A method and system for predicting complications after cardiothoracic surgery
Through the cardiothoracic surgery postoperative complication prediction system, a variety of monitoring and analysis modules are integrated, multi-dimensional data are integrated, and the risk of postoperative complications is evaluated, which solves the problem of inaccurate prediction in the existing technology and improves the prevention of postoperative complications. and management level reduces the probability of complications.
Patent Information
- Application Number
- CN202411288140.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-14
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-14
AI Technical Summary
The prediction of postoperative complications of cardiothoracic surgery is difficult to integrate multi-dimensional data such as patient's respiratory status, abnormal lung distribution and lung dilation ability in real time, resulting in difficult time discovering potential risks, affecting the treatment effect.
Design a cardiothoracic surgery postoperative complication prediction system, including postoperative data monitoring module, data processing module, lung preliminary analysis module, comprehensive analysis module and prediction management module, and collect data through a variety of monitoring equipment for pre-processing and quantityless The complication data set was constructed, the oxygenation factor and lung dilation ability coefficient were calculated, the risk index was evaluated, and the evaluation threshold was set for comparison and analysis were used to predict the degree of complication risk.
Multi-dimensional monitoring and accurate prediction of patients after cardiothoracic surgery has been achieved, the early detection and management of postoperative complications has been improved, the probability of complications has been reduced, and the treatment effect and recovery quality of patients have been improved.
Smart Images

Figure CN118800458B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical treatment, and in particular to a method and system for predicting postoperative complications of cardiothoracic surgery. Background Art
[0002] Cardiothoracic surgery is a complex and high-risk field in modern medicine, involving various surgical operations on the heart, lungs, and chest. With the continuous advancement of medical technology, the success rate of cardiothoracic surgery has gradually increased, but postoperative complications remain an important clinical challenge. Postoperative complications not only prolong the patient's hospitalization time, but may also cause more serious health problems and even threaten the patient's life. Therefore, early prediction and effective management of postoperative complications of cardiothoracic surgery have become the focus of clinical attention.
[0003] Acute respiratory distress syndrome (ARDS) is one of the possible complications after cardiothoracic surgery. Although a variety of methods have been used in clinical practice to assess the risk of postoperative complications, these methods often rely on doctors' experience and routine examination methods. Especially in the prediction of postoperative pulmonary complications, traditional methods are difficult to integrate multi-dimensional data such as the patient's respiratory status, abnormal lung distribution, and lung expansion capacity in real time and in multiple aspects, making it difficult to detect potential dangers in a timely manner, thus affecting the overall treatment effect. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for predicting postoperative complications of cardiothoracic surgery, which solves the problems in the above-mentioned background technology.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a cardiothoracic surgery postoperative complication prediction system, including a postoperative data monitoring module, a data processing module, a lung preliminary analysis module, a comprehensive analysis module and a prediction management module;
[0006] The postoperative data monitoring module is used to perform physical examinations on patients who have undergone cardiothoracic surgery, and obtain relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information of the patients through reports obtained from the physical examinations;
[0007] The data processing module is used to pre-process the relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information obtained in the postoperative data monitoring module, and construct a postoperative complication data set after dimensionless technology processing;
[0008] The lung preliminary analysis module is used to construct the patient's postoperative oxygenation factor Yhyz based on the relevant respiratory status data information, and preliminarily judge whether the current patient's lung oxygenation capacity is impaired according to the value of the oxygenation factor Yhyz. If impaired, a complication prediction instruction is issued externally;
[0009] The comprehensive analysis module is used to calculate the alveolar ventilation Ftqz and the lung expansion capacity coefficient Fkxs of each lung segment according to the relevant lung abnormality distribution data information and the relevant expansion degree data information after receiving the complication prediction instruction, and to fit and obtain the risk assessment index Fpzs in combination with the trained complication prediction model;
[0010] The prediction management module is used to pre-set an evaluation threshold W, and compare and analyze the evaluation threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient.
[0011] Preferably, the postoperative data monitoring module includes a physical examination unit and a collection unit;
[0012] The physical examination unit is used to examine the postoperative status of patients who have undergone cardiothoracic surgery through several sets of monitoring equipment, wherein the several sets of monitoring equipment include an arterial blood gas analyzer, a ventilator, a pulmonary function tester and a pressure sensor;
[0013] The acquisition unit is used to collect and record relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information of the patient according to the patient reports obtained by several groups of monitoring devices, wherein the relevant respiratory status data information includes the arterial blood oxygen partial pressure Dfy and the inspired oxygen concentration Xynd of the patient before, during and after the operation;
[0014] The relevant lung abnormality distribution data information includes the total ventilation volume Mzz and dead space ventilation Wtz per minute in each lung segment of the patient's lungs after surgery;
[0015] Relevant expansion degree data information includes tidal volume Cqz, positive end-expiratory pressure Hmzy and plateau pressure Pty.
