Layered control system and method for low body temperature risk in perioperative period and storage medium
By using a perioperative hypothermia risk stratification control system, AI is employed to assess and dynamically adjust ambient temperature and equipment power. This addresses the issue of insufficient human experience in the prevention and treatment of perioperative hypothermia, achieving fully automated and intelligent temperature management and improving prevention and treatment efficiency and accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-24
- Publication Date
- 2026-04-03
AI Technical Summary
Current technologies for the prevention and treatment of perioperative hypothermia mainly rely on human experience and decentralized management measures, which are highly subjective, have delayed responses, involve scattered interventions, are inefficient, and make it difficult to achieve precise and standardized prevention and treatment throughout the entire process.
A perioperative hypothermia risk stratification control system is provided, including a data acquisition and preprocessing module, an AI risk assessment module, a decision support and intervention module, and an intelligent control module. It assesses the risk level through automated data acquisition and machine learning, generates personalized intervention strategies based on the level, and dynamically adjusts the ambient temperature and equipment power to form a closed-loop management.
It achieves fully automated and intelligent low body temperature management, reduces the subjectivity of human intervention, improves response speed and intervention accuracy, and ensures the stability of body temperature and patient safety during the perioperative period.
Smart Images
Figure CN121789984A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of distributed real-time systems and deterministic network technology, and in particular to a perioperative hypothermia risk stratification control system, method, and computer-readable storage medium. Background Technology
[0002] Perioperative hypothermia (core body temperature below 36°C) is a common complication in patients undergoing general anesthesia, with an incidence rate as high as 7%-90%. Hypothermia can lead to an increased risk of surgical site infection, coagulation dysfunction, delayed drug metabolism, increased cardiovascular events, and postoperative shivering and discomfort, seriously affecting the quality of patient recovery and medical safety.
[0003] Currently, the prevention and treatment of perioperative hypothermia in clinical practice mainly relies on manual experience and decentralized management measures, all of which have certain limitations. These limitations can be summarized as: high subjectivity, delayed response, decentralized intervention, low efficiency, and difficulty in achieving precise and standardized prevention and treatment throughout the entire process.
[0004] Specifically, healthcare professionals use paper-based or simple electronic spreadsheets (such as the Predictors scale) to assess patient risk. This method is highly subjective, lacks consistent standards, and is prone to omissions or misjudgments due to busy schedules. Decentralized intervention: Warming measures (such as adjusting air conditioning or turning on warm blankets) require manual operation by healthcare professionals. Environmental temperature control, warming equipment management, and body temperature monitoring are disconnected, making it difficult to form a coordinated and precise warming plan. Reactive rather than preventative: Interventions are usually initiated only after a patient's body temperature has dropped, representing "post-event remediation" and lacking proactive prediction and preventative control of hypothermia. Process management relies heavily on human intervention: The entire process, from risk assessment to measure implementation, heavily depends on the conscientiousness and experience of healthcare professionals, lacking systematic and mandatory closed-loop management, resulting in low guideline compliance and unstable management effectiveness.
[0005] Therefore, existing technologies still need to be improved and developed. Summary of the Invention
[0006] The main objective of this invention is to provide a perioperative hypothermia risk stratification control system, method, and computer-readable storage medium, aiming to solve the problems in the existing technology of perioperative hypothermia prevention and control, which mainly rely on human experience and decentralized management measures, resulting in high subjectivity, delayed response, decentralized intervention, low efficiency, and difficulty in achieving precise and standardized prevention and control throughout the entire process.
[0007] To achieve the above objectives, the present invention provides a perioperative hypothermia risk stratification control system, which includes: a data acquisition and preprocessing module, an AI risk assessment module, a decision support and intervention module, and an intelligent control module, wherein the data acquisition and preprocessing module, the AI risk assessment module, the decision support and intervention module, and the intelligent control module are connected sequentially. The data acquisition and preprocessing module is used to automatically acquire the patient's preoperative indicator data from the hospital information system in the preoperative stage, and to continuously receive the patient's real-time vital signs data in the intraoperative stage. The AI risk assessment module is used to receive the preoperative indicator data, perform an initial calculation based on the preoperative indicator data, output an initial low body temperature risk level, and receive the real-time vital signs data. It also periodically combines the real-time vital signs data with the updated surgical data to perform a recalculation and output an updated risk level. The decision support and intervention module is used to receive the initial low body temperature risk level and the updated risk level, and generate and execute the preoperative pre-intervention strategy and the updated pre-intervention strategy according to the initial low body temperature risk level and the updated risk level, respectively. The intelligent control module is used to perform closed-loop regulation based on real-time body temperature data in the real-time vital signs data during the intraoperative phase.
