Post-anesthesia tracking and monitoring system

Abstract

This invention provides a post-anesthesia follow-up monitoring system for monitoring and early warning of post-brain surgery conditions. Specifically, it includes: a patient brain surgery information synchronization module, a follow-up monitoring task generation module, a complication early warning handling module, a quality control and data management module, a standardized electronic form module, a mobile follow-up monitoring module, a discharge remote follow-up monitoring module, and a system management module. The modules are divided into main modules and auxiliary modules. Each module in the main modules interacts bidirectionally with three or more other modules and performs core functions such as data processing, optimization, and parameter calculation. The auxiliary modules provide auxiliary functions such as data acquisition, task execution, and access control. This invention automates the follow-up monitoring process, significantly reducing post-operative PONV, respiratory depression, and nerve damage, and achieves full retention of perioperative anesthesia data, supporting departmental quality control, grade review, research projects, and thesis data.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical monitoring technology, and in particular to a post-anesthesia follow-up monitoring system. Background Technology

[0002] Traditional postoperative anesthesia follow-up relies on paper records, manual telephone follow-up, and manual data entry and statistics, which have drawbacks such as untimely follow-up, non-standard records, easy data loss, delayed early warning of complications, difficulty in ensuring follow-up rate, and inability to automatically extract research data. At the same time, perioperative anesthesia data is separated from ward data, making it impossible to form a closed-loop management of anesthesia safety.

[0003] There is an urgent need for a digital system capable of automatic tracking, real-time early warning, data sharing, and intelligent analysis to achieve full-cycle, traceable, and standardized management of post-anesthesia care from in-hospital to out-of-hospital care. This system should integrate clinical scenarios such as nerve block efficacy assessment, electrical impedance localization puncture complication tracking, pain management, and ERAS (Enhanced Recovery After Surgery) to improve post-operative safety management of regional and minimally invasive anesthesia. Summary of the Invention

[0004] Specifically, the present invention is achieved through the following technical solution:

[0005] This invention provides a post-anesthesia follow-up monitoring system for monitoring and early warning of post-brain surgery status. Specifically, it includes eight modules: a patient brain surgery information synchronization module, a follow-up monitoring task generation module, a complication early warning handling module, a quality control and data management module, a standardized electronic form module, a mobile follow-up monitoring module, a discharge remote follow-up monitoring module, and a system management module.

[0006] The above eight modules are divided into main modules and auxiliary modules. Each module in the main modules interacts bidirectionally with three or more modules and completes the core functions of data processing, optimization, and parameter calculation. These include the patient brain surgery information synchronization module, the tracking and monitoring task generation module, the complication early warning and handling module, and the quality control and data management module.

[0007] Among them, the patient brain surgery information synchronization module serves as the basic data source for all modules, enabling bidirectional data exchange with the other three main modules, pushing core basic data, and simultaneously receiving optimization parameters and configuration parameters;

[0008] The tracking and monitoring task generation module interacts bidirectionally with the other three main modules, receives basic data from the patient brain surgery information synchronization module, and generates personalized tracking and monitoring tasks.

[0009] The complication early warning and processing module interacts bidirectionally with two main modules and four auxiliary modules, pushes early warning data, and receives algorithm optimization parameters to dynamically calibrate early warning thresholds and optimize prediction models.

[0010] The quality control and data management module exchanges data bidirectionally with the other three main modules, receives various types of data, and performs aggregation, statistics, and analysis.

[0011] In addition, the auxiliary modules complete auxiliary functions such as data collection, task execution, and access control, including a standardized electronic form module, a mobile tracking and monitoring module, a discharge remote tracking and monitoring module, and a system management module.

[0012] Optionally, the standardized electronic form module interacts bidirectionally with only the two main modules, receives tracking and monitoring task instructions from the tracking and monitoring task generation module, and provides standardized forms for data entry.

[0013] The mobile tracking and monitoring module interacts bidirectionally with only two main modules, pushing tracking and monitoring data, task completion status, early warning and handling results, and providing feedback on the execution effect;

[0014] The discharge remote tracking and monitoring module interacts bidirectionally with only one main module to enable timely early warning and handling of abnormalities after discharge.

[0015] The system management module interacts bidirectionally with all four main modules, providing access control, system parameter configuration, and initial algorithm parameter settings for each main module.

[0016] Optionally, the tracking and monitoring task generation module, as the main module for executing the core algorithm, generates personalized tracking and monitoring plans based on basic data, automatically sorts tasks, and optimizes the tracking and monitoring strategy in conjunction with the early warning module. This module completes multi-dimensional data collection and preprocessing, postoperative tracking and monitoring risk quantification calculation, intelligent generation of personalized tracking and monitoring plans, automatic sorting of tracking and monitoring tasks, and task management.

[0017] The multi-dimensional data acquisition and preprocessing automatically collects multi-dimensional patient data, performs standardized preprocessing on the collected data, and outputs standardized data.

[0018] The postoperative follow-up monitoring risk quantification calculation uses standardized data from the multi-dimensional data collection and preprocessing as input parameters. Based on this data, a multi-dimensional risk assessment model is constructed to calculate the patient's postoperative follow-up monitoring risk score.

[0019] The personalized tracking and monitoring plan is generated intelligently. The input parameters are the corrected total risk score and risk level output by the postoperative tracking and monitoring risk quantification calculation. Combined with the patient's surgical type and anesthesia method, a personalized tracking and monitoring plan is generated and dynamically adjusted according to the patient's real-time recovery status.

[0020] Optionally, the multi-dimensional data acquisition and preprocessing specifically includes:

[0021] (1) First, set the scores for the fixed data in the multidimensional data, and then set the weights according to the importance to perform weighted calculations to obtain fixed quantitative values;

[0022] (2) Standardize the changing data, first filter out outliers, correct outliers to the maximum or minimum value within the range, and then use the standardization formula to output standardized continuous data, i.e. standard values, to ensure the accuracy of subsequent risk scoring algorithm calculations.

