Adaptive evaluation method and system for ventilation by combining ECMO with prone position

Abstract

The invention provides an adaptation evaluation method and system for ECMO combined prone position ventilation, based on ECMO application big data under prone position ventilation, a prone position adaptation prediction model for predicting the ECMO intervention opportunity under prone position ventilation is constructed and deployed, the model is used for monitoring breathing data of a patient with prone position ventilation in real time, and the ECMO intervention opportunity is evaluated in real time. Whether the prone position ventilation patient triggers an ECMO intervention condition or not is evaluated; if the ECMO intervention condition is triggered, ECMO equipment is activated to intervene in ventilation treatment of the prone position ventilation patient. Therefore, the prone position ventilation effect is evaluated through the model, a reference is provided for determining the ECMO application time so as to optimize the treatment strategy, the evaluation reference of the ECMO application time can be provided for nurses, the situation that the nurses lack prone position ventilation treatment experience and do not open ECMO equipment for ventilation for patients in time is avoided, and the treatment success rate of the patients is increased.

Description

Technical Field

[0001] The present invention relates to the field of clinical medical technology, and in particular to an adaptation assessment method for ECMO combined with prone position ventilation, an adaptation assessment system for ECMO combined with prone position ventilation, an electronic device and a computer-readable storage medium. Background Art

[0002] Extracorporeal life support (ECLS) was originally used for newborns and is now commonly used for pediatric and adult patients who require lung and cardiac support. ECLS systems provide hemodynamic stability and gas exchange. While it does not cure the disease, it can serve as a bridge to organ recovery or definitive treatment.

[0003] Prone ventilation is a method of improving ventilation function by changing the patient's body position. However, this type of prone ventilation often leads to some clinical indications, including:

[0004] 1. Severe acute respiratory distress syndrome (ARDS);

[0005] 2. Oxygenation index (PaO2 / FiO2) is less than 200 mmHg;

[0006] 3. After active respiratory support, the oxygenation index still does not improve;

[0007] 4. Restrictions on ventilator conditions;

[0008] 5. The patient is conscious and able to cooperate with prone ventilation;

[0009] 6. The patient has no contraindications to prone ventilation.

[0010] Therefore, when prone ventilation indications appear, ECLS care is required, and severe patients will be treated with extracorporeal membrane oxygenation (ECMO) equipment. However, the use of ECMO must meet certain conditions. For example, for patients with respiratory failure, the most common indicator is the oxygenation index (OI). If the OI is greater than 40, it means an 80% mortality rate. Newborns with an OI greater than 25% should consider using ECMO.

[0011] When prone ventilation indications occur, nurses and other personnel usually cannot know in time whether the patient's relevant indicator data meets the application conditions of ECMO, such as the oxygenation index. Unless critically ill patients such as ICUs are receiving intensive care through respiratory equipment, their clinical respiratory parameters can be monitored in real time, allowing nurses to intervene in the application of ECMO equipment in a timely manner. For most patients who adopt prone ventilation, nurses cannot directly use ECMO equipment. They need to undergo a series of evaluations before they can implement ECMO treatment under prone ventilation.

[0012] Because most patients who undergo prone ventilation do not show severe respiratory problems clinically, nurses are unable to judge whether the patients are suitable for supine ventilation and whether ECMO intervention is needed in a timely manner based on their clinical respiratory performance. Summary of the invention

[0013] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0014] On the one hand, a method for evaluating the adaptability of ECMO combined with prone position ventilation is provided, which is implemented by an electronic device and includes:

[0015] S1. Based on the big data of ECMO application under prone ventilation, a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation was constructed and deployed;

[0016] S2. Real-time monitoring of respiratory data of the patient in prone position ventilation, and analysis of the respiratory data to obtain real-time oxygenation index data;

[0017] S3, importing the oxygenation index data into the prone position adaptation prediction model, and using the prone position adaptation prediction model to evaluate whether the prone position ventilation patient triggers ECMO intervention conditions based on the oxygenation index data:

[0018] If an evaluation result that triggers ECMO intervention conditions is received, an activation command is issued to the ECMO device; otherwise, monitoring continues;

[0019] S4. The ECMO device responds to the activation instruction and intervenes in the ventilation treatment of the patient undergoing prone ventilation.

[0020] Optionally, in step S1, the ECMO application big data based on prone ventilation includes the occurrence of prone ventilation indications, oxygenation index data when ECMO is intervened, and the corresponding ECMO application time, wherein:

[0021] The oxygenation index data include arterial blood gas analysis data, blood oxygen saturation data and pulmonary artery compliance index data when prone ventilation indications occur, wherein:

[0022] The pulmonary artery compliance index data include: lung capacity data of the pulmonary artery under static compliance and / or dynamic compliance during prone position ventilation;

[0023] The blood oxygen saturation data includes: blood oxygen saturation value under prone position ventilation;

[0024] The arterial blood gas analysis data include: arterial oxygen partial pressure (PaO2), arterial carbon dioxide partial pressure (PaCO2), pH, actual bicarbonate (AB) and standard bicarbonate (SB) and excess base (BE) under prone ventilation.

