Intelligent medical care and patient information interaction system and method based on 5G technology

Abstract

The invention discloses an intelligent medical care and patient information interaction system and method based on the 5G technology, and the method comprises the steps: extracting a time queue of a heart rate value of a patient from the vital sign data of the patient, carrying out the variational mode decomposition of the heart rate value of the patient through employing the data analysis and coding technology based on artificial intelligence, and then, carrying out the recognition of the variational mode; frequency domain feature extraction is carried out on each decomposed patient heart rate intrinsic mode component, then feature hub sparse processing is carried out on each extracted feature, and full-band splicing is carried out according to each sparse patient heart rate intrinsic mode component frequency feature to intelligently generate an identification result whether an abnormal signal exists or not. And performing priority division on the vital sign data and the real-time audio and video data based on the recognition result. In this way, the dynamic mode of the heart rate change can be understood more comprehensively, and the problem that a traditional threshold value method is prone to generating false alarm or missing alarm is avoided.

Description

Technical Field

[0001] The present application relates to the field of intelligent information interaction technology, and more specifically, to a smart medical-nursing-patient information interaction system and method based on 5G technology. Background Art

[0002] With the continuous development of medical informatization, the importance of information exchange between doctors, nurses and patients has become increasingly prominent. Through information exchange, medical staff can obtain comprehensive and accurate vital signs data and disease change information of patients in a timely manner, so as to make more accurate diagnosis and treatment decisions.

[0003] However, under the traditional medical model, it is often inconvenient for medical staff to obtain patient information. For example, when patients are recovering at home or visiting primary medical institutions, it is difficult for doctors to obtain accurate and comprehensive vital signs data in a timely manner. Moreover, conventional vital signs measurements mostly rely on patients going to the hospital and being performed by medical staff using professional equipment. This method is not only time-consuming and labor-intensive, but also unable to achieve real-time monitoring. At the same time, in terms of information interaction, it is difficult for patients to feedback their own conditions to medical staff at any time, and medical staff cannot intuitively understand the patient's daily life scenes and real-time status, which limits the accuracy and timeliness of medical services.

[0004] Therefore, an optimized medical-nursing-patient information interaction solution is desired. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In the first aspect, a smart medical-nursing-patient information interaction system based on 5G technology is provided, which includes: A patient information data acquisition module, used to collect patient vital signs data using an IoT medical device worn by the patient, and to collect the patient's real-time audio and video data through a portable medical terminal; a data transmission module, used to transmit the patient's vital signs data and the real-time audio and video data to an edge server close to the patient using a 5G communication module; a data processing module, used to perform data preprocessing on the patient's vital signs data on the edge server to determine whether there is an abnormal signal, and transmit the patient's vital signs data and the real-time audio and video data to a cloud data center based on priority for access by medical staff; a data synchronization module, used to enable the medical staff to receive the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device; wherein the data processing module includes: a patient heart rate value extraction unit, configured to extract a time queue of a patient heart rate value from the patient vital sign data; A patient heart rate frequency feature extraction unit, used for performing variational modal decomposition and frequency domain feature extraction on the time queue of the patient's heart rate values ​​to obtain a set of frequency features of the patient's heart rate intrinsic modal components; A frequency feature sparseness unit, used for performing a feature hub-driven sparseness process on the set of frequency features of the intrinsic modal components of the patient's heart rate to obtain a sparse set of frequency features of the intrinsic modal components of the patient's heart rate; The recognition result generating unit is used to obtain the recognition result based on the set of the sparsed frequency characteristics of the intrinsic modal components of the patient's heart rate.

[0007] Optionally, the patient heart rate frequency feature extraction unit includes: a variational modal decomposition subunit, used to perform variational modal decomposition on the time queue of the patient heart rate values ​​to obtain a set of patient heart rate intrinsic modal components; an intrinsic modal feature extraction subunit, used to extract frequency domain features of each patient heart rate intrinsic modal component in the set of patient heart rate intrinsic modal components to obtain a set of patient heart rate intrinsic modal component frequency feature vectors as the set of patient heart rate intrinsic modal component frequency features.

[0008] Optionally, the intrinsic modal feature extraction subunit is used to: use a frequency domain feature extractor based on dilated convolution to extract frequency domain features of each patient's heart rate intrinsic modal component in the set of patient's heart rate intrinsic modal components to obtain a set of frequency feature vectors of the patient's heart rate intrinsic modal components.

[0009] Optionally, the frequency feature sparsification unit includes: a feature distribution hub search subunit, which is used to perform a feature distribution hub search on the set of the patient's heart rate intrinsic modal component frequency feature vectors to obtain the patient's heart rate intrinsic modal component frequency feature distribution hub coding vector; a frequency sparsification semantic fine-grained ablation factor calculation subunit, which is used to calculate the sparsification semantic fine-grained ablation factor of each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vectors based on the patient's heart rate intrinsic modal component frequency feature distribution hub coding vector to obtain a set of patient's heart rate intrinsic modal component frequency sparsification semantic fine-grained ablation factors; a sparsification subunit, which is used to perform feature sparsification on the set of the patient's heart rate intrinsic modal component frequency feature vectors based on the set of the patient's heart rate intrinsic modal component frequency sparsification semantic fine-grained ablation factors to obtain the set of the sparse patient's heart rate intrinsic modal component frequency feature vectors as the set of the sparse patient's heart rate intrinsic modal component frequency features.

[0010] Optionally, the frequency sparsification semantic fine-grained ablation factor calculation subunit is used to: respectively calculate the semantic fine-grained ablation factor of each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vector relative to the patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector to obtain a set of patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factors; perform gating function sparsification processing on each patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factor in the set of patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factors to obtain the set of patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors.

[0011] Optionally, the sparsification subunit is used to: use each patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factor in the set of patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors as a weight, and weight the set of patient's heart rate intrinsic modal component frequency feature vectors respectively to obtain the set of sparse patient's heart rate intrinsic modal component frequency feature vectors.