[0023] Optionally, the postoperative follow-up monitoring risk quantification calculation specifically includes:

[0024] (1) Determine the weight of each evaluation indicator: Use a combination of expert indicators and analytical indicators to reduce subjective bias. First, have multiple experts score each evaluation indicator by expert indicator weight, then calculate the weight of analytical indicators, and finally obtain the combined weight as the main parameter for scoring calculation;

[0025] (2) Quantitative scoring of indicators: Each evaluation indicator is graded and quantitatively scored, and the quantitative scores of each indicator are output;

[0026] (3) Risk score calculation: The total risk score for postoperative follow-up monitoring of patients is calculated using a weighted summation formula, and the total risk score is output;

[0027] (4) Risk decay correction: Add a time decay coefficient. The longer the postoperative time, the lower the risk. Correct the total risk score and output the corrected total risk score as the main input parameter for subsequent module data processing and calculation.

[0028] (5) Risk level classification: The risk level range is set based on actual clinical data. Different risk levels correspond to different tracking and monitoring strategies. The risk level is output as the input parameter for the intelligent generation of personalized tracking and monitoring plans.

[0029] Optionally, the intelligent generation of the personalized tracking and monitoring plan specifically includes:

[0030] (1) Generate and output the basic tracking and monitoring plan;

[0031] (2) Dynamic adjustment of tracking and monitoring frequency: Calculate the rate of change of indicators and calculate the new tracking and monitoring frequency as input parameters for automatic sorting of tracking and monitoring tasks.

[0032] Optionally, the complication early warning processing module monitors and tracks monitoring data and dynamically calibrates the early warning threshold to achieve early prediction, graded early warning, and closed-loop treatment of complications, thereby reducing the risk of complications. This module completes the construction of early warning indicators, abnormal data monitoring and analysis, early warning level calculation, and complication early warning.

[0033] The early warning indicators are constructed by taking historical tracking and monitoring data and complication case data as input parameters, constructing a multi-dimensional complication early warning indicator system, and training the optimal threshold for each early warning indicator.

[0034] The abnormal data monitoring and analysis system takes as input parameters the optimal threshold of the early warning indicator and real-time tracking and monitoring data. The system collects tracking and monitoring data, vital sign data, and test result data in real time, analyzes the data in real time, calculates the rate and magnitude of data change, and identifies abnormal data.

[0035] The warning level calculation takes abnormal data, abnormal trends, and the number of abnormal indicators as input parameters, uses a quantitative scoring method to calculate the warning level, outputs the warning level value, and simultaneously synchronizes the warning data to the mobile tracking and monitoring module and the quality control and data management module.

[0036] The complication early warning system uses abnormal data, abnormal trends, and historical patient tracking and monitoring data as input parameters to construct a complication prediction model and achieve complication early warning.

[0037] Optionally, the abnormal data monitoring and analysis specifically includes:

[0038] (1) Calculation of rate of change: The early warning indicators are continuously collected, and the linear regression algorithm is used. The data collected three times in a row and the corresponding time are input, the rate of change of the three times in a row is calculated, and the abnormal trend indicator is output.

[0039] (2) Calculation of change range: Calculate the absolute value of the difference between the current data and the previous data, the ratio of the current data to the previous data, determine abnormal data fluctuations, and output abnormal fluctuation indicators;

[0040] (3) Anomaly determination: Combine threshold, rate of change and magnitude of change for comprehensive determination. If a single data exceeds the calibrated threshold, or if the data shows an abnormal trend or abnormal fluctuation, it is determined to be abnormal data, and the abnormal data and anomaly identifier are output. If multiple warning indicators are abnormal at the same time, they are determined to be high-risk anomalies, and high-risk anomaly identifiers are output as input parameters for warning level calculation and complication prediction algorithm, triggering the warning process.

[0041] Optionally, the patient brain surgery information synchronization module, as the main data acquisition module, automatically captures and synchronizes basic data related to the patient, surgery, and anesthesia. This module completes data collection, real-time data synchronization, data supplementation, and data association and archiving.

[0042] The data collection is achieved by automatically collecting information through standard interfaces connecting to hospital information systems, electronic medical record systems, and surgical anesthesia systems.

[0043] The real-time data synchronization refers to synchronizing intraoperative data and patient condition changes to this system in real time.

[0044] The data supplementation refers to editing and supplementing the synchronized data;

[0045] The data association and archiving automatically associates and archives patient information, brain surgery information, anesthesia information, and follow-up monitoring records to form a personal post-anesthesia follow-up monitoring file for the patient.

[0046] Optionally, the quality control and data management module aggregates data from all modules for statistical analysis, outputs quality control reports and algorithm optimization parameters, and assists in the upgrade of each module. This module completes indicator statistics and report generation, data filtering and organization, and data compliance analysis;

[0047] The standardized electronic form module provides standardized tracking and monitoring forms that conform to clinical guidelines, simplifies data entry operations, and pushes standardized tracking and monitoring data to the main module;

[0048] The mobile tracking and monitoring module is adapted to clinical mobile scenarios, supports medical staff to enter tracking and monitoring data in real time at the bedside, receive and execute tracking and monitoring / early warning tasks, and provide feedback on the task execution effect;

[0049] The discharge remote tracking and monitoring module conducts post-discharge tracking and monitoring through multiple channels, collects remote tracking and monitoring data and abnormal patient feedback, and provides continuous data support for the main module's early warning algorithm.

[0050] The system management module is responsible for system permission allocation, parameter configuration, and log auditing, ensuring the safe and standardized operation of each module, and does not participate in core algorithm and data interaction.

[0051] The beneficial effects of this invention are as follows:

[0052] (1) The tracking and monitoring process is automated, reducing manual data entry and paper record work by more than 80%.

[0053] (2) Early detection and intervention of complications significantly reduce postoperative PONV, respiratory depression, nerve damage and other adverse events.

[0054] (3) To achieve full retention of perioperative anesthesia data, supporting departmental quality control, grade review, research projects and paper data support.