[0025] Optionally, S1. Based on the big data of ECMO application under prone ventilation, a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation is constructed and deployed, including:

[0026] Retrieve the ECMO application big data of several patients with prone ventilation from the background database;

[0027] Preprocessing the ECMO application big data;

[0028] Performing feature extraction on the preprocessed ECMO application big data to extract oxygenation index data characteristic values ​​when prone ventilation indications occur and ECMO intervention is required, as well as the corresponding ECMO application time, wherein the oxygenation index data characteristic values ​​include arterial blood gas analysis data characteristic values, blood oxygen saturation data characteristic values, and pulmonary artery compliance index data characteristic values;

[0029] Based on the characteristic values ​​of several oxygenation index data extracted by feature engineering and the corresponding ECMO application time, the corresponding feature set is constructed;

[0030] Dividing the feature set into a training set and a validation set according to a preset ratio;

[0031] Inputting the training set into a preset random forest model, allowing the model to learn the oxygenation index data characteristic values ​​when prone ventilation indications occur and ECMO intervention occurs, as well as the corresponding ECMO application time, so as to train and generate the prone adaptation prediction model;

[0032] The validation set is used to verify whether the prediction performance of the prone position adaptation prediction model for the ECMO intervention timing under prone position ventilation reaches a preset value:

[0033] If so, deploying and applying the prone position adaptation prediction model;

[0034] Otherwise, repeat the above steps to retrain and construct the prone position adaptation prediction model.

[0035] Optionally, S2, real-time monitoring of respiratory data of a patient in prone position ventilation, and parsing the respiratory data to obtain real-time oxygenation index data, including:

[0036] Preset oxygenation index for patients ventilated in prone position;

[0037] Collecting respiratory data of a patient in prone position ventilation in real time, and analyzing the respiratory data to obtain respiratory analysis data;

[0038] Based on a preset oxygenation index of a patient in prone position ventilation, data under the oxygenation index is filtered out from the respiratory analysis data to obtain real-time oxygenation index data;

[0039] The oxygenation index data is written into the HIS (Hospital Information System, HIS) medical record file of the prone ventilation patient in real time.

[0040] Optionally, in step S3, importing the oxygenation index data into the prone position adaptation prediction model includes:

[0041] Monitor the HIS system to determine whether the new oxygenation index data is written into the HIS medical record file of the patient with prone ventilation in the HIS system:

[0042] If yes, read the oxygenation index data written in real time into the HIS medical record file of the patient undergoing prone ventilation, and import the preset prone position adaptation prediction model;

[0043] Otherwise continue to monitor.

[0044] On the other hand, an adaptation assessment system for ECMO combined with prone position ventilation is provided, and the adaptation assessment system for ECMO combined with prone position ventilation is used to implement the adaptation assessment method for ECMO combined with prone position ventilation, and the system includes:

[0045] Respiratory monitoring system, used to monitor the respiratory data of patients undergoing prone ventilation in real time and report to the backend server;

[0046] Model management module, which is used to build and deploy a prone position adaptation prediction model for predicting the timing of ECMO intervention under prone position ventilation based on ECMO application big data under prone position ventilation;

[0047] A data analysis module, used to analyze the respiratory data to obtain real-time oxygenation index data;

[0048] The ECMO ventilation adaptation assessment module is used to import the oxygenation index data into the prone position adaptation prediction model, and the prone position adaptation prediction model assesses whether the prone ventilation patient triggers ECMO intervention conditions according to the oxygenation index data:

[0049] If an evaluation result that triggers ECMO intervention conditions is received, an activation command is issued to the ECMO device; otherwise, monitoring continues;

[0050] A message transmission module, used for transmitting the activation instruction to the ECMO device;

[0051] HIS system, used to record the HIS medical records of patients undergoing prone ventilation;

[0052] A monitoring module is used to monitor the HIS system and determine whether the HIS medical record file of the patient with prone ventilation in the HIS system has new oxygenation index data written into it: if so, read the oxygenation index data written into the HIS medical record file of the patient with prone ventilation in real time, and import the preset prone adaptation prediction model; otherwise, continue to monitor;

[0053] Backend database, used for backend database storage;

[0054] An ECMO device, configured to respond to the activation instruction and intervene in ventilation treatment of the prone ventilation patient;

[0055] The model management module, data analysis module, ECMO ventilation adaptation assessment module, message transmission module, HIS system, monitoring module and background database are all deployed on the background server;

[0056] The respiratory monitoring system and the ECMO device are respectively connected to the delivery background server for communication.

[0057] On the other hand, an electronic device is provided, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned methods for evaluating the adaptation of ECMO combined with prone ventilation is implemented.

[0058] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned ECMO combined with prone position ventilation adaptation assessment methods.

