A real-time anesthesia evaluation system fusing multi-channel signals of conductive needle

Abstract

The application discloses a kind of real-time anesthesia evaluation systems of fusion conductive needle multi-channel signal, it is related to medical information processing and anesthesia monitoring technical field, including acquisition classification module, and the physiological electrical signal of multiple channel is obtained by the multiple conductive needle electrodes arranged in the patient's body surface or tissue, and original anesthesia data is generated in combination with anesthesia parameters;Original anesthesia data is classified using content adaptive algorithm, and the classified anesthesia data group, including key indicators and trends, is obtained;Terminal adaptation module, according to the classified anesthesia data group, the processing capacity and display specification of terminal equipment are obtained from terminal equipment registration information, if the processing capacity and display specification are higher than preset threshold, then trend is distributed to terminal equipment, and the data packet for terminal is obtained;The real-time anesthesia evaluation system of fusion conductive needle multi-channel signal, efficient, accurate and intelligent real-time anesthesia evaluation in complex medical scene is realized.

Description

Technical Field

[0001] This invention relates to the field of signal pattern recognition technology, and more specifically to a real-time anesthesia assessment system that integrates multi-channel signals from conductive needles. Background Technology

[0002] In modern surgical procedures and intensive care, anesthesia assessment systems are responsible for real-time monitoring and comprehensive evaluation of the patient's anesthetic status. Their core function is to collect and analyze the patient's physiological signals to provide the medical team with reliable information on anesthetic depth, analgesic effect, and vital signs.

[0003] However, existing anesthesia assessment systems still have the following shortcomings: First, the signal acquisition methods are relatively simple, usually relying on a small number of electrodes, making it difficult to comprehensively reflect the complex physiological changes of patients, resulting in insufficient data integrity and anti-interference capabilities. Second, the data distribution mode lacks differentiation, generally adopting a single broadcast approach, causing high-performance terminals to be unable to obtain sufficiently detailed data, while low-performance terminals lag due to processing complex information, resulting in information mismatch. Furthermore, during the information distribution process, existing systems often fail to dynamically adjust based on user permissions. High-privilege users such as surgeons may not be able to obtain complete anesthesia parameters, while support staff receive redundant information, reducing team collaboration efficiency. In addition, in the complex operating room environment, limited network bandwidth and differences in terminal performance further amplify the problems, leading to data transmission delays, untimely updates of key indicators, and increased medical risks. Summary of the Invention

[0004] The purpose of this invention is to provide a real-time anesthesia assessment system that integrates multi-channel signals from conductive needles, thereby solving the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a real-time anesthesia assessment system integrating multi-channel signals from conductive needles, comprising: a data acquisition and classification module, which acquires multi-channel physiological electrical signals by deploying multiple conductive needle electrodes on the patient's body surface or in tissues, and generates raw anesthesia data by combining anesthesia parameters; a content adaptation algorithm is used to perform preliminary classification of the raw anesthesia data to obtain classified anesthesia data sets, including key indicators and trends; a terminal adaptation module, which, based on the classified anesthesia data sets, obtains the processing capacity and display specifications of the terminal device from the terminal device registration information, and determines that if the processing capacity and display specifications are higher than a preset threshold, the trend is allocated to the terminal device to obtain data packets for the terminal; a permission management module, which obtains the user's permission level on the terminal device, sorts the user's permission level using a priority ranking algorithm, determines that the terminal device corresponding to the user in the ranking result needs to receive comprehensive anesthesia parameters, and obtains a data subset with matching permissions; and a customized generation module, which generates data for the data set with matching permissions. Based on subsets and data packets, a content adaptation algorithm is used to generate customized data content. If the terminal device is portable and the permission level is auxiliary, the data content is simplified to key indicators to obtain a simplified data version. The broadcast module extracts the real-time update portion from the simplified data version and data packets, and broadcasts the real-time update portion to all connected terminals through the network protocol. If the broadcast delay exceeds a preset threshold, the key indicator portion is transmitted first to obtain the basic synchronization data after broadcast. The optimization scheduling module obtains the feedback response time of each terminal based on the basic synchronization data after broadcast, and uses a priority sorting algorithm to adjust the subsequent data push order. Terminals with response delays exceeding the preset threshold are only given the simplified data version to obtain the optimized push sequence. The data distribution module distributes the customized data content and the simplified data version to the corresponding terminals according to the optimized push sequence. If network fluctuations are detected during the distribution process, the system switches to the backup channel to transmit data and obtains the final distribution confirmation signal.

[0006] Preferably, the acquisition and classification module acquires multi-channel physiological electrical signals through multiple conductive needle electrodes deployed on the patient's body surface or tissues, and generates raw anesthesia data by combining anesthesia parameters; it then uses a content adaptation algorithm to perform preliminary classification of the raw anesthesia data, obtaining a classified anesthesia data set, including key indicators and trends. Specifically, this includes acquiring multi-channel physiological electrical signals through conductive needle electrodes, performing synchronous reading processing using a signal acquisition circuit to obtain a synchronous physiological electrical signal set; performing a fusion processing method on the synchronous physiological electrical signal set, combining it with anesthesia parameter data to generate raw anesthesia data; using a content adaptation classification algorithm to perform preliminary classification processing of the raw anesthesia data, obtaining a classified anesthesia data set; extracting key indicators from the classified anesthesia data set to generate a key indicator set; performing trend analysis on the key indicator set using a time series analysis algorithm to obtain trend feature data; if the trend feature data exceeds a preset threshold, using a support vector machine algorithm to detect anomalies in the trend feature data and determine abnormal states; and generating anesthesia state classification results based on the abnormal states to obtain the final anesthesia state assessment data.

[0007] Preferably, the terminal adaptation module, based on the categorized anesthesia data set, obtains the processing capability and display specifications of the terminal device from the terminal device registration information. If the processing capability and display specifications are higher than a preset threshold, the trend is assigned to the terminal device, and a data packet for the terminal is obtained. This includes obtaining the terminal device registration information from the anesthesia data set, extracting the processing capability and display specification data, and obtaining a terminal performance parameter set. If both the processing capability and display specifications in the terminal performance parameter set are higher than the preset threshold and preset standard, the trend data is formatted to obtain a standard trend data packet. Based on the standard trend data packet, a data compression algorithm is used to compress the data to obtain a compressed trend data packet. Keyframe data is extracted from the compressed trend data packet, and a time series segmentation algorithm is used to generate segmented trend data to obtain a segmented trend set. If the amount of data in any segment of the segmented trend set exceeds the terminal device's cache capacity, the segmented trend set is divided into blocks to obtain an adapted block data set. A terminal device adaptation data packet is generated from the adapted block data set, and the adaptation data packet is transmitted to the terminal using a data distribution protocol to obtain the terminal received data. Based on the terminal received data, the data is decompressed and reassembled on the terminal to obtain the terminal presentation trend data.

[0008] Preferably, the permission management module obtains the user's permission level on the terminal device, sorts the user permission levels using a priority sorting algorithm, determines the terminal device corresponding to the user in the sorting result that needs to receive general anesthesia parameters, and obtains a permission matching data subset including obtaining user permission level data from the terminal device identifier, sorting the permission level data using a quick sorting algorithm, and obtaining a list of authorized users; based on the list of authorized users, obtains the corresponding terminal device type, and determines if the terminal device type supports the bandwidth transmission protocol, then extracts the general anesthesia parameter requirements, obtaining an anesthesia parameter set; extracts the terminal processing capability requirements from the anesthesia parameter set, and uses a data filtering tool to match the terminal processing capabilities to obtain those that meet the high-performance requirements. The system retrieves a list of terminal devices; based on this list, it obtains permission matching data. If the permission matching data matches the user's permission level, the anesthesia parameter set is divided into packets to obtain packetized anesthesia data. Real-time transmission parameters are extracted from the packetized anesthesia data and compressed using a data compression algorithm to obtain compressed transmission data. Based on the compressed transmission data, the system obtains the display adaptation parameters for the terminal devices. If the display adaptation parameters match the terminal display specifications, the compressed transmission data is layered and encapsulated to obtain terminal display data. Update frequency parameters are extracted from the terminal display data. If the update frequency parameters match the timestamp of the anesthesia parameter set, the terminal display data is stored in the terminal device to obtain terminal storage data.

[0009] Preferably, the customized generation module uses a content adaptation algorithm to fuse and generate customized data content for the data subset and data package that match permissions. If the terminal device is portable and the permission level is auxiliary, the data content is simplified to key indicators. Obtaining the simplified data version includes: obtaining user permission level data from the terminal device identifier; classifying the permission levels using a data filtering tool to obtain a set of categorized permissions; extracting auxiliary permission level data from the set of categorized permissions; fusing the auxiliary permission data and data package using a content adaptation algorithm to obtain preliminary customized data; based on the preliminary customized data, if the terminal device is portable, extracting key indicator data to obtain a simplified indicator set; obtaining data display requirements from the simplified indicator set; performing structured processing on the simplified indicator set using a format conversion tool to obtain adapted display data; based on the adapted display data, if the terminal device display specifications support a resolution lower than a preset threshold, compressing the adapted display data to obtain compressed display data; extracting update frequency parameters from the compressed display data; verifying the update frequency parameters using a data verification tool to obtain verified data; and based on the verified data, if the data integrity meets a preset threshold, storing the verified data in the terminal device to obtain terminal storage data.

[0010] Preferably, the broadcast module extracts the real-time update portion from the simplified data version and data packets, broadcasts the real-time update portion to all connected terminals via a network protocol, and prioritizes the transmission of key indicator portions if the broadcast delay exceeds a preset threshold. The resulting basic synchronization data includes obtaining real-time update data from the basic synchronization data, segmenting the real-time update data using a data segmentation tool to obtain a fragmented data set; prioritizing the fragmented data set based on the segmented data set if the terminal connection status is high load, obtaining a sorted data set; and extracting key indicator data from the sorted data set, and compressing the key indicators using a data compression tool. The data is compressed to obtain compression index data. Based on the compression index data, if the data synchronization frequency exceeds a preset threshold, the transmission rate is adjusted through network protocol configuration to obtain adjusted transmission data. The data broadcast range is obtained from the adjusted transmission data, and a data distribution tool is used to distribute the adjusted transmission data in a targeted manner to obtain targeted distribution data. Based on the targeted distribution data, if the terminal connection status supports multi-threaded transmission, a parallel transmission tool is used to process the targeted distribution data to obtain parallel transmission data. Data integrity parameters are extracted from the parallel transmission data, and a data verification tool is used to verify the data integrity parameters to obtain verified synchronization data.

[0011] Preferably, the optimization scheduling module obtains the feedback response time of each terminal based on the basic synchronization data after broadcast, adjusts the subsequent data push order using a priority sorting algorithm, and determines that terminals with response delays exceeding a preset threshold only receive simplified data versions. Obtaining the optimized push sequence includes obtaining the terminal reception status from the optimized push sequence, classifying the terminal reception status using a status analysis tool to obtain a classified status set; based on the classified status set, determining if the terminal reception status has a bandwidth below a preset threshold, then using a data fragmentation tool to segment the push sequence to obtain a fragmented data set; extracting data priority parameters from the fragmented data set, and sorting the data priority parameters using a sorting algorithm to obtain a sorted data set.