[0055] (4) Unify tracking and monitoring standards to avoid problems such as missed visits, incorrect visits, and non-standard records.

[0056] (5) It can be connected to techniques such as nerve block, regional anesthesia, and electrical impedance localization puncture to achieve targeted tracking of postoperative effects and complications.

[0057] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0058] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0059] Figure 1 This is a schematic diagram of data interaction between the various modules of the system of the present invention. Detailed Implementation

[0060] The invention will now be described with reference to several embodiments. It should be understood that these embodiments are described only to enable those skilled in the art to better understand and thus implement the invention, and are not intended to imply any limitation on the scope of the invention.

[0061] Specific numerical values ​​or ranges may be referred to in the following description. It should be understood that these values ​​and ranges are merely exemplary and may be helpful in putting the ideas of the invention into practice. However, the description of these examples is not intended to limit the scope of the invention in any way. These values ​​or ranges may be set differently depending on the specific application scenario and requirements.

[0062] This invention provides a post-anesthesia follow-up monitoring system for monitoring and early warning of post-brain surgery status. Specifically, it includes eight modules: a patient brain surgery information synchronization module, a follow-up monitoring task generation module, a complication early warning and handling module, a quality control and data management module, a standardized electronic form module, a mobile follow-up monitoring module, a discharge remote follow-up monitoring module, and a system management module.

[0063] like Figure 1 As shown, the above eight modules are divided into main modules and auxiliary modules. Each module in the main modules interacts bidirectionally with three or more modules and completes the core functions of data processing, optimization, and parameter calculation. These include the patient brain surgery information synchronization module, the tracking and monitoring task generation module, the complication early warning and handling module, and the quality control and data management module.

[0064] Among them, the patient brain surgery information synchronization module serves as the basic data source for all modules, enabling bidirectional data exchange with the other three main modules. Specifically, it pushes core basic data such as patient basic information, surgical anesthesia information, and historical tracking and monitoring data to the tracking and monitoring task generation module, complication early warning and handling module, and quality control and data management module; at the same time, it receives optimization parameters from the quality control and data management module and configuration parameters from the system management module.

[0065] The tracking and monitoring task generation module interacts bidirectionally with the other three main modules: it receives basic data from the patient brain surgery information synchronization module to generate personalized tracking and monitoring tasks; it pushes task progress, task sorting, and other data to the quality control and data management module; it receives algorithm optimization parameters from the quality control and data management module to optimize task generation and sorting logic; and it achieves data exchange with the complication early warning processing module to synchronize tracking and monitoring tasks and early warning-related data.

[0066] The complication early warning and handling module interacts bidirectionally with two main modules and four auxiliary modules. Specifically, it receives basic data and historical complication data from the patient brain surgery information synchronization module, as well as relevant tracking and monitoring data; it pushes early warning data and early warning handling results to the quality control and data management module; it receives algorithm optimization parameters from the quality control and data management module, dynamically calibrates early warning thresholds, and optimizes prediction models; and it synchronizes data with the tracking and monitoring task generation module to adjust tracking and monitoring strategies in a coordinated manner.

[0067] The quality control and data management module communicates bidirectionally with the other three main modules. It receives various types of data (basic data, tracking and monitoring data, early warning data, task data, etc.) and summarizes, statistically analyzes, and pushes algorithm optimization parameters and quality control data to the other three main modules to assist in algorithm iteration and process optimization.

[0068] The auxiliary modules complete auxiliary functions such as data collection, task execution, and access control, including a standardized electronic form module, a mobile tracking and monitoring module, a discharge remote tracking and monitoring module, and a system management module.

[0069] The standardized electronic form module interacts bidirectionally with only two main modules: it receives tracking and monitoring task instructions from the tracking and monitoring task generation module and provides standardized forms for data entry; it pushes standardized tracking and monitoring data to the tracking and monitoring task generation module and the complication early warning and handling module; and it receives form configuration parameters from the main modules to adapt to the needs of the core modules.

[0070] The mobile tracking and monitoring module interacts bidirectionally with only two main modules: it receives tracking and monitoring tasks and priority instructions from the tracking and monitoring task generation module, and receives early warning information and handling suggestions from the complication early warning and handling module; it pushes tracking and monitoring data, task completion status, and early warning handling results to the two main modules and provides feedback on the execution effect. This helps the main modules to dynamically adjust tasks and early warning strategies.

[0071] The discharge remote tracking and monitoring module interacts bidirectionally with only one main module. It receives tracking and monitoring plans and warning instructions from the main module, collects post-discharge patient tracking and monitoring data and abnormal feedback data, and pushes relevant data to the main module to achieve timely warning and handling of post-discharge abnormalities.

[0072] The system management module interacts bidirectionally with all four main modules, providing access control, system parameter configuration, and initial algorithm parameter settings for each main module; it receives operation logs and algorithm running logs from each main module and performs system maintenance, but does not participate in core data interaction or algorithm logic, only providing basic data interaction.

[0073] Each of the eight modules mentioned above performs its corresponding function in the overall system and uses the data processing results as the basis for interaction with multiple modules.

[0074] 1. The patient brain surgery information synchronization module, as the main data acquisition module, automatically captures and synchronizes basic data related to the patient, surgery, and anesthesia. This module completes data collection, real-time data synchronization, data supplementation, and data association archiving.

[0075] The specific functions are as follows:

[0076] The data collection process involves connecting to the hospital information system, electronic medical record system, and surgical anesthesia system via standard interfaces. It automatically collects basic patient information (name, gender, age, hospital number, contact information, underlying diseases, etc.), brain surgery information (surgery name, surgery time, surgeon, surgery type, etc.), and anesthesia information (anesthesia method, anesthesiologist, nerve block method, intraoperative medication, intraoperative vital signs, intraoperative complications, etc.). No manual data entry is required from medical staff. This data serves as the foundational input for all subsequent algorithms (data preprocessing, risk scoring, early warning calculation, etc.).