[0059] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0060] Based on the big data of ECMO application under prone ventilation, the present invention constructs and deploys a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation, and uses the model to monitor the respiratory data of patients with prone ventilation in real time, and evaluate whether the patient with prone ventilation triggers the ECMO intervention conditions: if the ECMO intervention conditions are triggered, the ECMO device is activated to intervene in the ventilation treatment of the patient with prone ventilation. Therefore, the present invention evaluates the effect of prone ventilation through the model, provides a reference for determining the application timing of ECMO, optimizes the treatment strategy, and can provide nurses with an evaluation reference for the application timing of ECMO, so as to avoid nurses' lack of experience in prone ventilation treatment and failure to open the ECMO device for ventilation for patients in time, thereby improving the success rate of patient treatment.

[0061] By adopting the present invention, after prone ventilation, if the patient's oxygenation is still not ideal or the improvement is not obvious, prone ventilation is tended to be combined in the early stage of ECMO establishment to improve oxygenation and respiratory system compliance as early as possible, improve the treatment effect, and allow the patient who is undergoing prone ventilation to be treated with ECMO in time, avoiding early dyspnea caused by the indications of prone ventilation, thereby ensuring the safety of the patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0063] Figure 1 It is a flow chart of an adaptation assessment method for ECMO combined with prone position ventilation provided by an embodiment of the present invention;

[0064] Figure 2 It is a structural schematic diagram of an adaptation assessment method for ECMO combined with prone position ventilation provided by an embodiment of the present invention;

[0065] Figure 3 It is a block diagram of an adaptation assessment system for ECMO combined with prone position ventilation provided by an embodiment of the present invention;

[0066] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0068] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0069] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0070] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.

[0071] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0072] Prone ventilation patients: respiratory patients who receive respiratory medical treatment and care based on prone ventilation.

[0073] The embodiment of the present invention provides an adaptability assessment method for ECMO combined with prone position ventilation, which can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart of the adaptation assessment method for ECMO combined with prone position ventilation is shown. The processing flow of the method may include the following steps:

[0074] S1. Based on the big data of ECMO application under prone ventilation, a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation was constructed and deployed;

[0075] S2. Real-time monitoring of respiratory data of the patient in prone position ventilation, and analysis of the respiratory data to obtain real-time oxygenation index data;

[0076] S3, importing the oxygenation index data into the prone position adaptation prediction model, and using the prone position adaptation prediction model to evaluate whether the prone position ventilation patient triggers ECMO intervention conditions based on the oxygenation index data:

[0077] If an evaluation result that triggers ECMO intervention conditions is received, an activation command is issued to the ECMO device; otherwise, monitoring continues;

[0078] S4. The ECMO device responds to the activation instruction and intervenes in the ventilation treatment of the patient undergoing prone ventilation.

[0079] The present invention evaluates the effect of prone ventilation through a model, provides a reference for determining the timing of ECMO application, and optimizes the treatment strategy. It can provide nurses with an evaluation reference for the timing of ECMO application, avoid nurses' lack of experience in prone ventilation treatment and fail to open the ECMO device for ventilation for patients in time, and improve the success rate of patient treatment.

[0080] The following will be combined Figure 3 The application system of the present invention is shown to describe the implementation principle of the present invention.

[0081] The present invention mainly combines the respiratory monitoring system, ECMO equipment and the background server to perform ECMO intervention care and treatment on the prone ventilation patients (referred to as ventilation patients or patients). A corresponding prediction model is deployed on the background server, which can predict and analyze the oxygenation index data of the prone ventilation patients monitored by the respiratory monitoring system, and judge whether the prone ventilation patients trigger the activation conditions of ECMO. If the prone adaptation prediction model on the background monitors the current oxygenation index data of the ventilation patients and triggers the ECMO intervention conditions, the background server generates a corresponding activation instruction according to the prediction result output by the model, and sends it to the ECMO equipment. At this time, the ECMO equipment is started and activated to intervene in the respiratory treatment of the ventilation patients. Therefore, there is no need for nurses to combine the respiratory symptoms or other data of clinical patients to perform manual nursing assessment.

[0082] Through the present invention, the AI ​​model can intelligently analyze and predict whether ECMO treatment and assisted breathing should be performed on patients receiving ventilation in the prone position. This can avoid situations where the patient's ventilation treatment fails due to lack of experience or the need for real-time clinical evaluation by nurses, and can intelligently monitor the application conditions of ECMO intervention.

[0083] like Figure 2 As shown in the figure, the ECMO (extracorporeal membrane oxygenation) device drains the patient's blood outside the body, oxygenates it and removes carbon dioxide before returning it to the body. Its operating principle mainly includes three parts: blood pump, oxygenator, and heat exchanger. The function of the controller is to monitor and adjust the operating parameters of the ECMO system, such as blood flow rate, oxygen partial pressure, carbon dioxide partial pressure, etc., to ensure stable operation of the system and provide appropriate gas exchange. The controller can communicate data with the background server and execute instructions / information issued by the background server. It can be understood in combination with the background control principle of the existing ECMO.