[0012] Optionally, the recognition result generating unit includes: a vector splicing subunit, used to splice the set of the sparsed patient heart rate intrinsic modal component frequency feature vectors to obtain a patient heart rate full-band feature cascade coding vector; an abnormal signal judgment subunit, used to obtain a recognition result based on the patient heart rate full-band feature cascade coding vector, and the recognition result is used to indicate whether there is an abnormal signal; wherein, in response to the recognition result being the presence of an abnormal signal, the patient's vital signs data has a higher priority than the real-time audio and video data; in response to the recognition result being the absence of an abnormal signal, the real-time audio and video data has a higher priority than the patient's vital signs data.

[0013] Optionally, the abnormal signal judgment subunit is used to input the patient's heart rate full-band feature cascade coding vector into a classifier-based abnormal recognition module to obtain the recognition result, and the recognition result is used to indicate whether there is an abnormal signal.

[0014] In a second aspect, a smart medical-patient information interaction method based on 5G technology is provided, which includes: using an IoT medical device worn by a patient to collect patient vital signs data, and collecting the patient's real-time audio and video data through a portable medical terminal; using a 5G communication module to transmit the patient's vital signs data and the real-time audio and video data to an edge server close to the patient; on the edge server, pre-processing the patient's vital signs data to determine whether there is an abnormal signal, and transmitting the patient's vital signs data and the real-time audio and video data to a cloud data center based on priority for medical staff to access; enabling the medical staff to receive the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device; wherein, on the edge server, pre-processing the patient's vital signs data to determine whether there is an abnormal signal, and transmitting the patient's vital signs data and the real-time audio and video data to the cloud data center based on priority for medical staff to access, including: extracting a time sequence of patient heart rate values ​​from the patient vital sign data; Performing variational mode decomposition and frequency domain feature extraction on the time queue of the patient's heart rate values ​​to obtain a set of frequency features of intrinsic mode components of the patient's heart rate; Performing feature hub-driven sparse processing on the set of frequency features of the intrinsic modal components of the patient's heart rate to obtain a sparse set of frequency features of the intrinsic modal components of the patient's heart rate; Based on the set of the sparsed frequency features of the intrinsic modal components of the patient's heart rate, an identification result is obtained.

[0015] The above technical solution is adopted to first extract the time queue of the patient's heart rate value from the patient's vital signs data, and use artificial intelligence-based data analysis and coding technology to perform variational modal decomposition of the patient's heart rate value. Then, the frequency domain feature extraction is performed on each patient's heart rate intrinsic modal component after decomposition, and then each extracted feature is subjected to feature hub sparsification processing, so as to intelligently generate an identification result of whether there is an abnormal signal by splicing the full frequency band according to the frequency characteristics of each sparse patient's heart rate intrinsic modal component, and prioritize the vital signs data and real-time audio and video data based on the identification result. In this way, the dynamic pattern of heart rate changes can be more comprehensively understood, avoiding the problem of false alarms or omissions that are prone to occur in traditional threshold methods.

[0016] Other features and advantages of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a block diagram of a smart medical-nursing-patient information interaction system based on 5G technology according to an embodiment of the present application.

[0019] Figure 2 This is a block diagram of the data processing module in the smart medical-nursing-patient information interaction system based on 5G technology according to an embodiment of the present application.

[0020] Figure 3 This is a block diagram of the patient heart rate frequency feature extraction unit in the smart medical and patient information interaction system based on 5G technology according to an embodiment of the present application.

[0021] Figure 4 This is a block diagram of the frequency feature sparseness unit in the smart medical and patient information interaction system based on 5G technology according to an embodiment of the present application.

[0022] Figure 5 This is a flowchart of a smart medical-nursing-patient information interaction method based on 5G technology according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0024] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0025] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0026] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0027] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0028] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0029] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0030] Based on this, this application proposes a smart medical-patient information interaction system based on 5G technology, which integrates advanced technologies such as the Internet of Things and 5G communications to achieve real-time and accurate collection and transmission of patient vital signs data, so that medical staff can fully grasp the daily health status of patients, thereby building a highly intelligent, responsive, safe and reliable remote medical service interaction platform. Specifically including: a patient information data acquisition module, which is used to collect patient vital signs data using the Internet of Things medical device worn by the patient, and collect the patient's real-time audio and video data through a portable medical terminal; a data transmission module, which is used to use the 5G communication module to transmit the patient's vital signs data and the real-time audio and video data to an edge server close to the patient; a data processing module, which is used to pre-process the patient's vital signs data on the edge server to determine whether there is an abnormal signal, and transmit the patient's vital signs data and the real-time audio and video data to the cloud data center based on priority for access by medical staff; a data synchronization module, which is used to enable the medical staff to receive the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device.

[0031] It should be understood that in a telemedicine environment, timely detection of abnormal signals is crucial. The preprocessing step can quickly detect changes in key indicators such as heart rate, helping the system to respond immediately, while the reasonable allocation of priorities can avoid resource waste and ensure the timely transmission of key data. However, traditional methods usually rely on preset thresholds or rules to judge abnormalities, lacking understanding of complex and dynamically changing patterns, which can easily lead to false positives or missed reports. In addition, traditional data transmission is based on a fixed priority, which may result in the inability to transmit vital signs data in a timely manner when the patient's condition is urgent, while real-time audio and video data occupies a large amount of bandwidth resources, delaying the transmission and real-time processing of key information.