[0077] The real-time data synchronization refers to synchronizing intraoperative data and changes in the patient's condition (such as test results and adjustments to the diagnosis) to this system in real time, serving as dynamic input parameters for subsequent risk score calculations, etc.

[0078] The data supplementation involves editing and supplementing the synchronized data, with the editing operations automatically recorded in the log.

[0079] The data association and archiving automatically associates and archives patient information, brain surgery information, anesthesia information, and follow-up monitoring records to form a personal post-anesthesia follow-up monitoring file for the patient. This file serves as historical input data for subsequent algorithms (trend analysis, model training, complication prediction), optimizing data processing and calculation results. It also facilitates subsequent querying, tracing, and scientific research.

[0080] 2. The tracking and monitoring task generation module, as the main module for executing the core algorithm, generates personalized tracking and monitoring plans based on basic data, automatically sorts tasks, and optimizes the tracking and monitoring strategy in conjunction with the early warning module. This module completes multi-dimensional data collection and preprocessing, postoperative tracking and monitoring risk quantification calculation, intelligent generation of personalized tracking and monitoring plans, automatic sorting of tracking and monitoring tasks, and task management.

[0081] The specific functions are as follows:

[0082] The multi-dimensional data acquisition and preprocessing automatically collects multi-dimensional patient data, including preoperative baseline data (age, weight, type of underlying disease, liver and kidney function indicators, etc.), intraoperative data (operation duration, anesthesia method, dosage of anesthetic drugs, whether intraoperative complications occurred, etc.), and surgery-related data (operation type, surgical difficulty score, surgical site, etc.). The collected data undergoes standardized preprocessing, and the standardized data is output as input parameters for subsequent postoperative follow-up monitoring and risk quantification calculations. Specifically:

[0083] (1) First, set scores for fixed data (such as basic disease type and anesthesia method) in multidimensional data (such as hypertension = 2 points, diabetes = 3 points, general anesthesia = 3 points, local anesthesia = 1 point), and then set weights according to importance (such as basic disease weight 0.3, anesthesia method weight 0.2) to perform weighted calculation and obtain fixed quantitative values.

[0084] Fixed quantified value = score × weight.

[0085] (2) Standardize the variable data (such as age, weight, and operation time), first screen outliers, correct the outliers to the maximum value within the range, and then use the standardization formula to output standardized continuous data, i.e. standard values, to ensure the accuracy of subsequent risk scoring algorithm calculations.

[0086] The formula for filtering outliers is as follows:

[0087] Outliers > mean + 3 × standard deviation or outliers < mean - 3 × standard deviation

[0088] The mean is the arithmetic mean of a set of variable data (such as the age of all patients or the duration of all surgeries); the standard deviation is the standard deviation of a set of variable data based on the mean.

[0089] The standardized formula is:

[0090] Standard value = (Original value - Mean) ÷ Standard deviation

[0091] The original values ​​represent the changes before standardization.

[0092] The postoperative follow-up monitoring risk quantification calculation uses standardized data from the multi-dimensional data collection and preprocessing as input parameters. Based on this data, a multi-dimensional risk assessment model is constructed to calculate the patient's postoperative follow-up monitoring risk score. The specific calculation steps are as follows:

[0093] (1) Determine the weights of each assessment indicator: A combination of expert indicators and analytical indicators is used to reduce subjective bias. First, multiple experts score each assessment indicator (analysis of anesthetized patient status ASA, surgical difficulty, age, underlying diseases, intraoperative complications, etc.) with expert indicator weights (the total weight is 1, such as ASA grade 0.25, surgical difficulty 0.2, etc.). Then, the weights of analytical indicators are calculated to obtain the combined weights, which serve as the main parameters for scoring.

[0094] The formula for calculating the weights of the analytical indicators is as follows:

[0095] Analysis indicator weight = (1 - entropy value) ÷ sum of (1 - entropy values) of all indicators

[0096] Entropy = - (1 ÷ ln total number of samples) × (proportion of all samples × ln proportion)

[0097] Among them, the entropy calculation method is the most commonly used.

[0098] The formula for combined weights is:

[0099] Combined weight = 0.6 × expert indicator weight + 0.4 × analysis indicator weight

[0100] (0.6 / 0.4 are empirical coefficients and can be adjusted according to actual circumstances.)

[0101] (2) Quantitative scoring of indicators: Each evaluation indicator is graded and quantitatively scored (1-5 points, 1 point for low risk, 5 points for high risk). For example, ASA grade I = 1 point, II = 2 points, III = 3 points, IV = 4 points, V = 5 points; surgical difficulty score 1-5 points (1 point for routine minor surgery, 5 points for complex major surgery); age ≤ 40 years old = 1 point, 41-60 years old = 2 points, 61-70 years old = 3 points, 71-80 years old = 4 points, > 80 years old = 5 points, and the quantitative score of each indicator is output.

[0102] (3) Risk score calculation: The total risk score for postoperative follow-up monitoring is calculated using a weighted summation formula, and the total risk score is output. The specific formula is as follows:

[0103] Total Risk Score = (Score of Indicator 1 × Portfolio Weight 1) + (Score of Indicator 2 × Portfolio Weight 2) + ... + (Score of Indicator n × Portfolio Weight n)

[0104] Each indicator corresponds to a combined weight, such as indicator 1 corresponding to combined weight 1.

[0105] (4) Risk decay correction: A time decay coefficient is added. The longer the postoperative time, the lower the risk. The total risk score is corrected, and the corrected total risk score is output as the main input parameter for subsequent data processing and calculation. The specific formula is:

[0106] Corrected total risk score = total risk score × e^(-k × postoperative hours) (e≈2.718, k is the attenuation coefficient, such as 0.01)

[0107] (5) Risk level classification: Set the level range based on actual clinical data. For example, after correction, the risk is 1.0-2.0 as low risk, 2.1-3.5 as medium risk, and 3.6-5.0 as high risk. Different risk levels correspond to different tracking and monitoring strategies. Output the risk level as the input parameter for the intelligent generation of personalized tracking and monitoring plans.

