Intelligent prejudgment, prevention and control system for risk of complications of thoracic surgery

Abstract

The invention discloses a thoracic surgery complication risk intelligent pre-judgment and prevention and control system, and relates to the technical field of medical surgery, the system comprises a data acquisition module, an edge fusion module, a baseline storage module, a two-dimensional evaluation module and a prevention and control execution module; according to the invention, intraoperative multi-source dynamic data are comprehensively collected through the data acquisition module, after feature quantification is carried out on image data through the edge fusion module and normalization processing is carried out on other data, intraoperative real-time comprehensive data with unified dimensions are generated through fusion operation, and meanwhile, preoperative baseline data are called by relying on the baseline storage module; the two-dimensional evaluation module is used for correspondingly matching intraoperative comprehensive data with preoperative baseline data, and a pre-judgment result of a complication occurrence condition is formed through multi-step quantitative calculation, so that full-process collaboration from data acquisition, fusion processing to evaluation and pre-judgment is realized, the pre-judgment process depends on integration and multi-dimensional analysis of multi-source data, and the pre-judgment efficiency is improved. And a complete logic closed loop is formed.

Description

Technical Field

[0001] This invention relates to the field of medical surgical technology, specifically to an intelligent prediction and control system for the risk of complications in thoracic surgery. Background Technology

[0002] Thoracic surgery is a core medical approach for treating diseases of the lungs, esophagus, mediastinum, and other thoracic organs. Its operation area involves multiple key physiological structures, making the surgery highly complex. Acute complications such as pneumothorax and massive hemorrhage occur suddenly and progress rapidly, directly affecting the surgical process and the patient's life. If they are not detected and treated in time, they may lead to serious consequences such as organ damage and shock. These are key issues that need to be prevented in clinical thoracic surgery. The ability to predict and control complications directly affects the quality of medical care and the patient's prognosis.

[0003] However, existing technologies have some limitations in the prediction and prevention of complications in thoracic surgery. First, data collection lacks systematicity, with most data being single-type or scattered and acquired independently. Intraoperative multi-source dynamic data is not effectively integrated and standardized, making it impossible to uncover correlations between data. Second, there is a lack of a dedicated preoperative baseline data storage and retrieval mechanism, making it difficult to achieve accurate matching and analysis between intraoperative data and preoperative baseline information. Predictions rely solely on single-dimensional data, resulting in incomplete logic. Third, prevention and control measures lack tiered adaptability. Targeted early warning and equipment linkage plans are not developed based on differences in the probability of complication, and the intensity of early warnings and equipment control commands do not correspond to the risk level. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an intelligent prediction and prevention system for thoracic surgery complications. This invention comprehensively collects multi-source dynamic data during surgery through a data acquisition module. After feature quantification of the image data and normalization of other data by an edge fusion module, unified-dimensional real-time comprehensive intraoperative data is generated through fusion calculation. At the same time, the baseline storage module calls up the preoperative baseline data, and the dual-dimensional assessment module matches the comprehensive intraoperative data with the preoperative baseline data. Through multi-step quantitative calculation, a prediction result of the occurrence of complications is formed. This realizes the full-process collaboration from data acquisition, fusion processing to assessment and prediction, allowing the prediction process to form a complete logical closed loop based on the integration of multi-source data and multi-dimensional analysis.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an intelligent prediction and prevention system for complications in thoracic surgery, the system comprising: Data acquisition module: Collects multi-source dynamic data of patients through thoracoscopic image acquisition unit, respiratory motion sensor, anesthesia depth monitor and hemodynamic monitor, and transmits the multi-source dynamic data; Edge fusion module: Receives multi-source dynamic data, performs data fusion after format standardization, and generates real-time comprehensive intraoperative data; Baseline storage module: Constructs a storage database to store patients' preoperative baseline data, including preoperative imaging baseline, physiological parameter baseline, and medical history baseline; Dual-dimensional assessment module: Based on real-time comprehensive data during surgery and the patient's preoperative imaging baseline, physiological parameter baseline and medical history baseline, a dual-dimensional risk assessment model is constructed to predict the occurrence of acute complications and output the prediction results of the occurrence of acute complications. Prevention and control execution module: includes an audible and visual early warning unit and an equipment linkage unit. After receiving the prediction results, it activates the corresponding early warning mechanism and links the surgical equipment to perform prevention and control operations.