[0012] Preferably, the optimization scheduling module, based on the basic synchronization data after broadcast, obtains the feedback response time of each terminal, adjusts the subsequent data push order using a priority sorting algorithm, and determines that terminals with response delays exceeding a preset threshold only receive simplified data versions. Obtaining the optimized push sequence also includes, based on the sorted data set, determining if the push frequency exceeds a preset threshold, adjusting the transmission rate through the network protocol to obtain adjusted transmission data; obtaining the data broadcast range from the adjusted transmission data, distributing the adjusted transmission data using a targeted distribution tool to obtain targeted distribution data; determining, based on the targeted distribution data, if the terminal supports multi-threaded transmission, processing the targeted distribution data using a parallel transmission tool to obtain parallel transmission data; extracting transmission integrity parameters from the parallel transmission data, verifying the transmission integrity parameters using a verification tool to obtain verified synchronization data.

[0013] Preferably, the data distribution module, for the optimized push sequence, distributes customized data content and simplified data versions to corresponding terminals respectively. If network fluctuations are detected during the distribution process, it switches to a backup channel to transmit data. Obtaining the final distribution confirmation signal includes extracting terminal type parameters from the optimized push sequence, classifying terminal types using a classification tool to obtain a set of divided terminals; extracting terminal customization requirement parameters based on the divided terminal set, generating customized data content and simplified data versions using a content generation tool to obtain a generated data set; extracting data distribution priority from the generated data set, and sorting the data distribution priority using a priority sorting algorithm to obtain a sorted distribution sequence.

[0014] Preferably, the data distribution module, for the optimized push sequence, distributes customized data content and simplified data versions to corresponding terminals respectively. If network fluctuations are detected during the distribution process, it switches to a backup channel to transmit data. Obtaining the final distribution confirmation signal also includes: initiating the data distribution process according to the sorted distribution sequence; if network fluctuations are detected during the distribution process, switching to the backup channel using a channel switching tool to obtain the switched transmission channel; obtaining transmission status parameters from the switched transmission channel, analyzing the transmission status parameters using a status monitoring tool to obtain the analyzed transmission status; determining whether the transmission status is stable based on the analyzed transmission status, generating a distribution confirmation signal using a distribution confirmation tool to obtain the final distribution confirmation signal; extracting distribution completion parameters from the final distribution confirmation signal, storing the distribution completion parameters using a log recording tool to obtain the stored distribution record.

[0015] As can be seen from the above technical solution, the present invention has the following beneficial effects: This real-time anesthesia assessment system, integrating multi-channel conductive needle signals, acquires multi-channel physiological electrical signals by deploying multiple conductive needle electrodes on the patient's body surface or tissues. These signals are then fused with anesthesia parameters, significantly improving the comprehensiveness and anti-interference capabilities of signal acquisition. A content adaptation algorithm categorizes raw anesthesia data, generating key indicators and detailed trend data. Differentiated data delivery is then implemented based on terminal performance and display specifications, ensuring high-performance terminals receive detailed trend data while low-performance terminals receive only key indicators, preventing information overload. A priority ranking algorithm dynamically generates matching data subsets based on user permissions, allowing high-privilege users to receive complete anesthesia parameters while assisting users receive simplified data, thus achieving a reasonable match between information and responsibilities. In cases of network latency or fluctuations, the system prioritizes the transmission of key indicators and compensates using backup channels, ensuring real-time synchronization of key indicators and robust overall transmission. Furthermore, the system optimizes the data delivery order based on terminal feedback response time, ensuring slower-responding terminals receive only simplified data. This enhances real-time performance, system flexibility, and overall security and reliability, ultimately achieving efficient, accurate, and intelligent real-time anesthesia assessment in complex medical scenarios. Attached Figure Description

[0016] Figure 1 This is a connection diagram of the real-time anesthesia assessment system that integrates multi-channel signals from conductive needles according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1As shown, this invention provides a technical solution: a real-time anesthesia assessment system integrating multi-channel conductive needle signals, comprising a data acquisition and classification module, which acquires multi-channel physiological electrical signals by deploying multiple conductive needle electrodes on the patient's body surface or tissues, and generates raw anesthesia data by combining anesthesia parameters; a content adaptation algorithm is used to perform preliminary classification of the raw anesthesia data to obtain classified anesthesia data sets, including key indicators and trends; a terminal adaptation module, which, based on the classified anesthesia data sets, obtains the processing capacity and display specifications of the terminal device from the terminal device registration information, and determines that if the processing capacity and display specifications are higher than a preset threshold, the trend is assigned to the terminal device to obtain data packets for the terminal; a permission management module, which obtains the user's permission level on the terminal device, sorts the user's permission level using a priority ranking algorithm, determines that the terminal device corresponding to the user in the ranking result needs to receive comprehensive anesthesia parameters, and obtains a data subset with matching permissions; and a customized generation module, which generates data for the data subset with matching permissions. The system integrates and fused data packets using a content adaptation algorithm to generate customized data content. If the terminal device is portable and has an auxiliary access level, the data content is simplified to key indicators to obtain a simplified data version. The broadcast module extracts the real-time update portion from the simplified data version and data packets, and broadcasts the real-time update portion to all connected terminals via network protocol. If the broadcast delay exceeds a preset threshold, the key indicator portion is transmitted first to obtain the basic synchronization data after broadcast. The optimization scheduling module obtains the feedback response time of each terminal based on the basic synchronization data after broadcast, and uses a priority sorting algorithm to adjust the subsequent data push order. Terminals with response delays exceeding the preset threshold are only given the simplified data version to obtain an optimized push sequence. The data distribution module distributes the customized data content and the simplified data version to the corresponding terminals according to the optimized push sequence. If network fluctuations are detected during the distribution process, the system switches to a backup channel to transmit data and obtains a final distribution confirmation signal.

[0019] The core of this system lies in integrating conductive needle electrodes with intelligent signal processing and scheduling mechanisms to achieve accurate assessment and real-time transmission of anesthesia status through hardware and software collaboration. First, the acquisition and classification module uses multiple conductive needles 3 deployed on the patient's skin or tissues to continuously acquire multi-channel physiological electrical signals, including electromyography (EMG), electroencephalography (EEG), and skin resistance. This data is then combined with anesthesia parameters output by the anesthesia machine (such as inhaled anesthetic concentration and anesthesia depth indicators) to form raw anesthesia data. This data is filtered, denoised, and normalized by an embedded signal processing chip before being input into the content adaptation algorithm. The content adaptation algorithm, based on a preset model (e.g., a convolutional neural network based on time-series feature extraction), classifies the raw data into key indicators (such as heart rate variability and bispectral index) and trend data (such as the indicator's evolution curve over time). The terminal adaptation module reads the processor frequency, memory capacity, and display resolution of the currently connected terminal from the database to determine if it meets the minimum requirements for displaying trend data (e.g., display refresh rate ≥ 30Hz, processing power ≥ 500MIPS). If it does, the complete data packet is allocated to this type of terminal. The access control module determines user access levels through authentication (e.g., username and fingerprint recognition) and assigns data access permissions to all active users using a priority ranking algorithm (e.g., hierarchical task queue sorting), outputting a data subset matching the permissions (e.g., anesthesiologists receive all indicators, nurses receive only key indicators). The customization generation module calls the data fusion engine to integrate data packets with access matching results, generates adapted content through a graphics rendering engine, and further compresses graphics and data structures based on terminal type (e.g., portable terminals), retaining only key parameters to obtain a simplified version. The broadcast module monitors indicators in the data stream with a change frequency ≥ a threshold as "real-time updates" and transmits them to all terminals on the local area network via UDP broadcast protocol. When a delay exceeds a set limit (e.g., 100ms), it dynamically adjusts to send only key indicators, such as EEG indicators and heart rate, ensuring that critical clinical data arrives first. The optimization scheduling module obtains the feedback time difference of each terminal to the broadcast data by comparing timestamps and adjusts the subsequent data push order using the shortest feedback priority principle. For terminals with response delays exceeding a set limit (e.g., 200ms), only a simplified data version is pushed to avoid congestion affecting backbone node transmission. The data distribution module sends customized or simplified data to each terminal in batches based on the optimized push sequence. If a packet loss rate exceeding 3% or an average bandwidth fluctuation exceeding 10% is detected during the distribution process, it automatically switches to a backup channel (such as an LTE cellular network) and reconnects until a terminal confirmation signal is received. Through the above process, the system achieves a real-time, multi-terminal, permission-sensitive, and network-adaptive visualized assessment and distribution capability for anesthesia status.

[0020] This implementation enables efficient acquisition, dynamic classification, and differentiated distribution of multi-channel anesthesia signals. Its adaptation strategy based on terminal processing capabilities and user permissions significantly improves system response speed and user experience, ensuring that key indicators are prioritized for transmission even under conditions of communication latency or insufficient equipment capacity, thus enhancing the system's real-time performance and robustness. Simultaneously, the customized generation mechanism adjusts data complexity according to device type and permission level, reducing invalid data transmission and improving overall transmission efficiency. Multi-level scheduling and network fluctuation handling mechanisms further enhance the system's data reliability and transmission success rate in complex environments, adapting to the needs of multi-terminal collaboration in clinical settings.

[0021] The data acquisition and classification module acquires multi-channel physiological electrical signals through multiple conductive needle electrodes deployed on the patient's body surface or tissues, and combines this with anesthesia parameters to generate raw anesthesia data. A content-adaptive algorithm is used to initially classify the raw anesthesia data, resulting in classified anesthesia data sets, including key indicators and trends. Specifically, this involves acquiring multi-channel physiological electrical signals through conductive needle electrodes, synchronously reading and processing these signals using a signal acquisition circuit to obtain synchronous physiological electrical signal sets; fusing these synchronous physiological electrical signal sets with anesthesia parameter data to generate raw anesthesia data; using a content-adaptive classification algorithm to initially classify the raw anesthesia data, resulting in categorized anesthesia data sets; extracting key indicators from the categorized anesthesia data sets to generate a key indicator set; performing trend analysis on the key indicator set using a time series analysis algorithm to obtain trend feature data; if the trend feature data exceeds a preset threshold, a support vector machine algorithm is used to detect anomalies and determine abnormal states; and based on the abnormal states, anesthesia state classification results are generated, resulting in the final anesthesia state assessment data.

[0022] First, multiple conductive needle electrodes are inserted into the patient's skin or tissue. Each electrode is made of stainless steel or silver chloride, with a tip length of 10 mm and a diameter of 0.3 mm. The puncture depth is controlled between 5 and 10 mm. At least six conductive needles are inserted, usually an even number, to facilitate bipolar or differential signal acquisition. Each conductive needle is connected to a dedicated signal acquisition circuit via wires. This circuit includes a low-noise preamplifier, a bandpass filter, and a multi-channel analog-to-digital converter module. After system startup, the signal acquisition circuit synchronously reads and processes the analog electrical signals acquired by the conductive needles at a preset frequency. The preset sampling frequency is 1000 times per second, or once every 1 millisecond. To ensure synchronization of all channels, the system uses a unified master clock to control the startup time of each analog-to-digital converter channel and triggers simultaneous sampling of all channels via a control bus. The analog-to-digital converter (ADC) performs voltage-to-digital conversion on each input signal with a sampling precision of 10 bits. This means the input range of 0 to 1 volt is divided into 1024 digital levels. The converted digital signals are output uniformly and timestamped. Each sampled channel signal corresponds to a time point at the same moment, with a total of 1000 sets of synchronous signals acquired per second. Each set contains 6 channel voltage values, forming a synchronized physiological electrophysiological signal group. The data structure of this signal group is matrix-based, where rows represent time points and columns represent channels. To ensure signal quality, data cleaning is first performed: a 100-point window smoothing process is applied to each channel signal using a moving average method. This means that each data point is replaced by the average of its 50 preceding and following data points to suppress noise and occasional signal jumps.