[0032] Based on this, in the data processing module, the technical concept of the present application is to extract the time queue of the patient's heart rate value from the patient's vital signs data, use artificial intelligence-based data analysis and coding technology to perform variational modal decomposition of the patient's heart rate value, and then perform frequency domain feature extraction on each patient's heart rate intrinsic modal component after decomposition, and then perform feature hub sparsification processing on each extracted feature, so as to intelligently generate an identification result of whether there is an abnormal signal based on the frequency characteristics of each sparse patient's heart rate intrinsic modal component, and prioritize vital signs data and real-time audio and video data based on the identification result. The present application can more comprehensively understand the dynamic pattern of heart rate changes, avoiding the problem of false alarms or missed alarms that are prone to occur in traditional threshold methods. In addition, by dynamically adjusting the priority, the system ensures that the most important information is delivered first in an emergency, reducing the risk of delays.

[0033] Figure 1 is a block diagram of a smart medical-care-patient information interaction system based on 5G technology according to an embodiment of the present application, such as Figure 1 As shown, according to the embodiment of the present application, the smart medical-care-patient information interaction system 100 based on 5G technology includes: a patient information data acquisition module 110, which is used to collect patient vital signs data using an IoT medical device worn by the patient, and collect the patient's real-time audio and video data through a portable medical terminal; a data transmission module 120, which is used to use a 5G communication module to transmit the patient's vital signs data and the real-time audio and video data to an edge server close to the patient; a data processing module 130, which is used to perform data preprocessing on the patient's vital signs data on the edge server to determine whether there is an abnormal signal, and transmit the patient's vital signs data and the real-time audio and video data to a cloud data center based on priority for access by medical staff; a data synchronization module 140, which is used to enable the medical staff to receive the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device.

[0034] In the above-mentioned smart medical-nurse-patient information interaction system 100 based on 5G technology, the patient information data acquisition module 110 is used to collect the patient's vital signs data using the IoT medical device worn by the patient, and collect the patient's real-time audio and video data through a portable medical terminal. Specifically, in the technical solution of the present application, first, the time queue of the patient's heart rate value is extracted from the patient's vital signs data. It should be understood that heart rate is one of the key vital signs that reflects the patient's health status. It can sensitively reflect changes in the body's physiological state, such as exercise, mood swings, and the impact of disease. Among many vital signs data (such as body temperature, blood pressure, blood oxygen saturation, etc.), heart rate data plays an important role in assessing the patient's real-time health status, so extracting heart rate values ​​as the focus of analysis has strong data representativeness.

[0035] In the specific implementation process, the first thing involved is to use the medical devices worn by the patient to collect vital signs data. These devices include but are not limited to smart bracelets, smart watches, heart rate monitors, blood pressure monitors, blood glucose meters, etc. They are usually equipped with wireless communication functions (such as Bluetooth or Wi-Fi) and can directly connect to the patient's smartphone or other portable medical terminals. Such devices have built-in various sensors that can automatically measure and record the patient's key physiological parameters such as heart rate, blood oxygen saturation, body temperature, respiratory rate, etc. continuously or at a preset frequency. Taking the smart bracelet as an example, it uses photoplethysmography (PPG) technology to measure heart rate and combines accelerometers and other sensors to monitor physical activity levels. When a measurement is completed, the data obtained will be initially processed locally, such as compression and encryption, and then transmitted to the patient's portable medical terminal through a secure wireless communication protocol.

[0036] It is worth noting that during the entire data collection process, the system has taken a series of safeguards to ensure the quality of the data. On the one hand, the accuracy of the data provided by the device is maintained through a regular calibration mechanism, and users are guided to wear the device correctly and perform standardized tests according to the instructions. On the other hand, the application has a built-in anomaly detection function. Once data anomalies are found (such as signal loss, values ​​outside the normal range), the user will be immediately reminded to check the device settings or re-measure, and try to fix the problem.

[0037] In the above-mentioned smart medical-nurse-patient information interaction system 100 based on 5G technology, the data transmission module 120 is used to transmit the patient's vital signs data and the real-time audio and video data to an edge server close to the patient using a 5G communication module. In modern telemedicine services, 5G, as the fifth-generation mobile communication standard, provides unprecedented high-speed, low-latency and large-capacity connection capabilities, which make it an ideal data transmission medium. Especially in the smart medical scenario, the 5G network can support the simultaneous access of a large number of IoT devices and ensure stable performance even when the network load is high. Specifically, the 5G network can achieve a downlink rate of up to several Gb / s and a near real-time response time (milliseconds), which is crucial for transmitting high-definition video streams or large amounts of vital signs data. In addition, 5G introduces network slicing technology, which can allocate independent logical network resources according to the needs of different applications, thereby ensuring the quality of critical medical services.

[0038] The 5G communication module plays a bridging role in this process. It converts locally generated data packets into a format that can be transmitted on the 5G network and ensures that they arrive at their destination smoothly. This conversion process involves a series of complex protocol conversions and technical details, such as the operation of the TCP / IP protocol stack, the setting of QoS (Quality of Service) parameters, etc. At the same time, in order to ensure the security of the data, all transmitted data will be strictly encrypted, using protocols such as TLS (Transport Layer Security) to protect the data from eavesdropping and tampering. Once the data packets enter the 5G network, they will be quickly routed to the nearest base station and continue to move towards the target edge server from there. Thanks to the high reliability and wide coverage of the 5G network, stable data transmission can be achieved even in remote areas.

[0039] Unlike traditional cloud computing architecture, edge computing deploys computing resources closer to the data source, reducing the time it takes for data to travel back and forth to the cloud. For smart medical applications, this means faster data processing and lower latency, especially when instant feedback is required. Edge servers are usually located in telecommunications rooms inside or near hospitals. They are equipped with high-performance processors, ample storage space, and necessary network security facilities to cope with large-scale concurrent access and complex data processing tasks. When data packets from the 5G network arrive at the edge server, the server will perform preliminary verification and analysis on the data to confirm that its source is legitimate and the content is correct. Subsequently, based on pre-set rules or algorithms, the server will decide which data needs to be processed immediately and which can be temporarily archived for future reference. For example, if abnormal vital sign data is detected, the server will immediately forward the relevant information to the mobile devices of medical staff so that they can take timely action; for routine health monitoring data, it can be regularly summarized and analyzed to provide a basis for long-term health management. In addition, edge servers also undertake the task of docking with other medical information systems, such as electronic medical record systems (EMRs), hospital information systems (HISs), etc., thus building a complete remote medical service ecosystem.