[0006] Furthermore, the data acquisition module includes multi-source dynamic data such as real-time intraoperative thoracoscopic images, patient respiratory motion data, anesthesia depth monitoring indicators, and hemodynamic parameters. The real-time intraoperative thoracoscopic images are acquired through the thoracoscopic image acquisition unit; patient respiratory motion data are acquired through the respiratory motion sensor; anesthesia depth monitoring indicators, including BIS values, are acquired through the anesthesia depth monitor; and hemodynamic parameters, including cardiac output and central venous pressure, are acquired through the hemodynamic monitor.

[0007] Furthermore, in the edge fusion module, the steps for generating real-time intraoperative comprehensive data are as follows: extracting frames from the real-time stream of thoracoscopic images in the multi-source dynamic data, quantifying the anatomical structural features and dynamic features of the surgical area in the images, normalizing respiratory motion data, anesthesia depth monitoring indicators, and hemodynamic parameters, unifying the data dimensions, and performing fusion calculations on the processed data using a multi-source data weighted fusion formula to generate real-time intraoperative comprehensive data of a unified dimension.

[0008] Furthermore, in the edge fusion module, the multi-source data weighted fusion formula is: ,in, This refers to real-time comprehensive data values ​​during the operation; These are the normalized hemodynamic parameters. The data represents normalized respiratory movement. This is a normalized indicator for monitoring the depth of anesthesia. These are the quantified thoracoscopic image feature values; The weighting coefficients were determined using historical clinical data from thoracic surgery. .

[0009] Furthermore, in the baseline storage module, the preoperative imaging baseline stored in the storage database includes the anatomical parameters of the preoperative chest CT and MRI, the physiological parameter baseline includes the preoperative blood pressure, heart rate, and pulmonary function test results, and the medical history baseline includes the patient's underlying pulmonary disease, coagulation function status, and surgical history information.

[0010] Furthermore, the specific steps for predicting acute complications in the dual-dimensional assessment module are as follows: preoperative baseline data of the patient stored in the database is retrieved and matched with real-time comprehensive data during surgery; and the deviation of each matching parameter is calculated. With rate of change The risk correlation degree is calculated using the formula for calculating the risk correlation degree of complications. Based on risk correlation The comprehensive assessment score is calculated using a two-dimensional risk comprehensive assessment scoring formula. Then, the comprehensive evaluation score will be used. Substitute the formula for calculating the probability of complication in the two-dimensional risk assessment model to calculate the probability of complication. Based on the probability of complications This allows for the prediction of potential complications.

[0011] Furthermore, in the dual-dimensional assessment module, the formula for calculating the correlation between complication risk and risk is as follows: ,in, For risk correlation, This represents the difference between the intraoperative real-time value and the preoperative baseline value. The mean of preoperative baseline data. For the statistical period, The contribution coefficient was determined through an analysis of historical cases of acute complications in thoracic surgery.

[0012] Furthermore, in the dual-dimensional assessment module, the formula for the dual-dimensional risk comprehensive assessment score is: ,in, To comprehensively evaluate the score, For risk correlation, The risk association correction coefficient is determined based on individual characteristics such as patient age and type of underlying disease. The individual difference correction term is determined by calculating the standard deviation of the results of 3-4 preoperative basic physiological parameter tests, including blood pressure, heart rate, and respiratory movement.

[0013] Furthermore, in the dual-dimensional assessment module, the formula for calculating the probability of complication is: ,in, The probability of complications. To comprehensively evaluate the score, The threshold for assessing the risk of complications was determined by calibrating and using clinical data from thoracic surgery in multiple tertiary hospitals, based on the clinical critical value standards for two types of complications: pneumothorax and massive hemorrhage. when When, it is determined that there is a high probability of complications occurring; when When, it is determined to be a complication with a medium probability of occurrence; when When the probability of complications is low, the threshold for each probability interval is determined by the preoperative baseline data stored in the database.