[0016] Preferably, the data processing module includes a preprocessing unit and an integration unit;
[0017] The preprocessing unit is used to check and identify missing data, duplicate data and data formats in the relevant respiratory state data information, the relevant lung abnormality distribution data information and the relevant expansion degree data information, so as to remove duplicate data, correct data formats and fill in missing values;
[0018] The integration unit is used to standardize the preprocessed data information using dimensionless processing technology to eliminate dimensional differences between different data sources, and to combine multi-sensor data fusion technology to merge data from different data sources into a unified view to construct a postoperative complication data set.
[0019] Preferably, the lung preliminary analysis module includes a lung preliminary analysis unit and a preliminary diagnosis unit;
[0020] The lung preliminary analysis unit is used to construct the patient's postoperative oxygenation factor Yhyz according to the relevant respiratory state data information, which is specifically obtained in the following manner: ; In the formula, Dfy represents the partial pressure of oxygen in arterial blood, and Xynd represents the concentration of inspired oxygen.
[0021] Preferably, the preliminary diagnosis unit is used to analyze and obtain the patient's oxygenation factor Yhyz before and during surgery according to the method of obtaining the patient's postoperative oxygenation factor Yhyz in the lung preliminary analysis unit, and obtain the patient's overall average oxygenation factor Yhyz through a statistical algorithm. The oxygenation factor Yhyz of the patient after surgery is compared with the average oxygenation factor of the patient as a whole. A comparison is performed to preliminarily determine whether the patient's lung oxygenation capacity is impaired. The specific determination contents are as follows:
[0022] If the patient's oxygenation factor Yhyz does not exceed the patient's overall average oxygenation factor When the patient's lung oxygenation capacity is initially judged to be impaired, a complication prediction instruction will be issued;
[0023] If the patient's oxygenation factor Yhyz after surgery exceeds the patient's overall average oxygenation factor At this time, it was preliminarily determined that the patient's lung oxygenation capacity was not impaired.
[0024] Preferably, the comprehensive analysis module includes a local analysis unit, an expansion degree analysis unit and a comprehensive prediction unit;
[0025] The local analysis unit is used to calculate the alveolar ventilation Ftqz of each lung segment according to the relevant lung abnormality distribution data information after receiving the complication prediction instruction, which is specifically obtained in the following manner: ; In the formula, Mzz represents the total ventilation per minute; Wtz represents the dead space ventilation; the alveolar ventilation Ftqz of each lung segment is counted and the average alveolar ventilation is calculated .
[0026] Preferably, the expansion degree analysis unit is used to construct a lung expansion capacity coefficient Fkxs according to relevant expansion degree data information, which is specifically obtained in the following manner: ; In the formula, Cqz represents tidal volume, Pty represents plateau pressure, and Hmzy represents positive end-expiratory pressure.
[0027] Preferably, the comprehensive prediction unit is used to construct a complication prediction model using convolutional neural network technology, and the patient's postoperative oxygenation factor Yhyz, lung expansion capacity coefficient Fkxs and mean alveolar ventilation Input into the complication prediction model to fit and obtain the risk assessment index Fpzs, which is obtained by the following formula: ; In the formula, C represents the correction constant, and are weight values, among which, , ,and .
[0028] Preferably, the prediction management module includes a comparison unit;
[0029] The comparison unit is used to pre-set the evaluation threshold W, and compare and analyze the evaluation threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient. The specific contents are as follows:
[0030] If the risk assessment index Fpzs> assessment threshold W, it means that the patient has no risk of complications after surgery. At this time, the originally planned postoperative care measures will continue to be implemented, and regular monitoring every 5 hours will be maintained. On the premise of ensuring that the patient's condition is stable, the patient's mechanical ventilation support will be gradually reduced, and restorative treatment and rehabilitation training will be started. At the same time, the patient will be reviewed regularly;
[0031] If the risk assessment index Fpzs = assessment threshold W, it means that the patient is at risk of complications after surgery. At this time, the frequency of regular monitoring will be adjusted to once every 3 hours, and the treatment plan will be fine-tuned based on the real-time monitoring results. The specific fine-tuning equipment includes the ventilator;
[0032] If the risk assessment index Fpzs is less than the assessment threshold W, it means that the patient is at risk of complications after surgery. At this time, emergency medical intervention measures will be taken immediately. Specific emergency medical intervention measures include: adjusting mechanical ventilation parameters, increasing the intensity of oxygen therapy and using vasoactive drugs, and adjusting the frequency of regular monitoring to once every 2 hours.
[0033] A method for predicting postoperative complications of cardiothoracic surgery comprises the following steps:
[0034] Step 1: Perform a physical examination on the patient who has undergone cardiothoracic surgery, and obtain the patient's relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information through the report obtained from the physical examination;
[0035] Step 2: Preprocess the relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information obtained in the postoperative data monitoring module, and construct a postoperative complication data set after dimensionless technology processing;
[0036] Step 3: Based on the relevant respiratory status data information, construct the patient's postoperative oxygenation factor Yhyz, and based on the value of the oxygenation factor Yhyz, preliminarily determine whether the patient's current lung oxygenation capacity is impaired. If impaired, issue a complication prediction instruction;
[0037] Step 4: After receiving the complication prediction instruction, the alveolar ventilation Ftqz and lung expansion capacity coefficient Fkxs of each lung segment are calculated according to the relevant lung abnormality distribution data information and the relevant expansion degree data information, and the risk assessment index Fpzs is obtained by fitting in combination with the trained complication prediction model;
[0038] Step 5: pre-set the assessment threshold W, and compare and analyze the assessment threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient.
