A method and apparatus for integrating surgical bed measurement data

Abstract

This application belongs to the field of medical data processing technology and discloses a method and device for integrating measurement data from an operating table. The method includes: acquiring patient physiological data collected by multiple sensors on the operating table and preprocessing it to obtain preprocessed physiological data; dividing the preprocessed physiological data into key physiological data and non-key physiological data; monitoring the network status information of the wireless network, and dynamically adjusting the allocated bandwidth of key physiological data and non-key physiological data according to the network status information, the amount of data of key physiological data, and the amount of data of non-key physiological data to obtain a bandwidth allocation scheme; determining the data transmission strategy of key physiological data and non-key physiological data according to the bandwidth allocation scheme; and transmitting data to a remote medical center according to the data transmission strategy. Thus, in the mobile emergency unit environment with limited and unstable network bandwidth, the data transmission strategy can be dynamically adjusted to ensure the real-time performance and reliability of key physiological data transmission.

Description

Technical Field

[0001] This application relates to the field of medical data processing technology, and more specifically, to a method and apparatus for integrating operating table measurement data. Background Technology

[0002] In responding to public health emergencies, mobile emergency units have become a key means of rapidly increasing medical resources and have been widely adopted. These mobile units typically deploy multi-sensor operating tables to receive and initially assess patients' conditions. Especially in emergency scenarios, patients' conditions are often critical, generating an extremely large amount of data, including multiple real-time physiological parameters such as electrocardiogram, blood pressure, and blood oxygen saturation. Telemedicine centers need to monitor patient data from multiple operating tables in real time to guide on-site medical staff in treatment and make timely decisions regarding the need for emergency surgery. However, mobile emergency units typically have limited computing resources and heavily rely on temporarily deployed wireless networks for data transmission. This results in limited network bandwidth and susceptibility to external environmental interference, leading to complex and unstable network conditions.

[0003] Existing data integration systems face numerous technical challenges in mobile emergency units. First, the fluctuating bandwidth of wireless networks makes data transmission rates highly unstable, making it difficult to simultaneously guarantee the integrity and real-time nature of all medical data. Simply increasing data compression rates to adapt to limited bandwidth may lead to the loss of critical medical information, severely impacting the accuracy of remote diagnosis and even delaying patient assessment and treatment. Second, traditional static priority queue scheduling mechanisms may not effectively cope with the drastic changes in network bandwidth in emergency scenarios. Static priority allocation strategies cannot dynamically adjust the transmission priorities of various data types based on real-time network conditions, easily causing increased latency for high-priority data transmission and excessive bandwidth consumption by low-priority data, resulting in the unreliable and untimely transmission of critical emergency data.

[0004] Therefore, in special application scenarios such as mobile emergency units where computing resources are limited, wireless network environments are complex, and real-time data requirements are extremely high, there is an urgent need for an operating table measurement data integration system capable of dynamically adjusting data transmission strategies based on wireless network conditions and real-time data demands. This system needs to intelligently identify the priority of different types of medical data and dynamically adjust the transmission strategies for various data types based on real-time network bandwidth and congestion levels. This ensures that critical physiological data can be transmitted with low latency and high reliability, while also accommodating the transmission needs of non-critical physiological data, maintaining ease of use and operational stability, and ultimately effectively guaranteeing the real-time performance and accuracy of remote emergency care.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] The purpose of this application is to provide a method and apparatus for integrating operating table measurement data, which can dynamically adjust the data transmission strategy in a mobile emergency unit environment with limited and unstable network bandwidth, so as to ensure the real-time performance and reliability of key physiological data transmission.

[0007] In a first aspect, this application provides a method for integrating operating table measurement data for use in an edge computing module within a mobile emergency unit. The method includes the following steps:

[0008] A1. Acquire patient physiological data collected by multiple sensors on the operating table, perform preprocessing, and obtain preprocessed physiological data;

[0009] A2. Classify the preprocessed physiological data into key physiological data and non-key physiological data;

[0010] A3. Monitor the network status information of the wireless network, and dynamically adjust the bandwidth allocation for key and non-key physiological data based on the network status information, the amount of key physiological data, and the amount of non-key physiological data to obtain a bandwidth allocation scheme; the network status information includes available bandwidth and network congestion level.

[0011] A4. Based on the bandwidth allocation scheme, determine the data transmission strategy for critical physiological data and non-critical physiological data;

[0012] A5. Data is transmitted to the telemedicine center according to the data transmission strategy.

[0013] Furthermore, this application also proposes that step A2 includes:

[0014] A201. Extract the feature vectors of the preprocessed physiological data;

[0015] A202. For each preprocessed physiological data, calculate the similarity between the feature vector of the preprocessed physiological data and the typical feature vectors of the corresponding physiological data in the physiological data feature library, and obtain a similarity set; the physiological data feature library records the typical feature vectors of each physiological data under various emergency situations;

[0016] A203. Based on the similarity set of each preprocessed physiological data, the preprocessed physiological data are determined to be either key physiological data or non-key physiological data.

[0017] Furthermore, this application also proposes that step A201 includes:

[0018] For each preprocessed physiological data, a feature extraction algorithm based on wavelet transform is used to decompose it into wavelet coefficients in multiple frequency bands;

[0019] Calculate the energy value of the wavelet coefficients for each frequency band to form an energy feature vector;

[0020] The energy feature vector is normalized to obtain the feature vector of each preprocessed physiological data.

[0021] Furthermore, this application also proposes that step A203 includes:

[0022] If at least one similarity in the similarity set is greater than the preset similarity threshold, then the corresponding preprocessed physiological data is determined to be key physiological data.

[0023] If the similarity scores in the similarity set are all no greater than the preset similarity threshold, then the probability that the preprocessed physiological data is physiological data under unknown emergency conditions is estimated based on the similarity set.

[0024] If the probability is greater than the preset probability threshold, the corresponding preprocessed physiological data is determined to be critical physiological data; otherwise, the corresponding preprocessed physiological data is determined to be non-critical physiological data.

[0025] Furthermore, this application also proposes that if the similarity scores in the similarity set are all no greater than a preset similarity threshold, then the step of estimating the probability that the corresponding preprocessed physiological data is physiological data under an unknown emergency situation based on the similarity set includes:

[0026] Calculate the variance of the similarity set to obtain the similarity variance;

[0027] If the similarity variance is less than the preset variance threshold, the Bayesian estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set.

[0028] If the similarity variance is not less than the preset variance threshold, then the kernel density estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set.

[0029] Furthermore, this application also proposes that step A3 includes:

[0030] A301. Monitor the available bandwidth and network congestion level of the wireless network, and calculate the network quality score;

[0031] A302. Obtain the amount of key physiological and non-key physiological data, and calculate the ratio of key physiological data to non-key physiological data.

[0032] A303. Based on the network quality score and data volume ratio, a fuzzy control algorithm is used to determine the target bandwidth allocation ratio for critical physiological data and non-critical physiological data;

[0033] A304. Based on the target bandwidth allocation ratio and the available bandwidth of the wireless network, calculate the allocated bandwidth for critical physiological data and the allocated bandwidth for non-critical physiological data to obtain the bandwidth allocation scheme.

[0034] Furthermore, this application also proposes that step A303 includes:

[0035] Based on the network quality score and the data volume ratio, a fuzzy inference algorithm is used to select applicable fuzzy rules from the fuzzy rule base. The activation strength of each fuzzy rule is calculated based on the network quality score and the data volume ratio to obtain the fuzzy inference result. The fuzzy rule base contains multiple fuzzy rules, and the fuzzy rules define the mapping relationship between the network quality score, the data volume ratio, and the target bandwidth allocation ratio.