[0014] Furthermore, in the aforementioned prevention and control execution module, the specific steps for performing prevention and control operations are as follows: Based on the prediction results, the audible and visual warning unit activates the corresponding warning. When the probability of complications is low, a green indicator light and a low-frequency prompt sound are triggered. The equipment linkage unit does not send active control commands, but only provides real-time feedback on the operating status of each device. When the probability of complications is medium, a yellow indicator light and a medium-frequency prompt sound are triggered by the audible and visual warning unit. The equipment linkage unit sends a visual field clarity adjustment prompt command to the thoracoscopic surgery equipment, a parameter fluctuation warning feedback to the anesthesia machine, and a standby status confirmation command to the hemostasis equipment. When the probability of complications is high, a red indicator light, a high-frequency prompt sound, and a vibration reminder are triggered by the audible and visual warning unit. The equipment linkage unit sends a visual field magnification and focusing command to the thoracoscopic surgery equipment, an anesthesia dosage optimization command to the anesthesia machine, and a start-up standby command to the hemostasis equipment.

[0015] Compared with existing technologies, this intelligent prediction and prevention system for thoracic surgery complications has the following beneficial effects: I. This invention comprehensively collects multi-source dynamic data during surgery through a data acquisition module. After the edge fusion module performs feature quantification on the image data and normalization on other data, it generates unified-dimensional real-time comprehensive intraoperative data through fusion calculation. At the same time, it relies on the baseline storage module to call up preoperative baseline data. The dual-dimensional assessment module matches the comprehensive intraoperative data with the preoperative baseline data. Through multi-step quantitative calculation, it forms a prediction result of the occurrence of complications. This realizes the full-process collaboration from data acquisition and fusion processing to assessment and prediction, allowing the prediction process to rely on the integration of multi-source data and multi-dimensional analysis to form a complete logical closed loop.

[0016] Second, this invention outputs prediction results with different probabilities through a dual-dimensional assessment module, triggering a hierarchical response mechanism in the prevention and control execution module. Corresponding audible and visual warnings are activated for each prediction result. At the same time, the equipment linkage unit sends appropriate equipment control commands based on the prediction results, achieving precise adaptation between the warning mechanism and equipment linkage. This allows prevention and control operations to be adjusted in a targeted manner according to the differences in the probability of complication, making the warning information clearly distinguish the degree of risk. The equipment control commands are tailored to the actual needs of different risk scenarios, forming an efficient linkage between prediction results and prevention and control measures.

[0017] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 A flowchart of an intelligent prediction and control system for complications in thoracic surgery; Figure 2 A framework diagram of an intelligent prediction and control system for complications in thoracic surgery; Figure 3 This is a flowchart of a two-dimensional assessment module in an intelligent prediction and prevention system for complications in thoracic surgery. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] Example 1: During thoracoscopic radical resection surgery for early-stage non-small cell lung cancer patients, the data acquisition module continuously performs multi-source dynamic data acquisition. The thoracoscopic image acquisition unit captures real-time intraoperative thoracoscopic images, comprehensively recording the dynamic changes of anatomical structures such as lung tissue, blood vessels, and lymph nodes within the surgical area. A respiratory motion sensor continuously collects the patient's respiratory motion data, accurately capturing dynamic indicators such as respiratory rhythm and chest wall movement amplitude. An anesthesia depth monitor obtains real-time anesthesia depth monitoring indicators, including BIS value, to dynamically monitor the patient's anesthesia and sedation status. A hemodynamic monitor continuously collects hemodynamic parameters such as cardiac output and central venous pressure to comprehensively monitor intraoperative blood circulation function. All acquired multi-source dynamic data is transmitted in real-time to the edge fusion module, such as... Figure 1 As shown.

[0022] After receiving multi-source dynamic data from the data acquisition module, the edge fusion module first performs format standardization processing. For the real-time stream of thoracoscopic images, it performs frame extraction, quantifying anatomical features such as lung lesion boundaries and blood vessel orientation, as well as dynamic features of the surgical area such as surgical instrument trajectories. Simultaneously, it normalizes respiratory motion data, anesthesia depth monitoring indicators including BIS values, and hemodynamic parameters such as cardiac output and central venous pressure, unifying the data dimensions of various types of data and eliminating the influence of differences between different data types. Then, it uses a multi-source data weighted fusion formula to fuse the processed data, ultimately generating unified intraoperative real-time comprehensive data, providing standardized data support for subsequent risk assessment. The multi-source data weighted fusion formula is as follows: ,in, This refers to real-time comprehensive data values ​​during the operation; These are the normalized hemodynamic parameters. The data represents normalized respiratory movement. This is a normalized indicator for monitoring the depth of anesthesia. These are the quantified thoracoscopic image feature values; The weighting coefficients were determined using historical clinical data from thoracic surgery. .