[0039] The present invention provides a method and system for predicting complications after cardiothoracic surgery, which has the following beneficial effects:
[0040] The system can further realize multi-dimensional monitoring and accurate prediction of postoperative patients of cardiothoracic surgery by integrating postoperative data monitoring module, data processing module, lung preliminary analysis module, comprehensive analysis module and prediction management module. The postoperative data monitoring module collects data information related to respiratory status, abnormal lung distribution and lung expansion degree through comprehensive physical examination of patients, providing basic support for subsequent data processing and analysis. Through the lung preliminary analysis module, the system can construct the postoperative oxygenation factor Yhyz and preliminarily judge whether the patient's lung oxygenation capacity is impaired, ensuring timely issuance of complication prediction instructions, further enhancing the real-time response capability of the system. The comprehensive analysis module calculates the alveolar ventilation Ftqz and lung expansion capacity coefficient Fkxs of each lung segment in detail, combines the trained complication prediction model, and fits the risk assessment index Fpzs, further realizing high-precision prediction of postoperative complications. Finally, the prediction management module can comprehensively evaluate and predict the risk level of postoperative complications of patients by setting the evaluation threshold W, thereby providing strong support for clinical decision-making and improving the prevention and management level of postoperative complications as much as possible. In short, the application of this system can effectively reduce the probability of postoperative complications and improve the overall treatment effect of patients.
[0041] The system further realizes the accurate assessment of the patient's postoperative lung oxygenation capacity through the preliminary lung analysis module. The preliminary lung analysis unit constructs the postoperative oxygenation factor Yhyz by analyzing the relevant respiratory status data information to reflect the patient's lung oxygenation capacity in the postoperative stage. The preliminary diagnosis unit further uses statistical algorithms to analyze the patient's preoperative and intraoperative oxygenation factors to obtain the overall average oxygenation factor. By comparing the patient's postoperative oxygenation factor Yhyz with the patient's overall average oxygenation factor, the system can preliminarily determine whether the patient's lung oxygenation capacity is impaired. Through dynamic analysis and comparison of oxygenation factors, it can effectively warn of possible complications and provide data support for subsequent treatment and management, thereby optimizing the patient's postoperative care and management and improving the quality of postoperative recovery.
[0042] By analyzing the relevant abnormal lung distribution data information, the alveolar ventilation Ftqz of each lung segment can be calculated. Specifically, the calculation method of alveolar ventilation includes deducting the dead space ventilation Wtz from the total ventilation per minute Mzz to obtain the effective ventilation actually involved in gas exchange. This calculation can describe the effective ventilation of each lung segment in detail, and obtain the average alveolar ventilation by counting the alveolar ventilation of each lung segment. This detailed segment-level analysis allows doctors to accurately identify the uneven distribution of lung ventilation and provide reliable data support for further clinical decision-making. Combining the analysis results of these two units, the comprehensive prediction unit can integrate the data of alveolar ventilation and lung expansion capacity coefficient to provide a more comprehensive and accurate risk assessment of postoperative complications. The implementation of this module not only improves the ability to evaluate the postoperative lung function status, but also helps to timely discover potential complications and optimize the postoperative management strategy of patients. Through accurate data analysis and prediction, the pertinence and effectiveness of clinical intervention can be further improved, and the postoperative prognosis and quality of life of patients can be improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 This is a block diagram of a cardiothoracic surgery postoperative complication prediction system of the present invention;
[0044] Figure 2 The present invention is a flowchart of a method for predicting complications after cardiothoracic surgery. DETAILED DESCRIPTION
[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0046] Example 1: Please refer to Figure 1 , the present invention provides a cardiothoracic surgery postoperative complication prediction system, including a postoperative data monitoring module, a data processing module, a lung preliminary analysis module, a comprehensive analysis module and a prediction management module;
[0047] The postoperative data monitoring module is used to perform physical examinations on patients who have undergone cardiothoracic surgery, and obtain relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information of the patients through reports obtained from the physical examinations;
[0048] The data processing module is used to pre-process the relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information obtained in the postoperative data monitoring module, and construct a postoperative complication data set after dimensionless technology processing;
[0049] The lung preliminary analysis module is used to construct the patient's postoperative oxygenation factor Yhyz based on the relevant respiratory status data information, and preliminarily judge whether the current patient's lung oxygenation capacity is impaired according to the value of the oxygenation factor Yhyz. If impaired, a complication prediction instruction is issued externally;
[0050] The comprehensive analysis module is used to calculate the alveolar ventilation Ftqz and the lung expansion capacity coefficient Fkxs of each lung segment according to the relevant lung abnormality distribution data information and the relevant expansion degree data information after receiving the complication prediction instruction, and to fit and obtain the risk assessment index Fpzs in combination with the trained complication prediction model;
[0051] The prediction management module is used to pre-set an evaluation threshold W, and compare and analyze the evaluation threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient.