[0084] The construction and application of the AI ​​prediction model of the present invention will be further described below.

[0085] When prone ventilation indications occur, patients will have corresponding symptom data, and administrators, nurses, or monitoring systems (various clinical examination equipment / systems used by hospitals, etc.) can record the oxygenation index data and corresponding ECMO application time during ECMO intervention treatment of patients. Therefore, these historical application big data can be used to train and build corresponding application models for ECMO intervention prediction.

[0086] Interventional monitoring indicators: monitoring of oxygenation indicators such as arterial blood gas analysis, blood oxygen saturation, and pulmonary artery compliance, and evaluating the improvement of prone ventilation on patient oxygenation and respiratory system compliance.

[0087] Clinical judgment: Observe the improvement of the patient's symptoms, such as whether dyspnea is relieved, and any complications during prone ventilation.

[0088] ‌ECMO application timing judgment‌

[0089] ‌Poor oxygenation‌: After prone ventilation, if the patient's oxygenation is still not ideal or has not improved significantly, consider using ECMO in combination‌

[0090] Early combined use: It is preferred to combine prone ventilation with ECMO in the early stage of establishment to improve oxygenation and respiratory system compliance as early as possible and enhance the treatment effect.

[0091] The data collection and recording of the above-mentioned prone ventilation indications, oxygenation index data when ECMO is intervened, and the corresponding ECMO application time can be recorded by the system or the nurse, so the data can be collected from the background database or the work log of the nurse station.

[0092] Optionally, in step S1, the ECMO application big data based on prone ventilation includes the occurrence of prone ventilation indications, oxygenation index data when ECMO is intervened, and the corresponding ECMO application time, wherein:

[0093] The oxygenation index data include arterial blood gas analysis data, blood oxygen saturation data and pulmonary artery compliance index data when prone ventilation indications occur, wherein:

[0094] The pulmonary artery compliance index data include: lung capacity data of the pulmonary artery under static compliance and / or dynamic compliance during prone position ventilation;

[0095] The blood oxygen saturation data includes: blood oxygen saturation value under prone position ventilation;

[0096] The arterial blood gas analysis data include: arterial oxygen partial pressure (PaO2), arterial carbon dioxide partial pressure (PaCO2), pH, actual bicarbonate (AB) and standard bicarbonate (SB) and excess base (BE) under prone ventilation.

[0097] At the beginning of data collection, a large amount of ECMO application case data under prone ventilation was collected, including basic patient information, condition, ventilation duration, ECMO use timing, treatment results, etc. The collected basic patient information includes oxygenation index data when prone ventilation indications occur and the corresponding ECMO application time, oxygenation index data, including arterial blood gas analysis data, blood oxygen saturation data, and pulmonary artery compliance index data when prone ventilation indications occur.

[0098] The blood oxygen saturation value can be measured by a routine blood test or a nail oximeter.

[0099] Definition and significance of the index data of pulmonary artery compliance: Pulmonary compliance refers to the change in lung volume caused by a unit pressure change, reflecting the elasticity of lung tissue. It is an indicator of lung expandability. Good lung compliance means good elasticity, which is conducive to smooth respiratory function.

[0100] Normal value range:

[0101] Static compliance: The static lung compliance of a normal adult is generally 70-100 ml / cmH2O. Some people also believe that the normal value is 200mL / cmH2O or 0.2L / cmH2O, that is, for every increase of 1 cm of water column pressure, the lung capacity increases by 0.2 liters.

[0102] Dynamic compliance: The normal reference value is usually 0.20L / cmH2O. CLdyn20, CLdyn40, CLdyn60, etc. can be used to represent the dynamic compliance at different respiratory rates.

[0103] Arterial blood gas analysis data mainly includes the following key indicators:

[0104] Arterial oxygen partial pressure (PaO2): The normal value is 95-100 mmHg. A value below 60 mmHg may indicate respiratory failure, and a value below 30 mmHg may be life-threatening.

[0105] Arterial carbon dioxide partial pressure (PaCO2): The normal value is 35-45 mmHg. A value above 50 mmHg may indicate respiratory acidosis, and a value below 35 mmHg may indicate respiratory alkalosis.

[0106] pH: The normal value is 7.35-7.45. A pH value below 7.35 is acidosis, and a pH value above 7.45 is alkalosis.

[0107] ‌Actual bicarbonate (AB) and standard bicarbonate (SB): The normal value of AB is 21.4-27.3mmol / L, and the normal value of SB is 21.3-24.8mmol / L. The difference between the two can reflect respiratory or metabolic acid-base poisoning.

[0108] ‌Excess base (BE)‌: The normal value is -3-+3mmol / L. A positive value indicates base excess, and a negative value indicates base deficiency.

[0109] The above data collected during ECMO intervention are used as application data. The nurses will record these data systematically.