[0023] The baseline storage module's database pre-stores the patient's preoperative baseline data. Among them, the preoperative imaging baseline includes lung anatomical parameters obtained from the patient's preoperative chest CT and MRI examinations, such as lesion size, location, lobar volume, and mediastinal lymph node distribution; the physiological parameter baseline covers multiple measurements of preoperative blood pressure and heart rate, as well as indicators such as vital capacity and forced vital capacity obtained from pulmonary function tests; the medical history baseline records in detail whether the patient has underlying lung diseases such as chronic obstructive pulmonary disease and tuberculosis, coagulation function test results, and whether there is a history of chest surgery, providing a comprehensive preoperative reference for intraoperative risk comparison and assessment.

[0024] After the dual-dimensional assessment module is activated, it first constructs a dual-dimensional risk assessment model based on the correlation logic between the patient's preoperative imaging baseline, physiological parameter baseline, medical history baseline, and intraoperative real-time comprehensive data in the baseline storage module. Then, it calls up the patient's preoperative baseline data and performs parameter-by-parameter matching with the intraoperative real-time comprehensive data generated by the edge fusion module. Next, it calculates the deviation magnitude and rate of change of each matched parameter to quantify the degree of deviation and speed of change between the intraoperative data and the preoperative baseline. Finally, it uses the complication risk correlation formula, combined with the deviation magnitude and rate of change, to derive the risk correlation. The complication risk correlation formula is as follows: ,in, For risk correlation, This represents the difference between the intraoperative real-time value and the preoperative baseline value. The mean of preoperative baseline data. For the statistical period, The contribution coefficient was determined through historical analysis of acute complications in thoracic surgery cases. Subsequently, a risk correlation correction coefficient was determined based on individual characteristics such as patient age and whether they had underlying chronic lung diseases. Individual difference correction terms were calculated based on the results of 3-4 preoperative measurements of basic physiological parameters including blood pressure, heart rate, and respiratory movement. The comprehensive assessment score was then calculated using a two-dimensional risk comprehensive assessment scoring formula, which is as follows: ,in, To comprehensively evaluate the score, For risk correlation, The risk association correction coefficient is determined based on individual characteristics such as patient age and type of underlying disease. For the individual variability correction term, the standard deviation was calculated based on the results of 3-4 preoperative baseline physiological parameter tests (blood pressure, heart rate, and respiratory movement). Finally, the comprehensive assessment score was substituted into the complication probability calculation formula in the two-dimensional risk assessment model to calculate the complication probability. The complication probability calculation formula is as follows: ,in, The probability of complications. To comprehensively evaluate the score, The threshold for determining the risk of complications; when When, it is determined that there is a high probability of complications occurring; when When, it is determined to be a complication with a medium probability of occurrence; when At that time, it was determined that the probability of complications was low; the judgment thresholds corresponding to each probability interval can be dynamically adjusted through the preoperative baseline data stored in the database; it was calculated that the probability of complications for this patient was 78%, and it was determined that the probability of complications was high.

[0025] like Figure 2 As shown, after receiving the high-probability complication prediction result from the dual-dimensional assessment module, the prevention and control execution module immediately activates the highest-level prevention and control mechanism. The sound and light warning unit simultaneously triggers a constant red indicator light, continuous high-frequency prompts, and vibration alerts, sending an emergency warning signal to the surgical medical team, indicating the need to pay close attention to the risk of complications. The equipment linkage unit quickly sends a field-of-view magnification and focusing command to the thoracoscopic surgical equipment, helping medical staff to clearly observe key areas such as pulmonary vascular anastomosis sites and lesion resection wounds, and promptly identify potential risk points. It sends an anesthesia dosage optimization command to the anesthesia machine to ensure that the patient's anesthesia depth is maintained within a safe range, avoiding complications induced by excessively deep or shallow anesthesia. It sends an activation and standby command to the hemostasis equipment, putting the hemostasis instruments in an immediately operable state, so that in the event of sudden situations such as intraoperative bleeding, a rapid response and hemostasis operation can be carried out to minimize the risk of complications.