[0036] The fuzzy inference results are defuzzified by weighted average. Based on the activation intensity of each fuzzy rule and the corresponding target bandwidth allocation ratio, the target bandwidth allocation ratio of key physiological data and non-key physiological data is calculated.

[0037] Furthermore, this application also proposes that step A4 includes:

[0038] A401. Based on the bandwidth allocation scheme, determine the priority queues for critical physiological data and non-critical physiological data;

[0039] A402. For each priority queue, the data is divided into multiple data packets according to the amount of data to be transmitted in the queue, and the maximum transmission unit of each data packet is determined according to the bandwidth allocation scheme.

[0040] A403. According to the priority order of the priority queue, schedule the data packets in the queue, encapsulate the data packets according to the determined maximum transmission unit, use Turbo codes for error control encoding of data packets containing critical physiological data, and use parity check codes for error control encoding of data packets containing non-critical physiological data.

[0041] Furthermore, this application also proposes that, prior to step A401, the following step is included:

[0042] A400a. Calculate the ratio of the allocated bandwidth of non-critical physiological data to the amount of non-critical physiological data, denoted as the specific density;

[0043] A400b. If the specific density is lower than the preset specific density threshold, at least some non-critical physiological data are selected as replacement objects, and the feature data of the replacement objects are extracted to replace the corresponding non-critical physiological data so that the specific density is not lower than the preset specific density threshold.

[0044] Secondly, this application also proposes an operating table measurement data integration device for an edge computing module in a mobile emergency unit, the device comprising:

[0045] The data acquisition module is used to acquire patient physiological data collected by multiple sensors on the operating table, perform preprocessing, and obtain preprocessed physiological data.

[0046] The data classification module is used to classify the preprocessed physiological data into key physiological data and non-key physiological data.

[0047] The bandwidth allocation module is used to monitor the network status information of the wireless network. Based on the network status information, the amount of key physiological data and the amount of non-key physiological data, it dynamically adjusts the allocated bandwidth for key physiological data and non-key physiological data to obtain a bandwidth allocation scheme. The network status information includes available bandwidth and network congestion level.

[0048] The transmission strategy adjustment module is used to determine the data transmission strategy for critical physiological data and non-critical physiological data based on the bandwidth allocation scheme.

[0049] The data transmission module is used to transmit data to the telemedicine center according to the data transmission strategy.

[0050] Beneficial effects: The operating table measurement data integration method and device provided in this application, by classifying physiological data and dynamically adjusting bandwidth allocation and transmission strategies according to network status, has the advantages of realizing real-time and reliable transmission of key physiological data in mobile emergency units with limited network, thereby improving the efficiency and accuracy of telemedicine. Attached Figure Description

[0051] Figure 1 A flowchart illustrating the method for integrating operating table measurement data provided in this application embodiment.

[0052] Figure 2 A schematic diagram of the structure of the operating table measurement data integration device provided in the embodiments of this application.

[0053] Labeling Explanation: 1. Data Acquisition Module; 2. Data Classification Module; 3. Bandwidth Allocation Module; 4. Transmission Strategy Adjustment Module; 5. Data Transmission Module. Detailed Implementation

[0054] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0055] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0056] refer to Figure 1 This application proposes a method for integrating operating table measurement data for use in an edge computing module within a mobile emergency unit. The method includes the following steps:

[0057] A1. Acquire patient physiological data collected by multiple sensors on the operating table, perform preprocessing, and obtain preprocessed physiological data;

[0058] A2. Classify the preprocessed physiological data into key physiological data and non-key physiological data;

[0059] A3. Monitor the network status information of the wireless network, and dynamically adjust the bandwidth allocation for key and non-key physiological data based on the network status information, the amount of key physiological data, and the amount of non-key physiological data to obtain a bandwidth allocation scheme; the network status information includes available bandwidth and network congestion level.

[0060] A4. Based on the bandwidth allocation scheme, determine the data transmission strategy for critical physiological data and non-critical physiological data;

[0061] A5. Data is transmitted to the telemedicine center according to the data transmission strategy.

[0062] In step A1, multiple sensors on the operating table collect the patient's physiological data, including electrocardiogram (ECG), blood pressure, blood oxygen saturation, heart rate, and body temperature. An edge computing module processes the physiological data collected by the sensors locally before sending it to a remote medical center. Then, the raw physiological data undergoes preprocessing, which may include noise removal, interference filtering, data smoothing, baseline correction, outlier handling, data format conversion, and unit standardization, to improve data quality and lay the foundation for subsequent data classification and transmission. Different physiological data may employ different preprocessing methods. For example, ECG data is filtered using a bandpass filter to remove power frequency interference and electromyographic noise; blood pressure and heart rate data undergo outlier detection and correction to eliminate measurement errors; blood oxygen saturation data undergoes baseline drift correction; and body temperature data is smoothed to reduce noise interference. The preprocessed physiological data provides higher-quality and more standardized data input for subsequent steps.

[0063] In step A2, the preprocessed physiological data is classified to distinguish between critical and non-critical physiological data. This can be achieved by pre-setting classification rules or models based on characteristics such as the type and importance of the physiological data. For example, data directly reflecting a patient's vital signs, such as electrocardiogram (ECG) and blood pressure data, can be classified as critical physiological data, while data such as body temperature and heart rate can be classified as non-critical physiological data. The purpose of this classification is to allow for differentiated processing strategies for different types of data in subsequent steps, ensuring that critical data is prioritized for processing and transmission, while non-critical data is also transmitted if network resources permit, thus balancing data integrity and transmission efficiency. The data classification results provide a data foundation for subsequent dynamic bandwidth allocation and transmission strategy adjustments.

[0064] In step A3, the wireless network status is monitored, and the bandwidth allocation for critical and non-critical physiological data is dynamically adjusted based on the network status and the data volume of various types of data. This can be achieved using the following methods: First, network monitoring tools or protocols are used to monitor the available bandwidth and network congestion levels of the wireless network in real time. This network status information reflects the current quality and transmission capacity of the wireless network. Then, the size of the critical and non-critical physiological data to be transmitted is used as a reference for bandwidth allocation. The larger the data volume, the higher the bandwidth required. Furthermore, a dynamic bandwidth adjustment algorithm can be designed, such as a fuzzy control algorithm or a PID control algorithm, to dynamically calculate the bandwidth ratio or specific bandwidth value that should be allocated to critical and non-critical physiological data based on the network status information and data volume, generating a bandwidth allocation scheme. This bandwidth allocation scheme reflects how to rationally allocate limited bandwidth resources under the current network conditions to maximize the priority transmission of critical data while also considering the transmission needs of non-critical data. Dynamic bandwidth adjustment is the core of this scheme and can effectively cope with the instability of the wireless network environment.

[0065] In step A4, based on the bandwidth allocation scheme obtained in step A3, specific transmission strategies for critical and non-critical physiological data are determined. Specifically, the transmission strategy may include determining some or all of the following: the priority of various data types, packet size, transmission rate, encoding method, and retransmission mechanism. For example, a higher transmission priority can be set for critical physiological data to ensure priority transmission during network congestion; the packet size can be adjusted according to the allocated bandwidth to avoid excessively large packets causing congestion in low-bandwidth networks; different error control encoding methods can be used, such as using more reliable Turbo codes for critical data and less expensive parity check codes for non-critical data. The transmission strategy puts the bandwidth allocation scheme into practice, ensuring the reliability and real-time performance of data transmission through refined transmission control.