[0023] After cleaning, the system enters the fusion processing stage. First, at each time point, a weighted average is performed on the denoised data of all channels. The weights are initially set to equal weights, meaning the weight of each channel is the reciprocal of the total number of channels; for example, with 6 channels, the weight of each channel is 0.1667. For each time point, the voltage values ​​of the 6 channels are multiplied by their corresponding weights and then summed to obtain the fused voltage value for that time point. This generates 1000 fusion points per second, forming a fused one-dimensional signal data sequence. After completing the electrical signal fusion, the system synchronously acquires anesthesia parameter data. Anesthesia parameters include anesthetic concentration, respiratory rate, ventilation volume, and blood pressure. This data is read from the anesthesia device via a serial communication interface, updated once per second. After reading, the system binds each anesthesia parameter to its corresponding timestamp and interpolates to once per millisecond to ensure complete temporal alignment with the fused signal. Linear interpolation is used, meaning the values ​​between two reads are evenly divided to ensure time series consistency. The fused signal data and the anesthesia parameter data at the corresponding time points together form the original anesthesia data. The raw anesthesia data is stored in structured data format, with each time point corresponding to a fusion signal value and four anesthesia parameter values, forming a data table with five columns. This data table serves as the basic input data for subsequent classification and analysis algorithms, fully reflecting the patient's current multi-channel physiological state and anesthesia parameter information.

[0024] The system acquires raw anesthesia data consisting of fused signals and anesthesia parameters. The data structure is set in groups of 5 columns, each group per millisecond, containing one fused voltage value and four anesthesia parameter values. The system then uses a content-adaptive classification algorithm to perform preliminary classification of the raw anesthesia data. This classification algorithm is a rule-guided classification process with the following steps: Feature Extraction: The system divides the data into windows per second, with each window containing 1000 data points. For the fused voltage value within each window, the maximum and minimum values ​​are calculated, and the difference is taken to obtain the amplitude value of that window. Then, based on the difference between two consecutive fused points, the absolute value is calculated and averaged to obtain the average slope value. Simultaneously, the average value of each anesthesia parameter within each second is calculated as a parameter index under steady-state conditions. Rule Judgment: The data state for that second is determined according to a preset judgment threshold. The specific rules are as follows: If the amplitude value is less than 0.2 and the average slope value is less than 0.05, and none of the four anesthesia parameters exceed 20% of the average value of the previous 10 seconds, it is marked as "normal state"; if the amplitude is between 0.2 and 0.4, or the average slope is between 0.05 and 0.1, or any anesthesia parameter fluctuates by more than 20% but not more than 40%, it is marked as "fluctuating state"; if the amplitude is greater than 0.4, or the slope exceeds 0.1, or any anesthesia parameter fluctuates by more than 40%, it is marked as "abnormal state". Classification output: The classification results are appended to the raw data per second in the form of labels, forming a classified anesthesia data set. This data set retains the original data and the corresponding classification labels, with label values ​​of "normal", "fluctuating", or "abnormal".

[0025] After classification, the system extracts key indicators from the classified anesthesia data sets and generates a set of key indicators. Key indicators include the fusion voltage amplitude per second, average slope, anesthetic concentration, blood pressure, ventilation volume, and normalized rate of change of respiratory rate. The normalized rate of change is calculated by subtracting the historical 10-second mean from the current second value and then dividing by the historical 10-second standard deviation. The system uses these six indicators as a group of key indicators, recording 10 groups consecutively to form a key indicator set for trend analysis. In the trend analysis phase, the system performs time series trend calculations on each indicator in the key indicator set. Specifically, each value in the 10 data groups is arranged chronologically, and its corresponding time point number (e.g., from second 1 to second 10) is used as the independent variable. A linear regression is performed on the indicator values, and the slope of the fitted line is calculated. The absolute value of the slope is the trend velocity. This trend velocity represents the direction and speed of change of the indicator over the past 10 seconds. For example, if the slope of the amplitude trend is positive and the value is greater than 0.05 volts per second, it indicates that the amplitude is increasing rapidly. The system records the trend speed of six key indicators and forms trend feature data. This trend feature data serves as the direct input basis for subsequent anomaly detection, reflecting the continuous change patterns of key physiological signals and anesthesia parameters, and is an important technical support for judging the stability and risk trend of the patient's condition.

[0026] After the system completes trend analysis of the key indicator set and generates trend feature data, it immediately compares each trend feature data with its corresponding preset threshold. The trend feature data includes six items: the rate of change of fusion voltage amplitude, the rate of change of fusion voltage slope, the rate of change of anesthetic concentration, the rate of change of blood pressure, the rate of change of respiratory rate, and the rate of change of ventilation volume. Each data point represents the linear fit slope of that indicator over the past 10 seconds. Specific thresholds are set as follows: 0.05 volts per second for the rate of change of fusion voltage amplitude; 0.02 volts per second for the rate of change of fusion voltage slope; 0.1 volts per second for the rate of change of anesthetic concentration; 5 mmHg per second for the rate of change of blood pressure; 0.3 breaths per second for the rate of change of respiratory rate; and 20 ml per second for the rate of change of ventilation volume. These thresholds are derived from the 90th percentile of statistical analysis of 50 actual clinical anesthesia data cases to ensure sufficient sensitivity to abnormal states and avoid misinterpreting normal fluctuations. If any trend feature data exceeds its corresponding preset threshold, the system immediately initiates the anomaly detection process. Anomaly detection employs a Support Vector Machine (SVM) algorithm. This algorithm model is pre-built into the local embedded processor and is a binary classification linear kernel SVM. The training data comes from at least 10,000 sets of labeled historical anesthesia state data. The specific anomaly detection process is as follows: The system uses six current trend features as input features, standardizing them sequentially to normal distribution values ​​with a mean of 0 and a standard deviation of 1. These six standardized values ​​are then input into the SVM model. Internally, the SVM classifies the input data using a trained hyperplane. The classification result is either "abnormal state" or "non-abnormal state," with an output value of 0 or 1. If the SVM determines it to be an "abnormal state," the system marks that time point as abnormal and records all original data, key indicators, trend data, and the detection result. If it is determined to be a "non-abnormal state," it is recorded as a normal development process.

[0027] After obtaining the abnormal state detection results, the system generates an anesthesia state classification result based on the abnormal state and the classification state of the previous stage. The anesthesia state classification includes five categories: awake, induced, maintained, awake, and dangerous. The specific logic for state generation is as follows: If the support vector machine (SVM) determines an abnormal state for three consecutive time periods (i.e., 30 seconds), and the anesthetic concentration continuously increases while the fusion voltage amplitude significantly decreases, the system sets the state to "induced state"; if the SVM results remain non-abnormal and the key indicator trends are close to stable, the system sets it to "maintained state"; if the amplitude and slope trends continuously increase while the anesthetic concentration significantly decreases, the system sets it to "awakened state"; if the trend data fluctuates drastically and the SVM output is abnormal, the system sets it to "dangerous state"; if all trends are below the threshold and the SVM is non-abnormal, it sets it to "awake state". Finally, the system outputs the above states as "anesthesia state assessment data" for the current time point, which includes a timestamp, classification state label, trend data, and the original input.

[0028] The threshold for the rate of change of fusion voltage amplitude was set at 0.05 volts per second. This value was derived from the analysis of the trend of fusion voltage signals in 50 clinical anesthesia surgeries. The rate of change of this indicator within a 10-second sliding window under normal anesthesia conditions was statistically analyzed, and the 90th percentile was used as the upper limit for judgment. The threshold for the rate of change of fusion voltage slope was set at 0.02 volts per second, obtained in the same way, i.e., extracting the acceptable range of the maximum fluctuation under normal conditions from the slope fluctuation range. The threshold for the rate of change of anesthetic concentration was 0.1 volts per second. This value was obtained by analyzing the average rate of rise of the concentration change curve during the patient's transition from induction to maintenance anesthesia, and then adding one standard deviation to this value. The threshold for the rate of change of blood pressure was 5 mmHg per second, set based on the range analysis of blood pressure changes in any 10-second segment within a 10-minute continuous measurement. The threshold for the rate of change of respiratory rate was 0.3 breaths per second, derived from the maximum fluctuation range of intraoperative monitoring data during the patient's spontaneous breathing and mechanical ventilation phases. The threshold for the rate of change in ventilation volume is 20 ml per second, and its value is based on the minimum controllable ventilation fluctuation range of the mechanical ventilation equipment under normal adjustment conditions.

[0029] The terminal adaptation module, based on the categorized anesthesia data set, obtains the processing power and display specifications of the terminal devices from the terminal device registration information. If the processing power and display specifications exceed a preset threshold, the trend is assigned to the terminal device, obtaining a data packet for the terminal. This packet includes the terminal device registration information obtained from the anesthesia data set, the extraction of processing power and display specification data, and the resulting terminal performance parameter set. If both the processing power and display specifications in the terminal performance parameter set exceed the preset threshold and preset standard, the trend data is formatted to obtain a standard trend data packet. Based on the standard trend data packet, a data compression algorithm is used to compress the data, resulting in a compressed trend data packet. Keyframe data is extracted from the compressed trend data packet, and a time series segmentation algorithm is used to generate segmented trend data, resulting in a segmented trend set. If the data volume of any segment in the segmented trend set exceeds the terminal device's cache capacity, the segmented trend set is divided into blocks to obtain an adaptation block data set. A terminal device adaptation data packet is generated from the adaptation block data set and transmitted to the terminal using a data distribution protocol, resulting in the terminal received data. Based on the terminal received data, the data is decompressed and reassembled on the terminal to obtain the terminal presentation trend data.

[0030] When the terminal adaptation module runs, it first retrieves the terminal device registration information table from the terminal device identifier field in the categorized anesthesia data group. This table is generated when the terminal first connects to the system and is updated periodically, containing the terminal's unique identifier and a description of its current hardware capabilities. The system extracts two types of key data from the registration information: processing capability data and display specification data. Processing capability data includes three items: processor clock speed, memory capacity, and maximum number of threads, which are represented as integer values ​​when read by the system. The processor clock speed is in megahertz, the memory capacity is in megabytes, and the maximum number of threads is an integer. Display specification data includes three items: screen resolution, refresh rate, and color depth. Resolution is expressed in pixels as width multiplied by height, such as 1280 x 720; refresh rate is expressed in hertz; and color depth is expressed in bits, such as 24-bit color. After extracting the above six data items, the system combines them into a set of terminal performance parameters. The system sets a set of preset thresholds: processor clock speed no less than 1500 MHz, memory capacity no less than 2048 MB, maximum number of threads no less than 4; screen resolution no less than 1280 x 720 pixels, refresh rate no less than 30 Hz, color depth no less than 24 bits. These standards are based on actual terminal compatibility testing and determined through stability analysis of 100 terminals. The system compares each data point in the current terminal performance parameter set with the above standards item by item. Only when all six data points are greater than or equal to the standard values ​​is the terminal's performance deemed to meet the trend data display requirements and included in the "high-performance terminal" category. The system records the terminal's status and enters the trend data processing flow. Subsequently, the system calls the trend data content from the generated categorized anesthesia data group. The trend data contains continuous change sequences of key indicators such as fusion voltage, anesthetic concentration, and blood pressure changes over past periods. Each trend data point includes a timestamp, indicator name, and numerical field. The system formats this raw trend data and converts it into a standard trend data package. The formatting process includes three operations: field reordering, data type unification, and unit standardization. Field reordering places the timestamp first, the indicator name in the middle, and the numerical field last; data types are standardized by converting all numerical values ​​to floating-point format; units are standardized by converting all indicator units to a uniform standard, such as blood pressure to millimeters of mercury and anesthetic concentration to volume fraction percentage. The formatted standard trend data package is saved in a structured table format, generating one set of data per second, each set containing trend points corresponding to all key indicators. This data package serves as the basic data structure for subsequent compression, segmentation, and distribution, and is bound to the current terminal device identifier as its exclusive display data content.

