen saturation, time queue data of end-tidal carbon dioxide and time queue data of body temperature to obtain the patient's physiological monitoring parameter feature vector; S140, performing feature fine attribute alignment and combination based on scale standardization adjustment on the patient's electrocardiogram semantic feature vector and the patient's physiological monitoring parameter feature vector to obtain the patient's intraoperative monitoring index feature vector; S150, based on the patient's intraoperative monitoring index feature vector, obtaining the evaluation result of the patient's intraoperative physiological state.

[0030] In step S110, the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse blood oxygen saturation time queue data, end-tidal carbon dioxide time queue data and body temperature time queue data during the elderly hip replacement surgery are obtained. It should be understood that the electrocardiogram data mainly reflects the electrophysiological activity of the heart, including heart rate, heart rhythm (such as sinus rhythm, atrial fibrillation, etc.), ST segment changes, T wave morphology, etc. It can help identify risks such as arrhythmia and myocardial ischemia. Non-invasive blood pressure data mainly reflects changes in arterial blood pressure, including systolic pressure, diastolic pressure and mean arterial pressure. Blood pressure is an important indicator of the state of the circulatory system. Hypotension may lead to organ insufficiency, while hypertension may increase the risk of cardiovascular events. Blood pressure fluctuations during surgery need to be managed in time to maintain a stable circulation state. Pulse blood oxygen saturation data reflects the proportion of oxygenated hemoglobin in the blood, i.e. SpO2. Blood oxygen saturation reflects the oxygen carrying capacity of the blood. Low oxygen saturation may mean respiratory insufficiency or insufficient ventilation, which requires immediate treatment to prevent tissue hypoxia. End-tidal carbon dioxide data mainly reflects the patient's ventilation and perfusion, namely EtCO2. End-tidal carbon dioxide is an important indicator for evaluating the patient's respiratory and circulatory functions. Its changes can reflect pulmonary ventilation, alveolar ventilation / perfusion ratio, cardiac output, etc. In hip replacement surgery, monitoring of end-tidal carbon dioxide helps to timely detect risks such as respiratory depression and circulatory failure. Body temperature data reflects the patient's body temperature changes, including core body temperature and surface temperature. Body temperature is an important indicator for evaluating the patient's metabolic state and body temperature regulation function. In hip replacement surgery, the patient's body temperature may change due to surgical trauma, the use of anesthetic drugs, and the influence of ambient temperature. Fluctuations in body temperature may affect the patient's physiological functions, such as heart rate, blood pressure, and respiration. Therefore, monitoring body temperature helps to timely detect and deal with hypothermia or hyperthermia. In general, these indicators together reflect the patient's physiological state during hip replacement surgery. Through real-time monitoring and analysis of these indicators, potential unstable risks can be discovered in a timely manner, providing timely intervention basis for medical staff, thereby ensuring the safety of surgery and the patient's recovery.

[0031] In step S120, data analysis is performed on the patient's electrocardiogram data to obtain the patient's electrocardiogram semantic feature vector. Specifically, Figure 2 FIG. 1 is a flow chart of step S120 in the nursing method for elderly patients undergoing hip replacement according to an embodiment of the present application. Figure 2 As shown, the step S120 includes: S121, passing the patient's ECG data through an ECG image data feature encoder to obtain the patient's ECG semantic feature map; S122, performing mean pooling of each feature matrix along the channel dimension on the patient's ECG semantic feature map to obtain the patient's ECG semantic feature vector.

[0032] In step S121, the patient's electrocardiogram data is passed through an electrocardiogram image data feature encoder to obtain the patient's electrocardiogram semantic feature map. It should be understood that the electrocardiogram contains a large amount of complex electrical signal information, and directly analyzing the original data may not be efficient enough. In order to simplify the signal and highlight the key information, it is necessary to perform feature extraction processing on the patient's electrocardiogram data. By extracting features, abnormal conditions such as arrhythmia and myocardial ischemia can be more easily identified. And noise can be filtered to improve the accuracy and reliability of analysis. Specifically, the patient's electrocardiogram data is passed through an electrocardiogram image data feature encoder, where the electrocardiogram image data feature encoder is a deep separable convolutional neural network model. The deep separable convolution decomposes the standard convolution into deep convolution and point-by-point convolution, which greatly reduces the amount of calculation and the number of parameters and improves the processing speed. And at the same time, it can effectively capture the global pattern and local details in the electrocardiogram, thereby more accurately understanding the patient's heart condition.