[0108] The personalized tracking and monitoring plan is intelligently generated. The input parameters are the corrected total risk score and risk level output from the postoperative tracking and monitoring risk quantification calculation. Combined with the patient's surgical type and anesthesia method, a personalized tracking and monitoring plan is generated and dynamically adjusted according to the patient's real-time recovery. The specific rules and algorithm steps are as follows:

[0109] (1) Basic tracking and monitoring plan generation: For low-risk (1.0-2.0 points) patients, the tracking and monitoring time nodes are postoperative time, 24 hours postoperatively, 7 days postoperatively, etc., and the tracking and monitoring method is mainly online questionnaire tracking and monitoring; For medium-risk (2.1-3.5 points) patients, the tracking and monitoring time nodes are postoperative time, 6 hours postoperatively, 24 hours postoperatively, 48 hours postoperatively, 7 days postoperatively, etc., and the tracking and monitoring method is a combination of in-hospital bedside tracking and monitoring and online tracking and monitoring; For high-risk (3.6-5.0 points) patients, the tracking and monitoring time nodes are postoperative time, 3 hours postoperatively, 6 hours postoperatively, 12 hours postoperatively, 24 hours postoperatively, 48 hours postoperatively, 7 days postoperatively, 14 days postoperatively, etc., and the tracking and monitoring method is mainly in-hospital bedside tracking and monitoring and video tracking and monitoring, and a basic tracking and monitoring plan is generated.

[0110] (2) Dynamic adjustment of tracking and monitoring frequency: Calculate the rate of change of indicators and calculate the new tracking and monitoring frequency, which serves as the input parameter for automatically sorting tracking and monitoring tasks, thereby realizing the dynamic optimization of the tracking and monitoring plan.

[0111] The formula for calculating the new tracking and monitoring frequency is:

[0112] Monitoring frequency = (Current indicator value - Previous indicator value) ÷ Previous indicator value × 100%

[0113] The automatic sorting tracking and monitoring tasks take as input parameters the personalized tracking and monitoring plan intelligently generated and output, and the corrected total risk score. Combining optimal matching and multi-target priority sorting, intelligent task sorting is achieved, specifically:

[0114] A multi-objective optimization model is introduced, with input parameters including the corrected total risk score, the current number of tasks and the average number of tasks for medical staff, and the overdue time of tasks. It takes into account three dimensions: risk, workload, and timeliness, and calculates the ranking score. At the same time, the ranking score results are synchronized to the mobile tracking and monitoring module.

[0115] The specific formula is as follows:

[0116] Ranking score = 0.6 × Risk score + 0.2 × Workload score + 0.2 × Timeliness score

[0117] Wherein, the risk score = adjusted total risk score × 0.5;

[0118] Workload score = 1 ÷ (1 + |current number of tasks for medical staff - average number of tasks|);

[0119] Timeliness score = 1 ÷ (number of overdue hours + 1).

[0120] 3. The complication early warning and processing module monitors and tracks data and dynamically calibrates early warning thresholds to achieve early prediction, tiered early warning, and closed-loop management of complications, thereby reducing the risk of complications. This module completes the construction of early warning indicators, abnormal data monitoring and analysis, early warning level calculation, and complication early warning.

[0121] The specific functions are as follows:

[0122] The construction of the early warning indicators involves inputting historical tracking and monitoring data and complication case data to build a multi-dimensional complication early warning indicator system. The optimal threshold for each early warning indicator is trained by selecting postoperative tracking and monitoring data from anesthesia in the past three years and using a logistic regression algorithm with the early warning indicator data as input to train the optimal threshold for each early warning indicator, which serves as the input parameter for abnormal data monitoring and analysis.

[0123] The abnormal data monitoring and analysis uses the optimal threshold of the early warning indicator and real-time tracking and monitoring data as input parameters. The system collects tracking and monitoring data, vital sign data, and test result data in real time, analyzes the data in real time, calculates the rate and magnitude of data change, and identifies abnormal data. The specific calculation steps are as follows:

[0124] (1) Calculation of rate of change: Continuous data collection is performed on early warning indicators (heart rate, blood pressure, blood oxygen saturation, etc.). A linear regression algorithm is used, inputting three consecutive data collections and their corresponding times, to calculate the rate of change of the three consecutive data collections. If the absolute value of the rate of change exceeds a set threshold (e.g., heart rate change rate > 10 beats / minute·2 hours), it is determined to be an abnormal trend, and an abnormal trend indicator is output. The specific formula is as follows:

[0125] Rate of change = (3rd data acquisition - 1st data acquisition) ÷ (3rd acquisition time - 1st acquisition time).

[0126] (2) Calculation of variation range: Calculate the absolute value of the difference between the current collected data and the previous collected data, and the ratio of the current data to the previous data (percentage of variation range). If the variation range is >20%, it is judged as abnormal data fluctuation, and an abnormal fluctuation indicator is output. The specific formula is:

[0127] Change range = (|current data - previous data| ÷ previous data) × 100%.

[0128] (3) Anomaly determination: Combine threshold, rate of change and magnitude of change for comprehensive determination. If a single data exceeds the calibrated threshold, or if the data shows an abnormal trend or abnormal fluctuation, it is determined to be abnormal data, and the abnormal data and anomaly identifier are output. If multiple warning indicators are abnormal at the same time, they are determined to be high-risk anomalies, and high-risk anomaly identifiers are output as input parameters for warning level calculation and complication prediction algorithm, triggering the warning process.

[0129] The warning level calculation takes abnormal data, abnormal trends, and the number of abnormal indicators as input parameters, uses a quantitative scoring method to calculate the warning level, outputs the warning level value, and simultaneously synchronizes the warning data to the mobile tracking and monitoring module and the quality control and data management module.