[0052] During the operation of this system, through the synergy of multiple functional modules, the risk of complications after cardiothoracic surgery is further accurately predicted. First, the postoperative data monitoring module can obtain key data such as the patient's postoperative respiratory status, abnormal lung distribution and expansion degree in many aspects and accurately, providing sufficient data support for subsequent risk assessment. Through the preprocessing and dimensionless technical processing of the data processing module, these data are standardized and structured, and a postoperative complication data set is constructed, making subsequent analysis more efficient and reliable. Furthermore, the lung preliminary analysis module constructs the oxygenation factor Yhyz based on the patient's respiratory status data information, which can quickly and preliminarily determine the damage to the patient's lung oxygenation capacity. When potential risks are found, timely issuing complication prediction instructions is helpful for early intervention and preventing further development of complications; the comprehensive analysis module can further accurately fit the risk assessment index Fpzs by calculating the alveolar ventilation Ftqz and the lung expansion capacity coefficient Fkxs, and combining the trained prediction model, providing accurate complication risk quantification results. Finally, the predictive management module can effectively predict the risk level of postoperative complications in patients by comparing and analyzing the risk assessment index Fpzs with the preset assessment threshold W. This systematic prediction method can provide important decision-making basis in postoperative management, effectively reduce the probability of complications, and thus improve the quality of postoperative recovery of patients.
[0053] Example 2: Please refer to Figure 1,Specifically: the postoperative data monitoring module includes a physical examination unit and a collection unit;
[0054] The physical examination unit is used to examine the postoperative status of patients who have undergone cardiothoracic surgery through several sets of monitoring equipment, wherein the several sets of monitoring equipment include an arterial blood gas analyzer, a ventilator, a pulmonary function tester and a pressure sensor;
[0055] The acquisition unit is used to collect and record relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information of the patient according to the patient reports obtained by several groups of monitoring devices, wherein the relevant respiratory status data information includes the arterial blood oxygen partial pressure Dfy and the inspired oxygen concentration Xynd of the patient before, during and after the operation;
[0056] The relevant lung abnormality distribution data information includes the total ventilation volume Mzz and dead space ventilation Wtz per minute in each lung segment of the patient's lungs after surgery;
[0057] Relevant expansion degree data information includes tidal volume Cqz, positive end-expiratory pressure Hmzy and plateau pressure Pty.
[0058] The data processing module includes a preprocessing unit and an integration unit;
[0059] The preprocessing unit is used to check and identify missing data, duplicate data and data formats in the relevant respiratory state data information, relevant lung abnormality distribution data information and relevant expansion degree data information, so as to remove duplicate data, correct data formats and fill missing values, wherein the missing value filling methods include mean filling, median filling, interpolation filling and regression filling; converting data into a format suitable for further analysis, for example, data normalization, data encoding (such as converting categorical data into numerical data).
[0060] The integration unit is used to utilize dimensionless processing technology to standardize the preprocessed data information to eliminate dimensional differences between different data sources, and to combine multi-sensor data fusion technology to merge data from different data sources into a unified view to construct a postoperative complication data set, for example, to fuse data from different sensors together to provide more comprehensive data information.
[0061] In this embodiment, the system further realizes comprehensive and accurate monitoring of the patient's postoperative status through the design of the physical examination unit and the acquisition unit. The physical examination unit integrates several groups of monitoring equipment to ensure the dynamic tracking and inspection of the patient's preoperative, intraoperative and postoperative status, and can monitor and obtain multi-dimensional key data information in real time, which provides a solid data basis for predicting complications. In the data processing module, the preprocessing unit further ensures the integrity and accuracy of the data through a series of strict data cleaning and preprocessing steps, especially by filling missing values and data format standardization, further improving the quality of the data. The integration unit uses dimensionless processing technology and multi-sensor data fusion technology to standardize and integrate data from different data sources, eliminate dimensional differences, and construct a unified postoperative complication data set. This data set provides high-quality input for subsequent risk assessment and complication prediction, and improves the accuracy and reliability of the prediction model as much as possible. In short, the system can further effectively improve the prediction accuracy of postoperative complications, so as to help the medical team take corresponding preventive measures in time, reduce the incidence of postoperative complications, and thus improve the postoperative rehabilitation effect and overall treatment level of patients.
[0062] Example 3: Please refer to Figure 1 ,Specifically: the lung preliminary analysis module includes a lung preliminary analysis unit and a preliminary diagnosis unit;
[0063] The lung preliminary analysis unit is used to construct the patient's postoperative oxygenation factor Yhyz according to the relevant respiratory state data information, which is specifically obtained in the following manner: ; In the formula, Dfy represents the partial pressure of oxygen in arterial blood, and Xynd represents the concentration of inspired oxygen.
[0064] The arterial oxygen partial pressure Dfy mentioned above can be monitored and obtained by an arterial blood gas analyzer (ABG analyzer), specifically by drawing blood from the artery and sending the blood sample to the arterial blood gas analyzer for measurement;
[0065] Inhaled oxygen concentration Xynd can measure the oxygen concentration of the inhaled gas in real time through the ventilator's built-in oxygen concentration sensor.