[0031] After formatting and generating the standard trend data packet, the system enters the data compression stage. This process employs a two-step compression algorithm. The first step is differential encoding compression, which calculates the difference between each trend value in the standard trend data packet and its trend value at the previous time point, recording only the difference rather than the original absolute value, thus significantly reducing redundant information. For example, if the values ​​of an indicator in the 1st, 2nd, and 3rd seconds are 1.2, 1.25, and 1.3 respectively, they will be recorded as 0, 0.05, and 0.05 after compression. The second step is dictionary encoding compression. The system statistically analyzes the most frequent differences among the differential values, establishes an encoding dictionary, and replaces these frequent differences with short-bit codes to achieve further compression. The final output is a compressed trend data packet with an average compression rate of no less than 50%, meaning that if the original trend data volume per minute is 100 kilobytes, the compressed volume will not exceed 50 kilobytes.

[0032] The system then extracts keyframe data from the compressed trend data packet. A keyframe is defined as a representative moment in the trend data where an inflection point, extreme value, or abrupt change occurs. Specifically, the identification process involves performing a sliding window trend detection operation on the compressed data, with the window size set to 5 seconds. The system calculates the average slope and direction of change of the trend value within each window. When the direction reverses between two consecutive windows, or when the slope of the current window exceeds twice the slope of the previous window, the system marks the time point of the maximum or minimum trend value within that window as a keyframe. All keyframes are extracted chronologically, along with the original trend segments within a 2-second range above and below them as context, combining to form a keyframe extension segment set.

[0033] The system employs a time-series segmentation algorithm to divide the keyframe extension set into segments based on fixed time units, with a segmentation period of 5 seconds, ensuring the temporal integrity and logical continuity of each data segment. Each segment contains a set of trend points for all key indicators within that time period, maintaining a consistent data structure. The system statistically analyzes the size of each trend data segment in bytes to determine if its data volume exceeds the maximum cache capacity declared by the terminal device in its registration information. The cache capacity is measured in megabytes, and the system sets a threshold of 80% of the cache capacity. For example, if the terminal cache capacity is 8 megabytes, then the maximum allowed size for each trend data segment is 6.4 megabytes.

[0034] If the data volume of a segment in the segmented trend set exceeds the threshold, the system will perform block processing on that segment. The block processing method is based on the average number of data points, dividing the data into segments with each block not exceeding 80% of the maximum allowed capacity, automatically calculating the required number of blocks and numbering them. Each block maintains the original data order and is packaged into an independent structure, containing fields such as block number, segment identifier, data content, and checksum, forming an adapted block data group, providing standardized structural support for subsequent terminal data distribution, reception, caching, and reassembly.

[0035] After generating the adapted data chunks, the system enters the data encapsulation stage. The system performs terminal data packet encapsulation on each data chunk, constructing a terminal device-adapted data packet. The encapsulation structure includes a header and a body. The header contains a unique terminal device identifier, the trend segment number to which the current data packet belongs, the block sequence number within the trend segment, the total number of blocks, the data body length, and a checksum, used to identify data order, locate missing data, and ensure data integrity. The body is the data chunk content itself, stored as a compressed trend data sequence, maintaining the original temporal order. After encapsulation, the system calls the data distribution protocol to execute data packet transmission. By default, it uses the TCP protocol, which supports connection persistence and error retransmission mechanisms, and configures the transmission window size to 64 kilobytes, enabling the receive acknowledgment mechanism. During data packet transmission, if network fluctuations or packet loss occur, the protocol mechanism will automatically re-request unacknowledged data packets until all chunks of data are completely transmitted to the designated terminal. After completing the transmission of all chunks for each trend segment, the system records a distribution log and writes the sending status as "completed". After receiving the adaptation data packet, the receiving module first parses the packet header information and determines whether the current block can be cached based on the terminal's local buffer space. If the buffer space is sufficient, the system stores the data packet body content in the local buffer; if the space is insufficient, the system will pause receiving new data and resume receiving after the existing data has been processed. All block data is temporarily stored according to the block sequence number and a data integrity check is performed using the checksum in the packet header. If the check fails, the system automatically requests a retransmission of the block from the server. Once all block data under the same trend segment has been successfully received and verified, the system starts the decompression module to perform a two-step decompression process on the compressed data in the buffer. The first step is dictionary restoration, which replaces the dictionary encoding with the original difference values; the second step is difference restoration, which restores the difference sequence to the original trend value sequence. The original timestamp information is retained during the decompression process to ensure that the data points are arranged in the order of collection time.

[0036] After decompression, the system performs a trend data reorganization operation. This operation aligns the decompression results of all indicators by time, combining them into a multi-column time series data table. Each column represents a key indicator trend sequence, and each row corresponds to a specific time point. Subsequently, the system calls the trend rendering engine in the terminal display module to generate a trend curve graph in a graphical interface, displaying the changes of each key indicator over time within the current trend segment. The display update cycle is set to once per second, with the scrolling update display range being the most recent 60 seconds, ensuring that the terminal presents the anesthesia trend status in real time, smoothly, and continuously for clinical personnel to make intuitive judgments.

[0037] The processor clock speed threshold is set to 1500 MHz. This value was selected through statistical analysis of the minimum processing performance of current mainstream medical mobile terminals and fixed workstations, ensuring a response time of less than 200 milliseconds under high-concurrency data reception and graphics decoding loads. The memory capacity threshold is set to 2048 MB. This value is based on the memory requirements for decompression and caching operations, and can simultaneously buffer and decompress compressed data from at least three trend segments without memory overflow. The number of concurrent threads supported is no less than four, based on the fundamental requirement that the system's multi-threaded structure requires at least one main receiving thread, one decompression thread, one rendering thread, and one control thread to run in parallel. The resolution threshold is set to 1280 x 720 pixels, ensuring that the trend curves have sufficient clarity on the display terminal to show the detailed changes of key indicators. The refresh rate threshold is 30 Hz, set according to the minimum requirements for the trend data refresh cycle and the smoothness of human eye recognition. The color depth threshold is set to 24 bits to meet the accurate presentation of color changes and curve contrast in the graphic display of key trends.

[0038] The permission management module obtains the user's permission level on the terminal device, sorts the user permission levels using a priority sorting algorithm, and determines the terminal device corresponding to the user in the sorted result that needs to receive general anesthesia parameters. This results in a permission matching data subset, which includes user permission level data obtained from the terminal device identifier, sorted using a quicksort algorithm to obtain a list of authorized users. Based on the authorized user list, the module obtains the corresponding terminal device type. If the terminal device type supports the bandwidth transmission protocol, it extracts the general anesthesia parameter requirements, obtaining a set of anesthesia parameters. From the anesthesia parameter set, it extracts the terminal processing capability requirements and uses a data filtering tool to match the terminal processing capabilities, obtaining terminals that meet the high-performance requirements. Device list; Based on the terminal device list, obtain permission matching data. If the permission matching data matches the user's permission level, then the anesthesia parameter set is divided into packets to obtain packetized anesthesia data. Real-time transmission parameters are extracted from the packetized anesthesia data, and data compression algorithms are used to compress the real-time transmission parameters to obtain compressed transmission data. Based on the compressed transmission data, obtain the terminal device's display adaptation parameters. If the display adaptation parameters meet the terminal display specifications, then the compressed transmission data is layered and encapsulated to obtain terminal display data. Update frequency parameters are extracted from the terminal display data. If the update frequency parameters match the timestamp of the anesthesia parameter set, then the terminal display data is stored in the terminal device to obtain terminal storage data.

[0039] During system initialization or terminal deployment, the access control module first calls the user access authentication interface to extract the user identifier from the terminal device identifier and then obtains the current user's access level in the system through a access data table matching method. There are five access levels: Assisted Observation, Nursing Operation, On-Duty Doctor, Attending Physician, and Anesthesiologist. The system assigns a unique number to each access level, from 1 to 5, with higher numbers indicating higher access. During runtime, the system extracts the access numbers from all currently online terminals to form an access number array. This array is then sorted in descending order using a quicksort algorithm. The sorting algorithm employs a recursive three-way partitioning process, sequentially comparing the selected baseline value with the remaining values ​​and adjusting their positions. In typical scenarios with fewer than 1000 terminals, the sorting time does not exceed 50 milliseconds. After sorting, a list of authorized users is obtained, with each item in the list identifying a user and their access level.

[0040] The system further retrieves the terminal registration information table from each user's information in the authorized user list to extract the corresponding terminal device type. The device type field is identified by a string, with common types including fixed workstations, medical tablets, and portable monitoring terminals. The system compares the device types; if the device type contains the words "workstation" or "tablet," or if its network communication field shows support for any of the following protocols: Ethernet interface, 5G module, or WiFi 6, it is determined that the device supports the bandwidth transmission protocol. The determination is based on the network interface's ability to maintain an average stable bandwidth of at least 10 megabits per second during testing, sufficient to support the data transmission of complete anesthesia parameters.

[0041] For users meeting bandwidth requirements, the system determines the set of anesthesia parameters they need to access based on their permission level. When the permission level is 4 or 5 (i.e., attending physician and anesthesiologist levels), the system extracts the complete anesthesia parameter set, including 10 key parameters: EEG trend, voltage fusion signal, anesthetic concentration, blood pressure, heart rate, respiratory rate, carbon dioxide concentration, blood oxygen saturation, muscle relaxation monitoring indicators, and anesthesia status labels. Each parameter is updated between once and ten times per second. Each parameter is represented as a floating-point number, and the average data transmission volume is approximately 80 bytes per transmission. The system calculates the total data transmission requirement. If the average update frequency of the 10 parameters is 5 times per second, the total data load is 10 parameters per second multiplied by 5 times and then multiplied by 80 bytes, resulting in 4000 bytes, or 4 kilobytes per second.

[0042] The system uses this data load to assess the required terminal processing capacity and sets matching thresholds. The minimum processor clock speed is 1800 MHz, which is the minimum requirement to complete decompression, analysis, and visualization rendering of the above data within 500 milliseconds; the memory capacity requirement is no less than 2048 megabytes, which ensures that at least 3 minutes of historical data are cached while processing high-frequency data input; the concurrent thread capability is no less than 4, corresponding to data receiving threads, data parsing threads, graphics rendering threads, and status monitoring threads. These thresholds are derived from stress test results of different parameter combinations under simulated terminal operating conditions, obtained through experimental analysis on 20 real devices.