[0033] In step S122, the patient's ECG semantic feature map is mean-pooled along the channel dimension for each feature matrix to obtain the patient's ECG semantic feature vector. It should be understood that in order to reduce data complexity and make subsequent analysis and processing more efficient, a more concise feature vector can be extracted by performing a dimensionality reduction operation on the feature map. Specifically, the patient's ECG semantic feature map is mean-pooled along the channel dimension for each feature matrix to obtain the patient's ECG semantic feature vector. In this way, the information of each feature matrix can be aggregated to extract global features, and through the aggregation operation, the stability of the features can be enhanced, so that the model performs more consistently when facing different inputs.

[0034] In step S130, the time queue data of the patient's non-invasive blood pressure, the time queue data of the pulse oxygen saturation, the time queue data of the end-tidal carbon dioxide and the time queue data of the body temperature are analyzed to obtain the patient's physiological monitoring parameter feature vector. Specifically, Figure 3 FIG. 1 is a flow chart of step S130 in the method for caring for elderly patients undergoing hip replacement according to an embodiment of the present application. Figure 3 As shown, the step S130 includes: S131, arranging the time queue data of the non-invasive blood pressure, the time queue data of the pulse blood oxygen saturation, the time queue data of the end-tidal carbon dioxide and the time queue data of the body temperature into a patient physiological monitoring parameter input matrix according to the sample dimension; S132, passing the patient physiological monitoring parameter input matrix through a physiological monitoring parameter feature encoder to obtain a patient physiological monitoring parameter feature matrix; S133, expanding the patient physiological monitoring parameter feature matrix to obtain the patient physiological monitoring parameter feature vector.

[0035] In step S131, the time queue data of the non-invasive blood pressure, the time queue data of the pulse blood oxygen saturation, the time queue data of the end-tidal carbon dioxide and the time queue data of the body temperature are arranged into a patient physiological monitoring parameter input matrix according to the sample dimension. It should be understood that in order to integrate the time queue data of different physiological indicators together and facilitate subsequent analysis thereof, the time queue data of the non-invasive blood pressure, the time queue data of the pulse blood oxygen saturation, the time queue data of the end-tidal carbon dioxide and the time queue data of the body temperature need to be arranged according to the sample dimension. By arranging and integrating according to the sample dimension, the correlation and change trend between different indicators are easier to capture. In the arranged patient physiological monitoring parameter input matrix, each row usually represents the physiological monitoring data at a time point. That is to say, each row contains the values ​​of physiological indicators such as non-invasive blood pressure, pulse blood oxygen saturation, end-tidal carbon dioxide and body temperature collected at the same time point. These values ​​jointly reflect the physiological state of the patient at that time point. Each column represents the monitoring data of a physiological indicator at different time points. It is also worth mentioning that data in matrix form is easier to store and transmit, and can be conveniently saved in a database or transmitted to a remote server for further analysis and processing.

[0036] In step S132, the patient physiological monitoring parameter input matrix is ​​passed through a physiological monitoring parameter feature encoder to obtain a patient physiological monitoring parameter feature matrix. It should be understood that the patient physiological monitoring parameter input matrix contains a large amount of original data of physiological indicators. In order to identify the changing trends of different physiological parameters of the patient from these original data, it is necessary to extract features. And through feature extraction, the influence of redundant information in the original data can be reduced. Specifically, the patient physiological monitoring parameter input matrix is ​​input into the physiological monitoring parameter feature encoder. Here, the physiological monitoring parameter feature encoder is a convolutional neural network model in which adjacent layers use mutually transposed convolution kernels. Adjacent layers use mutually transposed convolution kernels to make the network structure symmetrical. This symmetry helps to maintain consistency between input and output data, thereby improving the stability and reliability of the model. And in this structure, the features of the previous layer can be effectively reused by the next layer. This helps to reduce redundant calculations and improve calculation efficiency, and also helps to extract deeper feature information. Through the processing of the physiological monitoring parameter feature encoder, the physiological monitoring data can be analyzed more comprehensively and the accuracy of understanding the patient's physiological state can be improved.