[0110] A respiratory monitoring system, such as a ventilator or other equipment / system, can be configured according to the data type, and this embodiment does not limit this. For blood oxygen data, the corresponding blood oximeter is used for detection and written into the case file together with the respiratory parameters. Therefore, the respiratory monitoring system herein includes respiratory monitoring equipment, blood oxygen monitoring equipment, etc., which are generally included in the respiratory monitoring system.

[0111] Optionally, S1. Based on the big data of ECMO application under prone ventilation, a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation is constructed and deployed, including:

[0112] Retrieve the ECMO application big data of several patients with prone ventilation from the background database;

[0113] Preprocessing the ECMO application big data;

[0114] Performing feature extraction on the preprocessed ECMO application big data to extract oxygenation index data characteristic values ​​when prone ventilation indications occur and ECMO intervention is required, as well as the corresponding ECMO application time, wherein the oxygenation index data characteristic values ​​include arterial blood gas analysis data characteristic values, blood oxygen saturation data characteristic values, and pulmonary artery compliance index data characteristic values;

[0115] Based on the characteristic values ​​of several oxygenation index data extracted by feature engineering and the corresponding ECMO application time, the corresponding feature set is constructed;

[0116] Dividing the feature set into a training set and a validation set according to a preset ratio;

[0117] Inputting the training set into a preset random forest model, allowing the model to learn the oxygenation index data characteristic values ​​when prone ventilation indications occur and ECMO intervention occurs, as well as the corresponding ECMO application time, so as to train and generate the prone adaptation prediction model;

[0118] The validation set is used to verify whether the prediction performance of the prone position adaptation prediction model for the ECMO intervention timing under prone position ventilation reaches a preset value:

[0119] If so, deploying and applying the prone position adaptation prediction model;

[0120] Otherwise, repeat the above steps to retrain and construct the prone position adaptation prediction model.

[0121] The steps of using the model to train feature data generally go through the following process:

[0122] 1. Data collection and preprocessing:

[0123] Collect big data on ECMO application of patients with prone ventilation. Patients with prone ventilation refer to respiratory patients who undergo prone ventilation. The big data on ECMO application for respiratory treatment under prone ventilation includes:

[0124] Data such as respiratory rate, tidal volume, minute ventilation, oxygen concentration, blood oxygen saturation, airway pressure, ventilator mode, ventilator alarm, and the timing of triggering ECMO for patients in prone position ventilation.

[0125] The timing of triggering ECMO is different for different patients, and the ECMO parameters shown by different patients are also different.

[0126] Therefore, it is necessary to preprocess ECMO application big data, including the following steps:

[0127] Clean the data to remove outliers and duplicate records.

[0128] Standardize data formats and unify data units.

[0129] Fill or delete missing values.

[0130] 2. Feature Engineering:

[0131] Select features related to the timing of ECMO application, such as patient age, weight, disease type, severity of the disease, triggered ECMO mode and corresponding timing, respiratory parameter characteristic values, etc. To extract the characteristic values ​​from ECMO application big data, you can refer to the following steps:

[0132] ‌Principal Component Analysis (PCA)‌: PCA is used to reduce the dimensionality of the data, improving the interpretation and readability of the data while ensuring that important features are correctly retained.

[0133] ‌Independent Component Analysis (ICA): Extracting independent components separately helps to reveal the underlying structure of the data.

[0134] ‌Kernel Principal Component Analysis (KPCA): For nonlinear data structures, the KPCA method can significantly improve prediction accuracy and generalization ability.

[0135] ‌Feature selection and optimization‌: Select the features that contribute most to the model from the extracted features to reduce the data dimension and improve the model training efficiency. At the same time, transform the features, such as scaling and normalization, to increase the convergence speed of the model.

[0136] Through the above methods, the characteristic values ​​in ECMO application big data can be effectively extracted, providing strong support for subsequent model construction and data analysis.

[0137] Create derived variables, such as the ratio of ventilation duration to disease severity.

[0138] Perform feature selection and dimensionality reduction, such as principal component analysis, feature importance assessment, etc.

[0139] 3. Model construction:

[0140] Choose an appropriate machine learning model such as random forest, gradient boosted tree, neural network, etc.

[0141] Divide the training set and test set for model training.

[0142] Adjust model parameters such as learning rate, number of iterations, tree depth, etc. to optimize model performance.

[0143] For example, the feature set is divided into a training set and a test set in a ratio of 3:7. The training set is input into the random forest model to allow the model to learn the oxygenation index data feature values ​​when prone ventilation indications occur and ECMO intervention occurs, as well as the corresponding ECMO application time, so as to train and generate the prone adaptation prediction model.

[0144] ‌The learning process of the random forest model for feature values‌:

[0145] 1. Build a decision tree:

[0146] ‌Random Sampling‌: Randomly select a certain number of samples from the original data set (with replacement) to provide different training data for each decision tree.

[0147] Feature selection: At each node, a subset of features is randomly selected and optimal partitioning is performed based on these features.

[0148] 2. Generate Random Forest:

[0149] ‌Repeat construction‌: By repeating the above steps, multiple decision trees are constructed to form a random forest.