[0043] The system then activates the data filtering module to compare the processor clock speed, memory capacity, and maximum number of threads of all currently registered terminal devices. Terminal devices meeting all the aforementioned performance thresholds are selected and compiled into a high-performance terminal list. For each terminal in this list, the system compares its current user permission number with the sorting result in the permission user list. If they match, it indicates that the terminal belongs to a high-privilege user and possesses high-performance receiving capabilities; the system confirms that it can receive complete anesthesia parameter data.

[0044] The system divides the complete anesthesia parameter set into packets, grouping each packet into one or two parameters. Each packet is no larger than 100 kilobytes and includes a parameter type identifier, start timestamp, end timestamp, total number of records, and data content field. The system extracts all fields with an update frequency higher than once per second from these packets, using this as the real-time transmission parameter set. The data in this set further undergoes compression processing, employing a sliding window-based differential coding combined with Huffman coding. In the compression step, every 10 consecutive data entries are used as a sliding window unit. The difference between each data entry and the previous one is calculated, and only the difference is recorded. Repeated differences are then replaced with variable-length coding using Huffman tree coding. The average compression ratio reaches 1:2, meaning that the original data is 4 kilobytes per second, while the compressed data is approximately 2 kilobytes per second.

[0045] The compressed data is then transmitted to the display adaptation module. The system reads the display adaptation parameters from the terminal registration information, including display resolution (in pixels), color depth (in bits), and maximum refresh rate (in Hertz). The criteria are a resolution of no less than 1280 x 720 pixels, a color depth of no less than 24 bits, and a refresh rate of no less than 30 Hertz. This standard is derived from the minimum requirements for the smoothness and visual recognition of the anesthesia curve. If the terminal meets all the requirements, the system performs a three-layer encapsulation on the compressed data. The first layer establishes a data group index according to parameter categories. The second layer adds a timestamp index for trend line coordinate generation. The third layer converts the data structure into a format supported by the terminal display interface, such as SVG or Canvas, ultimately generating the terminal display data.

[0046] The system further extracts the update frequency field from the displayed data and compares it with the original sampling timestamp field in the anesthesia parameter set. It uses a line-by-line matching method to check if the time intervals are the same. If all are identical, it is determined to be time-synchronized data. The system writes this synchronized data to the terminal's local cache by time period and stores it as a structured data file. The file structure includes the display segment number, update timestamp, display format version, and graphic reconstruction parameters, used for subsequent data retrieval or trend playback, forming the final version of the data stored on the terminal.

[0047] User access level thresholds are divided into five levels, set according to the user job responsibilities in the hospital information management system. These are: Assisted Observation Level, Nursing Operation Level, On-Duty Doctor Level, Attending Physician Level, and Anesthesiologist Level, corresponding to numbers 1 to 5. This ensures that only users with access levels 4 or 5 can access all anesthesia parameters. The bandwidth protocol support threshold is determined by network interface performance testing, requiring terminal devices to have at least one high-speed transmission capability, including Ethernet, 5G modules, or WiFi 6 protocols. In actual testing, all of these protocols supported a continuous data stream of at least 10 megabits per second. Regarding processing power, the minimum processor clock speed is set at 1800 MHz, derived from the minimum stable operating frequency required to process 10 anesthesia parameters per second and update each parameter 10 times in a simulation environment. The memory capacity threshold is set at 2048 megabytes to ensure that the system does not overflow when data decompression, graphics rendering, and historical caching are performed simultaneously. The minimum number of threads is required to be at least four, corresponding to the resource allocation for four parallel tasks: receiving, decompression, drawing, and interaction. In the display adaptation parameters, the resolution threshold is set to 1280 x 720 pixels, which is determined based on the minimum display requirements for displaying multiple indicators without overlap in the graphical interface according to the trend chart; the color depth is set to 24 bits, which can cover the color recognition requirements of conventional medical graphics; the refresh rate is not less than 30 Hz, which is derived from the minimum smooth frame rate standard for the human visual system to recognize dynamic change curves.

[0048] The customized generation module uses a content adaptation algorithm to fuse and generate customized data content for data subsets and data packages that match permissions. If the terminal device is portable and the permission level is auxiliary, the data content is simplified to key indicators. The simplified data version includes obtaining user permission level data from the terminal device identifier, classifying permission levels using a data filtering tool to obtain a categorized permission set; extracting auxiliary permission level data from the categorized permission set, and fusing the auxiliary permission data and data packages using a content adaptation algorithm to obtain preliminary customized data; based on the preliminary customized data, if the terminal device is portable, extracting key indicator data to obtain a simplified indicator set; obtaining data display requirements from the simplified indicator set, and using a format conversion tool to perform structured processing on the simplified indicator set to obtain adapted display data; based on the adapted display data, if the terminal device's display specifications support a resolution lower than a preset threshold, compressing the adapted display data to obtain compressed display data; extracting update frequency parameters from the compressed display data, and verifying the update frequency parameters using a data validation tool to obtain verified data; based on the verified data, if the data integrity meets a preset threshold, storing the verified data on the terminal device to obtain terminal storage data.

[0049] The customized generation module first extracts the user's bound permission level information from the unique identifier of the current terminal device. This permission level is written by the hospital information system through the permission synchronization interface. The data structure is an integer field with a value range of 1 to 5, corresponding to auxiliary permissions, nursing permissions, on-call doctor permissions, attending physician permissions, and anesthesiologist permissions, respectively. A level of 1 indicates auxiliary permissions. The module calls the system's data filtering tool to classify the permission level field of all online terminal users. The classification logic is to match records with a permission level field value equal to 1, extracting them to form a set of categorized permissions used to identify all auxiliary permission terminals.

[0050] The system then enters the content fusion processing stage. The content adaptation algorithm fuses the data packets corresponding to users with auxiliary permissions with their permission restrictions. The fusion method is a field-level merging strategy. First, the system iterates through all parameter fields in the data packet, including 10 parameters: heart rate, respiratory rate, blood pressure, blood oxygen saturation, anesthetic concentration, carbon dioxide concentration, EEG trend, fused electrical signals, muscle relaxation monitoring, and anesthesia status. Then, the set of parameters allowed to be accessed by users with permission level 1 is set as the filtering condition. The filtering result retains only four parameter fields: heart rate, respiratory rate, blood pressure, and anesthetic concentration; the remaining six data fields are discarded. This permission filtering strategy is set according to the data authorization level system for non-professional clinical terminals in the national medical information management regulations, ensuring that the data access scope is consistent with the permission level.

[0051] After the above fusion operation is completed, preliminary customized data is output. The data format uses a timestamp-indexed structured array, and each record contains a field name, corresponding value, unit symbol, and millisecond-level timestamp. The system then determines whether the current terminal device type is portable, based on the device type field in the terminal device identification information. If the value of this field contains any of the keywords "mobile," "handheld," or "portable," or if the screen size field is less than or equal to 6.5 inches, the system determines that the terminal is a portable device. This screen size threshold is set according to the maximum size specification standard for handheld devices to ensure accurate classification.

[0052] After the terminal is determined to be portable, the system continues to extract the aforementioned four key indicator fields from the initial customized data to form a simplified indicator set. Each parameter in this indicator set has the following attributes: heart rate in beats per minute, ranging from 30 to 180; respiratory rate in breaths per minute, ranging from 10 to 40; blood pressure in millimeters of mercury, with systolic pressure ranging from 90 to 180; and anesthetic concentration expressed as a volume fraction percentage, ranging from 0 to 6. All values ​​are stored as floating-point numbers, uniformly retaining two decimal places. Each data point is bound to a corresponding timestamp for synchronous calculation.

[0053] The system invokes a format conversion tool to perform structured processing on the simplified indicator set. The target format is a unified key-value pair structure, with each record including four fields: parameter name, parameter value, unit, and timestamp. The output format uses JSON encoding, and the size of each data point is controlled to not exceed 128 bytes to ensure stable decoding even in slow network environments. The system then reads the display specification fields from the terminal device registration information, including screen resolution width and height. If either resolution dimension is less than 800 pixels wide or 480 pixels high, it is considered a low-resolution terminal. This threshold is derived from the analysis of minimum display requirements for trend data visualization; it is known that only four data rows can be displayed at a height of 480 pixels, and exceeding this will result in content truncation.

[0054] The system performs compression processing on the data to be displayed. The compression algorithm includes field name replacement and unit merging strategies. Specifically, "heart_rate" is replaced with "HR", and "respiratory_rate" is replaced with "RR". At the same time, the unit identifier field and parameter name are merged to form a compound key, such as "HR_mmHg". The parameter value type is changed from 2 floating-point precision to 1 digit, that is, the value is retained to one decimal place. After the above compression, a single data entry is controlled within 64 bytes.

[0055] After compressing and displaying the data, the system extracts the update frequency parameter field. Each parameter is bound to an update cycle attribute, in seconds, during generation. The system reads the update frequency field value and uses a data validation tool to verify its validity and consistency. The validity standard is that the value is an integer, ranging from 1 to 10, derived from the typical refresh cycle setting of the core parameters for anesthesia monitoring in the system. The system then compares the update frequency value with the timestamp field one by one, checking whether all data points increase according to a uniform periodic pattern. If all data meets the above conditions, it is determined to be validated data.

[0056] After verification, the data enters the integrity verification process. The system performs field integrity checks on the data structure, checking whether each record contains four items: parameter name, value, unit, and timestamp. If any field is missing, the record is considered incomplete. Next, time continuity is calculated, checking if the timestamps of 100 consecutive records are arithmetic progressions and if the difference equals the update cycle, with a maximum allowed time drift of 1 millisecond. The data missing rate is calculated as the number of incomplete records divided by the total number of records, with a system threshold of 1%. Data exceeding this threshold is considered incomplete. If all three conditions are met, the system writes the verified data to the terminal device's local cache and saves it in a structured data file format. This file structure includes a file header, data segment index, field structure definitions, and actual value segments, serving as the data source for terminal display and retrieval.

[0057] The broadcast module extracts real-time updates from simplified data versions and data packets, broadcasts these updates to all connected terminals via network protocols, and prioritizes the transmission of key indicator data if the broadcast delay exceeds a preset threshold. The resulting basic synchronization data includes real-time update data obtained from the basic synchronization data. This data is then segmented using a data segmentation tool to obtain a fragmented data set. Based on the fragmented data set, if the terminal connection status is high load, the fragmented data set is prioritized to obtain a sorted data set. Key indicator data is extracted from the sorted data set and compressed using a data compression tool. Compression processing is performed to obtain compression index data. Based on the compression index data, if the data synchronization frequency exceeds a preset threshold, the transmission rate is adjusted through network protocol configuration to obtain adjusted transmission data. The data broadcast range is obtained from the adjusted transmission data, and a data distribution tool is used to distribute the adjusted transmission data in a targeted manner to obtain targeted distribution data. Based on the targeted distribution data, if the terminal connection status supports multi-threaded transmission, a parallel transmission tool is used to process the targeted distribution data to obtain parallel transmission data. Data integrity parameters are extracted from the parallel transmission data, and a data verification tool is used to verify the data integrity parameters to obtain verified synchronization data.