[0066] In step A5, according to the data transmission strategy determined in step A4, the classified physiological data is transmitted to the telemedicine center. Specifically, wireless communication technologies such as Wi-Fi, 4G / 5G, etc., can be used to send the data to the telemedicine center's server via a wireless network. During data transmission, the transmission strategy determined in step A4 must be strictly followed, such as sending data packets in priority order and using a set encoding method to ensure that the data is transmitted according to the expected strategy. After receiving the physiological data transmitted from the operating table, the telemedicine center allows medical staff to monitor the patient's physiological status in real time, providing data support for remote diagnosis and treatment. Data transmission is the ultimate goal of this solution, realizing the remote sharing and application of physiological data.

[0067] Specifically, this solution aims to address the limitations and instability of wireless network bandwidth in specific application scenarios of mobile emergency units. First, step A1 completes initial data collection and preprocessing, preparing a high-quality data source for subsequent operations. Step A2 classifies the data, distinguishing between critical and non-critical data, a prerequisite for differentiated processing. Step A3 is the core step, dynamically adjusting the bandwidth allocation scheme based on real-time network status monitoring and data volume. This dynamic adjustment mechanism enables the system to adapt to fluctuating wireless network environments, ensuring that critical data receives sufficient bandwidth resources for priority transmission under any network conditions. Step A4 translates the bandwidth allocation scheme into specific transmission strategies, refining data transmission control, such as priority queues, packet size, and encoding methods, ensuring the executability and effectiveness of the transmission strategy. Finally, step A5 transmits the data to the telemedicine center, realizing the final application value of the data.

[0068] Through the above technical solution, this application achieves low-latency, high-reliability transmission of critical physiological data under conditions of limited and unstable wireless network bandwidth in mobile emergency units, while also considering the transmission needs of non-critical physiological data. The dynamic bandwidth allocation mechanism can adaptively adjust according to changes in network conditions, ensuring that critical data is always transmitted with priority under various network conditions. This improves the reliability and real-time performance of data transmission, provides stable and reliable data support for telemedicine, and enhances the efficiency and accuracy of remote emergency care.

[0069] Furthermore, this application also proposes that step A2 includes:

[0070] A201. Extract the feature vectors of the preprocessed physiological data;

[0071] A202. For each preprocessed physiological data, calculate the similarity between the feature vector of the preprocessed physiological data and the typical feature vectors of the corresponding physiological data in the physiological data feature library, and obtain a similarity set; the physiological data feature library records the typical feature vectors of each physiological data under various emergency situations;

[0072] A203. Based on the similarity set of each preprocessed physiological data, the preprocessed physiological data are determined to be either key physiological data or non-key physiological data.

[0073] Step A201 involves extracting feature vectors from the preprocessed physiological data. This can be achieved using wavelet transform-based feature extraction algorithms, decomposing the data into wavelet coefficients across multiple frequency bands. Alternatively, Fourier transform, discrete cosine transform, or other methods can be employed to convert the physiological data from the time domain to the frequency domain or other domains, extracting feature vectors that represent the characteristics of the physiological data, such as energy features, frequency features, and statistical features. The dimension and type of the feature vectors can be selected and adjusted according to the specific physiological data type and classification requirements to balance computational complexity and classification performance.

[0074] In step A202, the physiological data feature library can be pre-constructed by collecting and organizing a large amount of physiological data, covering physiological data features under various emergency conditions, such as abnormally rapid heart rate and abnormally high blood pressure. Typical feature vectors can be extracted and filtered from this data and stored in the feature library. Similarity calculation can use methods such as cosine similarity, Euclidean distance, and Pearson correlation coefficient to quantify the degree of similarity between the physiological data to be classified and the typical feature vectors in the feature library. The similarity set refers to the numerical set obtained by calculating the similarity between a physiological data point and multiple typical feature vectors of the same type of physiological data in the feature library. This set can comprehensively reflect the degree of association between the physiological data and various emergency conditions.

[0075] In step A203, a similarity threshold can be set, for example, 0.8. If at least one similarity score in the similarity set is greater than this threshold, the corresponding physiological data is determined to be critical physiological data, indicating that the physiological data is very similar to the typical characteristics of a certain emergency situation and needs to be prioritized for processing and transmission. If none of the similarities in the similarity set are greater than the preset similarity threshold, the probability that the corresponding preprocessed physiological data belongs to an unknown emergency situation can be further estimated based on the similarity set. For example, by counting the number of similarities in the similarity set that are greater than a certain small threshold (such as 0.6), or by calculating the mean and variance of the similarity set, the likelihood (i.e., probability) of the physiological data belonging to an unknown emergency situation can be comprehensively assessed. If the probability is greater than the preset probability threshold, it is also determined to be critical physiological data to address potential risks. Otherwise, it is determined to be non-critical physiological data. Thus, automatic classification of physiological data can be achieved, distinguishing between critical and non-critical data, providing a basis for subsequent bandwidth allocation and transmission strategy adjustments.

[0076] Furthermore, steps A201 and A202 are prerequisites for step A203. Step A201 quantifies the physiological data, step A202 analyzes the correlation between the physiological data and known emergency situations, and step A203 completes the classification decision based on the first two steps. Through the synergistic effect of steps A201, A202, and A203, a physiological data classification method based on feature vector similarity is provided, achieving effective classification of physiological data and solving the problem of how to effectively identify key physiological data mentioned in the background technology, thus providing a data foundation for subsequent data transmission strategies.

[0077] Through the above technical solution, this application achieves effective classification of preprocessed physiological data, distinguishing between critical and non-critical physiological data. This provides a data foundation for dynamically adjusting bandwidth allocation and data transmission strategies based on network status and data importance. Thus, in a mobile emergency unit environment with limited network coverage, it can prioritize the transmission of critical physiological data, improve the transmission efficiency and reliability of medical data, and provide strong support for remote medical diagnosis and treatment.

[0078] In some possible implementations, step A201 includes:

[0079] For each preprocessed physiological data, a feature extraction algorithm based on wavelet transform is used to decompose it into wavelet coefficients in multiple frequency bands;

[0080] Calculate the energy value of the wavelet coefficients for each frequency band to form an energy feature vector;

[0081] The energy feature vector is normalized to obtain the feature vector of each preprocessed physiological data.

[0082] In this process, a wavelet transform-based feature extraction algorithm is executed for each preprocessed physiological data. This decomposes the physiological data into wavelet coefficients across multiple frequency bands. As a preferred implementation, the wavelet transform algorithm can be Daubechies wavelet, Symlets wavelet, or Coiflets wavelet, etc. Furthermore, the number of decomposition levels can be adjusted according to actual application requirements. For example, using a 5-level wavelet decomposition can decompose the physiological data into an approximate coefficient band containing low-frequency information and five detail coefficient bands containing high-frequency information. Subsequently, the energy value is calculated for the wavelet coefficients of each frequency band. Specifically, the energy value of the wavelet coefficients in each frequency band can be obtained by calculating the sum of the squares of all wavelet coefficients within that band. This forms an energy feature vector. Each element in the energy feature vector represents the energy value of the wavelet coefficients in the corresponding frequency band, reflecting the energy distribution of the physiological data across different frequency bands. Finally, to eliminate differences in dimensions and numerical ranges between different physiological data, the energy feature vector is normalized. For example, a minimum-maximum normalization method can be used to scale each element of the energy feature vector to between 0 and 1. Through the above steps, the feature vectors of the preprocessed physiological data are obtained, providing quantitative features for the subsequent classification of physiological data.