[0066] The preliminary diagnosis unit is used to analyze and obtain the patient's oxygenation factor Yhyz before and during surgery according to the method of obtaining the patient's postoperative oxygenation factor Yhyz in the lung preliminary analysis unit, and obtain the patient's overall average oxygenation factor Yhyz through a statistical algorithm. The oxygenation factor Yhyz of the patient after surgery is compared with the average oxygenation factor of the patient as a whole. A comparison is performed to preliminarily determine whether the patient's lung oxygenation capacity is impaired. The specific determination contents are as follows:
[0067] If the patient's oxygenation factor Yhyz does not exceed the patient's overall average oxygenation factor When the patient's lung oxygenation capacity is initially judged to be impaired, a complication prediction instruction will be issued;
[0068] If the patient's oxygenation factor Yhyz after surgery exceeds the patient's overall average oxygenation factor At this time, it was preliminarily determined that the patient's lung oxygenation capacity was not impaired.
[0069] In this embodiment, the lung preliminary analysis unit is responsible for constructing the patient's postoperative oxygenation factor Yhyz based on the relevant respiratory status data information. Specifically, this method can accurately evaluate the patient's postoperative oxygenation status, further ensuring the scientificity and effectiveness of the monitoring results. The preliminary diagnosis unit obtains the patient's overall average oxygenation factor by statistically analyzing the preoperative and intraoperative oxygenation factors. Then, the patient's postoperative oxygenation factor Yhyz was compared with the overall average oxygenation factor A comparison is made to preliminarily determine the patient's lung oxygenation capacity. If the postoperative oxygenation factor Yhyz is lower than the overall average level, this indicates that the patient's lung oxygenation capacity may be impaired, and the system will issue a complication prediction instruction so that timely intervention measures can be taken. On the contrary, if the postoperative oxygenation factor Yhyz is higher than or equal to the overall average level, it is preliminarily judged that the lung oxygenation capacity is normal and no additional intervention is required. Through this systematic analysis process, possible oxygenation capacity problems after surgery can be quickly and accurately identified, thereby improving the timeliness and accuracy of complication predictions and ensuring the patient's safety and rehabilitation quality after surgery. This method can not only detect potential complication risks in advance, but also provide clinicians with valuable decision-making basis, which helps to optimize treatment plans and improve patient prognosis.
[0070] Example 4: Please refer to Figure 1 ,Specifically: the comprehensive analysis module includes a local analysis unit, an expansion degree analysis unit and a comprehensive prediction unit;
[0071] The local analysis unit is used to calculate the alveolar ventilation Ftqz of each lung segment according to the relevant lung abnormality distribution data information after receiving the complication prediction instruction, which is specifically obtained in the following manner: ; In the formula, Mzz represents the total ventilation per minute; Wtz represents the dead space ventilation, which refers to the airway ventilation that does not participate in gas exchange;
[0072] The total ventilation volume per minute Mzz mentioned above can be recorded in real time by a ventilator or calculated by a pulmonary function tester. The total ventilation volume per minute Mzz is usually the product of tidal volume multiplied by respiratory rate;
[0073] Dead space ventilation Wtz can be obtained by measuring the amount of anatomical dead space (for example, by measuring the Bohr dead space formula) and functional dead space. The anatomical dead space can be measured by nitrogen washout technology (such as the Fowler method), while the functional dead space needs to be obtained through further evaluation of body exchange efficiency.
[0074] Among them, anatomical dead space refers to the part of the airway between the nasal cavity, oral cavity and terminal bronchioles. Although these areas participate in the entry and exit of air (ventilation), they do not have the ability of gas exchange, which means that the inhaled oxygen will not enter the blood in these areas, and carbon dioxide will not be discharged from the blood.
[0075] The alveolar ventilation Ftqz of each lung segment refers to the effective ventilation entering the alveoli per minute;
[0076] The alveolar ventilation Ftqz of each lung segment is counted and the average alveolar ventilation is calculated. .
[0077] The expansion degree analysis unit is used to construct the lung expansion capacity coefficient Fkxs according to the relevant expansion degree data information, which is specifically obtained in the following manner: ; In the formula, Cqz represents tidal volume, Pty represents plateau pressure, and Hmzy represents positive end-expiratory pressure.
[0078] The tidal volume Cqz mentioned above can be directly measured by a ventilator or a pulmonary function tester for each breath;
[0079] Plateau pressure Pty refers to the airway pressure measured after a brief pause in airflow during mechanical ventilation. The ventilator can automatically record this value at the end of inspiration pause to assess the elastic compliance of the lungs.
[0080] The positive end-expiratory pressure Hmzy can be monitored and displayed in real time through a pressure sensor.