[0058] The broadcast module first receives simplified data versions and complete data packets. At the beginning of each broadcast cycle, the system compares and filters these two types of data, using a timestamp matching mechanism to identify data entries that have been updated within the last second. Specifically, it reads the timestamp of each data record, calculates the difference between the current system time and that timestamp (in milliseconds), and considers any difference less than or equal to 1000 milliseconds as "real-time updated data," forming a real-time updated data set. The system then inputs this real-time updated data set into a data fragmentation tool, which segments the data according to a maximum byte limit of 512 bytes. This maximum byte limit is based on the typical maximum transmission unit of network protocols and terminal unpacking capability tests. If a record's structure exceeds 512 bytes, the tool splits it into multiple ordered sub-fragments, each with a sequence number tag, timestamp, and field structure header, generating a fragmented data set.

[0059] Next, the system determines whether it is under high load based on the current connection status of each connected terminal. The connection status assessment includes three metrics: queue length (in terms of entries), average response latency (in milliseconds), and network bandwidth utilization (percentage). These metrics are obtained in real time by reading the system communication buffer, log statistics, and network interface monitoring. If any metric exceeds a threshold—queue length exceeding 50 entries, average response latency exceeding 200 milliseconds, or bandwidth utilization exceeding 80%—the system determines it to be under high load. These thresholds are based on system load test data during 200 actual surgeries to ensure the system can still respond promptly under high concurrency environments.

[0060] Under high load, the system performs priority sorting on the sharded dataset. The sorting rule is to prioritize key indicators before general indicators. Key indicators include heart rate, blood pressure, respiratory rate, and anesthetic concentration. The sorting algorithm uses a stable sorting method. First, key indicator shards are marked as priority 1, and non-key indicator shards are marked as priority 2. Then, they are sorted in descending order of update timestamp. In the case of the same priority, the newer data is placed first. After sorting, a sorted dataset is generated.

[0061] From the sorted dataset, the system extracts a subset of key indicator data (i.e., those fragmented records with priority 1) and sends these records to the data compression tool. The compression tool performs compression using strategies such as field name abbreviation (e.g., "heart_rate" is abbreviated to "HR", "resp_rate" is abbreviated to "RR"), numerical precision reduction (from retaining two decimal places to one decimal place), and unit merging (e.g., "mmHg" is retained as the abbreviation "mmHg"). The size of a single compressed record is controlled within 64 bytes, keeping the overall compression rate at around 60%.

[0062] Subsequently, the system records the synchronization frequency of the compressed key metrics, i.e., the number of records per unit time (1 second). If this synchronization frequency exceeds the system's set threshold (100 records per second), the system controls the transmission rate by adjusting network protocol parameters. Specific adjustments include reducing the TCP window size (e.g., from the default 64 kilobytes to 32 kilobytes), reducing the maximum transmission unit (MTU, e.g., from 1500 bytes to 512 bytes), and modifying the acknowledgment retransmission strategy (e.g., extending the ACK waiting time to 50 milliseconds), to alleviate network congestion and ensure timely delivery of critical data. This threshold of 100 records per second is set as an empirical upper limit based on the maximum stable broadcast rate observed by the system in multiple surgical communication tests.

[0063] After rate adjustment, the compression index data is labeled with broadcast metadata such as terminal identifier and area identifier, and the system determines the broadcast range. Based on the connection status, location identifier, and online status in the terminal registry, the system uses a data distribution tool to send the adjusted transmission data only to terminal devices that are online, have unrestricted receiving capabilities, and are located in the current surgical area, thus forming targeted data distribution. The filtering rules are: the terminal online flag is true, the receiving capability flag is not "restricted," and the area label matches the broadcast area.

[0064] After receiving the targeted data, the system determines whether each target terminal supports multi-threaded transmission. The support standard is determined by the thread support value field in the registration information; if this field value is greater than 1, the terminal supports multi-threaded transmission. For supported terminals, the system splits their corresponding targeted data into multiple sub-streams and assigns an independent transmission thread to each sub-stream, sending them concurrently to reduce overall transmission latency. Transmission scheduling uses a thread pool, with each thread responsible for sending one sub-stream of data. The sending order is synchronously controlled to ensure consistent segment order.

[0065] After parallel transmission concludes, the system extracts data integrity parameters from the feedback received from each terminal, including: packet loss rate (formulated as the percentage of failed received packets / expected received packets), retransmission count (statistics from the system's internal retransmission log), checksum consistency flag (the result of comparing the CRC checksum of each segment), and field integrity rate (i.e., the percentage of actual received fields / original number of fields). The system calls a data verification tool to verify each item: if the packet loss rate is less than 1%, the retransmission count is less than 3, all segment checksums pass the comparison, and the field integrity rate is 100%, then it is determined to be verified synchronized data. The data that passes the verification is marked as the final basic synchronization data output by the broadcast module for use by downstream modules.

[0066] The real-time data extraction window is set to 1 second, determined based on the frequency of changes in key physiological indicators during actual anesthesia, ensuring the system can capture sufficiently timely and valuable data changes. The maximum number of bytes per data fragment is set to 512 bytes. This value is determined based on the Maximum Transmission Unit (MTU) limits of mainstream network protocols (such as TCP / IP) and terminal processing capabilities testing, to avoid fragment loss or terminal decoding failures during network transmission. The thresholds for determining high terminal load include three aspects: a queue length greater than 50 entries, a response latency exceeding 200 milliseconds, and bandwidth utilization exceeding 80%. These three indicators are performance limits summarized after concurrent connection tests on various terminal types, ensuring the system still possesses flow control capabilities under extreme communication pressure. The threshold for key indicator synchronization frequency is set to 100 entries per second. This value references the balance between the generation frequency of monitoring data and the terminal processing speed in typical surgical scenarios, avoiding network congestion caused by frequent synchronization. In the data integrity assessment, the packet loss rate is less than 1%, the number of retransmissions does not exceed 3, the field integrity rate is 100%, and the checksum is consistent. The above standards refer to the definition of high reliability transmission in the national communication protocol standard and are adapted and adjusted according to the real-time requirements of this system to ensure that the broadcast synchronous data has high integrity and reliability.

[0067] The optimization scheduling module, based on the basic synchronization data after broadcast, obtains the feedback response time of each terminal, and uses a priority sorting algorithm to adjust the subsequent data push order. Terminals with response delays exceeding a preset threshold are only given a simplified data version. The resulting optimized push sequence includes obtaining the terminal reception status from the optimized push sequence, classifying the terminal reception status using a status analysis tool to obtain a classified status set; based on the classified status set, if a terminal's reception status is below a preset threshold bandwidth, a data fragmentation tool is used to segment the push sequence to obtain a fragmented data set; data priority parameters are extracted from the fragmented data set, and... The sorting algorithm sorts the data priority parameters to obtain a sorted data set. Based on the sorted data set, if the push frequency exceeds a preset threshold, the transmission rate is adjusted through the network protocol to obtain adjusted transmission data. The data broadcast range is obtained from the adjusted transmission data, and a targeted distribution tool is used to distribute the adjusted transmission data to obtain targeted distribution data. Based on the targeted distribution data, if the terminal supports multi-threaded transmission, a parallel transmission tool is used to process the targeted distribution data to obtain parallel transmission data. Transmission integrity parameters are extracted from the parallel transmission data, and a verification tool is used to verify the transmission integrity parameters to obtain verified synchronization data.

[0068] The optimization scheduling module first records the response time data of all terminals after the basic synchronization data broadcast is completed. It measures the single response time in milliseconds by sending data request packets with timestamps and receiving acknowledgment response packets from the terminals. The system executes this process multiple times to obtain the average response time. If the average exceeds a preset response latency threshold of 300 milliseconds, the terminal is marked as a high-latency terminal and set to receive only simplified data versions.

[0069] The system then extracts terminal reception status parameters from the optimized push sequence, mainly including three items: current reception bandwidth, remaining cache capacity, and data processing rate. These parameters are monitored in real time by the network management module, read from the system cache management interface to read remaining memory, and calculated by the decoder data output rate interface to measure the amount of data processed per second. These parameters are then input into a status analysis tool for structured processing. The system has a pre-defined set of classification rules: if the bandwidth is less than 200 kilobytes per second, the remaining cache space is less than 20% of the terminal's total cache capacity, or the data processing rate is less than 64 frames per second, the terminal's reception status is determined to be low-performance, and it is classified into the weak connection device set.

[0070] For terminals in a weakly connected device set, their originally planned push data will enter a data fragmentation process. The system calls a data fragmentation tool to divide the complete data content into segments with a maximum size of 512 bytes. Each segment is automatically appended with metadata such as a segment identifier, the original total number of segments, and the current segment number for subsequent sorting and reassembly. The fragmentation strategy prioritizes field structure and key indicators, ensuring that critical data is prioritized and packaged into the first or highest-priority segments. Simultaneously, the compressed structure of each segment retains complete field boundaries to prevent field breaks. The final fragmented data set will serve as the actual push data content for that terminal and proceed to the next push process. This mechanism ensures that core anesthesia monitoring data can still be effectively acquired even under conditions of limited terminal performance, improving the system's resilience and real-time performance.

[0071] After the system segments the data, it attaches a priority label to each segment. This priority parameter consists of two dimensions: data type weight and data generation time weight. The data type weight is set based on the actual clinical urgency of anesthesia monitoring. For example, key indicators such as "blood pressure," "heart rate," and "respiratory rate" are weighted at 3, while auxiliary indicators such as "body temperature" and "historical records of anesthetic drugs" are weighted at 1. The data generation time weight is calculated based on the difference between the data's timestamp and the current system time. The closer the time is to the present, the higher the weight, with a linear decreasing weight from 3 to 1. For example, a generation time within 1 second of the present has a weight of 3, 1 to 3 seconds has a weight of 2, and more than 3 seconds has a weight of 1.

[0072] The system uses a priority scoring function to sum the two weights mentioned above to generate a comprehensive priority score. Then, a stable sorting algorithm (such as merge sort) is used to sort the sharded data set from high to low according to the comprehensive priority score, ensuring that high-priority data can be pushed first, forming a sorted data set.

[0073] After prioritizing the data, the system immediately calculates the total data volume and number of fragments in the sorted dataset within the current push cycle, and then calculates the current push frequency, including "total data pushed per second" and "number of fragments pushed per second." If the "total data pushed per second" exceeds 512 kilobytes or the "number of fragments pushed per second" exceeds 100, the current push frequency is considered to have exceeded the preset transmission limit. These two thresholds are determined based on the maximum network throughput and decoding processing capabilities supported by most portable terminals in actual clinical use, where 512 kilobytes is the maximum network capacity test limit and 100 fragments is the maximum stable decoding capability of the decoder.

[0074] When the system detects that the push frequency exceeds any threshold, it immediately initiates a network protocol configuration adjustment procedure, which includes the following steps: First, the TCP transmission window is adjusted to 32 kilobytes to slow down the single data flow rate; second, the size of a single data packet is limited to within 512 bytes to reduce the risk of data packet fragmentation; then, the ACK response waiting time is increased to 50 milliseconds to reduce acknowledgment pressure; finally, congestion control strategies are enabled to dynamically monitor network load and adjust the sending rate in real time. These adjustment parameters are completed by calling the system's underlying network protocol interface, ensuring that the adjustment process does not affect data integrity. The final adjusted transmission data will be distributed at a secure and optimized rate, effectively reducing the risk of data loss and network congestion, and improving the overall system transmission efficiency.