[0040] In the above-mentioned smart medical-care-patient information interaction system 100 based on 5G technology, the data processing module 130 is used to pre-process the patient's vital signs data on the edge server to determine whether there are abnormal signals, and transmit the patient's vital signs data and the real-time audio and video data to the cloud data center based on priority for medical staff to access. Figure 2 : is a block diagram of the data processing module in the smart medical-nursing-patient information interaction system based on 5G technology according to an embodiment of the present application, such as Figure 2 As shown, the data processing module 130 includes: a patient heart rate value extraction unit 131, which is used to extract a time queue of patient heart rate values ​​from the patient's vital signs data; a patient heart rate frequency feature extraction unit 132, which is used to perform variational modal decomposition and frequency domain feature extraction on the time queue of the patient's heart rate values ​​to obtain a set of frequency features of the patient's heart rate inherent modal components; a frequency feature sparsification unit 133, which is used to perform feature hub-driven sparsification processing on the set of frequency features of the patient's heart rate inherent modal components to obtain a set of sparsed frequency features of the patient's heart rate inherent modal components; and a recognition result generation unit 134, which is used to obtain a recognition result based on the sparsed set of frequency features of the patient's heart rate inherent modal components.

[0041] In a specific embodiment of the present application, Figure 3: is a block diagram of the patient heart rate frequency feature extraction unit in the smart medical-nursing-patient information interaction system based on 5G technology according to an embodiment of the present application, such as Figure 3 As shown, the patient heart rate frequency feature extraction unit 132 includes: a variational modal decomposition subunit 1321, used for performing variational modal decomposition on the time queue of the patient heart rate values ​​to obtain a set of patient heart rate intrinsic modal components; an intrinsic modal feature extraction subunit 1322, used for extracting frequency domain features of each patient heart rate intrinsic modal component in the set of patient heart rate intrinsic modal components to obtain a set of patient heart rate intrinsic modal component frequency feature vectors as the set of patient heart rate intrinsic modal component frequency features.

[0042] Specifically, in the variational modal decomposition subunit 1321, it is considered that the heart rate signal is complex and changeable in actual situations. It not only contains information about the normal rhythm of the heart, but may also be interfered by various factors, such as breathing, body movement, muscle electrical activity, environmental noise, and the combined influence of various physiological and pathological factors. There may be a lot of potential information related to the patient's health status hidden in the heart rate signal, and this information may be distributed in different frequency ranges. For example, the low-frequency component of the heart rate signal may be related to the balance of the sympathetic and parasympathetic nerves of the autonomic nervous system, and the high-frequency component may be related to phenomena such as respiratory sinus arrhythmia. Therefore, in order to more clearly separate these frequency components related to the patient's health status and reduce the interference of noise, the present application performs variational modal decomposition on the time queue of the patient's heart rate value to obtain a set of intrinsic modal components of the patient's heart rate. It should be understood that variational modal decomposition can effectively solve the problem of modal mixing. It is based on the variational principle and adaptively decomposes complex signals into intrinsic modal components with different center frequencies, so that each component can more accurately represent a specific frequency component or signal mode. In this way, the time sequence of heart rate values ​​can be decomposed into a set of intrinsic modal components of the patient's heart rate, each of which represents a different frequency range or signal mode. In other words, the complex heart rate signal is decomposed into relatively simple sub-signals with clear physical or physiological meanings, so that each sub-signal can be analyzed separately in the future to more accurately analyze the potential information in each component.

[0043] Specifically, the intrinsic modal feature extraction subunit 1322 is used to extract the frequency domain features of each patient's heart rate intrinsic modal component in the set of the patient's heart rate intrinsic modal components using a frequency domain feature extractor based on dilated convolution to obtain a set of frequency feature vectors of the patient's heart rate intrinsic modal components. In order to further capture and mine the key features in different frequencies to reflect different physiological phenomena and pathological states, for example, for subtle fluctuations in the heart rate signal, it may appear as energy changes of certain frequencies in the frequency domain, and these changes may be early signs of some diseases. Based on this, the present application extracts the frequency domain features of each patient's heart rate intrinsic modal component in the set of the patient's heart rate intrinsic modal components using a frequency domain feature extractor based on dilated convolution to obtain a set of frequency feature vectors of the patient's heart rate intrinsic modal components. It should be understood that the conventional convolution operation has a limited receptive field when processing heart rate signals, and may ignore some long-distance information associations. The dilated convolution can expand the receptive field without increasing the amount of computation, and can capture a wider range of information. For data with time series characteristics such as heart rate signals, it can better understand the long-term dependencies of signals, which helps to extract frequency domain features more comprehensively, especially for the extraction of heart rate intrinsic modal component features with periodicity and long-range correlation.

[0044] In a specific embodiment of the present application, Figure 4 : is a block diagram of the frequency feature sparse unit in the smart medical-nurse-patient information interaction system based on 5G technology according to an embodiment of the present application, such as Figure 4 As shown, the frequency feature sparseness unit 133 includes: a feature distribution hub search subunit 1331, which is used to perform a feature distribution hub search on the set of the patient's heart rate intrinsic modal component frequency feature vectors to obtain the patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector; a frequency sparse semantic fine-grained ablation factor calculation subunit 1332, which is used to calculate each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vectors based on the patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector. There are sparse semantic fine-grained ablation factors of modal component frequency feature vectors to obtain a set of sparse semantic fine-grained ablation factors of the patient's heart rate intrinsic modal component frequency; a sparse subunit 1333 is used to perform feature sparsification on the set of patient's heart rate intrinsic modal component frequency feature vectors based on the set of sparse semantic fine-grained ablation factors of the patient's heart rate intrinsic modal component frequency to obtain the set of sparse patient's heart rate intrinsic modal component frequency feature vectors as the set of sparse patient's heart rate intrinsic modal component frequency features.