[0081] In this embodiment, the system further realizes an in-depth evaluation of the postoperative patient's lung status through a comprehensive analysis module. The comprehensive analysis module includes a local analysis unit, an expansion degree analysis unit and a comprehensive prediction unit, which can analyze the postoperative patient's lung status in many aspects and accurately, thereby providing strong support for clinical decision-making. After receiving the complication prediction instruction, the local analysis unit calculates the alveolar ventilation Ftqz of each lung segment based on the relevant lung abnormality distribution data information, and obtains the effective alveolar ventilation of each lung segment by deducting the dead space ventilation from the total ventilation volume, and further calculates the average alveolar ventilation. This method provides an accurate quantification of the alveolar ventilation effect, and can identify the specific lung segment ventilation deficiency, thereby providing a scientific basis for the evaluation and improvement of local lung function. The expansion degree analysis unit constructs the lung expansion capacity coefficient Fkxs through the relevant expansion degree data information, thereby evaluating the expansion capacity of the lungs during the breathing process. This analysis can provide important information about the compliance and functional status of the lungs, which helps to understand the efficiency and degree of lung expansion of the patient. By combining the analysis results of local alveolar ventilation and lung expansion capacity coefficient, the comprehensive analysis module can evaluate the patient's lung status in multiple dimensions, which not only improves the accuracy of predicting postoperative complications, but also provides detailed data support for further intervention and treatment. Through accurate evaluation of alveolar ventilation and expansion capacity, the system can better guide clinical management, optimize the postoperative recovery process, and improve the overall health of patients.
[0082] Example 5: Please refer to Figure 1 Specifically: the comprehensive prediction unit is used to construct a complication prediction model using convolutional neural network technology, and the patient's postoperative oxygenation factor Yhyz, lung expansion capacity coefficient Fkxs and mean alveolar ventilation Input into the complication prediction model to fit and obtain the risk assessment index Fpzs, which is obtained by the following formula: ; In the formula, C represents the correction constant, and are weight values, among which, , ,and .
[0083] Convolutional neural network technology is used to build an initial model, and the initial model is trained and tested with a complication data set. The trained initial model is used as a state recognition model, and feature information in the state recognition model is obtained. The state recognition model is trained and tested with the obtained feature information. Combined with complication prediction instructions, the trained state recognition model is used as a complication prediction model.
[0084] The prediction management module includes a comparison unit;
[0085] The comparison unit is used to pre-set the evaluation threshold W, and compare and analyze the evaluation threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient. The specific contents are as follows:
[0086] If the risk assessment index Fpzs> assessment threshold W, it means that the patient has no risk of complications after surgery. At this time, the originally planned postoperative care measures will continue to be implemented, and regular monitoring every 5 hours will be maintained. On the premise of ensuring that the patient's condition is stable, the patient's mechanical ventilation support will be gradually reduced, and restorative treatment and rehabilitation training will be started. At the same time, the patient will be reviewed regularly to ensure that the risk assessment index continues to be higher than the threshold;
[0087] If the risk assessment index Fpzs = assessment threshold W, it means that the patient is at risk of complications after surgery. At this time, the frequency of regular monitoring will be adjusted to once every 3 hours to timely monitor changes in the patient's condition, especially the dynamic changes in key indicators such as lung function and oxygenation level. The treatment plan will be fine-tuned based on the real-time monitoring results. Specific fine-tuning equipment includes ventilators;
[0088] If the risk assessment index Fpzs is less than the assessment threshold W, it means that the patient is at risk of complications after surgery. At this time, emergency medical intervention measures will be taken immediately. Specific emergency medical intervention measures include: adjusting mechanical ventilation parameters, increasing the intensity of oxygen therapy and using vasoactive drugs, and adjusting the frequency of regular monitoring to once every 2 hours.
[0089] In this embodiment, the comprehensive prediction unit and prediction management module in the system further improve the prediction accuracy of postoperative complications and the personalized care level of patients through convolutional neural network technology and detailed risk assessment mechanism. The complication prediction model constructed by the comprehensive prediction unit using convolutional neural network technology can comprehensively consider the patient's postoperative oxygenation factor Yhyz, lung expansion capacity coefficient Fkxs and mean alveolar ventilation, and calculate an accurate risk assessment index Fpzs. This deep learning model performs weighted processing on different physiological indicators through dynamically adjusted weights and correction constants C, thereby providing a multi-dimensional and accurate postoperative complication risk assessment. This data-driven prediction method can effectively identify potential risk points and provide clinicians with a reliable decision-making basis. The prediction management module compares and analyzes the risk assessment index Fpzs with the preset assessment threshold W through the comparison unit, and can take corresponding nursing measures at different risk levels. Specifically: If the risk assessment index Fpzs> assessment threshold W, it means that the patient has a low risk of postoperative complications. In this case, the system recommends continuing the original postoperative care measures, maintaining regular monitoring every 5 hours, gradually reducing mechanical ventilation support, and starting restorative treatment and rehabilitation training, while regularly reviewing. This approach can effectively optimize the patient's recovery process and promote the rehabilitation process while ensuring that the condition is stable. If the risk assessment index Fpzs=assessment threshold W, it means that the patient has a certain risk of complications after surgery. The system will adjust the frequency of regular monitoring to once every 3 hours to ensure timely tracking of changes in the patient's condition, especially key indicators such as lung function and oxygenation level, and fine-tune the treatment plan. This measure can respond to potential problems in a timely manner and reduce the probability of complications. If the risk assessment index Fpzs<assessment threshold W, it means that the patient has a high risk of postoperative complications. At this time, the system will immediately take emergency medical intervention measures, including adjusting mechanical ventilation parameters, increasing oxygen therapy intensity, and using vasoactive drugs, while increasing the monitoring frequency to once every 2 hours. This emergency response mechanism can quickly respond to acute complications and ensure the safety of patients. Overall, the implementation of this system has improved the efficiency of predicting and managing postoperative complications. Through accurate risk assessment and timely adjustment measures, it has further improved the quality and safety of postoperative care for patients and effectively reduced the possibility of complications.