[0008] Optionally, in the perioperative hypothermia risk stratification control system, the data acquisition and preprocessing module includes: a preoperative acquisition unit, a data preprocessing unit, and an intraoperative acquisition unit. The preoperative data collection unit is used to automatically extract the patient's preoperative indicator data from the hospital's anesthesia system and electronic medical records through a standardized interface after the patient has completed identity verification and been connected to vital sign monitoring in the operating room. The preoperative preprocessing unit is used to receive the preoperative indicator data, preprocess the preoperative indicator data to obtain preprocessed data, and integrate and transform the preprocessed data to obtain numerical features that can be processed by the AI model. The intraoperative data acquisition unit is used to continuously receive real-time vital sign data of the patient through a vital sign monitor after the patient enters anesthesia, and to continuously receive updated surgical data of the patient from the operating room.
[0009] Optionally, in the perioperative hypothermia risk stratification control system, the AI risk assessment module includes: a preoperative risk assessment unit and an intraoperative risk assessment unit. The preoperative risk assessment unit is used to receive the numerical features, input the numerical features into the trained random forest model for calculation, and output the patient's initial hypothermia risk level. The intraoperative risk assessment unit is used to receive the real-time vital signs data and the updated surgical data, input the real-time vital signs data and the updated surgical data into the random forest model for recalculation, and output the patient's updated risk level.
[0010] Optionally, in the perioperative hypothermia risk stratification control system, the decision support and intervention module includes: a preoperative decision intervention unit and an intraoperative decision intervention unit; The preoperative decision-making intervention unit is used to receive the initial low body temperature risk level. When the initial low body temperature risk level is high, it sends a first instruction to the inflatable warmer in the operating room via the MQTT protocol and a second instruction to the air conditioning system. The first instruction is to control the inflatable warmer to start running before anesthesia induction and to perform preoperative warming for a preset duration. The second instruction is to set and lock the operating room temperature at a first preset air conditioning temperature. The preoperative decision-making intervention unit is also used to notify the nurse to prepare and turn on the warming device when the initial low body temperature risk level is medium risk, and to notify the nurse to use thicker bedding and remind the nurse to pay attention to changes in the patient's body temperature when the initial low body temperature risk level is low risk. The intraoperative decision-making intervention unit is used to receive the updated risk level and generate an updated pre-intervention strategy based on the updated risk level.
[0011] Optionally, in the perioperative hypothermia risk stratification control system, the intelligent control module includes: a first strategy control unit and a second strategy control unit.
[0012] The first strategy control unit is used to increase the power of the heating blanket to a preset power and raise the temperature of the air conditioning system to a second preset air conditioning temperature when the real-time body temperature data in the real-time vital signs data is lower than a first preset temperature. The first strategy control unit is also used to trigger a high-level alarm and send a suggestion to the nurse to start fluid warming when the real-time body temperature data is lower than the second preset temperature; The second strategy control unit is used to perform PID calculations based on the difference between the real-time body temperature data and the preset target body temperature data, and to dynamically adjust the output power of the heating device based on the calculation results.
[0013] Optionally, the perioperative hypothermia risk stratification control system further includes a postoperative feedback optimization module. The postoperative feedback optimization module is used to automatically generate a perioperative temperature management report after the patient's surgery and archive the perioperative temperature management report into the patient's medical record. The perioperative temperature management report includes the initial risk level, the intervention measures taken, and the temperature change curve. The postoperative feedback optimization module is also used to collect perioperative temperature management reports from all anonymous patients and to periodically train and optimize the AI model of the AI risk assessment module using these reports.
[0014] Furthermore, to achieve the above objectives, the present invention also provides a perioperative hypothermia risk stratification control method based on a perioperative hypothermia risk stratification control system, the method comprising: The data acquisition and preprocessing module automatically acquires the patient's preoperative indicator data from the hospital information system during the preoperative stage and continuously receives the patient's real-time vital signs data during the intraoperative stage. The AI risk assessment module receives the preoperative indicator data, performs an initial calculation based on the preoperative indicator data, outputs the initial hypothermia risk level, and receives the real-time vital signs data. The AI risk assessment module periodically combines the real-time vital signs data with the updated surgical data to recalculate and output the updated risk level. The decision support and intervention module receives the initial low body temperature risk level and the updated risk level, and generates and executes the preoperative pre-intervention strategy and the updated pre-intervention strategy based on the initial low body temperature risk level and the updated risk level, respectively. The intelligent control module performs closed-loop regulation based on the real-time body temperature data in the real-time vital signs data during the intraoperative phase.
[0015] Optionally, the perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system includes preoperative indicator data including static data and dynamic data. The static data includes age, gender, weight, height, anesthesiologist association classification, and comorbidities; The dynamic data includes preoperative core body temperature, type of surgery to be performed, estimated surgery duration, and planned anesthesia method.
[0016] Optionally, in the perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system, the real-time vital signs data include heart rate data, blood pressure data, respiratory rate data, real-time body temperature data, and blood oxygen saturation data. The updated surgical data includes vital sign monitoring data, physiological metabolic status data, equipment feedback data, and surgical duration data.