[0045] Furthermore, considering that the frequency feature set of the intrinsic modal component of the patient's heart rate may contain a large amount of data. There may be redundant information in these data, for example, there may be a high correlation between certain frequency features, or the importance of some features is low in the entire diagnostic process. In order to retain the most critical features for judging whether the patient's heart rate is abnormal or not, remove the influence of redundant information, and obtain a more representative representation, in the technical solution of the present application, the set of the frequency features of the intrinsic modal component of the patient's heart rate is subjected to a feature hub-driven sparse processing to obtain a sparse set of the frequency features of the intrinsic modal component of the patient's heart rate. In particular, the feature hub refers to those feature points or feature subsets that play a key role in the entire feature set. They are crucial to accurately judge the patient's heart rate state and whether there are abnormal signals. Therefore, by performing a feature hub-driven sparse processing, insignificant features can be removed, focusing on the key elements that best represent the heart rate characteristics, so as to more accurately reflect the core characteristics and potential abnormalities of the patient's heart rate.

[0046] Specifically, in the feature distribution hub search subunit 1331, the set of the patient's heart rate intrinsic modal component frequency feature vectors is first input into the feature distribution hub search network, which uses an adaptive learning strategy to explore and encode the key points or regions of the feature distribution in the set of the patient's heart rate intrinsic modal component frequency feature vectors to obtain the patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector. This dynamicity ensures that the network can flexibly identify and encode key features according to the changes in the input set of the patient's heart rate intrinsic modal component frequency feature vectors. In particular, in the technical solution of the present application, the patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector not only relies on static statistical characteristics, but may also combine high-order statistical information or nonlinear patterns to capture more complex distribution structures in the set of the patient's heart rate intrinsic modal component frequency feature vectors, aiming to find the key feature points in the low-dimensional embedding space that can best represent the original high-dimensional data.

[0047] The process can be expressed as follows: ; in, is the set of frequency eigenvectors of the intrinsic modal components of the patient's heart rate, , , , and are respectively the first, second, and third frequency characteristic vectors of the patient's heart rate intrinsic modal component set. , and The frequency eigenvector of the intrinsic modal component of the patient's heart rate, is the transpose operation of the vector, for and The inverse matrix of the covariance matrix between is the number of vectors in the set of frequency eigenvectors of the intrinsic modal components of the patient's heart rate minus one, is the first in the set of frequency characteristic weight coefficients of the intrinsic modal component of the patient's heart rate The frequency characteristic weight coefficient of the intrinsic modal component of the patient's heart rate, represents the exponential function value with the natural constant e as the base, is the number of weight coefficients in the set of weight coefficients of the frequency characteristics of the intrinsic modal components of the patient's heart rate, It is the hub encoding vector of the frequency characteristic distribution of the intrinsic modal components of the patient's heart rate.

[0048] The modal component frequency feature distribution hub encoding vector.

[0049] Specifically, in the frequency sparsification semantic fine-grained ablation factor calculation subunit 1332, after obtaining the feature distribution hub coding vector, the semantic fine-grained ablation factor relative to the patient's heart rate intrinsic modal component frequency feature distribution hub coding vector is calculated for each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vectors. The role of the semantic fine-grained ablation factor is to measure the relationship between the patient's heart rate intrinsic modal component frequency feature vector and the patient's heart rate intrinsic modal component frequency feature distribution hub coding vector, which draws on the idea of ​​contrastive learning to enhance the model's ability to distinguish important features. The semantic fine-grained ablation factor can be used as a feedback signal to guide model training, so that the network can gradually optimize the feature selection strategy in the iterative process, and the final set of the patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factors reflects the correlation or difference between each patient's heart rate intrinsic modal component frequency feature vector and the patient's heart rate intrinsic modal component frequency feature distribution hub coding vector, as well as the degree of influence on the overall expression when a specific feature is removed.

[0050] The process can be expressed as follows: ; in, is the Euclidean norm of the vector, is the adjustment parameter, To obtain the maximum value, is the first in the set of semantic fine-grained ablation factors of the intrinsic modal component frequency of the patient's heart rate. The semantic fine-grained ablation factor of the intrinsic modal component frequency of each patient’s heart rate, represents the exponential function value with the natural constant e as the base, is the first in the set of frequency eigenvectors of the intrinsic modal component of the patient's heart rate The frequency eigenvector of the intrinsic modal component of the patient's heart rate, It is the hub encoding vector of the frequency characteristic distribution of the intrinsic modal components of the patient's heart rate.

[0051] Subsequently, each patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factor in the set of the patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factors is input into the gating function-based sparsification module to obtain the set of the patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors. The gating function plays a crucial role here, determining which features should be retained and which should be suppressed. The gating mechanism allows the network to automatically adjust parameters during the learning process, thereby achieving dynamic sparsification. Sparsification is not only a process of reducing the number of redundant features, but more importantly, it is to improve the model interpretability and reduce the risk of overfitting. In the technical solution of the present application, other prior knowledge (specifically, the setting of thresholds) is also incorporated into the design of the sparsification module based on the gating function to simulate the information screening mechanism in the brain.

[0052] The process can be expressed as follows: ; in, is the first in the set of semantic fine-grained ablation factors of the intrinsic modal component frequency of the patient's heart rate. The semantic fine-grained ablation factor of the intrinsic modal component frequency of each patient’s heart rate, is a masking operation, is the preset threshold, It is the first in the set of semantic fine-grained ablation factors of the frequency sparseness of the intrinsic modal components of the patient's heart rate. Semantic fine-grained ablation factors for frequency sparsification of intrinsic modal components of heart rate of each patient.