[0130] The specific formula for calculating the warning level is as follows:

[0131] Warning level = Highest score of a single abnormal indicator + Number of abnormal indicators × 0.5 + Trend score

[0132] (No abnormal trend = 0 points, abnormal trend = 1 point, high-risk abnormal trend = 2 points)

[0133] In this system, combined with dynamic scoring of the degree of abnormality, each early warning indicator is quantitatively scored (1-3 points) according to the degree of abnormality (slight abnormality, moderate abnormality, severe abnormality). For example, a slight abnormality in heart rate (80-100 beats / minute, exceeding the threshold but not reaching the dangerous range) = 1 point, a moderate abnormality (101-120 beats / minute) = 2 points, and a severe abnormality (>120 beats / minute or <60 beats / minute) = 3 points. The scores of each abnormal indicator are output.

[0134] The complication early warning system uses abnormal data, abnormal trends, and historical patient tracking and monitoring data as input parameters to construct a complication prediction model and achieve complication early warning. The specific steps are as follows:

[0135] Using postoperative anesthesia follow-up monitoring data from the past three years as training samples, and with patient basic information, intraoperative data, and postoperative follow-up monitoring data as features, a prediction model was trained. The prediction model was used to calculate the probability of patients developing specific complications. Combined with clinical intervention needs, if the probability is ≥45%, an early warning will be triggered, and the predicted probability and intervention suggestions will be output to achieve early intervention for complications.

[0136] 4. The quality control and data management module aggregates data from all modules for statistical analysis, outputs quality control reports and algorithm optimization parameters, and assists in the upgrade of each module. This module completes indicator statistics and report generation, data filtering and organization, and data compliance analysis.

[0137] The specific functions are as follows:

[0138] Indicator statistics and report generation: Input parameters include tracking and monitoring task completion data from the tracking and monitoring task module, tracking and monitoring record data from the form module, mobile terminal module, and remote tracking and monitoring module, early warning and treatment data from the complication early warning module, and patient satisfaction data from the remote tracking and monitoring module. The system automatically calculates relevant quality control indicators for post-anesthesia tracking and monitoring without manual statistics. At the same time, the statistical data can be used to evaluate the effectiveness of the algorithm model and output various standardized reports, which are synchronized to the system management module and department management terminal.

[0139] Data filtering and organization: Supports filtering of tracking and monitoring data, quality control data, and research data by time range, surgical type, anesthesia method, patient age, department, attending physician, and other methods. The high-quality data filtered can be used as input parameters for training and optimizing algorithm models, and the output is the filtered data / dataset. Simultaneously, the system automatically collects, cleans, and structures all tracking and monitoring data, extracts important data required for research (such as patient basic information, surgical and anesthesia information, postoperative tracking and monitoring data, complication data, etc.), and forms a standardized research dataset.

[0140] Data compliance analysis: Input parameters include tracking and monitoring data, quality control data, and industry standard data for different time periods, different doctors, and different surgical types. Using a comparative analysis algorithm, the system identifies management weaknesses by analyzing data differences, optimizes postoperative tracking and monitoring processes and management plans, and uses the comparative data as input for adjusting and optimizing algorithm parameters. The system outputs comparative analysis results and automatically records tracking and monitoring related operations and data to form a complete tracking and monitoring file, outputting compliance reports and related data.

[0141] 5. The standardized electronic form module provides standardized tracking and monitoring forms that conform to clinical standards, simplifies data entry operations, and pushes standardized tracking and monitoring data to the main module.

[0142] The specific functions are as follows:

[0143] The design incorporates a standardized form structure, with core post-anesthesia follow-up monitoring content including a vital signs assessment form, a pain rating form, a neurological abnormality assessment form, and a complication assessment form. The output is standardized follow-up monitoring data, providing a unified format of input data for complication early warning, risk scoring, and quality control statistics.

[0144] In addition, the system enables form association and data recording. The tracking and monitoring form automatically associates with the patient's basic information, brain surgery information, and anesthesia information, eliminating the need for repeated data entry. Subsequent tracking and monitoring forms can automatically associate with previous tracking and monitoring records, making it easier for medical staff to compare the patient's recovery status and identify abnormal changes. At the same time, the system supports uploading photos, videos, and audio recordings (such as photos of puncture sites and audio recordings of the patient's pain feedback) into the tracking and monitoring form, outputting tracking and monitoring data in multiple formats.

[0145] 6. The mobile tracking and monitoring module is adapted to clinical mobile scenarios, supports medical staff to enter tracking and monitoring data in real time at the bedside, receive and execute tracking and monitoring / early warning tasks, and provide feedback on the task execution results.

[0146] The specific functions are as follows:

[0147] An input system was built to enable offline tracking and monitoring, supporting installation and use on mobile apps, tablets, and other terminals. It allows for real-time input and output of tracking and monitoring data, providing real-time input for anomaly detection algorithms and risk scoring. Simultaneously, it supports filling in tracking and monitoring records and uploading photos / videos offline, automatically synchronizing to the system upon network connection. This resolves network instability issues in areas such as wards, ensuring uninterrupted tracking and monitoring and preventing data loss.

[0148] It enables real-time ward rounds and early warnings. When medical staff are monitoring patients at the bedside, the system outputs real-time and accurate monitoring data based on vital signs, pain scores, and complication status. This data serves as input parameters for anomaly detection and trend analysis algorithms, providing timely data support for real-time early warnings and trend analysis. Simultaneously, input parameters include the warning level, abnormal data, and treatment suggestions. The mobile terminal receives warning messages in real time, allowing medical staff to quickly view warning details, rapidly fill in treatment results, and output warning and treatment outcome data.

[0149] 7. The discharge remote tracking and monitoring module conducts post-discharge tracking and monitoring through multiple channels, collects remote tracking and monitoring data and patient abnormal feedback, and provides continuous data support for the main module's early warning algorithm.