[0017] In addition, to achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a perioperative hypothermia risk stratification control program, and the perioperative hypothermia risk stratification control program, when executed by a processor, implements the steps of the perioperative hypothermia risk stratification control method as described above.
[0018] In this invention, the system includes: a data acquisition and preprocessing module, an AI risk assessment module, a decision support and intervention module, and an intelligent control module. The data acquisition and preprocessing module automatically acquires the patient's preoperative indicator data from the hospital information system during the preoperative stage and continuously receives the patient's real-time vital sign data during the intraoperative stage. The AI risk assessment module receives the preoperative indicator data, performs an initial calculation based on the preoperative indicator data, outputs an initial low body temperature risk level, and receives real-time vital sign data, periodically combining the real-time vital sign data with updated surgical data to perform a recalculation and output an updated risk level. The decision support and intervention module receives the initial low body temperature risk level and the updated risk level, and generates and executes preoperative and updated preoperative intervention strategies based on the initial and updated risk levels, respectively. The intelligent control module performs closed-loop control based on real-time body temperature data from the real-time vital sign data during the intraoperative stage. This invention utilizes machine learning algorithms to intelligently analyze multiple preoperative indicators of the patient and automatically outputs low, medium, and high risk levels. Based on risk level, the system automatically generates personalized intervention suggestions and dynamically adjusts the ambient temperature and equipment power according to the patient's real-time body temperature, forming a fully automated and intelligent closed-loop management system. Attached Figure Description
[0019] Figure 1 This is the overall architecture diagram of the perioperative hypothermia risk stratification control system of the present invention; Figure 2 This is a flowchart of a preferred embodiment of the perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system of the present invention. Detailed Implementation
[0020] This application provides a method and related equipment for risk stratification and control of perioperative hypothermia. To make the purpose, technical solution, and effects of this application clearer and more explicit, the following detailed description is provided with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.
[0021] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.
[0022] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.
[0023] The perioperative hypothermia risk stratification control system described in the preferred embodiment of the present invention, such as... Figure 1 As shown, the perioperative hypothermia risk stratification control system includes: a data acquisition and preprocessing module, an AI risk assessment module, a decision support and intervention module, and an intelligent control module, which are connected sequentially. The data acquisition and preprocessing module is used to automatically acquire the patient's preoperative indicator data from the hospital information system in the preoperative stage and to continuously receive the patient's real-time vital signs data during the intraoperative stage. The AI risk assessment module is used to receive the preoperative indicator data, perform an initial calculation based on the preoperative indicator data, output an initial low body temperature risk level, and receive the real-time vital signs data. It also periodically combines the real-time vital signs data with the updated surgical data to perform a recalculation and output an updated risk level. The decision support and intervention module is used to receive the initial low body temperature risk level and the updated risk level, and generate and execute the preoperative pre-intervention strategy and the updated pre-intervention strategy according to the initial low body temperature risk level and the updated risk level, respectively. The intelligent control module is used to perform closed-loop regulation based on real-time body temperature data in the real-time vital signs data during the intraoperative phase.
[0024] It is understood that the present invention relates to an artificial intelligence-based perioperative hypothermia risk stratification control system. This system integrates with the hospital's hand anesthesia system to achieve intelligent preoperative risk assessment, real-time intraoperative monitoring and automatic control, forming a closed-loop management.
[0025] This system mainly comprises four core modules: data acquisition and preprocessing, AI risk assessment, decision support and intervention recommendation, and intelligent control. These modules work collaboratively to achieve full automation from data input to equipment control. Specifically, the system digitizes a hypothermia risk assessment scale and embeds it into the hospital's surgical anesthesia system. Utilizing machine learning algorithms, it intelligently analyzes multiple preoperative indicators of patients, automatically outputting low, medium, and high risk levels. Based on the risk level, the system automatically generates personalized intervention recommendations and intelligently controls the operating room's air conditioning system and heating equipment (such as inflatable ventilators) through IoT technology. It dynamically adjusts the ambient temperature and equipment power according to the patient's real-time body temperature, forming a fully automated and intelligent closed-loop management system encompassing "assessment-decision-intervention-feedback."
[0026] Furthermore, the data acquisition and preprocessing module includes: a preoperative acquisition unit, a data preprocessing unit, and an intraoperative acquisition unit; The preoperative data collection unit is used to automatically extract the patient's preoperative indicator data from the hospital's anesthesia system and electronic medical records through a standardized interface after the patient has completed identity verification and been connected to vital sign monitoring in the operating room. The preoperative preprocessing unit is used to receive the preoperative indicator data, preprocess the preoperative indicator data to obtain preprocessed data, and integrate and transform the preprocessed data to obtain numerical features that can be processed by the AI model. The intraoperative data acquisition unit is used to continuously receive real-time vital sign data of the patient through a vital sign monitor after the patient enters anesthesia, and to continuously receive updated surgical data of the patient from the operating room.