[0090] Example 6: Please refer to Figure 2 Specifically: A method for predicting complications after cardiothoracic surgery, comprising the following steps:
[0091] Step 1: Perform a physical examination on the patient who has undergone cardiothoracic surgery, and obtain the patient's relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information through the report obtained from the physical examination;
[0092] Step 2: Preprocess the relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information obtained in the postoperative data monitoring module, and construct a postoperative complication data set after dimensionless technology processing;
[0093] Step 3: Based on the relevant respiratory status data information, construct the patient's postoperative oxygenation factor Yhyz, and based on the value of the oxygenation factor Yhyz, preliminarily determine whether the patient's current lung oxygenation capacity is impaired. If impaired, issue a complication prediction instruction;
[0094] Step 4: After receiving the complication prediction instruction, the alveolar ventilation Ftqz and lung expansion capacity coefficient Fkxs of each lung segment are calculated according to the relevant lung abnormality distribution data information and the relevant expansion degree data information, and the risk assessment index Fpzs is obtained by fitting in combination with the trained complication prediction model;
[0095] Step 5: pre-set the assessment threshold W, and compare and analyze the assessment threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient.
[0096] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A cardiothoracic surgery postoperative complication prediction system, It is characterized in that It includes postoperative data monitoring module, data processing module, lung preliminary analysis module, comprehensive analysis module and prediction management module; The postoperative data monitoring module is used to perform physical examinations on patients who have undergone cardiothoracic surgery, and obtain relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information of the patients through reports obtained from the physical examinations; The data processing module is used to pre-process the relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information obtained in the postoperative data monitoring module, and construct a postoperative complication data set after dimensionless technology processing; The lung preliminary analysis module is used to construct the patient's postoperative oxygenation factor Yhyz based on the relevant respiratory status data information, and preliminarily judge whether the current patient's lung oxygenation capacity is impaired according to the value of the oxygenation factor Yhyz. If impaired, a complication prediction instruction is issued externally; The comprehensive analysis module is used to calculate the alveolar ventilation Ftqz and the lung expansion capacity coefficient Fkxs of each lung segment according to the relevant lung abnormality distribution data information and the relevant expansion degree data information after receiving the complication prediction instruction, and to fit and obtain the risk assessment index Fpzs in combination with the trained complication prediction model; The comprehensive analysis module includes a local analysis unit, an expansion degree analysis unit and a comprehensive prediction unit; The local analysis unit is used to calculate the alveolar ventilation Ftqz of each lung segment according to the relevant lung abnormality distribution data information after receiving the complication prediction instruction, which is specifically obtained in the following manner: ; Where, Mzz represents the total ventilation per minute; Wtz represents the dead space ventilation; The alveolar ventilation Ftqz of each lung segment is counted and the average alveolar ventilation is calculated. ; The comprehensive prediction unit is used to construct a complication prediction model using convolutional neural network technology, and the patient's postoperative oxygenation factor Yhyz, lung expansion capacity coefficient Fkxs and mean alveolar ventilation Input into the complication prediction model to fit and obtain the risk assessment index Fpzs, which is obtained by the following formula: ; In the formula, C represents the correction constant, and All are weight values; The prediction management module is used to pre-set an evaluation threshold W, and compare and analyze the evaluation threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient.
2. A cardiothoracic surgery postoperative complication prediction system according to claim 1, It is characterized in that The postoperative data monitoring module includes a physical examination unit and a collection unit; The physical examination unit is used to examine the postoperative status of patients who have undergone cardiothoracic surgery through several sets of monitoring equipment, wherein the several sets of monitoring equipment include an arterial blood gas analyzer, a ventilator, a pulmonary function tester and a pressure sensor; The acquisition unit is used to collect and record relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information of the patient according to the patient reports obtained by several groups of monitoring devices, wherein the relevant respiratory status data information includes the arterial blood oxygen partial pressure Dfy and the inspired oxygen concentration Xynd of the patient before, during and after the operation; The relevant lung abnormality distribution data information includes the total ventilation volume Mzz and dead space ventilation Wtz per minute in each lung segment of the patient's lungs after surgery; Relevant expansion degree data information includes tidal volume Cqz, positive end-expiratory pressure Hmzy and plateau pressure Pty.