[0150] The specific functions are as follows:

[0151] This system enables multi-mode remote tracking and monitoring, supporting three methods: push tracking and monitoring questionnaires via various social media tools, voice tracking and monitoring, and video visits. Specifically, the tracking and monitoring questionnaire is automatically pushed; input parameters include the post-discharge tracking and monitoring plan and the tracking and monitoring time nodes. Upon patient discharge, the system automatically binds the patient's account and automatically pushes the tracking and monitoring questionnaire at the corresponding time nodes according to the generated post-discharge tracking and monitoring plan. After the patient completes the questionnaire, the data is automatically synchronized to the system, outputting the patient's completed tracking and monitoring data. Voice tracking and monitoring uses tracking and monitoring questions as input parameters. The system automatically recognizes the patient's voice responses, converts them into text records, automatically records them, and reminds medical staff to manually supplement the tracking and monitoring, outputting complete tracking and monitoring data. Video visits allow medical staff to connect with patients via mobile devices to view the patient's recovery status in real time. Video recordings are automatically archived in the patient's tracking and monitoring file, outputting video tracking and monitoring data and evaluation records.

[0152] When monitoring the safety of the rehabilitation process, the input parameters are the patient's surgery type, anesthesia method, and recovery trend. Personalized rehabilitation guidance (such as dietary suggestions and medication use) is automatically pushed to the patient. At the same time, the effectiveness of the rehabilitation guidance can be fed back through subsequent tracking and monitoring data, and the relevant data on rehabilitation effect are output for the optimization of data and calculation processes.

[0153] 8. The system management module is responsible for system permission allocation, parameter configuration and log auditing, ensuring the safe and standardized operation of each module, and does not participate in core algorithm and data interaction.

[0154] The specific functions are as follows:

[0155] It supports hierarchical access control: adopts a hierarchical control mode, supports custom roles (such as anesthesiologist, nurse, head nurse, department director, quality control officer, system administrator), and finely allocates operation permissions (such as data viewing permission, input permission, editing permission, deletion permission, report export permission, etc.). The input parameters are the user's position and department requirements, and the output is the permission allocation result.

[0156] User management is supported: users can be added, deleted, edited, and disabled; user login passwords and permissions can be set; user login logs can be recorded for easy system management and security auditing; at the same time, algorithm-related operation permissions (such as parameter viewing, early warning threshold adjustment, etc.) can be assigned according to user positions; the input parameters are user information and position requirements, and the output is the user configuration result.

[0157] The embodiments described above are merely illustrative of several implementations of this disclosure, and their descriptions are specific and detailed, but should not be construed as limiting the scope of protection of this invention in any way. The description of preferred embodiments involves combinations of features, which may exist independently or in combination; this invention is not particularly limited to the preferred embodiments. The scope of this invention is defined by the claims.

[0158] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A post-anesthesia follow-up monitoring system for monitoring and early warning of post-brain surgery status, specifically comprising: Specifically, it includes eight modules: patient brain surgery information synchronization module, tracking and monitoring task generation module, complication early warning and handling module, quality control and data management module, standardized electronic form module, mobile tracking and monitoring module, discharge remote tracking and monitoring module, and system management module; The above eight modules are divided into main modules and auxiliary modules. Each module in the main modules interacts bidirectionally with three or more modules and completes the core functions of data processing, optimization, and parameter calculation. These include the patient brain surgery information synchronization module, the tracking and monitoring task generation module, the complication early warning and handling module, and the quality control and data management module. Among them, the patient brain surgery information synchronization module serves as the basic data source for all modules, enabling bidirectional data exchange with the other three main modules, pushing core basic data, and simultaneously receiving optimization parameters and configuration parameters; The tracking and monitoring task generation module interacts bidirectionally with the other three main modules, receives basic data from the patient brain surgery information synchronization module, and generates personalized tracking and monitoring tasks. The complication early warning and processing module interacts bidirectionally with two main modules and four auxiliary modules, pushes early warning data, and receives algorithm optimization parameters to dynamically calibrate early warning thresholds and optimize prediction models. The quality control and data management module exchanges data bidirectionally with the other three main modules, receives various types of data, and performs aggregation, statistics, and analysis. In addition, the auxiliary modules complete auxiliary functions such as data collection, task execution, and access control, including a standardized electronic form module, a mobile tracking and monitoring module, a discharge remote tracking and monitoring module, and a system management module.

2. The tracking and monitoring system according to claim 1, characterized in that, The standardized electronic form module interacts bidirectionally with only two main modules, receiving tracking and monitoring task instructions from the tracking and monitoring task generation module and providing standardized forms for data entry. The mobile tracking and monitoring module interacts bidirectionally with only two main modules, pushing tracking and monitoring data, task completion status, early warning and handling results, and providing feedback on the execution effect; The discharge remote tracking and monitoring module interacts bidirectionally with only one main module to enable timely early warning and handling of abnormalities after discharge. The system management module interacts bidirectionally with all four main modules, providing access control, system parameter configuration, and initial algorithm parameter settings for each main module.

3. The tracking and monitoring system according to claim 1, characterized in that, The tracking and monitoring task generation module, as the main module for executing the core algorithm, generates personalized tracking and monitoring plans based on basic data, automatically sorts tasks, and optimizes tracking and monitoring strategies in conjunction with the early warning module. This module completes multi-dimensional data acquisition and preprocessing, postoperative follow-up monitoring risk quantification calculation, intelligent generation of personalized follow-up monitoring plans, automatic sorting of follow-up monitoring tasks, and task management. The multi-dimensional data acquisition and preprocessing automatically collects multi-dimensional patient data, performs standardized preprocessing on the collected data, and outputs standardized data. The postoperative follow-up monitoring risk quantification calculation uses standardized data from the multi-dimensional data collection and preprocessing as input parameters. Based on this data, a multi-dimensional risk assessment model is constructed to calculate the patient's postoperative follow-up monitoring risk score. The personalized tracking and monitoring plan is generated intelligently. The input parameters are the corrected total risk score and risk level output by the postoperative tracking and monitoring risk quantification calculation. Combined with the patient's surgical type and anesthesia method, a personalized tracking and monitoring plan is generated and dynamically adjusted according to the patient's real-time recovery status.