[0075] After generating the adjusted transmission data, the system first invokes the network topology identification program to determine the data broadcast range based on the IP address, MAC address, and registration area information of the currently connected terminals, combined with the network access layer structure diagram. The data broadcast range is defined as the set of all terminals in the network node area to which the current surgical operation belongs that are "online" and "receiveable". The system reads the location information and current connection status of each terminal, removes terminals that are disconnected or inactive, retains target terminals with receiving capabilities, and uses the identifiers of these terminals as the broadcast target list, forming the targeted distribution targets for this push.

[0076] Subsequently, the system inputs the adjusted transmission data into the directional distribution tool. After the distribution tool reads the broadcast target list, it encapsulates data packets one by one according to the terminal ID, and binds each data packet with its unique target terminal address and session identifier. The data encapsulation process also includes information such as timestamps, shard sequence numbers, CRC check codes, etc., which are used for subsequent verification and recombination. The directional distribution tool uses a reliable transmission protocol to send data packets one by one and records the sending status of each terminal in real time.

[0077] During the directional distribution process, the system checks the capability parameters of each terminal, including the "number of thread supports", which is specified in the device description document during the terminal registration phase. If a certain terminal is marked as supporting multi-threaded transmission, the system will start the parallel transmission tool to perform thread splitting on the distributed data for it. The parallel transmission tool evenly divides the data content of each terminal into multiple sub-data blocks, and the number of sub-data blocks is equal to the number of thread supports. Subsequently, an independent transmission queue is configured for each thread, and the data is sent to the terminal in parallel using an asynchronous communication mechanism. Each thread carries meta-information such as an independent identification number, shard sequence number, transmission order, etc., to ensure that the terminal can perform data recombination after receiving.

[0078] After the transmission ends, the system immediately extracts the transmission integrity parameters from the information returned by the terminal, mainly including four indicators: packet loss rate, number of retransmissions, field integrity rate, and CRC check consistency. The packet loss rate is calculated by the difference between the total number of sent shards recorded by the system and the number of shards confirmed to be received by the terminal. If the number of lost shards does not exceed 1% of the total number of shards, it is judged as qualified; the number of retransmissions is automatically counted by the system side, and if it does not exceed 3 times, it is judged as normal; the field integrity rate requires the terminal to feedback whether all fields in each shard are complete and correct, and 100% is considered up to standard; the CRC check consistency is obtained by the terminal returning the recalculated CRC value of each shard and comparing it with the original CRC value, and all being the same is regarded as the data not being damaged. The verification tool makes a joint judgment on the above four parameters. If all meet the standards, the output result is "verified synchronous data", and this data is marked as displayable or further used for anesthesia evaluation and analysis, completing the data transmission link closed-loop.

[0079] The data distribution module, based on the optimized push sequence, distributes customized data content and simplified data versions to corresponding terminals. If network fluctuations are detected during distribution, it switches to a backup channel for data transmission. Obtaining a final distribution confirmation signal involves: extracting terminal type parameters from the optimized push sequence and classifying terminals using a classification tool to obtain a segmented terminal set; extracting terminal customization requirement parameters from the segmented terminal set and generating customized data content and simplified data versions using a content generation tool to obtain a generated data set; and extracting data distribution priorities from the generated data set and prioritizing data distribution using a priority sorting algorithm. The data is sorted to obtain a sorted distribution sequence. Based on the sorted distribution sequence, the data distribution process is initiated. If network fluctuations are detected during the distribution process, the data is switched to a backup channel using a channel switching tool, resulting in a switched transmission channel. Transmission status parameters are obtained from the switched transmission channel, and a status monitoring tool is used to analyze these parameters, resulting in an analyzed transmission status. Based on the analyzed transmission status, if the transmission status is stable, a distribution confirmation signal is generated using a distribution confirmation tool, resulting in a final distribution confirmation signal. Distribution completion parameters are extracted from the final distribution confirmation signal, and a log recording tool is used to store these parameters, resulting in a stored distribution record.

[0080] First, upon receiving the optimized push sequence, the system immediately initiates the terminal identification process, extracting the identification information of each terminal associated with the push sequence, including parameters such as the terminal's unique identifier, device model, network interface type, operating system type, screen size, and processor performance level. These terminal type parameters are all derived from the content uploaded and stored in the terminal information database when the terminal device registers with the system or connects for the first time. The system accesses these parameters by indexing them using the terminal's unique identifier.

[0081] Next, the system uses a classification tool to perform structured processing on the extracted terminal type parameters, specifically including the following steps: First, based on the device model and operating system fields, terminals are divided into two main categories: "mobile portable terminals" and "fixed terminals." Second, network performance is determined based on the network interface type; for example, terminals supporting Gigabit Ethernet are labeled as high-bandwidth terminals, while those supporting Wi-Fi or 4G are labeled as medium-to-low-bandwidth terminals. Third, based on processor performance level and memory size, terminals are categorized as "high-performance" or "low-performance." The system uses rule-based matching to complete the terminal type classification, with each terminal assigned to a specific terminal set, and each set assigned a unique type label.

[0082] The classification tool outputs a "separated terminal set" data structure, which includes a list of terminal IDs corresponding to each terminal type, device attribute details, and capability parameters required for subsequent data distribution. This structure is used for matching and distributing customized and simplified data versions, ensuring data compatibility with terminal capabilities and laying the foundation for subsequent distribution processing. This step ensures that the system can implement differentiated data push strategies based on terminal performance differences, improving data transmission efficiency and terminal reception success rate.

[0083] After classifying the terminal sets, the system first extracts customized requirement parameters from each category. These parameters mainly include three types of information: First, permission level parameters, used to determine whether the terminal user is an attending physician, anesthesiologist, or surgical assistant; permission information is determined by the user's authentication level after logging into the system. Second, data display capability parameters, including the terminal's screen resolution, supported graphics processing capabilities, and display refresh rate, obtained from terminal registration information and the system detection module. Third, network processing capability parameters, including the terminal's maximum supported receiving rate, buffer capacity, and average response latency, obtained by the system through historical records from the data distribution module and real-time network testing.

[0084] Based on these three types of parameters, the system invokes the content generation tool to initiate the data content generation process. First, for terminals with an "advanced" permission level (such as the attending physician's terminal), it generates complete, customized data content including all anesthesia parameters, trend data, historical records, and abnormal warnings. For terminals with an "auxiliary" permission level (such as anesthesia assistants or mobile terminals), it generates a simplified data version containing only core key indicators, such as heart rate, blood pressure, and respiratory rate. The content generation tool automatically filters data fields, determines the data refresh frequency and display format according to the parameter settings, and finally outputs a structured generated dataset. Each data entry includes information such as the target terminal identifier, data content type, data size, and expected transmission frequency.

[0085] Subsequently, the system extracts the distribution priority of each data item from the generated dataset. The data distribution priority parameter is calculated based on a weighted average of the following factors: first, the user's permission level, with higher permissions assigned a priority weight of 3 and secondary permissions a weight of 1; second, the data's real-time requirements, with key indicator data assigned a weight of 3 and secondary indicator data a weight of 1; and third, the terminal's connection status, with a weight of 2 if the network is stable and the response time is less than 100 milliseconds, otherwise 1. The system sums these three weights to obtain a total priority score, with a higher score indicating a higher priority for the data in the push sequence.

[0086] The system uses a priority sorting algorithm (such as quicksort or heapsort) to sort all data entries from highest to lowest priority, forming the final sorted distribution sequence. This sequence serves as the input order for the data distribution process, ensuring that critical data is prioritized for sending to terminals with high privileges, good network connectivity, and strong real-time requirements. This optimizes the efficiency and reliability of data push, while also ensuring the system's robustness in resource-constrained or network-unstable situations.

[0087] After prioritizing the distribution sequence, the system immediately initiates the data distribution process, pushing customized data content or simplified data versions to the corresponding terminals sequentially according to the order defined in the sorted distribution sequence. Before distributing each data item, the system first calls the network connection detection module to perform link detection on the current main channel, including indicators such as latency, packet loss rate, and bandwidth stability. Under normal circumstances, the system pushes data item by item through the main communication channel (such as the main Wi-Fi channel or Ethernet channel). Each time a data is pushed, the system records the sending timestamp, waits for the terminal's response confirmation, calculates the round-trip time of the data packet, and monitors whether retransmission occurs.

[0088] During data distribution, if the system detects network fluctuations—that is, if any network performance indicator exceeds a preset threshold in three consecutive data transmissions—it automatically triggers a channel switching mechanism. Specific criteria for judging network fluctuations include, but are not limited to: average latency exceeding 100 milliseconds, packet loss rate exceeding 2%, more than two consecutive retransmissions, and a short-term drop in network bandwidth exceeding 30%. When any of these criteria are triggered, the system immediately invokes the channel switching tool to check the availability of a preset backup communication channel. The backup channel can be a second network interface, a cellular data link, or a local subnet channel. If the backup channel is available, the system performs a switching action, directing the current data stream to the backup channel and re-encapsulating the current data packets for re-pushing through the new channel.

[0089] After the switchover is complete, the system obtains real-time transmission status parameters from the switched transmission channel, including key indicators such as data throughput, retransmission rate, connection loss rate, and average packet interval. The status monitoring tool monitors these parameters in real time and compares them with standard transmission performance indicators. If the monitoring results indicate that the transmission status is "good"—meaning the throughput is greater than the originally set 500 kilobytes per second, the retransmission rate is less than 1%, and the connection remains stable for more than 5 seconds without packet loss—the system determines that the backup channel's transmission status is "stable" and can continue to be used for subsequent data distribution tasks. Otherwise, the system will reassess other available channels or, if system policy allows, switch back to the primary channel to retry, ensuring that data can continuously and reliably reach the target terminal under network fluctuations. Through this scheme, the system can achieve a highly robust data transmission guarantee mechanism under dynamic network conditions.

[0090] After the status monitoring tool completes real-time analysis of the switched transmission channel, the system compares the analysis results with a set of preset stability standards. Specifically, if the analyzed transmission status meets the following three conditions: first, there is no interruption for 10 consecutive seconds during data transmission, and the latency of each packet is less than 50 milliseconds; second, there is no retransmission of 50 consecutively sent data packets; and third, the response confirmation time after the data packet arrives at the terminal does not exceed 100 milliseconds, then the system determines the current channel's transmission status as "stable".

[0091] Once the transmission status is confirmed to be stable, the system initiates the distribution confirmation process. The distribution confirmation tool retrieves the terminal response log from the current transmission status to verify whether the terminal has successfully received the distributed data content and checks whether the integrity of the data packet meets the verification standards. The verification standards include a 100% data packet integrity arrival rate, no missing bytes in the data content after comparison, and a 100% terminal decoding success rate. After the above conditions are met, the distribution confirmation tool immediately generates a distribution confirmation signal, which includes the following information fields: terminal identifier, distributed data packet number, confirmation timestamp, confirmation status (completed), channel type (primary or backup), and data content type (customized or simplified).

[0092] Subsequently, the system extracts a set of distribution completion parameters from the final distribution confirmation signal. These parameters include: distribution start and end times, total distribution duration, whether channel switching was triggered, channel type used for distribution, total number of data packets, number of successfully received packets, number of retransmissions, and final confirmation result. The system calls the logging tool to write this set of distribution completion parameters into the log system in a structured data format and simultaneously stores it in the background database. Database fields include terminal number, task number, distribution status, network status indicators, anomaly flags, and time records.