[0027] Understandably, once the patient has completed identity verification and is connected to vital sign monitoring in the operating room, the preoperative data acquisition unit automatically extracts patient information from the hospital's anesthesia system (AIMS) and electronic medical record (EMR) via standardized interfaces (such as HL7 and FHIR). The extracted data fields include: age, gender, weight, height (for calculating BMI), Association of Anesthesiologists (ASA) classification, and comorbidities (such as diabetes mellitus and hypothyroidism).
[0028] The preoperative preprocessing unit integrates the above data and converts it into numerical features that can be processed by the AI model. Preprocessing includes handling missing values and data standardization. The intraoperative acquisition unit is used to continuously monitor the patient's real-time vital signs data (measured via an esophageal or nasopharyngeal probe) after the patient enters anesthesia, and transmits the data to this system in real time via a device interface (such as a serial port or network).
[0029] Specifically, the preoperative data collection unit's innovation lies in automation and standardization. Instead of relying on manual data entry, it automatically retrieves data from the hospital's anesthesia system and electronic medical record system via a standardized interface after the patient enters the operating room, completes identity verification, and connects to monitoring equipment, ensuring the timeliness and accuracy of the data.
[0030] Data preprocessing unit: This is a crucial step in data usability. After receiving the raw data, it performs preprocessing operations such as cleaning, normalization, and handling missing values. It then integrates and transforms the processed data to form structured numerical features, providing high-quality input that can be directly processed by subsequent AI models.
[0031] Intraoperative data acquisition unit: Responsible for continuous data acquisition during the intraoperative phase. It not only acquires real-time data from vital sign monitors, but also receives updated surgical data (such as the duration of the procedure) from the operating room environment (such as the anesthesia machine and surgical information system), providing real-time evidence for dynamic risk assessment.
[0032] Furthermore, the AI risk assessment module includes: a preoperative risk assessment unit and an intraoperative risk assessment unit; The preoperative risk assessment unit is used to receive the numerical features, input the numerical features into the trained random forest model for calculation, and output the patient's initial hypothermia risk level. The intraoperative risk assessment unit is used to receive the real-time vital signs data and the updated surgical data, input the real-time vital signs data and the updated surgical data into the random forest model for recalculation, and output the patient's updated risk level.
[0033] Understandably, before the surgery begins, the preoperative risk assessment unit uses a trained random forest model to analyze the preprocessed static and planned characteristics (numerical features) of the patient, outputting an initial hypothermia risk level (e.g., high, medium, low) to provide a basis for preoperative intervention. During the surgery, the intraoperative risk assessment unit periodically (e.g., every 5 minutes) inputs the latest real-time vital signs and surgical progress data into the same or continuously optimized random forest model for recalculation. This allows the risk assessment to dynamically adjust as the surgery progresses, outputting an updated risk level to address unexpected situations or deviations from the original plan during the surgery.
[0034] It should be noted that the AI risk assessment model in this invention is not limited to random forest or XGBoost, but may also employ deep learning models (such as LSTM for analyzing time series data), ensemble learning algorithms, or more advanced machine learning algorithms that emerge in the future. The model can be designed as an online learning model, capable of continuous iterative optimization based on the data continuously generated by the hospital.
[0035] Furthermore, the decision support and intervention module includes: a preoperative decision intervention unit and an intraoperative decision intervention unit; The preoperative decision-making intervention unit is used to receive the initial low body temperature risk level. When the initial low body temperature risk level is high, it sends a first instruction to the inflatable warmer in the operating room via the MQTT protocol and a second instruction to the air conditioning system. The first instruction is to control the inflatable warmer to start running before anesthesia induction and to perform preoperative warming for a preset duration. The second instruction is to set and lock the operating room temperature at a first preset air conditioning temperature. The preoperative decision-making intervention unit is also used to notify the nurse to prepare and turn on the warming device when the initial low body temperature risk level is medium risk, and to notify the nurse to use thicker bedding and remind the nurse to pay attention to changes in the patient's body temperature when the initial low body temperature risk level is low risk. The intraoperative decision-making intervention unit is used to receive the updated risk level and generate an updated pre-intervention strategy based on the updated risk level.
[0036] In this embodiment, the preoperative decision intervention unit in the decision support and intervention module automatically executes pre-intervention measures based on the initial low body temperature risk level: High-risk patients: The system sends a command to the inflatable warmer in the operating room via the MQTT protocol (a lightweight Internet of Things communication protocol), instructing it to start operating before anesthesia induction for at least 10 minutes of "preoperative pre-warming"; simultaneously, it sends a command to the air conditioning system to set and lock the operating room temperature at 23°C. Medium-risk patients: The system recommends and prepares the warmer, which is then turned on after confirmation by the nurse. Low-risk patients: The system recommends using thicker bedding and reminds the nurse to monitor body temperature changes.