[0083] Specifically, in emergency scenarios, the collected physiological data contains rich physiological feature information. If an inappropriate feature extraction method is used, this feature information cannot be effectively extracted, thus affecting the accuracy of subsequent physiological data classification. The wavelet transform-based feature extraction algorithm proposed in this application can effectively solve this problem. First, through wavelet transform, the physiological data is decomposed into wavelet coefficients in multiple frequency bands, realizing time-frequency domain analysis of the physiological data and enabling a more comprehensive extraction of the feature information contained within the physiological data. Then, by calculating the energy value of the wavelet coefficients in each frequency band, an energy feature vector is constructed. This energy feature vector effectively represents the energy distribution of the physiological data in different frequency bands, thereby highlighting the key features of the physiological data. Finally, by normalizing the energy feature vector, the differences in dimensions and numerical ranges between different physiological data are eliminated, improving the accuracy and stability of subsequent data classification. Therefore, high-quality feature data can be provided for subsequent physiological data classification, thereby improving the accuracy of physiological data classification in emergency scenarios.

[0084] In some specific implementations, for ECG data, a 5-level wavelet decomposition can first be performed using the db4 wavelet basis to obtain one approximation coefficient CA5 and five detail coefficients CD1-CD5. Then, the energy values ​​of the wavelet coefficients in the six frequency bands CA5 and CD1-CD5 are calculated respectively, forming a 6-dimensional energy feature vector. Further, the energy feature vector is normalized using a minimum-maximum normalization method, scaling each element in the vector to the [0,1] interval. For example, the original energy feature vector is [100,25,10,5,2,1], and the normalized energy feature vector can be [1,0.25,0.1,0.05,0.02,0.01]. Thus, the feature vector of the ECG signal is obtained and used for subsequent physiological data classification. A similar method can be used to extract feature vectors from other physiological data such as blood pressure, blood oxygen, heart rate, and body temperature.

[0085] In some possible real-time methods, step A203 includes:

[0086] If at least one similarity in the similarity set is greater than the preset similarity threshold, then the corresponding preprocessed physiological data is determined to be key physiological data.

[0087] If the similarity scores in the similarity set are all no greater than the preset similarity threshold, then the probability that the preprocessed physiological data is physiological data under unknown emergency conditions is estimated based on the similarity set.

[0088] If the probability is greater than the preset probability threshold, the corresponding preprocessed physiological data is determined to be critical physiological data; otherwise, the corresponding preprocessed physiological data is determined to be non-critical physiological data.

[0089] The preset similarity threshold is a numerical standard used to initially determine the level of similarity. This threshold can be set according to specific application scenarios and data characteristics; for example, it can be set to 0.8, 0.9, etc. Setting the similarity threshold requires a balance between the accuracy and sensitivity of data classification. If at least one similarity value in the similarity set is greater than the preset similarity threshold, the corresponding preprocessed physiological data is determined to be key physiological data. This means that when the preprocessed physiological data is sufficiently similar to at least one typical feature vector in the physiological data feature library, it is considered key physiological data even if the specific emergency type cannot be completely determined. This processing method can quickly identify physiological data highly related to known emergency situations, ensuring a rapid response to known emergency situations.

[0090] Specifically, if the similarity scores in the similarity set are all below a preset similarity threshold, the probability that the preprocessed physiological data corresponds to physiological data under an unknown emergency situation is estimated based on the similarity set. This means that when the similarity between the preprocessed physiological data and all typical feature vectors in the physiological data feature library is not high, considering the possibility of "unknown emergency situations" not covered by the feature library, the likelihood of it being physiological data under an unknown emergency situation is further assessed. Probability estimation can employ various statistical methods, such as Bayesian estimation and kernel density estimation. The similarity values ​​in the similarity set can be used as input parameters for probability estimation.

[0091] The preset probability threshold is a numerical standard used to judge the probability of an unknown emergency. This threshold can be set according to the tolerance for the risk of misjudgment, for example, it can be set to 0.6, 0.7, etc. The setting of the probability threshold affects the sensitivity of identifying unknown emergencies. If the probability is greater than the preset probability threshold, the corresponding preprocessed physiological data is determined as critical physiological data; otherwise, the corresponding preprocessed physiological data is determined as non-critical physiological data. This means that the category of preprocessed physiological data is finally determined based on the estimated probability of the unknown emergency and the probability threshold. Even if the similarity is not high, if the probability of the unknown emergency is high, it is still determined as critical physiological data, reflecting the prediction and prevention of potential risks. Conversely, if the probability is not high, it is determined as non-critical physiological data, avoiding data misjudgment caused by oversensitivity.

[0092] Specifically, when classifying preprocessed physiological data, this scheme first calculates the similarity set between its feature vector and typical feature vectors in the physiological data feature library. If a high similarity exists in the similarity set, it is quickly identified as key physiological data, ensuring rapid identification of known emergency situations. When all similarities are low, the scheme does not directly classify it as non-key data, but further estimates the probability of it being an unknown emergency. By introducing a probability estimation mechanism, the potential risks of the data can be assessed more comprehensively, avoiding the omission of key data due to the limitations of the feature library. This approach improves the intelligence and robustness of data classification while ensuring accuracy, especially in complex and ever-changing emergency scenarios, enabling more accurate identification of key physiological data and providing a more reliable data foundation for subsequent data transmission strategy adjustments and telemedicine decision-making.

[0093] Through the above technical solution, this application can further accurately determine whether the corresponding preprocessed physiological data is key physiological data when the similarity is not high, thereby avoiding misjudging potential key physiological data as non-key physiological data, ensuring the accuracy of data classification, and providing a more reliable data foundation for subsequent data transmission strategy adjustments and telemedicine decision-making.

[0094] In some preferred embodiments, if the similarity scores in the similarity set are all no greater than a preset similarity threshold, the step of estimating the probability that the corresponding preprocessed physiological data is physiological data under an unknown emergency situation based on the similarity set includes:

[0095] Calculate the variance of the similarity set to obtain the similarity variance;

[0096] If the similarity variance is less than the preset variance threshold, the Bayesian estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set.

[0097] If the similarity variance is not less than the preset variance threshold, then the kernel density estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set.

[0098] The variance of the similarity set can be calculated using the sample variance formula. For example, for a set {s1,s2,...,sn} containing n similarity values, where s1 to sn are the similarity values ​​in the similarity set, the similarity variance can be calculated as: V=Σ(si-μ)^2 / (n-1), where V is the variance of the similarity set, si is the i-th variance in the similarity set, and μ is the mean of the similarity set.

[0099] The preset variance threshold is a reference value used to determine the degree of dispersion of the similarity set. It can be preset through experimental analysis or statistical methods, depending on the actual application scenario and data characteristics.

[0100] Bayesian estimation is a parameter estimation method based on Bayesian theory. Specific Bayesian estimation models such as Naive Bayes and Gaussian Bayes can be used. In this scheme, Bayesian estimation is used to estimate the probability of physiological data in unknown emergency situations when the similarity variance is small, i.e., when the similarity values ​​are relatively concentrated. Bayesian estimation combines prior knowledge and sample data to obtain a posterior probability estimate, thus providing a more accurate probability estimate. The specific calculation process is existing technology and will not be detailed here.