[0053] Specifically, in the sparse subunit 1333, based on the set of the sparse semantic fine-grained ablation factors of the patient's heart rate intrinsic modal component frequency, the set of the patient's heart rate intrinsic modal component frequency feature vectors is feature-sparsed to obtain the set of the sparse patient's heart rate intrinsic modal component frequency feature vectors. In a specific example, the set of the patient's heart rate intrinsic modal component frequency feature vectors is weighted with the sparse semantic fine-grained ablation factors of the patient's heart rate intrinsic modal component frequency as weights, thereby constructing the sparse patient's heart rate intrinsic modal component frequency feature vector set. Using the patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors as weights for weighted synthesis essentially maps the importance of features to a numerical range, so that those features that are considered more valuable get higher weights, while less important features have lower weights. The set of the sparse patient's heart rate intrinsic modal component frequency feature vectors processed in this way should have better interpretability and higher expressiveness, because these features have been carefully selected and quantitatively evaluated. The weighted synthesized set of sparse patient heart rate intrinsic modal component frequency feature vectors can be directly used for downstream tasks such as classification, clustering or regression analysis, or may be used as an intermediate layer output for use by a higher-level model.

[0054] The process can be expressed as follows: ; in, is the first in the set of the frequency feature vectors of the intrinsic modal component of the patient's heart rate after the sparse The sparsed frequency eigenvector of the intrinsic modal component of the patient’s heart rate, is the set of frequency eigenvectors of the intrinsic modal components of the patient's heart rate after the sparse treatment, It is the first in the set of semantic fine-grained ablation factors of the frequency sparseness of the intrinsic modal components of the patient's heart rate. The frequency sparse semantic fine-grained ablation factor of the intrinsic modal component of the heart rate of each patient, is the first in the set of frequency eigenvectors of the intrinsic modal component of the patient's heart rate The frequency eigenvector of the intrinsic modal components of the patient's heart rate.

[0055] In one embodiment of the present application, the recognition result generating unit 134 includes: a vector splicing subunit 1341, which is used to splice the set of the sparsed patient heart rate intrinsic modal component frequency feature vectors to obtain a patient heart rate full-band feature cascade coding vector; an abnormal signal judgment subunit 1342, which is used to obtain a recognition result based on the patient heart rate full-band feature cascade coding vector, and the recognition result is used to indicate whether there is an abnormal signal; wherein, in response to the recognition result being the presence of an abnormal signal, the patient's vital signs data has a higher priority than the real-time audio and video data; in response to the recognition result being the absence of an abnormal signal, the real-time audio and video data has a higher priority than the patient's vital signs data.

[0056] Specifically, in the vector splicing subunit 1341, it is considered that each intrinsic modal component frequency feature vector of the patient's heart rate represents the key frequency domain feature information in a specific frequency domain. However, a single intrinsic modal component frequency feature vector can only reflect the characteristics of the heart rate signal in a specific frequency range or mode. Therefore, in the technical solution of the present application, the set of the sparse patient heart rate intrinsic modal component frequency feature vectors is spliced ​​to integrate information from different frequency ranges and characteristic modes, thereby comprehensively covering the full-band characteristics of the heart rate signal and obtaining the patient's heart rate full-band characteristic cascade coding vector. That is, different intrinsic modal components correspond to the low-frequency, medium-frequency and high-frequency parts respectively, and after splicing, they can comprehensively reflect the changes in the heart rate in the entire frequency range, thereby more completely representing the essential characteristics of the original heart rate signal, thereby significantly improving the ability to recognize abnormal signals.

[0057] Specifically, in the abnormal signal judgment subunit 1342, the patient's heart rate full-band feature cascade coding vector is input into the abnormal recognition module based on the classifier to obtain a recognition result, and the recognition result is used to indicate whether there is an abnormal signal, wherein in response to the recognition result that an abnormal signal exists, the priority of the patient's vital signs data is higher than the real-time audio and video data; in response to the recognition result that there is no abnormal signal, the priority of the real-time audio and video data is higher than the patient's vital signs data. That is, the patient's heart rate full-band feature cascade coding vector obtained by full-band splicing using the set of the sparse patient's heart rate intrinsic modal component frequency feature vectors is classified and processed, so as to use the powerful pattern recognition ability of the classifier to accurately judge whether the heart rate data is abnormal. That is, when the recognition result is that an abnormal signal exists, the priority of the patient's vital signs data is increased, which can ensure that medical staff can obtain the most critical physiological data in the first time, so as to make diagnosis and treatment decisions quickly. For example, when a patient experiences an emergency such as arrhythmia, vital signs data such as heart rate and blood pressure are crucial for judging the severity of the disease and taking emergency measures. Prioritizing the transmission of this data can buy precious time for treatment for the patient.

[0058] In a preferred example, inputting the patient's heart rate full-band feature cascade coding vector into a classifier-based abnormality recognition module to obtain a recognition result includes: Determine the eigenvalue mean corresponding to the patient's heart rate full-band feature cascade coding vector and the standard deviation of the eigenvalues ; The point-wise subtraction vector of the patient's heart rate full-band feature concatenated coding vector and the eigenvalue mean and the eigenvalue standard deviation are multiplied to obtain the first patient's heart rate full-band feature concatenated coding fair target vector: ; in, is the patient's heart rate full-band feature cascade encoding vector, is the mean of the eigenvalues, is the eigenvalue standard deviation, is the first patient's heart rate full-band feature concatenation encoding fair target vector, It is subtracted by position point. It is the point product by position.

[0059] The point-wise multiplication of the patient's heart rate full-band feature cascade coding vector and the eigenvalue standard deviation vector and the eigenvalue mean is performed to obtain a second patient's heart rate full-band feature cascade coding fair target vector: ; in, is the patient's heart rate full-band feature cascade encoding vector, is the mean of the eigenvalues, is the eigenvalue standard deviation, is the fair target vector of the second patient's heart rate full-band feature concatenation encoding, It is subtracted by position point. It is the point product by position.