[0037] In step S133, the patient physiological monitoring parameter characteristic matrix is ​​expanded to obtain the patient physiological monitoring parameter characteristic vector. It should be understood that in matrix operations, especially when involving operations of large amounts of data, the computational complexity is often high. Therefore, in the technical solution of the present application, it is necessary to expand the patient physiological monitoring parameter characteristic matrix. By expanding the characteristic matrix into a vector, an efficient vector operation algorithm can be used for calculation, thereby improving the computational efficiency.

[0038] In step S140, the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector are aligned and combined based on the feature fine attributes adjusted by scale standardization to obtain the patient's intraoperative monitoring index feature vector. It should be understood that the ECG semantic feature vector can reflect the patient's cardiac electrophysiological condition, such as heart rate variability, arrhythmia, etc., and is an important indicator for evaluating cardiovascular health. The physiological monitoring parameter feature vector contains other important physiological information of the patient, such as blood pressure, blood oxygen saturation, body temperature, etc. These parameters are crucial for evaluating the patient's overall physiological condition. In elderly hip replacement surgery, patients may face a variety of physiological risks, such as cardiovascular events, respiratory problems, etc., which may occur simultaneously or affect each other. In order to provide more comprehensive information and improve the accuracy of intraoperative physiological monitoring of patients, it is necessary to fuse the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector.

[0039] In particular, considering that the patient's ECG semantic feature vector is derived from the ECG data, in which the ECG signal has a specific amplitude range, the feature extraction process mainly focuses on the abstract representation of the dimensions related to the electrical characteristics of the ECG waveform, such as the morphology and rhythm. The ECG semantic feature vector formed after the encoder and pooling, the numerical range of each element is based on the quantization and compression of these ECG specific attributes, and will fall within a specific interval that is compatible with the ECG signal itself and the processing method. The patient's physiological monitoring parameter feature vector is formed by the time series arrangement of different physiological parameters, and then after the feature encoder, expansion and other operations. Its scale range is a comprehensive combination of the changes in the original magnitude of these different physiological parameters and the subsequent processing. In other words, the difference in data sources leads to different scale ranges of the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector. Different scale ranges will lead to the misestimation of the importance of the feature during the fusion process, which will lead to the ECG-related features and other physiological monitoring parameter features that should be comprehensively considered and treated equally when judging whether the patient's intraoperative physiological state is stable. Participate in the decision-making according to their true contribution to the physiological state, and then have a negative impact on the evaluation results of the physiological stability state. Therefore, in the technical solution of the present application, the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector are aligned and combined based on the feature fine attributes adjusted based on scale standardization to obtain the patient's intraoperative monitoring index feature vector.

[0040] Specifically, in an embodiment of the present application, the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector are subjected to feature fine attribute alignment and combination based on scale normalization adjustment to obtain a patient intraoperative monitoring index feature vector, including: performing normalization modulation based on vector norming on the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector to obtain a pre-aligned patient ECG semantic feature vector and a pre-aligned patient physiological monitoring parameter feature vector; performing intrinsic feature extraction on the pre-aligned patient ECG semantic feature vector and the pre-aligned patient physiological monitoring parameter feature vector to obtain a first patient ECG semantic principal component feature vector and a second patient ECG semantic principal component feature vector; performing intrinsic fine-grained alignment and fusion on the first patient ECG semantic principal component feature vector and the second patient ECG semantic principal component feature vector to obtain the patient intraoperative monitoring index feature vector.