[0101] Kernel density estimation is a nonparametric estimation method used to estimate the probability density function of a random variable. Specifically, kernel functions such as the Gaussian kernel function and the Epanechnikov kernel function can be used. In this scheme, kernel density estimation is used to estimate the probability of physiological data in unknown emergency situations when the similarity variance is large, i.e., when the similarity values ​​are relatively dispersed. Kernel density estimation does not rely on specific assumptions about the data distribution and can adapt well to various data distributions, thus providing a more robust probability estimation. The specific calculation process is existing technology and will not be detailed here.

[0102] Specifically, estimating probabilities using Bayesian or kernel density estimation methods involves calculating the probability that the preprocessed physiological data represents physiological data under unknown emergency conditions, based on the chosen estimation method and the specific values ​​of the similarity set. For example, the probability value P_bayes can be obtained using Bayesian estimation, while the probability value P_kde can be obtained using kernel density estimation.

[0103] Specifically, this solution refines the step of "if the similarity scores in the similarity set are all no greater than the preset similarity threshold, then estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set".

[0104] First, the variance of the similarity set is calculated to obtain the similarity variance. The similarity variance reflects the dispersion of the similarity values ​​within the set. When the similarity variance is small, it indicates that the similarity values ​​in the set are relatively concentrated. This may mean that the overall similarity between the preprocessed physiological data and the typical feature vectors in the physiological data feature library is not high, but relatively stable. In this case, the Bayesian estimation method can effectively utilize this relatively concentrated similarity information, thus estimating the probability more accurately. Conversely, when the similarity variance is large, it indicates that the similarity values ​​in the set are relatively dispersed. This may mean that the similarity between the preprocessed physiological data and the typical feature vectors in the physiological data feature library fluctuates greatly. In this case, the kernel density estimation method is used. Since the kernel density estimation method does not make specific assumptions about the data distribution, it can better adapt to the dispersed similarity values, thus estimating the probability more robustly.

[0105] By dynamically selecting an appropriate probability estimation method based on the magnitude of the similarity variance, the probability of physiological data belonging to an unknown emergency situation can be more accurately assessed even when the similarity is low. This dynamic selection mechanism fully considers the distribution characteristics of the similarity set, improves the accuracy and reliability of probability estimation, and thus enhances the accuracy of data classification. This provides more precise data support for the formulation of subsequent data transmission strategies, ultimately ensuring the real-time performance and accuracy of remote emergency care.

[0106] Through the above technical solution, this application can dynamically select either Bayesian estimation or kernel density estimation method to estimate the probability of physiological data in unknown emergency situations based on the variance of the similarity set, thereby improving the accuracy and robustness of probability estimation, and thus enhancing the accuracy of data classification, providing more accurate data support for the formulation of subsequent data transmission strategies.

[0107] In some implementations, step A3 includes:

[0108] A301. Monitor the available bandwidth and network congestion level of the wireless network, and calculate the network quality score;

[0109] A302. Obtain the amount of key physiological and non-key physiological data, and calculate the ratio of key physiological data to non-key physiological data.

[0110] A303. Based on the network quality score and data volume ratio, a fuzzy control algorithm is used to determine the target bandwidth allocation ratio for critical physiological data and non-critical physiological data;

[0111] A304. Based on the target bandwidth allocation ratio and the available bandwidth of the wireless network, calculate the allocated bandwidth for critical physiological data and the allocated bandwidth for non-critical physiological data to obtain the bandwidth allocation scheme.

[0112] In step A301, the network quality score is calculated using available bandwidth and network congestion level. Specifically, available bandwidth can be directly obtained from network monitoring tools, and network congestion level can be assessed using parameters such as packet loss rate and latency (i.e., by monitoring packet loss rate and latency, network congestion level is assessed using packet loss rate and latency, for example, by weighting packet loss rate and latency to obtain network congestion level). The network quality score can be designed as a weighted combination of available bandwidth and network congestion level; for example, the higher the available bandwidth and the lower the network congestion level, the higher the network quality score. Specifically, the network quality score can be calculated using the following formula: Network Quality Score = α × Available Bandwidth Score + β × (1 - Network Congestion Level Score), where α and β are weighting coefficients used to adjust the relative importance of available bandwidth and network congestion level in the network quality score. The available bandwidth score and network congestion level score are obtained by converting the original data into a standardized score between 0 and 1 (this can be done using a linear scaling method or the Sigmoid function) for unified calculation.

[0113] In step A302, the data volume ratio is calculated by dividing the data volume of key physiological data by the data volume of non-key physiological data.

[0114] In step A303, a fuzzy control algorithm is used to determine the target bandwidth allocation ratio. The fuzzy rule base of the fuzzy control algorithm predefines the mapping relationship between network quality score, data volume ratio, and target bandwidth allocation ratio. For example, when the network quality score and data volume ratio are both high, the fuzzy rules may set a higher target bandwidth allocation ratio. The fuzzy inference process activates the corresponding fuzzy rules based on the current network quality score and data volume ratio, and performs fuzzy inference to obtain the fuzzy inference result. Defuzzification is then used to convert the fuzzy inference result into a precise target bandwidth allocation ratio value.

[0115] In step A304, the allocated bandwidth for critical physiological data and non-critical physiological data is calculated based on the target bandwidth allocation ratio and the available bandwidth of the wireless network. For example, if the target bandwidth allocation ratios for critical physiological data and non-critical physiological data are 70% and 30% respectively, and the available bandwidth of the wireless network is 10 Mbps, then the allocated bandwidth for critical physiological data is 7 Mbps, and the allocated bandwidth for non-critical physiological data is 3 Mbps.

[0116] Specifically, this scheme first involves step A301, which monitors the network status and calculates a network quality score, quantifying the current network condition. Second, step A302 acquires the data volume of various data types and calculates their ratios, reflecting the data volume relationships between different data types. Then, step A303 introduces a fuzzy control algorithm to determine the target bandwidth allocation ratio for critical and non-critical physiological data based on the network quality score and the data volume ratio. The fuzzy control algorithm effectively handles the uncertainty and fuzziness of network status and data volume ratios, thus allowing for a more reasonable determination of the bandwidth allocation ratio. Finally, step A304 calculates the specific bandwidth allocation for each type of data based on the target bandwidth allocation ratio and available bandwidth, resulting in a bandwidth allocation scheme. Through these steps, step A3 dynamically adjusts the bandwidth allocation scheme according to changes in network status and data volume, thereby more effectively responding to changes in network status and data volume and obtaining a more reasonable bandwidth allocation scheme.

[0117] In some specific implementations, it is assumed that the available bandwidth of the wireless network where the mobile emergency unit is located is 5 Mbps, and the network congestion level is rated as medium. In step A301, the network quality score is calculated to be 60 points (out of 100). Simultaneously, in step A302, the data volume of critical physiological data is 2 MB, and the data volume of non-critical physiological data is 1 MB, with a data volume ratio calculated to be 2. In step A303, based on the network quality score of 60 points and the data volume ratio of 2, the fuzzy control algorithm determines the target bandwidth allocation ratios for critical and non-critical physiological data to be 70% and 30% respectively by consulting the fuzzy rule base and using fuzzy inference. The fuzzy rule base may contain the following rule: if the network quality score is medium and the data volume ratio is high, then the target bandwidth allocation ratio is higher. In step A304, based on the target bandwidth allocation ratios of 70% and 30%, and the available bandwidth of 5 Mbps, the allocated bandwidth for critical physiological data is calculated to be 3.5 Mbps, and the allocated bandwidth for non-critical physiological data is calculated to be 1.5 Mbps, thus obtaining the bandwidth allocation scheme. This bandwidth allocation scheme ensures that critical physiological data is allocated higher bandwidth to guarantee its priority transmission, while non-critical physiological data can also obtain a certain amount of bandwidth for transmission. This achieves the goal of balancing the transmission needs of both critical and non-critical physiological data under conditions of limited network resources.