[0060] After the bit-by-bit inverse of the second patient heart rate full-band feature cascade coding fair target vector is dot-multiplied by the first patient heart rate full-band feature cascade coding fair target vector, the bit-by-bit logarithm with base 2 is taken to obtain the patient heart rate full-band feature cascade coding information correction vector: ; in, is the first patient's heart rate full-band feature concatenation encoding fair target vector, is the fair target vector of the second patient's heart rate full-band feature concatenation encoding, is the bit-wise inverse of the fair target vector of the second patient's heart rate full-band feature concatenation encoding, is the correction vector of the patient's heart rate full-band feature cascade coding information, It is the point product by position.

[0061] The eigenvalue mean Divide by the standard deviation of the eigenvalue The square root of the quotient is multiplied by the weight hyperparameter, and then the quotient is multiplied by the patient's heart rate full-band feature cascade coding information correction vector to obtain the optimized patient heart rate full-band feature cascade coding vector: ; in, is the correction vector of the patient's heart rate full-band feature cascade coding information, is the optimized patient heart rate full-band feature cascade encoding vector, is the mean of the eigenvalues, is the eigenvalue standard deviation, is the weight hyperparameter, It is added by location point.

[0062] The optimized patient heart rate full-band feature cascade coding vector is input into a classifier-based abnormality recognition module to obtain a recognition result.

[0063] Here, since each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vectors respectively represents the frequency domain characteristics of the local time variational mode of the patient's heart rate value, when the features are thinned out, the frequency domain feature population attributes of different time variational modes will have local representation fairness differences based on different feature sparsification levels, thereby affecting the cascade coding feature distribution inclusiveness of the spliced ​​patient's heart rate full-band feature cascade coding vector, and reducing the accuracy of the recognition result obtained by inputting the classifier-based abnormality recognition module.

[0064] Therefore, taking into account the attribute level fairness differences of the data population corresponding to the sequence fusion features of the patient's heart rate full-band feature cascade coding vector, in order to improve the inclusiveness of the patient's heart rate full-band feature cascade coding vector under the feature distribution diversity, the crossover probability value constraint based on the patient's heart rate full-band feature cascade coding vector is used as the interactive fairness target representation to correct the group feature information interaction propagation of the patient's heart rate full-band feature cascade coding vector, and the unified statistical feature response interaction based on the patient's heart rate full-band feature cascade coding vector is used as the feature distribution multi-level fairness target bias to achieve a robust distribution fairness unified representation of the patient's heart rate full-band feature cascade coding vector, form a fair collaboration paradigm under the feature distribution framework of the patient's heart rate full-band feature cascade coding vector, and improve the accuracy of the recognition results obtained by inputting the classifier-based abnormality recognition module.

[0065] In one embodiment of the present application, Figure 5 Flow chart of the intelligent medical-patient information interaction method based on 5G technology according to an embodiment of the present application, as shown in Figure 5 As shown, according to the embodiment of the present application, the smart medical-patient information interaction method based on 5G technology includes: S210, using the Internet of Things medical device worn by the patient to collect the patient's vital signs data, and collecting the patient's real-time audio and video data through a portable medical terminal; S220, using the 5G communication module to transmit the patient's vital signs data and the real-time audio and video data to an edge server close to the patient; S230, on the edge server, pre-processing the patient's vital signs data to determine whether there is an abnormal signal, and transmitting the patient's vital signs data and the real-time audio and video data to a cloud data center based on priority for medical staff to access; S240, enabling the medical staff to receive the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device; wherein, S230, on the edge server, pre-processing the patient's vital signs data to determine whether there is an abnormal signal, and transmitting the patient's vital signs data and the real-time audio and video data to the cloud data center based on priority for medical staff to access, including: S231, extracting a time queue of a patient's heart rate value from the patient's vital sign data; S232, performing variational mode decomposition and frequency domain feature extraction on the time queue of the patient's heart rate values ​​to obtain a set of frequency features of intrinsic mode components of the patient's heart rate; S233, performing feature hub-driven sparse processing on the set of frequency features of the intrinsic modal components of the patient's heart rate to obtain a sparse set of frequency features of the intrinsic modal components of the patient's heart rate; S234, obtaining a recognition result based on the set of sparsed frequency features of the intrinsic modal components of the patient's heart rate.

[0066] Those skilled in the art will appreciate that the specific operations of each step in the above-mentioned smart medical-nurse-patient information interaction method based on 5G technology have been referenced above. Figures 1 to 4 It has been introduced in detail in the description of the smart medical and patient information interaction system 100 based on 5G technology, and therefore, its repeated description will be omitted.

[0067] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0068] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0069] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A smart medical-nurse-patient information interaction system based on 5G technology, characterized in that: include: A patient information data collection module is used to collect patient vital sign data using an IoT medical device worn by the patient, and to collect real-time audio and video data of the patient through a portable medical terminal; A data transmission module, used to transmit the patient's vital sign data and the real-time audio and video data to an edge server close to the patient using a 5G communication module; A data processing module, configured to perform data preprocessing on the patient's vital signs data on the edge server to determine whether there are abnormal signals, and transmit the patient's vital signs data and the real-time audio and video data to a cloud data center based on priority for access by medical staff; A data synchronization module is used to enable the medical staff to receive the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device; wherein the data processing module includes: a patient heart rate value extraction unit, configured to extract a time queue of a patient heart rate value from the patient vital sign data; A patient heart rate frequency feature extraction unit, used for performing variational modal decomposition and frequency domain feature extraction on the time queue of the patient's heart rate values ​​to obtain a set of frequency features of the patient's heart rate intrinsic modal components; A frequency feature sparseness unit, used for performing a feature hub-driven sparseness process on the set of frequency features of the intrinsic modal components of the patient's heart rate to obtain a sparse set of frequency features of the intrinsic modal components of the patient's heart rate; The recognition result generating unit is used to obtain the recognition result based on the set of the sparsed frequency characteristics of the intrinsic modal components of the patient's heart rate.