[0041] In the embodiment of the present application, specifically, another implementation expression of performing feature fine attribute alignment and combination based on scale standardization adjustment on the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector to obtain the patient intraoperative monitoring index feature vector can be: processing the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector according to the following formula to obtain the patient intraoperative monitoring index feature vector; wherein the formula is:

[0042]

[0043] Among them, V1 represents the patient's ECG semantic feature vector, V2 represents the patient's physiological monitoring parameter feature vector, and v 1i represents the feature value of the i-th position of the patient's ECG semantic feature vector, v 2i represents the eigenvalue of the ith position of the patient's physiological monitoring parameter eigenvector, exp represents the natural exponential function, represents the square of the bi-norm of the vector, V1′ represents the pre-aligned patient ECG semantic feature vector, V2′ represents the pre-aligned patient physiological monitoring parameter feature vector, PCA(·) represents the intrinsic feature extraction, U1 represents the sequence of the first intrinsic decomposition vector, Λ1 represents the first diagonal matrix, λ 11 and λ 1m denote the eigenvalues ​​of the first and mth positions of the first diagonal matrix, respectively. T denotes the transpose of the vector. concat{·;·} denotes vector concatenation. v 11 、v 12 、v 1m The first, second and mth eigenvectors of the sequence of the first patient's ECG semantic principal component eigenvectors are represented respectively, V k represents the semantic principal component eigenvector of the first patient’s ECG, U2 represents the sequence of the second eigendecomposition vector, Λ2 represents the second diagonal matrix, λ 21 and λ 2m denote the eigenvalues ​​of the first and mth positions of the second diagonal matrix, respectively, v 21 、v 22 、v 2m Respectively represent the first, second and mth eigenvectors of the sequence of the second patient's ECG semantic principal component eigenvector, V s represents the semantic principal component eigenvector of the second patient’s ECG, represents positional subtraction, ⊙ represents positional multiplication, Indicates positional addition, Conv 1×1 represents point convolution, f vp (V k ,V s ) represents the feature vector of the patient’s intraoperative monitoring indicators.

[0044] More specifically, in an embodiment of the present application, the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector are normalized and modulated based on vector norming to obtain pre-aligned patient ECG semantic feature vector and pre-aligned patient physiological monitoring parameter feature vector, including: creating a new Spring Boot project and adding required dependencies in the Spring Boot project; creating a service class in the Spring Boot project, the service class is used to implement the normalization modulation logic of the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector; creating a controller class in the Spring Boot project, the controller class is used to provide a REST API interface for the front-end to call to realize data reception and processing; packaging the Spring Boot project and deploying the packaged application on the server.

[0045] In an embodiment of the present application, the code for performing normalization modulation based on vector norming on the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector to obtain a pre-aligned patient ECG semantic feature vector and a pre-aligned patient physiological monitoring parameter feature vector may be implemented as follows:

[0046]

[0047]

[0048]

[0049] Here, the advantages of using the Spring mechanism to deploy and implement "normalization modulation based on vector norming of the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector to obtain pre-aligned patient ECG semantic feature vector and pre-aligned patient physiological monitoring parameter feature vector" include: a. Spring Boot simplifies Spring-based application development. By providing default configuration, Spring applications can be quickly started and run, and it has microservice architecture support, which is convenient for rapid deployment and expansion; b. Spring Security provides powerful security and authentication mechanisms to protect applications from security threats; c. The modular nature of the Spring framework makes the code clearer, easier to maintain and test; d. Spring has a large community that provides a large number of documents, tutorials and third-party libraries to facilitate solving problems encountered in development; e. Spring provides strong testing support, including integration testing and unit testing, making testing simpler and more automated; f. Spring MVC provides a flexible error handling mechanism that can centrally handle errors and exceptions; g. The Spring framework provides a variety of performance optimization methods, such as asynchronous processing, caching, etc., which can improve application performance.

[0050] That is, in the technical solution of the present application, the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector are subjected to feature fine attribute alignment based on scale standardization adjustment to obtain the patient's intraoperative monitoring index feature vector, which first performs normalization modulation based on vector norming on the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector to obtain a pre-aligned patient ECG semantic feature vector and a pre-aligned patient's physiological monitoring parameter feature vector. This process is not just for simply adjusting the scale, but an important technology in feature engineering, which aims to solve the problem of inconsistent scales between features in the data set. By calculating the modulus of the feature vector and dividing each vector by its own modulus, it can be ensured that all feature vectors have the same length, thereby eliminating the impact of scale differences between different feature vectors. Normalization processing not only contributes to numerical stability in subsequent processing, but also makes the comparison between feature vectors more fair and reasonable, laying the foundation for subsequent feature alignment. In machine learning models, especially algorithms such as neural networks, the scale of the input features directly affects the speed and direction of weight updates, so normalization can be regarded as a regularization method that helps prevent overfitting and also helps avoid the problem of gradient disappearance or explosion.