[0093] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A real-time anesthesia assessment system integrating multi-channel signals from a conductive needle, characterized in that, include: The data acquisition and classification module acquires multi-channel physiological electrical signals by placing multiple conductive needle electrodes on the patient's body surface or in the tissues, and generates raw anesthesia data by combining anesthesia parameters. The original anesthesia data were initially classified using a content adaptation algorithm, resulting in classified anesthesia data groups, including key indicators and trends. The terminal adaptation module obtains the processing capability and display specifications of the terminal device from the terminal device registration information based on the classified anesthesia data group. If the processing capability and display specifications are higher than the preset threshold, the trend is allocated to the terminal device to obtain the data packet for the terminal. The permission management module obtains the user's permission level on the terminal device, sorts the user's permission level by priority sorting algorithm, determines the terminal device corresponding to the user in the sorting result that needs to receive the general anesthesia parameters, and obtains the data subset for permission matching; The customized generation module uses a content adaptation algorithm to fuse and generate customized data content for data subsets and data packets that match permissions. If the terminal device is portable and the permission level is auxiliary, the data content is simplified to key indicators to obtain a simplified data version. The broadcast module extracts the real-time update portion from the simplified data version and data packet, broadcasts the real-time update portion to all connected terminals through the network protocol, and determines if the broadcast delay exceeds a preset threshold, then prioritizes the transmission of key indicator portions to obtain the basic synchronization data after broadcast. The scheduling module is optimized. Based on the basic synchronization data after broadcast, the feedback response time of each terminal is obtained. The priority sorting algorithm is used to adjust the order of subsequent data pushes. Terminals with response delays exceeding a preset threshold are only given a simplified data version, thus obtaining an optimized push sequence. The data distribution module distributes customized data content and simplified data versions to the corresponding terminals based on the optimized push sequence. If network fluctuations are detected during the distribution process, it switches to the backup channel to transmit data and obtains the final distribution confirmation signal. The customized generation module uses a content adaptation algorithm to fuse and generate customized data content for data subsets and data packets that match permissions. If the terminal device is portable and the permission level is auxiliary, the data content is simplified to key indicators, resulting in a simplified data version including: User permission level data is obtained from the terminal device identifier, and the permission levels are classified using a data filtering tool to obtain a set of categorized permissions; Auxiliary permission level data is extracted from the category permission set, and the auxiliary permission data and data package are fused using a content adaptation algorithm to obtain preliminary customized data; Based on the initial customized data, if the terminal device is portable, then key indicator data is extracted to obtain a simplified indicator set; The data display requirements are obtained from a simplified indicator set. A format conversion tool is used to perform structured processing on the simplified indicator set to obtain data suitable for display. Based on the adaptation display data, if it is determined that the terminal device display specifications support a resolution lower than a preset threshold, the adaptation display data is compressed to obtain compressed display data. The update frequency parameter is extracted from the compressed display data, and the update frequency parameter is verified using a data verification tool to obtain the verified data. Based on the verified data, if the data integrity meets the preset threshold, the verified data is stored in the terminal device to obtain the terminal stored data.

2. The real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 1, characterized in that: The acquisition and classification module acquires multi-channel physiological electrical signals by deploying multiple conductive needle electrodes on the patient's body surface or in the tissues, and generates raw anesthesia data by combining anesthesia parameters. The original anesthesia data was initially classified using a content adaptation algorithm, resulting in classified anesthesia data sets, including key indicators and trends, specifically: Multi-channel physiological electrical signals are acquired using conductive needle electrodes, and synchronous reading and processing are performed using a signal acquisition circuit to obtain a synchronous physiological electrical signal group. A method for fusing synchronous physiological electrical signals and combining them with anesthesia parameter data is used to generate raw anesthesia data. The original anesthesia data were initially classified using a content-adaptive classification algorithm to obtain categorized anesthesia data groups. Key indicators were extracted from the categorized anesthesia dataset to generate a set of key indicators. Trend analysis is performed on the set of key indicators, and time series analysis algorithms are used to obtain trend characteristic data; If the trend feature data exceeds the preset threshold, the support vector machine algorithm is used to detect anomalies in the trend feature data and determine the abnormal state. Anesthesia status classification results are generated based on abnormal conditions, and the final anesthesia status assessment data is obtained.

3. The real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 1, characterized in that: The terminal adaptation module, based on the categorized anesthesia data group, obtains the processing capacity and display specifications of the terminal device from the terminal device registration information. If the processing capacity and display specifications are higher than a preset threshold, the module allocates the trend to the terminal device, obtaining data packets for the terminal including: The terminal device registration information was obtained from the anesthesia data set, and the processing capacity and display specification data were extracted to obtain the terminal performance parameter set. If the terminal's performance parameters, centralized processing capability, and display specifications are all higher than the preset threshold and preset standard, then the trend data is formatted to obtain a standard trend data package. Based on the standard trend data packet, a data compression algorithm is used to compress the data to obtain a compressed trend data packet; Keyframe data is extracted from compressed trend data packets, and segmented trend data is generated using a time series segmentation algorithm to obtain a segmented trend set; If the amount of data in any segment of the segmented trend set exceeds the cache capacity of the terminal device, the segmented trend set is divided into blocks to obtain an adapted block data group. The terminal device adaptation data packet is generated by adapting the block data group, and the adaptation data packet is transmitted to the terminal using the data distribution protocol to obtain the terminal received data. Based on the data received by the terminal, the data is decompressed and reassembled on the terminal to obtain the trend data presented by the terminal.

4. The real-time anesthesia assessment system integrating multi-channel signals from conductive needles according to claim 1, characterized in that: The permission management module obtains the user's permission level on the terminal device, sorts the user's permission levels using a priority ranking algorithm, determines the terminal device corresponding to the user in the ranking result that needs to receive general anesthesia parameters, and obtains a data subset for permission matching including: User permission level data is obtained from the terminal device identifier, and the permission level data is sorted using a quicksort algorithm to obtain a list of authorized users; Based on the list of authorized users, obtain the corresponding terminal device type. If the terminal device type supports the bandwidth transmission protocol, extract the general anesthesia parameter requirements and obtain the anesthesia parameter set. The terminal processing capability requirements are extracted from the anesthesia parameter set, and the terminal processing capabilities are matched using data filtering tools to obtain a list of terminal devices that meet the high-performance requirements. Based on the list of terminal devices, obtain permission matching data. If the permission matching data matches the user's permission level, then the anesthesia parameter set is subdivided to obtain subdivided anesthesia data. Real-time transmission parameters are extracted from the subcontracted anesthesia data, and data compression algorithms are used to compress the real-time transmission parameters to obtain compressed transmission data. Based on the compressed transmission data, the display adaptation parameters of the terminal device are obtained. If the display adaptation parameters meet the terminal display specifications, the compressed transmission data is then layered and encapsulated to obtain the terminal display data. The update frequency parameter is extracted from the terminal display data. If the update frequency parameter matches the timestamp of the anesthesia parameter set, the terminal display data is stored in the terminal device to obtain the terminal stored data.

5. The real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 1, characterized in that: The broadcast module extracts the real-time update portion from the simplified data version and data packets, broadcasts the real-time update portion to all connected terminals via network protocol, and determines whether the broadcast delay exceeds a preset threshold, prioritizing the transmission of key indicator portions. The resulting basic synchronization data after broadcast includes: Real-time updated data is obtained from basic synchronized data, and data sharding tools are used to divide the real-time updated data to obtain a sharded data set; Based on the fragmented data set, if the terminal connection status is high load, the fragmented data set is prioritized and sorted to obtain the sorted data set; Key indicator data is extracted from the sorted dataset, and the key indicator data is compressed using a data compression tool to obtain compressed indicator data. Based on the compression index data, if the data synchronization frequency exceeds the preset threshold, the transmission rate is adjusted through network protocol configuration to obtain the adjusted transmission data. The data broadcast range is obtained from the adjusted transmission data, and the adjusted transmission data is distributed in a targeted manner using a data distribution tool to obtain targeted distribution data; Based on the targeted distribution data, if the terminal connection status supports multi-threaded transmission, then a parallel transmission tool is used to process the targeted distribution data to obtain parallel transmission data. Data integrity parameters are extracted from parallel transmitted data, and data verification tools are used to verify the data integrity parameters to obtain verified synchronized data.

6. The real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 1, characterized in that: The optimization scheduling module obtains the feedback response time of each terminal based on the basic synchronization data after broadcast, adjusts the subsequent data push order using a priority sorting algorithm, and determines that terminals with response delays exceeding a preset threshold only receive simplified data versions, thus obtaining the optimized push sequence including: The terminal reception status is obtained from the optimized push sequence, and the terminal reception status is classified using a status analysis tool to obtain a set of classified statuses. Based on the classified state set, if the terminal's receiving state is below the preset threshold bandwidth, then the data fragmentation tool is used to segment the push sequence to obtain the fragmented data set. Extract data priority parameters from the partitioned dataset, sort the data priority parameters using a sorting algorithm, and obtain the sorted dataset.

7. The real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 6, characterized in that: The optimized scheduling module obtains the feedback response time of each terminal based on the basic synchronization data after broadcast, adjusts the subsequent data push order using a priority sorting algorithm, and determines that terminals with response delays exceeding a preset threshold only receive simplified data versions. The optimized push sequence also includes: Based on the sorted data set, if the push frequency exceeds a preset threshold, the transmission rate is adjusted through the network protocol to obtain the adjusted transmission data. The data broadcast range is obtained from the adjusted transmission data, and the adjusted transmission data is distributed using a targeted distribution tool to obtain targeted distribution data; Based on the targeted distribution data, if the terminal supports multi-threaded transmission, then a parallel transmission tool is used to process the targeted distribution data to obtain parallel transmission data. Transmission integrity parameters are extracted from parallel transmission data, and verification tools are used to verify the transmission integrity parameters to obtain verified synchronization data.

8. The real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 1, characterized in that: The data distribution module, for the optimized push sequence, distributes customized data content and simplified data versions to the corresponding terminals respectively. If network fluctuations are detected during the distribution process, it switches to a backup channel to transmit data. Obtaining the final distribution confirmation signal includes: Terminal type parameters are extracted from the optimized push sequence, and a classification tool is used to divide the terminal types to obtain the divided terminal set; Based on the segmented terminal set, extract the terminal customization requirement parameters, and use the content generation tool to generate customized data content and simplified data versions to obtain the generated data set; Extract the data distribution priority from the generated dataset, and sort the data distribution priority using a priority sorting algorithm to obtain the sorted distribution sequence.

9. A real-time anesthesia assessment system integrating multi-channel signals from a conductive needle according to claim 8, characterized in that: The data distribution module, for the optimized push sequence, distributes customized data content and simplified data versions to the corresponding terminals respectively. If network fluctuations are detected during distribution, it switches to a backup channel for data transmission. Obtaining the final distribution confirmation signal also includes: Based on the sorted distribution sequence, the data distribution process is initiated. If network fluctuations are detected during the distribution process, the data is switched to a backup channel using a channel switching tool to obtain the switched transmission channel. The transmission status parameters are obtained from the switched transmission channel, and the transmission status parameters are analyzed using a status monitoring tool to obtain the analyzed transmission status. Based on the analyzed transmission status, if the transmission status is stable, a distribution confirmation signal is generated through the distribution confirmation tool to obtain the final distribution confirmation signal. The distribution completion parameters are extracted from the final distribution confirmation signal, and the distribution completion parameters are stored using a logging tool to obtain the stored distribution record.

Images