[0150] Comprehensive prediction: For classification problems, the final prediction result is obtained by voting; for regression problems, the final prediction result is obtained by averaging.

[0151] 3. Feature Importance Assessment:

[0152] In random forests, the importance of each feature can be evaluated based on how much contribution it makes in the decision tree.

[0153] Contribution measurement indicators include the Gini index, out-of-bag data error rate, etc.

[0154] Through this process, the random forest model is able to effectively learn and use the feature values ​​to make predictions.

[0155] 4. Model Evaluation:

[0156] Use the test set to evaluate model performance, such as accuracy, recall, F1 value, etc.

[0157] Cross-validation was performed to verify the generalization ability of the model.

[0158] 5. Model deployment:

[0159] Deploy the trained model to the actual application environment.

[0160] Develop a user interface to facilitate doctors to enter patient information and obtain ECMO application timing prediction results.

[0161] Models are updated in real time to adapt to new data and clinical practice.

[0162] Therefore, the present invention also performs data preprocessing, data feature engineering and model training on ECMO application big data, and screens out the indications for prone ventilation, oxygenation index data during ECMO intervention and the corresponding ECMO application time from the big data for basic processing of model training.

[0163] Through feature engineering, the characteristic values ​​of oxygenation index data when prone ventilation indications occur and ECMO intervention occurs, as well as the corresponding characteristic values ​​of ECMO application time, are extracted to construct the corresponding feature set (composed of characteristic data of several patients).

[0164] Subsequently, the random forest model is used to train and learn the characteristic values ​​of oxygenation index data when prone ventilation indications and ECMO intervention occur, as well as the corresponding ECMO application time, so as to train and generate the prone adaptation prediction model; the validation set is used to verify whether the prediction performance of the prone adaptation prediction model for the timing of ECMO intervention under prone ventilation reaches the preset value (set by the administrator).

[0165] Here, the preset value is used to determine whether the ECMO intervention time (ECMO application time or ECMO application timing) predicted and output by the model based on the indicator characteristic data of the patients in the validation set is close to the preset value, and the error can be kept within ±5min, or lower. By adjusting the training data of the model, the prediction accuracy of the model can be higher, for example, the prediction error of the ECMO intervention time can be kept within ±1min, and it can be deployed in the background and used to predict the ECMO intervention timing of ventilated patients.

[0166] The specific model can be implemented, for example, by using accuracy verification, and the administrator can perform specific operations.

[0167] The present invention utilizes the function of Random Forests:

[0168] Because Random Forests is an integrated learning method that improves the stability and accuracy of the model by building multiple decision trees. It is suitable for classification and regression problems and can handle high-dimensional data and missing values.

[0169] Optionally, S2, real-time monitoring of respiratory data of a patient in prone position ventilation, and parsing the respiratory data to obtain real-time oxygenation index data, including:

[0170] Preset oxygenation index for patients ventilated in prone position;

[0171] Collecting respiratory data of a patient in prone position ventilation in real time, and analyzing the respiratory data to obtain respiratory analysis data;

[0172] Based on a preset oxygenation index of a patient in prone position ventilation, data under the oxygenation index is filtered out from the respiratory analysis data to obtain real-time oxygenation index data;

[0173] The oxygenation index data is written into the HIS medical record file of the patient undergoing prone ventilation in real time.

[0174] Oxygenation index, see the data under the various indicators mentioned above, such as arterial blood gas analysis, blood oxygen saturation and pulmonary artery compliance.

[0175] After the system collects data, it can perform data analysis, format processing, data classification, etc., filter out the data under the oxygenation index from the respiratory analysis data, and obtain real-time oxygenation index data. These data can be written into the case files of the HIS system in real time, which is convenient for real-time recording of HIS medical records of patients with prone ventilation, and provide a data basis for subsequent model recognition.

[0176] Optionally, in step S3, importing the oxygenation index data into the prone position adaptation prediction model includes:

[0177] Monitor the HIS system to determine whether the new oxygenation index data is written into the HIS medical record file of the patient with prone ventilation in the HIS system:

[0178] If yes, read the oxygenation index data written in real time into the HIS medical record file of the patient undergoing prone ventilation, and import the preset prone position adaptation prediction model;

[0179] Otherwise continue to monitor.

[0180] The present invention deploys a monitoring program or monitoring module on the HIS system, which is mainly used to monitor whether new oxygenation index data of the prone ventilation patient is written into the HIS system. When new written data is found, the oxygenation index data written in real time in the HIS medical record file of the prone ventilation patient is read, and the preset prone adaptation prediction model is imported, so that the model can timely predict and analyze the patient's monitoring data and output the corresponding prediction results.

[0181] The monitoring module can be used:

[0182] Set up a monitoring program in the background to monitor whether new data appears in the HIS system:

[0183] Determine monitoring requirements: clarify the type of data, frequency, and trigger conditions that need to be monitored.