[0051] Next, the pre-aligned patient ECG semantic feature vector and the pre-aligned patient physiological monitoring parameter feature vector are subjected to intrinsic feature extraction to obtain a first patient ECG semantic principal component feature vector and a second patient ECG semantic principal component feature vector. That is, after the normalization modulation is completed, the pre-aligned patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector are subjected to intrinsic feature extraction. As an unsupervised linear dimensionality reduction method, intrinsic feature extraction is widely used in data compression and feature extraction. This process determines the principal components by finding the direction of the maximum variance of the data. These principal components are new coordinate axes in the original feature space and can capture the structure of the data with minimal information loss. It is worth mentioning that by selecting the eigenvectors corresponding to the first few maximum eigenvalues, the dimension of the feature space can be effectively reduced while maintaining the key information of the data. This dimensionality reduction can not only accelerate the model training process, but also help the model to generalize better and reduce the impact of noise, thereby improving the performance of the model.

[0052] Finally, the first patient's ECG semantic principal component feature vector and the second patient's ECG semantic principal component feature vector are intrinsically aligned and fused to obtain the patient's intraoperative monitoring index feature vector. The goal of this step is to find the best matching point or area between the two target analysis raw data principal component feature vectors to achieve alignment between the two. Fine-grained alignment goes beyond the traditional rough fusion methods such as simple weighted averaging or splicing, and needs to consider the time series characteristics or spatial distribution characteristics of the feature vector. The fusion after alignment is not a simple numerical operation, but requires the design of a specific mechanism to determine how to combine the information of the two feature vectors. This content-based intelligent fusion method can more effectively utilize multi-source information and improve the model's representation ability and decision-making quality.

[0053] In step S150, based on the patient's intraoperative monitoring index feature vector, an evaluation result of the patient's intraoperative physiological state is obtained. Specifically, in an embodiment of the present application, based on the patient's intraoperative monitoring index feature vector, an evaluation result of the patient's intraoperative physiological state is obtained, including: passing the patient's intraoperative monitoring index feature vector through a physiological monitoring evaluator based on a classifier to obtain the evaluation result, and the evaluation result is used to indicate whether the patient's intraoperative physiological state is stable. It should be understood that a classifier is a machine learning model that, after training and optimization, can identify key information in the patient's intraoperative monitoring index feature vector, thereby evaluating the patient's physiological state. That is, through a physiological monitoring evaluator based on a classifier, an automated and intelligent evaluation of the patient's intraoperative physiological state can be achieved. This evaluation method can quickly and accurately analyze the intraoperative monitoring index feature vector, avoiding the tediousness and subjectivity of manual evaluation. The results obtained by the evaluation can provide doctors with real-time information about the patient's intraoperative physiological state, thereby guiding surgical operations. For example, when the evaluation results show that the patient's physiological state is unstable, the doctor may need to suspend the operation, adjust the anesthesia dose, or take other emergency measures.

[0054] In summary, the nursing method for elderly patients undergoing hip replacement based on the embodiment of the present application is explained, which determines whether the patient's intraoperative physiological state is stable by performing real-time analysis on the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data, and body temperature time queue data during the elderly hip replacement operation. In this way, a more objective evaluation result can be obtained, reducing the risk of subjective judgment, and based on real-time analysis, potential unstable risks can be discovered, thereby providing timely intervention basis for medical staff.