4. The tracking and monitoring system according to claim 3, characterized in that, The multi-dimensional data acquisition and preprocessing specifically includes: (1) First, set the scores for the fixed data in the multidimensional data, and then set the weights according to the importance to perform weighted calculations to obtain fixed quantitative values; (2) Standardize the changing data, first filter out outliers, correct outliers to the maximum or minimum value within the range, and then use the standardization formula to output standardized continuous data, i.e. standard values, to ensure the accuracy of subsequent risk scoring algorithm calculations.

5. The tracking and monitoring system according to claim 3, characterized in that, The specific calculation of postoperative follow-up monitoring risk quantification is as follows: (1) Determine the weight of each evaluation indicator: combine expert indicators and analytical indicators to reduce subjective bias; first, have multiple experts score the weight of each evaluation indicator, then calculate the weight of the analytical indicator, and finally obtain the combined weight as the main parameter for scoring calculation. (2) Quantitative scoring of indicators: Each evaluation indicator is graded and quantitatively scored, and the quantitative scores of each indicator are output; (3) Risk score calculation: The total risk score for postoperative follow-up monitoring of patients is calculated using a weighted summation formula, and the total risk score is output; (4) Risk decay correction: Add a time decay coefficient. The longer the postoperative time, the lower the risk. Correct the total risk score and output the corrected total risk score as the main input parameter for subsequent module data processing and calculation. (5) Risk level classification: The risk level range is set based on actual clinical data. Different risk levels correspond to different tracking and monitoring strategies. The risk level is output as the input parameter for the intelligent generation of personalized tracking and monitoring plans.

6. The tracking and monitoring system according to claim 3, characterized in that, The intelligent generation of the personalized tracking and monitoring plan specifically involves: (1) Generate and output the basic tracking and monitoring plan; (2) Dynamic adjustment of tracking and monitoring frequency: Calculate the rate of change of indicators and calculate the new tracking and monitoring frequency as input parameters for automatic sorting of tracking and monitoring tasks.

7. The tracking and monitoring system according to claim 1, characterized in that, The complication early warning and processing module monitors and tracks data and dynamically calibrates the early warning threshold to achieve early prediction, graded early warning and closed-loop treatment of complications, thereby reducing the risk of complications. This module completes the construction of early warning indicators, abnormal data monitoring and analysis, early warning level calculation and complication early warning. The early warning indicators are constructed by taking historical tracking and monitoring data and complication case data as input parameters, constructing a multi-dimensional complication early warning indicator system, and training the optimal threshold for each early warning indicator. The abnormal data monitoring and analysis system takes as input parameters the optimal threshold of the early warning indicator and real-time tracking and monitoring data. The system collects tracking and monitoring data, vital sign data, and test result data in real time, analyzes the data in real time, calculates the rate and magnitude of data change, and identifies abnormal data. The warning level calculation takes abnormal data, abnormal trends, and the number of abnormal indicators as input parameters, uses a quantitative scoring method to calculate the warning level, outputs the warning level value, and simultaneously synchronizes the warning data to the mobile tracking and monitoring module and the quality control and data management module. The complication early warning system uses abnormal data, abnormal trends, and historical patient tracking and monitoring data as input parameters to construct a complication prediction model and achieve complication early warning.

8. The tracking and monitoring system according to claim 7, characterized in that, The abnormal data monitoring and analysis specifically includes: (1) Calculation of rate of change: The early warning indicators are continuously collected, and the linear regression algorithm is used. The data collected three times in a row and the corresponding time are input, the rate of change of the three times in a row is calculated, and the abnormal trend indicator is output. (2) Calculation of change range: Calculate the absolute value of the difference between the current data and the previous data, the ratio of the current data to the previous data, determine abnormal data fluctuations, and output abnormal fluctuation indicators; (3) Anomaly detection: Combine threshold, rate of change and magnitude of change for comprehensive judgment. If a single data exceeds the calibrated threshold, or if the data shows an abnormal trend or abnormal fluctuation, it is judged as abnormal data and the abnormal data and anomaly identifier are output. If multiple warning indicators show abnormalities simultaneously, they are judged as high-risk abnormalities, and a high-risk abnormality identifier is output as an input parameter for the warning level calculation and complication prediction algorithm, triggering the warning process.

9. The tracking and monitoring system according to claim 8, characterized in that, The patient brain surgery information synchronization module, as the main data acquisition module, automatically captures and synchronizes basic data related to the patient, surgery, and anesthesia; this module completes data collection, real-time data synchronization, data supplementation, and data association archiving. The data collection is achieved by automatically collecting information through standard interfaces connecting to hospital information systems, electronic medical record systems, and surgical anesthesia systems. The real-time data synchronization refers to synchronizing intraoperative data and patient condition changes to this system in real time. The data supplementation refers to editing and supplementing the synchronized data; The data association and archiving automatically associates and archives patient information, brain surgery information, anesthesia information, and follow-up monitoring records to form a personal post-anesthesia follow-up monitoring file for the patient.

10. The tracking and monitoring system according to claim 8, characterized in that, The quality control and data management module aggregates data from all modules for statistical analysis, outputs quality control reports and algorithm optimization parameters, and assists in the upgrade of each module; this module completes indicator statistics and report generation, data filtering and organization, and data compliance analysis. The standardized electronic form module provides standardized tracking and monitoring forms that conform to clinical guidelines, simplifies data entry operations, and pushes standardized tracking and monitoring data to the main module; The mobile tracking and monitoring module is adapted to clinical mobile scenarios, supports medical staff to enter tracking and monitoring data in real time at the bedside, receive and execute tracking and monitoring / early warning tasks, and provide feedback on the task execution effect; The discharge remote tracking and monitoring module conducts post-discharge tracking and monitoring through multiple channels, collects remote tracking and monitoring data and abnormal patient feedback, and provides continuous data support for the main module's early warning algorithm. The system management module is responsible for system permission allocation, parameter configuration, and log auditing, ensuring the safe and standardized operation of each module, and does not participate in core algorithm and data interaction.

Images