[0037] Understandably, the preoperative decision-making intervention unit implements differentiated pre-intervention strategies based on the initial risk level. In high-risk situations, proactive and mandatory pre-warming measures are adopted. Instructions are sent to the heating equipment via the MQTT protocol, causing it to start operating and perform timed pre-warming before anesthesia induction; simultaneously, the operating room air conditioning temperature is automatically set and locked to create the optimal environment for the patient upon entry. In medium-risk situations, preparatory measures are taken, notifying nurses to prepare equipment and optimizing resource allocation. In low-risk situations, basic monitoring and reminder measures are implemented, focusing on nurses' personal attention and basic warming.
[0038] The intraoperative decision-making intervention unit within the decision support and intervention module is responsible for generating and executing updated intervention strategies based on the dynamically updated risk level during surgery, ensuring that the measures match the current risk. The AI risk assessment module is not idle during surgery. The system periodically (e.g., every 30 minutes) dynamically updates the risk level by incorporating new data such as real-time body temperature trends, actual surgical duration, and the amount of fluid already administered. For example, if a patient preoperatively assessed as medium-risk experiences significant intraoperative blood loss, the intraoperative decision-making intervention unit may upgrade their risk level to high-risk and accordingly escalate the intervention measures.
[0039] Furthermore, the intelligent control module includes: a first strategy control unit and a second strategy control unit.
[0040] The first strategy control unit is used to increase the power of the heating blanket to a preset power and raise the temperature of the air conditioning system to a second preset air conditioning temperature when the real-time body temperature data in the real-time vital signs data is lower than a first preset temperature. The first strategy control unit is also used to trigger a high-level alarm and send a suggestion to the nurse to start fluid warming when the real-time body temperature data is lower than the second preset temperature; The second strategy control unit is used to perform PID calculations based on the difference between the real-time body temperature data and the preset target body temperature data, and to dynamically adjust the output power of the heating device based on the calculation results.
[0041] Understandably, in closed-loop feedback control, the intelligent control module takes the lead. The intelligent control module compares the real-time body temperature with the set target body temperature (e.g., 36.5°C) and executes the following strategies based on the comparison result.
[0042] First strategy (rule-based regulation): If the patient's body temperature is below 36°C, the system will simultaneously increase the power of the heating blanket (e.g., from "medium" to "high") and appropriately raise the air conditioning temperature. If the body temperature is below 35.5°C, the system will trigger a high-level alarm and recommend starting liquid warming.
[0043] The second strategy (model-based advanced control): A better solution is to use a PID control algorithm. The system takes the difference between the real-time body temperature and the target body temperature as input, and through PID calculation, dynamically and smoothly adjusts the output power of the heating device to avoid temperature overshoot or fluctuations, thereby achieving accurate and stable body temperature maintenance.
[0044] As can be seen, the intelligent control module implements two specific strategies for closed-loop regulation: The first strategy control unit (threshold-based stepped regulation): This is a rule-based rapid response mechanism. When the real-time body temperature is lower than the first preset temperature (warning threshold), the system automatically increases the heating intensity (increases the power of the heating blanket) and raises the ambient temperature. When the body temperature is lower than the second preset temperature (alarm threshold), the system not only triggers a high-level audible and visual alarm but also proactively pushes suggestions to the nurse to start intravenous fluid warming, achieving human-machine collaborative intervention. The second strategy control unit (PID algorithm-based precise closed-loop control): This is a more advanced automated control method that simulates fine-tuning by humans. The system uses the difference between the real-time body temperature and the preset target body temperature as input, and performs calculations through the PID (proportional-integral-derivative) control algorithm to dynamically and smoothly adjust the output power of the heating equipment, ensuring that the patient's body temperature is stably maintained near the target value, avoiding overshoot or regulation lag.
[0045] It should be noted that, in addition to the operating room air conditioner and the inflatable warming blanket, the present invention can be further extended to connect to and control the following devices: infusion warming device: automatically sets the liquid warming temperature according to the patient's body temperature and risk level; respiratory gas warming and humidifying device: automatically adjusts the gas temperature in the anesthesia breathing circuit; medical incubator: used for preheating the body surface covering before surgery.
[0046] Furthermore, the perioperative hypothermia risk stratification control system also includes: a postoperative feedback optimization module; The postoperative feedback optimization module is used to automatically generate a perioperative temperature management report after the patient's surgery and archive the perioperative temperature management report into the patient's medical record. The perioperative temperature management report includes the initial risk level, the intervention measures taken, and the temperature change curve. The postoperative feedback optimization module is also used to collect perioperative temperature management reports from all anonymous patients and to periodically train and optimize the AI model of the AI risk assessment module using these reports.