[0118] In some preferred embodiments, step A303 includes:

[0119] Based on the network quality score and the data volume ratio, a fuzzy inference algorithm is used to select applicable fuzzy rules from the fuzzy rule base. The activation strength of each fuzzy rule is calculated based on the network quality score and the data volume ratio to obtain the fuzzy inference result. The fuzzy rule base contains multiple fuzzy rules, and the fuzzy rules define the mapping relationship between the network quality score, the data volume ratio, and the target bandwidth allocation ratio.

[0120] The fuzzy inference results are defuzzified by weighted average. Based on the activation intensity of each fuzzy rule and the corresponding target bandwidth allocation ratio, the target bandwidth allocation ratio of key physiological data and non-key physiological data is calculated.

[0121] The construction method of the fuzzy rule base can be predefined. Specifically, the fuzzy rule base can contain multiple fuzzy rules, each defining a mapping relationship between network quality score, data volume ratio, and target bandwidth allocation ratio. For example, a fuzzy rule could be: if the network quality score is high and the data volume ratio is low, then the target bandwidth allocation ratio is high. The selection of fuzzy rules can be based on expert experience or experimental data. The fuzzy inference algorithm can be any of several fuzzy inference methods, such as the Mamdani fuzzy inference algorithm or the Takagi-Sugeno fuzzy inference algorithm. The activation intensity can be calculated based on the antecedents of the fuzzy rules and the membership functions of the input variables. Weighted average defuzzification is a commonly used defuzzification method, which calculates the weighted average of the activation intensity of each fuzzy rule and the corresponding target bandwidth allocation ratio to obtain the final target bandwidth allocation ratio.

[0122] Specifically, addressing the broadness issue of fuzzy control algorithms, step A303 employs a fuzzy inference algorithm. This algorithm establishes a fuzzy rule base, explicitly defining the mapping relationship between network quality scores, data volume ratios, and target bandwidth allocation ratios. When making bandwidth allocation ratio decisions, applicable fuzzy rules are first selected from the fuzzy rule base. Then, the activation strength of each fuzzy rule is calculated based on the network quality score and data volume ratio to obtain the fuzzy inference result. Finally, a weighted average defuzzification process yields the precise target bandwidth allocation ratio. This method, through a specific fuzzy inference algorithm and fuzzy rule base, achieves a precise mapping from network status and data volume to bandwidth allocation ratios, improving the rationality and effectiveness of bandwidth allocation, thereby enhancing the efficiency and reliability of data transmission. In scenarios like mobile emergency units, characterized by limited resources, complex network environments, and extremely high real-time requirements, the data transmission strategy can be dynamically adjusted based on wireless network conditions and data real-time needs. This ensures low-latency, high-reliability transmission of critical real-time physiological data while also accommodating the transmission needs of non-real-time data, maintaining system usability and operational stability, and guaranteeing the real-time performance and accuracy of remote emergency care.

[0123] Furthermore, this application also proposes that step A4 includes:

[0124] A401. Based on the bandwidth allocation scheme, determine the priority queues for critical physiological data and non-critical physiological data;

[0125] A402. For each priority queue, the data is divided into multiple data packets according to the amount of data to be transmitted in the queue, and the maximum transmission unit of each data packet is determined according to the bandwidth allocation scheme.

[0126] A403. According to the priority order of the priority queue, schedule the data packets in the queue, encapsulate the data packets according to the determined maximum transmission unit, use Turbo codes for error control encoding of data packets containing critical physiological data, and use parity check codes for error control encoding of data packets containing non-critical physiological data.

[0127] In step A401, the determination of priority queues is based on the bandwidth allocation scheme. Specifically, a corresponding number of priority queues are established according to the bandwidth allocation ratio between critical physiological data and non-critical physiological data in the bandwidth allocation scheme. For example, if critical physiological data is allocated a higher proportion of bandwidth in the allocation scheme, then critical physiological data is placed in a higher priority queue, while non-critical physiological data is placed in a lower priority queue. This ensures that critical physiological data enjoys priority during transmission.

[0128] In step A402, the packet segmentation and maximum transmission unit (MTU) determination are performed independently for each priority queue. For higher priority queues, data is segmented into multiple packets, and the MTU for each packet is determined based on the bandwidth allocated to critical physiological data. When the allocated bandwidth is high, the MTU can be set to a larger value to improve transmission efficiency. Conversely, when the allocated bandwidth is low, the MTU is set to a smaller value to adapt to the current bandwidth conditions. A similar data segmentation and MTU determination process is performed for lower priority queues, but the MTU determination takes into account the bandwidth allocated to non-critical physiological data. In this way, the packet size can be dynamically adjusted according to the bandwidth allocation for different priority queues, thereby optimizing network bandwidth utilization.

[0129] In step A403, data packet scheduling is performed according to the priority queue order. Data packets in higher priority queues are scheduled for transmission first, while data packets in lower priority queues are scheduled only after the higher priority queues have been transmitted. During data packet encapsulation, data packets are encapsulated according to the maximum transmission unit determined in step A402, ensuring that the data packet size does not exceed the network's maximum allowed transmission unit. Furthermore, different error control coding methods are used for data packets of different priorities. For data packets containing critical physiological data, Turbo codes are used for error control coding. Turbo codes are a high-performance channel coding technique that provides strong error correction capabilities, effectively ensuring the reliability of critical physiological data transmission. For data packets containing non-critical physiological data, parity check codes, which have lower computational complexity, are used for error control coding, providing basic error detection capabilities while reducing system computational overhead. Thus, while ensuring reliable transmission of critical data, the transmission requirements of non-critical data and the effective utilization of system resources are also considered.

[0130] Specifically, addressing the issue of low data transmission efficiency and reliability in scenarios with limited bandwidth and complex network conditions, such as mobile emergency units, this application proposes a data transmission strategy that utilizes techniques such as priority queuing, dynamic maximum transmission unit adjustment, and differentiated error control coding to achieve prioritized and reliable transmission of critical physiological data under limited bandwidth. First, by establishing a priority queuing, critical physiological data is ensured to enjoy higher priority during data transmission, gaining priority in transmission opportunities, thereby reducing transmission latency and guaranteeing data real-time performance. Second, by dynamically adjusting the maximum transmission unit according to the bandwidth allocation scheme, the packet size can be matched to the current network bandwidth conditions. When bandwidth is sufficient, packets with larger maximum transmission units are used for transmission, improving transmission efficiency; when bandwidth is limited, packets with smaller maximum transmission units are used for transmission, avoiding network congestion and optimizing bandwidth utilization. Furthermore, different error control coding methods are used for critical and non-critical physiological data, reducing the coding overhead of non-critical data while ensuring the reliability of critical data, thus balancing data transmission reliability and system resource consumption. By comprehensively applying the above strategies, the integrated transmission performance of operating table measurement data can be improved in resource-constrained environments such as mobile emergency units, providing more timely and reliable data support for telemedicine.