[0184] Select monitoring technology: Select appropriate monitoring technology according to system architecture and requirements, such as message queues, Webhooks, etc.

[0185] Deploy a listening program: Deploy a listening program on the server to implement data monitoring and parsing functions.

[0186] Data synchronization: When new data is detected, the data is synchronized to the target system or storage.

[0187] Data Verification: Verify the received data to ensure data integrity and accuracy.

[0188] Exception handling: Set up exception handling mechanisms, such as solutions for data errors, system failures, etc.

[0189] Security assurance: Ensure the security of monitoring procedures to prevent data leakage and illegal access.

[0190] Logging: Record the operation log of the monitoring program to facilitate subsequent troubleshooting and performance optimization.

[0191] Monitoring program maintenance: Regularly maintain and update the monitoring program to ensure its stable operation.

[0192] Continuous optimization: Based on actual operating conditions and business needs, the monitoring program is continuously optimized to improve efficiency and accuracy.

[0193] Here, patients can collect data of various indicators according to the preset sampling frequency, such as every 2-5 minutes, so the collected data is processed and written into the medical record file of the HIS system. It can be monitored by the program to find out whether the patient has new monitoring data, and let the model predict and analyze the new data in time to determine whether the patient triggers the ECMO intervention conditions. The system will activate ECMO according to the prediction results and intervene in the treatment of ventilated patients in time.

[0194] By adopting the present invention, the patient does not need to be under ECMO care in real time, thus avoiding the continuous generation of expensive ECMO care costs.

[0195] The connection of ECMO to the patient is performed by nurses and the like, and will not be elaborated in this embodiment.

[0196] Therefore, with the present invention, if the patient's oxygenation is still not ideal or has not improved significantly after prone ventilation, prone ventilation is tended to be combined in the early stage of ECMO establishment to improve oxygenation and respiratory system compliance as early as possible, improve the treatment effect, and allow the patient who is undergoing prone ventilation to be treated with ECMO in a timely manner, avoiding early dyspnea caused by the indications of prone ventilation, thereby ensuring the safety of the patient.

[0197] Figure 3 1 is a block diagram of an adaptation assessment system for ECMO combined with prone position ventilation according to an exemplary embodiment, and the system is used for an adaptation assessment method for ECMO combined with prone position ventilation. Figure 3 , the system comprises:

[0198] Respiratory monitoring system, used to monitor the respiratory data of patients undergoing prone ventilation in real time and report to the backend server;

[0199] Model management module, which is used to build and deploy a prone position adaptation prediction model for predicting the timing of ECMO intervention under prone position ventilation based on ECMO application big data under prone position ventilation;

[0200] A data analysis module, used to analyze the respiratory data to obtain real-time oxygenation index data;

[0201] The ECMO ventilation adaptation assessment module is used to import the oxygenation index data into the prone position adaptation prediction model, and the prone position adaptation prediction model assesses whether the prone ventilation patient triggers ECMO intervention conditions according to the oxygenation index data:

[0202] If an evaluation result that triggers ECMO intervention conditions is received, an activation command is issued to the ECMO device; otherwise, monitoring continues;

[0203] A message transmission module, used for transmitting the activation instruction to the ECMO device;

[0204] HIS system, used to record the HIS medical records of patients undergoing prone ventilation;

[0205] A monitoring module is used to monitor the HIS system and determine whether the HIS medical record file of the patient with prone ventilation in the HIS system has new oxygenation index data written into it: if so, read the oxygenation index data written into the HIS medical record file of the patient with prone ventilation in real time, and import the preset prone adaptation prediction model; otherwise, continue to monitor;

[0206] Backend database, used for backend database storage;

[0207] An ECMO device, configured to respond to the activation instruction and intervene in ventilation treatment of the prone ventilation patient;

[0208] The model management module, data analysis module, ECMO ventilation adaptation assessment module, message transmission module, HIS system, monitoring module and background database are all deployed on the background server;

[0209] The respiratory monitoring system and the ECMO device are respectively connected to the delivery background server for communication.

[0210] Please understand the above system composition and interactive application in conjunction with the principles of the above method steps, and will not be repeated here.

[0211] Figure 4 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, the electronic device 410 may include a first processor 2001 .

[0212] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0213] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0214] Combine the following Figure 4 The components of the electronic device 410 are described in detail:

[0215] The first processor 2001 is the control center of the electronic device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or may be application specific integrated circuits (ASICs), or may be one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs).

[0216] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0217] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.

[0218] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 4 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0219] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0220] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0221] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0222] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0223] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0224] It should be noted that Figure 4 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than those shown in the figure, or combine certain components, or arrange the components differently.

[0225] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the adaptation assessment method for ECMO combined with prone position ventilation described in the above method embodiment, which will not be repeated here.