[0047] In this embodiment, after the surgery, the postoperative feedback optimization module automatically generates a perioperative temperature management report, including the initial risk level, intervention measures taken, temperature change curve, etc., and archives it in the medical record. Simultaneously, the postoperative feedback optimization module also collects all anonymized patient data, intervention measures, and results (whether hypothermia occurred) for periodic retraining and optimization of the AI model, enabling the model to become increasingly accurate with data accumulation and develop self-evolution capabilities.
[0048] The above results enhance the system's data backtracking and self-evolution capabilities through the introduction of a postoperative feedback optimization module: Report generation and archiving: After surgery, the system automatically summarizes key data from the entire perioperative period (initial risks, measures taken, temperature curves, etc.), generates a structured report, and archives it into the electronic medical record, completing the information loop and providing data support for quality improvement. Continuous model optimization: The system anonymizes and collects all patients' temperature management reports to form a training dataset for regular retraining and optimization of the core AI risk assessment model. This allows the system to continuously learn and improve using real clinical data, and its risk prediction capabilities can continuously improve with increased usage time, demonstrating true intelligence and adaptive characteristics.
[0049] Furthermore, such as Figure 2 As shown, based on the above-mentioned perioperative hypothermia risk stratification control system, the present invention also provides a perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system, the perioperative hypothermia risk stratification control method comprising: Step S10: The data acquisition and preprocessing module automatically acquires the patient's preoperative indicator data from the hospital information system during the preoperative stage, and continuously receives the patient's real-time vital signs data during the intraoperative stage.
[0050] Specifically, after a patient enters the operating room, the system automatically retrieves the patient's static and dynamic data from hospital information systems (such as HIS, EMR, and anesthesia clinical information systems) through standardized interfaces (such as HL7 and FHIR). This process eliminates the need for manual data entry, ensuring the accuracy, completeness, and timeliness of the data, and providing input for the next step of immediate risk assessment.
[0051] The preoperative data includes both static and dynamic data. Static data includes age, gender, weight, height, anesthesiology association classification, and comorbidities. Dynamic data includes preoperative core body temperature, type of surgery to be performed, estimated surgery duration, and planned anesthesia method. Real-time vital signs data includes heart rate, blood pressure, respiratory rate, real-time body temperature, and blood oxygen saturation.
[0052] Furthermore, once surgery begins, the system establishes a connection with vital signs monitors, anesthesia machines, and other equipment, receiving real-time, uninterrupted data streams of vital signs such as heart rate, blood pressure, and core body temperature. This continuous data stream is a prerequisite for the system to achieve dynamic monitoring and real-time control.
[0053] Step S20: The AI risk assessment module receives the preoperative indicator data, performs an initial calculation based on the preoperative indicator data, outputs the initial low body temperature risk level, and receives the real-time vital signs data.
[0054] Specifically, the system inputs preprocessed preoperative data (such as age, ASA classification, preoperative body temperature, and estimated operation duration) into a pre-trained AI model (random forest model). Based on patterns learned from historical big data, the model makes a prospective, initial prediction of the likelihood of perioperative hypothermia in patients.
[0055] The model's calculations categorize the initial hypothermia risk level as "high," "medium," "low," or other graded levels. This level serves as the direct basis for initiating differentiated pre-interventions, making resource allocation and pre-operative preparation more targeted and predictive.
[0056] Step S30: The AI risk assessment module periodically combines the real-time vital signs data with the updated surgical data to recalculate and output the updated risk level.
[0057] Understandably, the system does not perform a single preoperative assessment, but rather periodically reassesses at fixed time intervals (e.g., every 5-10 minutes). Each assessment incorporates the latest real-time vital signs data (such as current body temperature and heart rate trends) and updated surgical data as new input features. The updated surgical data includes vital sign monitoring data, physiological metabolic status data, device feedback data, and surgical duration data.
[0058] Furthermore, the AI model recalculates by incorporating these real-time intraoperative variables. This allows the risk level to be dynamically adjusted based on the patient's actual physiological response and the evolution of the surgical process. For example, even if the preoperative assessment indicates low risk, if massive intraoperative bleeding occurs or the surgical time significantly exceeds expectations, the system will promptly raise the risk level and trigger corresponding intervention strategy adjustments.
[0059] Step S40: The decision support and intervention module receives the initial low body temperature risk level and the updated risk level, and generates and executes the preoperative pre-intervention strategy and the updated pre-intervention strategy according to the initial low body temperature risk level and the updated risk level, respectively.
[0060] In this embodiment, the abstract risk level is transformed into specific, executable clinical instructions. Based on the initial risk level, the system automatically invokes preset rules. For high-risk patients, the system automatically executes the instructions: activate the heating equipment for preoperative warming, and adjust and lock the room temperature. This reflects the proactive preventative nature of the method. When an updated risk level is received, the system immediately generates a new intervention strategy. For example, if the risk level increases, it may automatically increase the heating intensity or issue an alert to medical staff suggesting supplementary interventions. This ensures that interventions are always matched to the patient's real-time risk status, achieving precise and timely intervention.