[0131] In some specific implementations, critical physiological data can be pre-prioritized to have a higher priority than non-critical physiological data. After the bandwidth allocation scheme is determined, data packets of critical physiological data are placed in a priority 1 queue, and data packets of non-critical physiological data are placed in a priority 2 queue. Assume the bandwidth allocated to critical physiological data is 5 Mbps, and the bandwidth allocated to non-critical physiological data is 2 Mbps. For the priority 1 queue, if the amount of data to be transmitted in the queue is 10 MB, this 10 MB of data is divided into, for example, 1000 data packets. Based on the 5 Mbps bandwidth, the maximum transmission unit (MTU) of each data packet is determined, for example, set to 625 bytes. For the priority 2 queue, if the amount of data to be transmitted in the queue is 5 MB, data packets are also divided, and the MTU is determined based on the 2 Mbps bandwidth, for example, set to 250 bytes. During data transmission, the system prioritizes the data packets in the priority 1 queue for transmission. Data packets in the priority 1 queue are encoded using Turbo codes, and data packets in the priority 2 queue are encoded using parity-check codes. Therefore, given the limited total bandwidth, critical physiological data, due to its higher priority and bandwidth allocation, receives priority transmission and reliable protection. While non-critical physiological data has a lower priority, it still receives bandwidth for transmission, ensuring data integrity.

[0132] Preferably, before step A401, the following step may be included:

[0133] A400a. Calculate the ratio of the allocated bandwidth of non-critical physiological data to the amount of non-critical physiological data, denoted as the specific density;

[0134] A400b. If the specific density is lower than the preset specific density threshold, at least some non-critical physiological data are selected as replacement objects, and the feature data of the replacement objects are extracted to replace the corresponding non-critical physiological data so that the specific density is not lower than the preset specific density threshold.

[0135] Specific density refers to the ratio of allocated bandwidth to the amount of non-critical physiological data, used to measure the bandwidth resources available per unit of data volume. Specifically, it can be calculated by dividing the bandwidth allocated to non-critical physiological data by the amount of non-critical physiological data. The allocated bandwidth can be obtained according to a bandwidth allocation scheme, and the data volume can be obtained by statistically analyzing the size of the non-critical physiological data.

[0136] The preset specific density threshold is a pre-defined reference value used to determine whether the specific density is too low. It can be set according to the actual application scenario and requirements. Setting the preset specific density threshold requires a balance between data transmission efficiency and data integrity. If the threshold is set too high, it may lead to excessive data replacement and loss of too much information; if the threshold is set too low, it may not effectively solve the problem of low efficiency in transmitting non-critical physiological data.

[0137] Selecting at least some non-critical physiological data as replacement targets means that when the specific density is lower than a preset specific density threshold, a portion or all of the non-critical physiological data is selected for replacement. Specifically, the selection strategy for replacement targets can be determined based on factors such as the amount and importance of the non-critical physiological data. For example, non-critical physiological data with a larger amount of data or relatively less important data can be prioritized as replacement targets.

[0138] The process of extracting feature data from the object to be replaced to replace corresponding non-critical physiological data involves processing the object, extracting feature data that represents the object's main information and has a smaller data volume, and then using this feature data to replace the original non-critical physiological data. Specifically, feature data extraction methods can include data compression algorithms, feature extraction algorithms, or data simplification algorithms. For example, for waveform data, feature parameters such as peaks, troughs, and frequencies can be extracted; for image data, feature parameters such as texture, color, and contour can be extracted. Data compression algorithms can employ wavelet transform, discrete cosine transform, and other methods. Feature extraction algorithms can employ principal component analysis, linear discriminant analysis, and other methods. Data simplification algorithms can employ downsampling, data smoothing, and other methods. By replacing non-critical physiological data with feature data, the amount of data can be effectively reduced, and the specific density can be improved.

[0139] Specifically, ensuring that the specific density is not lower than the preset specific density threshold means adjusting the amount of non-critical physiological data through data replacement operations so that the recalculated specific density value reaches or exceeds the preset specific density threshold. In other words, if the specific density is still lower than the preset specific density threshold after one data replacement, the data replacement operation can be repeated until the specific density meets the requirement.

[0140] Furthermore, steps A400a and A400b are executed before step A401 determines the priority queue. Their purpose is to compare and adjust the density before dividing the priority queue, thereby optimizing data transmission efficiency. By calculating the density ratio first and determining whether data replacement is necessary based on the ratio, the potential transmission efficiency problem caused by excessive non-critical physiological data can be addressed in advance, laying the foundation for subsequent priority queue division and data transmission strategy formulation. Steps A400a and A400b, combined with step A401, further enable dynamic adjustment of data volume based on bandwidth allocation, making the data transmission strategy more refined and intelligent.

[0141] Specifically, this solution addresses the efficiency issue of non-critical physiological data transmission by adding density calculation and data replacement steps. First, step A400a calculates the density of non-critical physiological data to assess its transmission efficiency under the current bandwidth allocation. If the density is below a preset threshold, it indicates that the amount of non-critical physiological data is too large relative to the allocated bandwidth, potentially affecting transmission efficiency. In this case, step A400b initiates a data replacement mechanism, selecting a portion of the non-critical physiological data as replacement targets, extracting their feature data for replacement, thereby reducing the data volume and improving the density. Through data replacement, the data volume of non-critical physiological data can be effectively reduced while ensuring that the main information is not lost, thus improving transmission efficiency and preventing non-critical physiological data from consuming excessive bandwidth resources and affecting overall data integration efficiency. After density adjustment, the non-critical physiological data then enters the subsequent priority queue division and data transmission process, ensuring efficient transmission of both critical and non-critical physiological data under limited bandwidth conditions, ultimately improving the overall performance of the operating table measurement data integration method in the mobile emergency unit.

[0142] Based on the above technical solution, this application further considers the problem of data transmission efficiency of non-critical physiological data on the basis of claim 8. By calculating the density and data replacement operation, it effectively solves the problem of low data transmission efficiency when the bandwidth allocated to non-critical physiological data is too low relative to its data volume. It ensures that non-critical physiological data can be transmitted efficiently under limited bandwidth conditions, avoids it from becoming a data transmission bottleneck, and ultimately improves the efficiency and real-time performance of the entire data integration method.

[0143] refer to Figure 2 This application also proposes an operating table measurement data integration device for an edge computing module in a mobile emergency unit, the device comprising:

[0144] Data acquisition module 1 is used to acquire patient physiological data collected by multiple sensors on the operating table, perform preprocessing, and obtain preprocessed physiological data (for details, refer to step A1 above).

[0145] Data classification module 2 is used to classify the preprocessed physiological data into key physiological data and non-key physiological data (for details, refer to step A2 above).

[0146] The bandwidth allocation module 3 is used to monitor the network status information of the wireless network. Based on the network status information, the amount of key physiological data and the amount of non-key physiological data, it dynamically adjusts the allocated bandwidth for key physiological data and non-key physiological data to obtain a bandwidth allocation scheme. The network status information includes available bandwidth and network congestion level (refer to step A3 above for details).

[0147] Transmission strategy adjustment module 4 is used to determine the data transmission strategy for critical physiological data and non-critical physiological data according to the bandwidth allocation scheme (for details, refer to step A4 above).

[0148] Data transmission module 5 is used to transmit data to the telemedicine center according to the data transmission strategy (refer to step A5 above for details).