2. The smart medical-nurse-patient information interaction system based on 5G technology according to claim 1 is characterized in that: The patient heart rate frequency feature extraction unit comprises: a variational modal decomposition subunit, configured to perform variational modal decomposition on the time sequence of the patient's heart rate values ​​to obtain a set of intrinsic modal components of the patient's heart rate; The intrinsic modal feature extraction subunit is used to extract the frequency domain features of each patient's heart rate intrinsic modal component in the set of the patient's heart rate intrinsic modal components to obtain a set of patient's heart rate intrinsic modal component frequency feature vectors as the set of patient's heart rate intrinsic modal component frequency features.

3. The intelligent medical-nursing-patient information interaction system based on 5G technology according to claim 2 is characterized in that: The intrinsic modal feature extraction subunit is used to: use a frequency domain feature extractor based on dilated convolution to extract the frequency domain features of each patient's heart rate intrinsic modal component in the set of patient's heart rate intrinsic modal components to obtain a set of patient's heart rate intrinsic modal component frequency feature vectors.

4. The smart medical-nursing-patient information interaction system based on 5G technology according to claim 3 is characterized in that: The frequency feature sparseness unit comprises: A feature distribution hub search subunit, used for performing a feature distribution hub search on the set of the patient's heart rate intrinsic modal component frequency feature vectors to obtain a patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector; A frequency sparse semantic fine-grained ablation factor calculation subunit, used to calculate the sparse semantic fine-grained ablation factor of each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vectors based on the patient's heart rate intrinsic modal component frequency feature distribution hub coding vector to obtain a set of patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors; The sparsification subunit is used to perform feature sparsification on the set of frequency feature vectors of the intrinsic modal components of the patient's heart rate based on the set of semantic fine-grained ablation factors of the sparseness of the frequency of the intrinsic modal components of the patient's heart rate to obtain a set of sparsed frequency feature vectors of the intrinsic modal components of the patient's heart rate as the set of sparsed frequency features of the intrinsic modal components of the patient's heart rate.

5. The smart medical-nurse-patient information interaction system based on 5G technology according to claim 4 is characterized in that: The frequency sparse semantic fine-grained ablation factor calculation subunit is used to: Respectively calculating the semantic fine-grained ablation factor of each patient's heart rate intrinsic modal component frequency feature vector in the set of the patient's heart rate intrinsic modal component frequency feature vector relative to the patient's heart rate intrinsic modal component frequency feature distribution hub encoding vector to obtain a set of patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factors; A gating function is performed on each patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factor in the set of the patient's heart rate intrinsic modal component frequency semantic fine-grained ablation factors to obtain the set of the patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors.

6. The intelligent medical-nursing-patient information interaction system based on 5G technology according to claim 5 is characterized in that: The sparsification subunit is used to: use each patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factor in the set of patient's heart rate intrinsic modal component frequency sparse semantic fine-grained ablation factors as a weight, and weight the set of patient's heart rate intrinsic modal component frequency feature vectors respectively to obtain the sparse set of patient's heart rate intrinsic modal component frequency feature vectors.

7. The intelligent medical-nursing-patient information interaction system based on 5G technology according to claim 6 is characterized in that: The recognition result generating unit comprises: A vector concatenation subunit, used for concatenating the set of the sparsed patient heart rate intrinsic modal component frequency feature vectors to obtain a full-band feature cascade coding vector of the patient's heart rate; An abnormal signal judging subunit is used to obtain a recognition result based on the full-band feature cascade coding vector of the patient's heart rate, and the recognition result is used to indicate whether there is an abnormal signal; Wherein, in response to the recognition result that an abnormal signal exists, the priority of the patient's vital sign data is higher than the real-time audio and video data; In response to the recognition result that there is no abnormal signal, the real-time audio and video data has a higher priority than the patient's vital sign data.

8. The smart medical-nurse-patient information interaction system based on 5G technology according to claim 7 is characterized in that: The abnormal signal judgment subunit is used to input the patient's heart rate full-band feature cascade coding vector into the classifier-based abnormal recognition module to obtain the recognition result, and the recognition result is used to indicate whether there is an abnormal signal.

9. A smart medical-nurse-patient information interaction method based on 5G technology, characterized in that: include: The patient's vital signs data are collected using the IoT medical device worn by the patient, and the patient's real-time audio and video data are collected through a portable medical terminal; Using a 5G communication module to transmit the patient's vital sign data and the real-time audio and video data to an edge server close to the patient; On the edge server, preprocessing the patient's vital sign data to determine whether there is an abnormal signal, and transmitting the patient's vital sign data and the real-time audio and video data to a cloud data center based on priority for access by medical staff; The medical staff receives the patient's vital signs data and the real-time audio and video data synchronized by the cloud data center in real time through a mobile device; wherein, on the edge server, the patient's vital signs data is pre-processed to determine whether there is an abnormal signal, and the patient's vital signs data and the real-time audio and video data are transmitted to the cloud data center based on priority for access by medical staff, including: extracting a time sequence of patient heart rate values ​​from the patient vital sign data; Performing variational mode decomposition and frequency domain feature extraction on the time queue of the patient's heart rate values ​​to obtain a set of frequency features of intrinsic mode components of the patient's heart rate; Performing feature hub-driven sparse processing on the set of frequency features of the intrinsic modal components of the patient's heart rate to obtain a sparse set of frequency features of the intrinsic modal components of the patient's heart rate; Based on the set of the sparsed frequency features of the intrinsic modal components of the patient's heart rate, an identification result is obtained.

Images