[0061] Step S50: The intelligent control module performs closed-loop regulation based on the real-time body temperature data in the real-time vital signs data during the intraoperative stage.
[0062] Specifically, the system uses real-time body temperature data as the core feedback signal and operates in two modes: first, a stepped adjustment based on preset thresholds (automatically increasing heating when the body temperature falls below a certain value); second, a more advanced PID closed-loop algorithm control, where the system continuously calculates the difference between the target body temperature and the actual body temperature, and dynamically and smoothly adjusts the output power of the heating device through the algorithm. The core of this step lies in forming an automatic "monitoring-calculation-adjustment" closed loop, eliminating the need for medical staff to continuously manually adjust equipment parameters, thus striving to stabilize the patient's body temperature within the target range, greatly reducing the clinical workload and improving the accuracy and stability of control.
[0063] Through the above specific implementation, the present invention has the following beneficial effects: (1) From “people looking for things” to “things looking for people”: The system automatically triggers all processes, freeing nurses from tedious recording, judgment and manual operation.
[0064] (2) From “static assessment” to “dynamic prediction”: AI models can integrate multiple risk factors, provide accurate predictions that far exceed human experience, and continuously update during the operation.
[0065] (3) From “open-loop reminder” to “closed-loop control”: The system no longer just issues alarms, but directly drives the equipment to perform interventions, forming an autonomous closed loop of “perception-decision-execution-re-perception”, which ensures the mandatory nature and consistency of intervention measures and greatly improves the compliance of clinical guidelines.
[0066] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a perioperative hypothermia risk stratification control program, and the perioperative hypothermia risk stratification control program, when executed by a processor, implements the steps of the perioperative hypothermia risk stratification control method as described above.
[0067] In summary, this invention provides a perioperative hypothermia risk stratification control system, method, and computer-readable storage medium. The perioperative hypothermia risk stratification control system includes: a data acquisition and preprocessing module, an AI risk assessment module, a decision support and intervention module, and an intelligent control module. The data acquisition and preprocessing module automatically acquires the patient's preoperative indicator data from the hospital information system during the preoperative stage and continuously receives the patient's real-time vital sign data during the intraoperative stage. The AI risk assessment module receives the preoperative indicator data, performs an initial calculation based on the preoperative indicator data, outputs an initial hypothermia risk level, and receives real-time vital sign data, periodically combining the real-time vital sign data with updated surgical data to perform a recalculation and output an updated risk level. The decision support and intervention module receives the initial hypothermia risk level and the updated risk level, and generates and executes preoperative and updated preoperative intervention strategies based on the initial and updated risk levels, respectively. The intelligent control module performs closed-loop regulation based on real-time temperature data from the real-time vital sign data during the intraoperative stage. This invention utilizes machine learning algorithms to intelligently analyze multiple preoperative indicators of patients and automatically outputs low, medium, and high risk levels. Based on the risk level, the system automatically generates personalized intervention suggestions and dynamically adjusts the ambient temperature and equipment power according to the patient's real-time body temperature, forming a fully automated and intelligent closed-loop management system.
[0068] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.
[0069] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0070] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.
Claims
1. A perioperative hypothermia risk stratification control system, characterized in that, The perioperative hypothermia risk stratification control system includes: a data acquisition and preprocessing module, an AI risk assessment module, a decision support and intervention module, and an intelligent control module, wherein the data acquisition and preprocessing module, the AI risk assessment module, the decision support and intervention module, and the intelligent control module are connected in sequence. The data acquisition and preprocessing module is used to automatically acquire the patient's preoperative indicator data from the hospital information system in the preoperative stage, and to continuously receive the patient's real-time vital signs data in the intraoperative stage. The AI risk assessment module is used to receive the preoperative indicator data, perform an initial calculation based on the preoperative indicator data, output an initial low body temperature risk level, and receive the real-time vital signs data. It also periodically combines the real-time vital signs data with the updated surgical data to perform a recalculation and output an updated risk level. The decision support and intervention module is used to receive the initial low body temperature risk level and the updated risk level, and generate and execute the preoperative pre-intervention strategy and the updated pre-intervention strategy according to the initial low body temperature risk level and the updated risk level, respectively. The intelligent control module is used to perform closed-loop regulation based on real-time body temperature data in the real-time vital signs data during the intraoperative phase.
2. The perioperative hypothermia risk stratification control system according to claim 1, characterized in that, The data acquisition and preprocessing module includes: a preoperative acquisition unit, a data preprocessing unit, and an intraoperative acquisition unit; The preoperative data collection unit is used to automatically extract the patient's preoperative indicator data from the hospital's anesthesia system and electronic medical records through a standardized interface after the patient has completed identity verification and been connected to vital sign monitoring in the operating room. The preoperative preprocessing unit is used to receive the preoperative indicator data, preprocess the preoperative indicator data to obtain preprocessed data, and integrate and transform the preprocessed data to obtain numerical features that can be processed by the AI model. The intraoperative data acquisition unit is used to continuously receive real-time vital sign data of the patient through a vital sign monitor after the patient enters anesthesia, and to continuously receive updated surgical data of the patient from the operating room.