[0149] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0150] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0151] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0152] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0153] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for integrating operating table measurement data, used in an edge computing module within a mobile emergency unit, characterized in that, The steps of this method include: A1. Acquire patient physiological data collected by multiple sensors on the operating table, perform preprocessing, and obtain preprocessed physiological data; A2. Classify the preprocessed physiological data into key physiological data and non-key physiological data; A3. Monitor the network status information of the wireless network, and dynamically adjust the allocated bandwidth for key physiological data and non-key physiological data based on the network status information, the amount of key physiological data, and the amount of non-key physiological data to obtain a bandwidth allocation scheme; the network status information includes available bandwidth and network congestion level. A4. Based on the bandwidth allocation scheme, determine the data transmission strategy for critical physiological data and non-critical physiological data; A5. Data is transmitted to the telemedicine center according to the aforementioned data transmission strategy; Step A2 includes: A201. Extract the feature vectors of the preprocessed physiological data; A202. For each preprocessed physiological data, calculate the similarity between the feature vector of the preprocessed physiological data and the typical feature vectors of the corresponding physiological data in the physiological data feature library, and obtain a similarity set; the physiological data feature library records the typical feature vectors of each physiological data under various emergency situations; A203. Based on the similarity set of each preprocessed physiological data, the preprocessed physiological data are determined to be either key physiological data or non-key physiological data; Step A203 includes: If at least one similarity in the similarity set is greater than the preset similarity threshold, then the corresponding preprocessed physiological data is determined to be key physiological data. If the similarity scores in the similarity set are all no greater than the preset similarity threshold, then the probability that the preprocessed physiological data is physiological data under unknown emergency conditions is estimated based on the similarity set. If the probability is greater than the preset probability threshold, the corresponding preprocessed physiological data is determined to be critical physiological data; otherwise, the corresponding preprocessed physiological data is determined to be non-critical physiological data. The step of estimating the probability that the preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set if the similarity scores in the similarity set are all no greater than a preset similarity threshold includes: Calculate the variance of the similarity set to obtain the similarity variance; If the similarity variance is less than the preset variance threshold, the Bayesian estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set. If the similarity variance is not less than the preset variance threshold, then the kernel density estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set.

2. The method for integrating operating table measurement data according to claim 1, characterized in that, Step A201 includes: For each preprocessed physiological data, a feature extraction algorithm based on wavelet transform is used to decompose it into wavelet coefficients in multiple frequency bands; Calculate the energy value of the wavelet coefficients for each frequency band to form an energy feature vector; The energy feature vector is normalized to obtain the feature vector of each preprocessed physiological data.

3. The method for integrating operating table measurement data according to claim 1, characterized in that, Step A3 includes: A301. Monitor the available bandwidth and network congestion level of the wireless network, and calculate the network quality score; A302. Obtain the amount of key physiological and non-key physiological data, and calculate the ratio of key physiological data to non-key physiological data. A303. Based on the network quality score and data volume ratio, a fuzzy control algorithm is used to determine the target bandwidth allocation ratio for critical physiological data and non-critical physiological data; A304. Based on the target bandwidth allocation ratio and the available bandwidth of the wireless network, calculate the allocated bandwidth for critical physiological data and the allocated bandwidth for non-critical physiological data to obtain the bandwidth allocation scheme.

4. The method for integrating operating table measurement data according to claim 3, characterized in that, Step A303 includes: Based on the network quality score and the data volume ratio, a fuzzy inference algorithm is used to select applicable fuzzy rules from the fuzzy rule base. The activation strength of each fuzzy rule is calculated based on the network quality score and the data volume ratio to obtain the fuzzy inference result. The fuzzy rule base contains multiple fuzzy rules, and the fuzzy rules define the mapping relationship between the network quality score, the data volume ratio, and the target bandwidth allocation ratio. The fuzzy inference results are defuzzified by weighted average. Based on the activation intensity of each fuzzy rule and the corresponding target bandwidth allocation ratio, the target bandwidth allocation ratio of key physiological data and non-key physiological data is calculated.

5. The method for integrating operating table measurement data according to claim 1, characterized in that, Step A4 includes: A401. Based on the bandwidth allocation scheme, determine the priority queues for critical physiological data and non-critical physiological data; A402. For each priority queue, the data is divided into multiple data packets according to the amount of data to be transmitted in the queue, and the maximum transmission unit of each data packet is determined according to the bandwidth allocation scheme. A403. According to the priority order of the priority queue, schedule the data packets in the queue, encapsulate the data packets according to the determined maximum transmission unit, use Turbo codes for error control encoding of data packets containing critical physiological data, and use parity check codes for error control encoding of data packets containing non-critical physiological data.

6. The method for integrating operating table measurement data according to claim 5, characterized in that, Before step A401, the following steps are also included: A400a. Calculate the ratio of the allocated bandwidth of non-critical physiological data to the amount of non-critical physiological data, denoted as the specific density; A400b. If the specific density is lower than a preset specific density threshold, at least some non-critical physiological data are selected as replacement objects, and feature data of the replacement objects are extracted to replace the corresponding non-critical physiological data, so that the specific density is not lower than the preset specific density threshold.

7. An operating table measurement data integration device for use as an edge computing module in a mobile emergency unit, characterized in that, The device includes: The data acquisition module is used to acquire patient physiological data collected by multiple sensors on the operating table, perform preprocessing, and obtain preprocessed physiological data. The data classification module is used to classify the preprocessed physiological data into key physiological data and non-key physiological data. The bandwidth allocation module is used to monitor the network status information of the wireless network, and dynamically adjust the allocated bandwidth for key physiological data and non-key physiological data based on the network status information, the amount of key physiological data, and the amount of non-key physiological data to obtain a bandwidth allocation scheme; the network status information includes available bandwidth and network congestion level. The transmission strategy adjustment module is used to determine the data transmission strategy for critical physiological data and non-critical physiological data according to the bandwidth allocation scheme. The data transmission module is used to transmit data to the remote medical center according to the data transmission strategy. When the data classification module classifies the preprocessed physiological data into critical and non-critical physiological data, it performs the following steps: A201. Extract the feature vectors of the preprocessed physiological data; A202. For each preprocessed physiological data, calculate the similarity between the feature vector of the preprocessed physiological data and the typical feature vectors of the corresponding physiological data in the physiological data feature library, and obtain a similarity set; the physiological data feature library records the typical feature vectors of each physiological data under various emergency situations; A203. Based on the similarity set of each preprocessed physiological data, the preprocessed physiological data are determined to be either key physiological data or non-key physiological data; Step A203 includes: If at least one similarity in the similarity set is greater than the preset similarity threshold, then the corresponding preprocessed physiological data is determined to be key physiological data. If the similarity scores in the similarity set are all no greater than the preset similarity threshold, then the probability that the preprocessed physiological data is physiological data under unknown emergency conditions is estimated based on the similarity set. If the probability is greater than the preset probability threshold, the corresponding preprocessed physiological data is determined to be critical physiological data; otherwise, the corresponding preprocessed physiological data is determined to be non-critical physiological data. The step of estimating the probability that the preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set if the similarity scores in the similarity set are all no greater than a preset similarity threshold includes: Calculate the variance of the similarity set to obtain the similarity variance; If the similarity variance is less than the preset variance threshold, the Bayesian estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set. If the similarity variance is not less than the preset variance threshold, then the kernel density estimation method is used to estimate the probability that the corresponding preprocessed physiological data is physiological data under unknown emergency conditions based on the similarity set.

Images