[0055] Figure 4 FIG. 1 is a system block diagram of a system for caring for elderly patients undergoing hip replacement according to an embodiment of the present application. Figure 4As shown, according to the embodiment of the present application, the elderly hip replacement patient care system 100 includes: a patient intraoperative monitoring data acquisition module 110, which is used to obtain the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data and body temperature time queue data during the elderly hip replacement operation; a patient intraoperative electrocardiogram data analysis module 120, which is used to analyze the patient's electrocardiogram data to obtain the patient's electrocardiogram semantic feature vector; a patient intraoperative physiological monitoring data analysis module 130, which is used to analyze the patient's non-invasive blood pressure The time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data and body temperature time queue data are analyzed to obtain the patient's physiological monitoring parameter feature vector; the patient's intraoperative monitoring indicator feature fusion module 140 is used to align the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector based on scale standardization adjustment to obtain the patient's intraoperative monitoring indicator feature vector; the patient's intraoperative physiological state evaluation module 150 is used to obtain the evaluation result of the patient's intraoperative physiological state based on the patient's intraoperative monitoring indicator feature vector.

[0056] Here, those skilled in the art will appreciate that the specific functions and operations of the various units and modules in the elderly hip replacement patient care system 100 have been described in detail above. Figures 1 to 3 The present invention has been described in detail in the description of the nursing methods for elderly hip replacement patients, and therefore, its repeated description will be omitted.

[0057] In summary, the elderly hip replacement patient care system 100 based on the embodiment of the present application is explained, which determines whether the patient's intraoperative physiological state is stable by performing real-time analysis on the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data, and body temperature time queue data during the elderly hip replacement surgery. In this way, a more objective evaluation result can be obtained, reducing the risk of subjective judgment, and based on real-time analysis, potential unstable risks can be discovered, thereby providing timely intervention basis for medical staff.

[0058] As described above, the elderly hip replacement patient care system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for elderly hip replacement patient care. In one example, the elderly hip replacement patient care system 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the elderly hip replacement patient care system 100 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the elderly hip replacement patient care system 100 can also be one of the many hardware modules of the wireless terminal.

[0059] Alternatively, in another example, the elderly hip replacement patient care system 100 and the wireless terminal may also be separate devices, and the elderly hip replacement patient care system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

Claims

1. A method for nursing elderly patients undergoing hip replacement surgery, characterized in that: include: Obtain the electrocardiogram data, time-cohort data of non-invasive blood pressure, time-cohort data of pulse oxygen saturation, time-cohort data of end-tidal carbon dioxide, and time-cohort data of body temperature of patients during hip replacement surgery in the elderly; Performing data analysis on the electrocardiogram data of the patient to obtain the patient's electrocardiogram semantic feature vector; Performing data analysis on the time queue data of the patient's non-invasive blood pressure, the time queue data of the pulse oxygen saturation, the time queue data of the end-tidal carbon dioxide, and the time queue data of the body temperature to obtain a characteristic vector of the patient's physiological monitoring parameters; Performing feature fine attribute alignment and combination based on scale standardization adjustment on the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector to obtain a patient's intraoperative monitoring index feature vector; Based on the patient's intraoperative monitoring index feature vector, an evaluation result of the patient's intraoperative physiological state is obtained.

2. The nursing method for elderly patients undergoing hip replacement according to claim 1, characterized in that: Performing data analysis on the electrocardiogram data of the patient to obtain the patient's electrocardiogram semantic feature vector includes: Passing the electrocardiogram data of the patient through an electrocardiogram image data feature encoder to obtain an electrocardiogram semantic feature map of the patient; The patient's ECG semantic feature map is subjected to mean pooling of each feature matrix along the channel dimension to obtain the patient's ECG semantic feature vector.

3. The nursing method for elderly patients undergoing hip replacement according to claim 2, characterized in that: The time queue data of the patient's non-invasive blood pressure, the time queue data of the pulse oxygen saturation, the time queue data of the end-tidal carbon dioxide, and the time queue data of the body temperature are analyzed to obtain a characteristic vector of the patient's physiological monitoring parameters, including: Arranging the time queue data of the non-invasive blood pressure, the time queue data of the pulse blood oxygen saturation, the time queue data of the end-tidal carbon dioxide and the time queue data of the body temperature into a patient physiological monitoring parameter input matrix according to the sample dimension; Passing the patient physiological monitoring parameter input matrix through a physiological monitoring parameter feature encoder to obtain a patient physiological monitoring parameter feature matrix; The patient physiological monitoring parameter feature matrix is ​​expanded to obtain the patient physiological monitoring parameter feature vector.