3. The perioperative hypothermia risk stratification control system according to claim 2, characterized in that, The AI risk assessment module includes: a preoperative risk assessment unit and an intraoperative risk assessment unit; The preoperative risk assessment unit is used to receive the numerical features, input the numerical features into the trained random forest model for calculation, and output the patient's initial hypothermia risk level. The intraoperative risk assessment unit is used to receive the real-time vital signs data and the updated surgical data, input the real-time vital signs data and the updated surgical data into the random forest model for recalculation, and output the patient's updated risk level.
4. The perioperative hypothermia risk stratification control system according to claim 3, characterized in that, The decision support and intervention module includes: a preoperative decision intervention unit and an intraoperative decision intervention unit; The preoperative decision-making intervention unit is used to receive the initial low body temperature risk level. When the initial low body temperature risk level is high, it sends a first instruction to the inflatable warmer in the operating room via the MQTT protocol and a second instruction to the air conditioning system. The first instruction is to control the inflatable warmer to start running before anesthesia induction and to perform preoperative warming for a preset duration. The second instruction is to set and lock the operating room temperature at a first preset air conditioning temperature. The preoperative decision-making intervention unit is also used to notify the nurse to prepare and turn on the warming device when the initial low body temperature risk level is medium risk, and to notify the nurse to use thicker bedding and remind the nurse to pay attention to changes in the patient's body temperature when the initial low body temperature risk level is low risk. The intraoperative decision-making intervention unit is used to receive the updated risk level and generate an updated pre-intervention strategy based on the updated risk level.
5. The perioperative hypothermia risk stratification control system according to claim 1, characterized in that, The intelligent control module includes: a first strategy control unit and a second strategy control unit. The first strategy control unit is used to increase the power of the heating blanket to a preset power and raise the temperature of the air conditioning system to a second preset air conditioning temperature when the real-time body temperature data in the real-time vital signs data is lower than a first preset temperature. The first strategy control unit is also used to trigger a high-level alarm and send a suggestion to the nurse to start fluid warming when the real-time body temperature data is lower than the second preset temperature; The second strategy control unit is used to perform PID calculations based on the difference between the real-time body temperature data and the preset target body temperature data, and to dynamically adjust the output power of the heating device based on the calculation results.
6. The perioperative hypothermia risk stratification control system according to claim 1, characterized in that, The perioperative hypothermia risk stratification control system also includes: a postoperative feedback optimization module; The postoperative feedback optimization module is used to automatically generate a perioperative temperature management report after the patient's surgery and archive the perioperative temperature management report into the patient's medical record. The perioperative temperature management report includes the initial risk level, the intervention measures taken, and the temperature change curve. The postoperative feedback optimization module is also used to collect perioperative temperature management reports from all anonymous patients and to periodically train and optimize the AI model of the AI risk assessment module using these reports.
7. A perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system according to any one of claims 1-6, characterized in that, The perioperative hypothermia risk stratification control method includes: The data acquisition and preprocessing module automatically acquires the patient's preoperative indicator data from the hospital information system during the preoperative stage and continuously receives the patient's real-time vital signs data during the intraoperative stage. The AI risk assessment module receives the preoperative indicator data, performs an initial calculation based on the preoperative indicator data, outputs the initial hypothermia risk level, and receives the real-time vital signs data. The AI risk assessment module periodically combines the real-time vital signs data with the updated surgical data to recalculate and output the updated risk level. The decision support and intervention module receives the initial low body temperature risk level and the updated risk level, and generates and executes the preoperative pre-intervention strategy and the updated pre-intervention strategy based on the initial low body temperature risk level and the updated risk level, respectively. The intelligent control module performs closed-loop regulation based on the real-time body temperature data in the real-time vital signs data during the intraoperative phase.
8. The perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system according to claim 7, characterized in that, The preoperative indicator data includes: static data and dynamic data; The static data includes age, gender, weight, height, anesthesiologist association classification, and comorbidities; The dynamic data includes preoperative core body temperature, type of surgery to be performed, estimated surgery duration, and planned anesthesia method.
9. The perioperative hypothermia risk stratification control method based on the perioperative hypothermia risk stratification control system according to claim 7, characterized in that, The real-time vital signs data include heart rate data, blood pressure data, respiratory rate data, real-time body temperature data, and blood oxygen saturation data. The updated surgical data includes vital sign monitoring data, physiological metabolic status data, equipment feedback data, and surgical duration data.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a perioperative hypothermia risk stratification control program, which, when executed by a processor, implements the steps of the perioperative hypothermia risk stratification control method as described in any one of claims 7-9.