[0026] In summary, in the scenario of thoracoscopic radical surgery for early-stage non-small cell lung cancer, the data acquisition module acquires and transmits real-time intraoperative thoracoscopic images, respiratory motion data, anesthesia depth monitoring indicators including BIS values, and hemodynamic parameters such as cardiac output. The edge fusion module, after format standardization, feature quantification, and normalization, generates real-time comprehensive intraoperative data through a multi-source data weighted fusion formula. The baseline storage module provides preoperative imaging, physiological parameters, and medical history baseline data. The dual-dimensional assessment module first constructs a dual-dimensional risk assessment model, then, by matching data, calculating the deviation amplitude and rate of change, and combining the complication risk correlation calculation formula, the dual-dimensional risk comprehensive assessment score formula, and the complication probability calculation formula in the dual-dimensional risk assessment model, it determines high-probability complications. The prevention and control execution module triggers corresponding early warnings and links relevant surgical equipment to perform corresponding prevention and control operations to achieve risk control.

[0027] Example 2: During lobectomy in patients with unilateral benign lung lesions, the data acquisition module performs multi-source dynamic data acquisition. The thoracoscopic imaging unit acquires real-time intraoperative thoracoscopic images, clearly showing the dynamic changes in lung tissue, bronchi, and surrounding blood vessels within the surgical area. A respiratory motion sensor continuously collects respiratory motion data, accurately capturing dynamic parameters such as chest wall movement and lung expansion and contraction during respiration. An anesthesia depth monitor collects anesthesia depth indicators, including BIS values, to monitor the stability of the patient's intraoperative anesthesia. A hemodynamic monitor continuously collects hemodynamic parameters such as cardiac output and central venous pressure to dynamically assess the patient's intraoperative blood circulation. All acquired multi-source dynamic data is transmitted in real-time to the edge fusion module.

[0028] After receiving the transmitted multi-source dynamic data, the edge fusion module first performs format standardization processing, extracts frames from the real-time thoracoscopic image stream, and quantifies features such as lung lobe anatomical details and the dynamics of tissues around the surgical incision. It then normalizes respiratory motion data, anesthesia depth monitoring indicators including BIS values, and hemodynamic parameters such as cardiac output and central venous pressure, unifying the dimensions of various data types. Finally, a multi-source data weighted fusion formula is used to fuse the processed data, successfully generating unified intraoperative real-time comprehensive data, providing integrated data support for subsequent risk assessment. The multi-source data weighted fusion formula is as follows: .

[0029] The baseline storage module's database pre-stores the patient's preoperative baseline data. The preoperative imaging baseline includes anatomical parameters of the lung lobes obtained from preoperative chest CT and MRI examinations, such as lung lobe size, bronchial branch morphology, and vascular distribution. The physiological parameter baseline covers multiple preoperative measurements of blood pressure and heart rate, as well as pulmonary function test results, such as ventilation and gas exchange function indicators. The medical history baseline records in detail information such as whether the patient has underlying lung diseases, coagulation function test results, and past surgical history, providing comprehensive preoperative data for intraoperative risk assessment.

[0030] The dual-dimensional assessment module first uses the patient's preoperative imaging baseline, physiological parameter baseline, and medical history baseline as a foundation, and combines this with the dynamic changes in real-time intraoperative data to construct a dual-dimensional risk assessment model. Then, it calls upon the patient's preoperative baseline data and matches it with the real-time intraoperative data generated by the edge fusion module. Figure 3 As shown, the deviation magnitude and rate of change of each matching parameter were calculated to clarify the differences and trends between intraoperative data and preoperative baseline; the risk correlation degree was obtained through the complication risk correlation degree calculation formula to quantify the degree of correlation between changes in each parameter and the risk of complications. The complication risk correlation degree calculation formula is as follows: Subsequently, risk correlation correction coefficients were determined based on individual characteristics such as patient age and the presence of underlying diseases. Individual difference correction items were determined based on the results of 3-4 preoperative baseline physiological parameter tests (blood pressure, heart rate, and respiratory movement). The comprehensive assessment score was calculated using the two-dimensional risk comprehensive assessment scoring formula, which is as follows: The comprehensive assessment score is then substituted into the complication probability calculation formula in the two-dimensional risk assessment model to calculate the complication probability. The complication probability calculation formula is as follows: ;when When, it is determined that there is a high probability of complications occurring; when When, it is determined to be a complication with a medium probability of occurrence; when At that time, it was determined that the probability of complications was low; the judgment thresholds corresponding to each probability interval can be dynamically adjusted through the preoperative baseline data stored in the storage database; the calculated probability of complications for this patient was 52%, and according to the judgment criteria, it was determined that the probability of complications was medium.