4. The nursing method for elderly patients undergoing hip replacement according to claim 3, characterized in that: The electrocardiogram image data feature encoder is a deep separable convolutional neural network model, and the physiological monitoring parameter feature encoder is a convolutional neural network model in which adjacent layers use mutually transposed convolution kernels.

5. The nursing method for elderly patients undergoing hip replacement according to claim 4, characterized in that: The patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector are aligned and combined based on scale standardization adjustment to obtain the patient's intraoperative monitoring index feature vector, including: Performing normalization modulation based on vector norming on the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector to obtain a pre-aligned patient ECG semantic feature vector and a pre-aligned patient physiological monitoring parameter feature vector; Performing intrinsic feature extraction on the pre-aligned patient ECG semantic feature vector and the pre-aligned patient physiological monitoring parameter feature vector to obtain a first patient ECG semantic principal component feature vector and a second patient ECG semantic principal component feature vector; The first patient's ECG semantic principal component feature vector and the second patient's ECG semantic principal component feature vector are intrinsically aligned and fused to obtain the patient's intraoperative monitoring index feature vector.

6. The nursing method for elderly patients undergoing hip replacement according to claim 5, characterized in that: The patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector are subjected to normalization modulation based on vector norming to obtain a pre-aligned patient ECG semantic feature vector and a pre-aligned patient physiological monitoring parameter feature vector, including: Create a new Spring Boot project and add the required dependencies in the said Spring Boot project; Creating a service class in the Spring Boot project, wherein the service class is used to implement the normalized modulation logic of the patient's ECG semantic feature vector and the patient's physiological monitoring parameter feature vector; Create a controller class in the Spring Boot project, which is used to provide a REST API interface for the front-end to call to receive and process data; Package the Spring Boot project and deploy the packaged application to the server.

7. The nursing method for elderly patients undergoing hip replacement according to claim 6, characterized in that: Based on the patient's intraoperative monitoring index feature vector, an evaluation result of the patient's intraoperative physiological state is obtained, including: passing the patient's intraoperative monitoring index feature vector through a classifier-based physiological monitoring evaluator to obtain the evaluation result, and the evaluation result is used to indicate whether the patient's intraoperative physiological state is stable.

8. A system for caring for elderly patients undergoing hip replacement surgery, characterized in that: include: The patient intraoperative monitoring data acquisition module is used to obtain the patient's electrocardiogram data, non-invasive blood pressure time queue data, pulse oxygen saturation time queue data, end-tidal carbon dioxide time queue data and body temperature time queue data during the elderly hip replacement surgery; The patient's intraoperative electrocardiogram data analysis module is used to analyze the patient's electrocardiogram data to obtain the patient's electrocardiogram semantic feature vector; A patient intraoperative physiological monitoring data analysis module is used to analyze the time queue data of the patient's non-invasive blood pressure, the time queue data of the pulse oxygen saturation, the time queue data of the end-tidal carbon dioxide, and the time queue data of the body temperature to obtain a patient physiological monitoring parameter feature vector; A patient intraoperative monitoring indicator feature fusion module is used to align and combine the patient ECG semantic feature vector and the patient physiological monitoring parameter feature vector based on scale standardization adjustment to obtain a patient intraoperative monitoring indicator feature vector; The patient's intraoperative physiological state assessment module is used to obtain the assessment result of the patient's intraoperative physiological state based on the patient's intraoperative monitoring index feature vector.

9. The elderly hip replacement patient care system according to claim 8, characterized in that: The patient intraoperative electrocardiogram data analysis module comprises: A patient electrocardiogram data feature encoding unit, used for passing the patient's electrocardiogram data through an electrocardiogram data feature encoder to obtain a patient electrocardiogram semantic feature map; The patient ECG image data feature dimensionality reduction unit is used to perform mean pooling of each feature matrix along the channel dimension of the patient ECG semantic feature map to obtain the patient ECG semantic feature vector.

10. The elderly hip replacement patient care system according to claim 9, characterized in that: The patient's intraoperative physiological state assessment module is used to: pass the patient's intraoperative monitoring index feature vector through a classifier-based physiological monitoring evaluator to obtain the assessment result, and the assessment result is used to indicate whether the patient's intraoperative physiological state is stable.

Images