3. A cardiothoracic surgery postoperative complication prediction system according to claim 1, It is characterized in that The data processing module includes a preprocessing unit and an integration unit; The preprocessing unit is used to check and identify missing data, duplicate data and data formats in the relevant respiratory state data information, the relevant lung abnormality distribution data information and the relevant expansion degree data information, so as to remove duplicate data, correct data formats and fill in missing values; The integration unit is used to standardize the preprocessed data information using dimensionless processing technology to eliminate dimensional differences between different data sources, and to combine multi-sensor data fusion technology to merge data from different data sources into a unified view to construct a postoperative complication data set.
4. A cardiothoracic surgery postoperative complication prediction system according to claim 2, It is characterized in that The lung preliminary analysis module includes a lung preliminary analysis unit and a preliminary diagnosis unit; The lung preliminary analysis unit is used to construct the patient's postoperative oxygenation factor Yhyz according to the relevant respiratory state data information, which is specifically obtained in the following manner: ; Where Dfy represents the partial pressure of oxygen in arterial blood, and Xynd represents the concentration of inspired oxygen.
5. A cardiothoracic surgery postoperative complication prediction system according to claim 4, It is characterized in that The preliminary diagnosis unit is used to analyze and obtain the patient's oxygenation factor Yhyz before and during surgery according to the method of obtaining the patient's postoperative oxygenation factor Yhyz in the lung preliminary analysis unit, and obtain the patient's overall average oxygenation factor Yhyz through a statistical algorithm. The oxygenation factor Yhyz of the patient after surgery is compared with the average oxygenation factor of the patient as a whole. A comparison is performed to preliminarily determine whether the patient's lung oxygenation capacity is impaired. The specific determination contents are as follows: If the patient's oxygenation factor Yhyz does not exceed the patient's overall average oxygenation factor When the patient's lung oxygenation capacity is initially judged to be impaired, a complication prediction instruction will be issued; If the patient's oxygenation factor Yhyz after surgery exceeds the patient's overall average oxygenation factor At this time, it was preliminarily determined that the patient's lung oxygenation capacity was not impaired.
6. A cardiothoracic surgery postoperative complication prediction system according to claim 5, It is characterized in that The expansion degree analysis unit is used to construct the lung expansion capacity coefficient Fkxs according to the relevant expansion degree data information, which is specifically obtained in the following manner: ; Where Cqz represents tidal volume, Pty represents plateau pressure, and Hmzy represents positive end-expiratory pressure.
7. A cardiothoracic surgery postoperative complication prediction system according to claim 1, It is characterized in that The prediction management module includes a comparison unit; The comparison unit is used to pre-set the evaluation threshold W, and compare and analyze the evaluation threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient. The specific contents are as follows: If the risk assessment index Fpzs> assessment threshold W, it means that the patient has no risk of complications after surgery. At this time, the originally planned postoperative care measures will continue to be implemented, and regular monitoring every 5 hours will be maintained. On the premise of ensuring that the patient's condition is stable, the patient's mechanical ventilation support will be gradually reduced, and restorative treatment and rehabilitation training will be started. At the same time, the patient will be reviewed regularly; If the risk assessment index Fpzs = assessment threshold W, it means that the patient is at risk of complications after surgery. At this time, the frequency of regular monitoring will be adjusted to once every 3 hours, and the treatment plan will be fine-tuned based on the real-time monitoring results. The specific fine-tuning equipment includes the ventilator; If the risk assessment index Fpzs is less than the assessment threshold W, it means that the patient is at risk of complications after surgery. At this time, emergency medical intervention measures will be taken immediately. Specific emergency medical intervention measures include: adjusting mechanical ventilation parameters, increasing the intensity of oxygen therapy and using vasoactive drugs, and adjusting the frequency of regular monitoring to once every 2 hours.
8. A method for predicting postoperative complications of cardiothoracic surgery, used to implement a system for predicting postoperative complications of cardiothoracic surgery as claimed in any one of claims 1 to 7. It is characterized in that The following steps are involved: Step 1: Perform a physical examination on the patient who has undergone cardiothoracic surgery, and obtain the patient's relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information through the report obtained from the physical examination; Step 2: Preprocess the relevant respiratory status data information, relevant lung abnormality distribution data information and relevant expansion degree data information obtained in the postoperative data monitoring module, and construct a postoperative complication data set after dimensionless technology processing; Step 3: Based on the relevant respiratory status data information, construct the patient's postoperative oxygenation factor Yhyz, and based on the value of the oxygenation factor Yhyz, preliminarily determine whether the patient's current lung oxygenation capacity is impaired. If impaired, issue a complication prediction instruction; Step 4: After receiving the complication prediction instruction, the alveolar ventilation Ftqz and lung expansion capacity coefficient Fkxs of each lung segment are calculated according to the relevant lung abnormality distribution data information and the relevant expansion degree data information, and the risk assessment index Fpzs is obtained by fitting in combination with the trained complication prediction model; Step 5: pre-set the assessment threshold W, and compare and analyze the assessment threshold W with the risk assessment index Fpzs to comprehensively predict the risk level of postoperative complications of the current patient.
Citation Information
Cited By
Thoracic surgery patient postoperative remote monitoring method and system based on Internet of Things
CN121096617A