[0031] Upon receiving the predicted result of a moderate probability complication, the prevention and control execution module initiates corresponding early warning and equipment linkage operations. The audible and visual warning unit triggers a flashing yellow indicator light and a cyclical mid-frequency prompt tone, issuing a moderate risk warning to medical staff and prompting them to strengthen intraoperative monitoring. The equipment linkage unit sends a visual field clarity adjustment prompt to the thoracoscopic surgical equipment, helping medical staff optimize the surgical field of view, observe the surgical operation area more clearly, and reduce the operational risks caused by blurred vision. It sends parameter fluctuation warning feedback to the anesthesia machine, reminding the anesthesiologist to pay attention to changes in anesthesia parameters, adjust the anesthesia plan in a timely manner, and maintain the patient's vital signs. It sends a standby status confirmation command to the hemostasis equipment to verify the operating status of the hemostasis equipment, ensuring that it is in good standby condition and can be quickly put into use when needed, effectively responding to potential bleeding risks and ensuring the safe and orderly conduct of the surgery.

[0032] In summary, in the scenario of unilateral lobectomy for benign lung lesions, the data acquisition module collects and transmits multi-source dynamic data during the operation, covering key information such as real-time thoracoscopic imaging and respiratory motion data; the edge fusion module processes the data and generates unified-dimensional real-time comprehensive data during the operation using a multi-source data weighted fusion formula; the baseline storage module provides comprehensive preoperative baseline data support for the patient; the dual-dimensional assessment module first constructs a dual-dimensional risk assessment model based on the preoperative baseline data and the real-time comprehensive data during the operation, and then, through data matching and parameter calculation, determines that the probability of complication is moderate; the prevention and control execution module then initiates corresponding early warnings, sends corresponding instructions to the thoracoscopic surgical equipment, anesthesia machine, and hemostasis equipment, and provides real-time feedback on equipment status or adjusts parameters to ensure surgical safety and reduce the risk of complications.

[0033] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A smart prediction and control system for the risk of complications in thoracic surgery, characterized in that, The system includes: Data acquisition module: Collects multi-source dynamic data of patients through thoracoscopic image acquisition unit, respiratory motion sensor, anesthesia depth monitor and hemodynamic monitor, and transmits the multi-source dynamic data; Edge fusion module: Receives multi-source dynamic data, performs data fusion after format standardization, and generates real-time comprehensive intraoperative data; Baseline storage module: Constructs a storage database to store patients' preoperative baseline data, including preoperative imaging baseline, physiological parameter baseline, and medical history baseline; Dual-dimensional assessment module: Based on real-time comprehensive data during surgery and the patient's preoperative imaging baseline, physiological parameter baseline and medical history baseline, a dual-dimensional risk assessment model is constructed to predict the occurrence of acute complications and output the prediction results of the occurrence of acute complications. Prevention and control execution module: includes an audible and visual early warning unit and an equipment linkage unit. After receiving the prediction results, it activates the corresponding early warning mechanism and links the surgical equipment to perform prevention and control operations.

2. The intelligent prediction and prevention system for thoracic surgery complications according to claim 1, characterized in that, The data acquisition module includes multi-source dynamic data such as real-time intraoperative thoracoscopic images, patient respiratory motion data, anesthesia depth monitoring indicators, and hemodynamic parameters. Real-time intraoperative thoracoscopic images are acquired through the thoracoscopic image acquisition unit; patient respiratory motion data is acquired through the respiratory motion sensor; anesthesia depth monitoring indicators, including BIS value, are acquired through the anesthesia depth monitor; and hemodynamic parameters, including cardiac output and central venous pressure, are acquired through the hemodynamic monitor.

3. The intelligent prediction and prevention system for thoracic surgery complications according to claim 1, characterized in that, In the edge fusion module, the steps for generating real-time comprehensive intraoperative data are as follows: extracting frames from the real-time stream of thoracoscopic images in multi-source dynamic data, quantifying the anatomical structural features and dynamic features of the surgical area in the images, normalizing respiratory motion data, anesthesia depth monitoring indicators, and hemodynamic parameters, unifying the data dimensions, and performing fusion calculations on the processed data using a multi-source data weighted fusion formula to generate real-time comprehensive intraoperative data of a unified dimension.

4. The intelligent prediction and prevention system for thoracic surgery complications according to claim 3, characterized in that, In the edge fusion module, the multi-source data weighted fusion formula is: ,in, This refers to real-time comprehensive data values ​​during the operation; These are the normalized hemodynamic parameters. The data represents normalized respiratory movement. This is a normalized indicator for monitoring the depth of anesthesia. These are the quantified thoracoscopic image feature values; These are the weighting coefficients.

5. The intelligent prediction and prevention system for thoracic surgery complications according to claim 1, characterized in that, The baseline storage module stores preoperative imaging baselines in the database, including anatomical parameters of preoperative chest CT and MRI, physiological parameter baselines, preoperative blood pressure, heart rate, and pulmonary function test results, and medical history baselines, including the patient's underlying pulmonary disease, coagulation function status, and surgical history information.

6. The intelligent prediction and prevention system for thoracic surgery complications according to claim 1, characterized in that, In the dual-dimensional assessment module, the specific steps for predicting acute complications are as follows: preoperative baseline data of the patient stored in the database is retrieved and matched with real-time comprehensive data during surgery; and the deviation of each matching parameter is calculated. With rate of change The risk correlation degree is calculated using the formula for calculating the risk correlation degree of complications. Based on risk correlation The comprehensive assessment score is calculated using a two-dimensional risk comprehensive assessment scoring formula. Then, the comprehensive evaluation score will be used. Substitute the formula for calculating the probability of complication in the two-dimensional risk assessment model to calculate the probability of complication. Based on the probability of complications This allows for the prediction of potential complications.

7. The intelligent prediction and prevention system for thoracic surgery complications according to claim 6, characterized in that, In the dual-dimensional assessment module, the formula for calculating the correlation between complication risk and risk is as follows: ,in, For risk correlation, This represents the difference between the intraoperative real-time value and the preoperative baseline value. The mean of preoperative baseline data. For the statistical period, This is the contribution coefficient.

8. The intelligent prediction and prevention system for thoracic surgery complications according to claim 7, characterized in that, In the dual-dimensional assessment module, the formula for the comprehensive risk assessment score is as follows: ,in, To comprehensively evaluate the score, For risk correlation, This is a risk correlation correction coefficient. This is an individual difference correction item.

9. The intelligent prediction and prevention system for thoracic surgery complications according to claim 8, characterized in that, In the dual-dimensional assessment module, the formula for calculating the probability of complication is as follows: ,in, The probability of complications. To comprehensively evaluate the score, Threshold for determining the risk of complications; when When, it is determined that there is a high probability of complications occurring; when When, it is determined to be a complication with a medium probability of occurrence; when At that time, it was determined that the probability of complications was low.

10. The intelligent prediction and prevention system for thoracic surgery complications according to claim 1, characterized in that, In the aforementioned prevention and control execution module, the specific steps for performing prevention and control operations are as follows: Based on the prediction results, the audible and visual warning unit activates the corresponding warning. When the probability of complications is low, a green indicator light and a low-frequency prompt sound are triggered. The equipment linkage unit does not send active control commands, but only provides real-time feedback on the operating status of each device. When the probability of complications is medium, a yellow indicator light and a medium-frequency prompt sound are triggered by the audible and visual warning unit. The equipment linkage unit sends a visual field clarity adjustment prompt command to the thoracoscopic surgery equipment, a parameter fluctuation warning feedback to the anesthesia machine, and a standby status confirmation command to the hemostasis equipment. When the probability of complications is high, a red indicator light, a high-frequency prompt sound, and a vibration reminder are triggered by the audible and visual warning unit. The equipment linkage unit sends a visual field magnification and focusing command to the thoracoscopic surgery equipment, an anesthesia dosage optimization command to the anesthesia machine, and a start-up standby command to the hemostasis equipment.

Images