[0226] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0227] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0228] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0229] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0230] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0231] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0232] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0233] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0234] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0235] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0236] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0237] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0238] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for evaluating the adaptability of ECMO combined with prone position ventilation, characterized in that: The method comprises: S1. Based on the big data of ECMO application under prone ventilation, a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation was constructed and deployed; S2. Real-time monitoring of respiratory data of the patient in prone position ventilation, and analysis of the respiratory data to obtain real-time oxygenation index data; S3, importing the oxygenation index data into the prone position adaptation prediction model, and using the prone position adaptation prediction model to evaluate whether the prone position ventilation patient triggers ECMO intervention conditions based on the oxygenation index data: If an evaluation result that triggers ECMO intervention conditions is received, an activation command is issued to the ECMO device; otherwise, monitoring continues; S4. The ECMO device responds to the activation instruction and intervenes in the ventilation treatment of the patient undergoing prone ventilation.

2. The ECMO combined with prone position ventilation adaptation assessment method according to claim 1, characterized in that: In step S1, the ECMO application big data based on prone ventilation includes the indications for prone ventilation, the oxygenation index data when ECMO is involved, and the corresponding ECMO application time, wherein: The oxygenation index data include arterial blood gas analysis data, blood oxygen saturation data and pulmonary artery compliance index data when prone ventilation indications occur, wherein: The pulmonary artery compliance index data include: lung capacity data of the pulmonary artery under static compliance and / or dynamic compliance during prone position ventilation; The blood oxygen saturation data includes: blood oxygen saturation value under prone position ventilation; The arterial blood gas analysis data include: arterial oxygen partial pressure (PaO2), arterial carbon dioxide partial pressure (PaCO2), pH, actual bicarbonate (AB) and standard bicarbonate (SB) and excess base (BE) under prone ventilation.

3. The ECMO combined with prone position ventilation adaptation assessment method according to claim 1, characterized in that: S1. Based on the big data of ECMO application under prone ventilation, a prone adaptation prediction model for predicting the timing of ECMO intervention under prone ventilation was constructed and deployed, including: From the background database, retrieve the ECMO application big data of several patients with prone ventilation; Preprocessing the ECMO application big data; Performing feature extraction on the preprocessed ECMO application big data to extract oxygenation index data characteristic values ​​when prone ventilation indications occur and ECMO intervention is required, as well as the corresponding ECMO application time, wherein the oxygenation index data characteristic values ​​include arterial blood gas analysis data characteristic values, blood oxygen saturation data characteristic values, and pulmonary artery compliance index data characteristic values; Based on the characteristic values ​​of several oxygenation index data extracted by feature engineering and the corresponding ECMO application time, the corresponding feature set is constructed; Dividing the feature set into a training set and a validation set according to a preset ratio; Inputting the training set into a preset random forest model, allowing the model to learn the oxygenation index data characteristic values ​​when prone ventilation indications occur and ECMO intervention occurs, as well as the corresponding ECMO application time, so as to train and generate the prone adaptation prediction model; The validation set is used to verify whether the prediction performance of the prone position adaptation prediction model for the ECMO intervention timing under prone position ventilation reaches a preset value: If so, deploying and applying the prone position adaptation prediction model; Otherwise, repeat the above steps to retrain and construct the prone position adaptation prediction model.

4. The ECMO combined with prone position ventilation adaptation assessment method according to claim 3, characterized in that: S2. Real-time monitoring of respiratory data of the patient in prone position ventilation, and analysis of the respiratory data to obtain real-time oxygenation index data, including: Preset oxygenation index for patients ventilated in prone position; Collecting respiratory data of a patient in prone position ventilation in real time, and analyzing the respiratory data to obtain respiratory analysis data; Based on a preset oxygenation index of a patient in prone position ventilation, data under the oxygenation index is filtered out from the respiratory analysis data to obtain real-time oxygenation index data; The oxygenation index data is written into the HIS medical record file of the patient undergoing prone ventilation in real time.

5. The ECMO combined with prone position ventilation adaptation assessment method according to claim 4, characterized in that: In step S3, the oxygenation index data is imported into the prone position adaptation prediction model, including: Monitor the HIS system to determine whether the new oxygenation index data is written into the HIS medical record file of the patient with prone ventilation in the HIS system: If yes, read the oxygenation index data written in real time into the HIS medical record file of the patient undergoing prone ventilation, and import the preset prone position adaptation prediction model; Otherwise continue to monitor.

6. An adaptation assessment system for ECMO combined with prone position ventilation, the adaptation assessment system for ECMO combined with prone position ventilation is used to implement the adaptation assessment method for ECMO combined with prone position ventilation as claimed in any one of claims 1 to 5, characterized in that: The system comprises: Respiratory monitoring system, used to monitor the respiratory data of patients undergoing prone ventilation in real time and report to the backend server; Model management module, which is used to build and deploy a prone position adaptation prediction model for predicting the timing of ECMO intervention under prone position ventilation based on ECMO application big data under prone position ventilation; A data analysis module, used to analyze the respiratory data to obtain real-time oxygenation index data; The ECMO ventilation adaptation assessment module is used to import the oxygenation index data into the prone position adaptation prediction model, and the prone position adaptation prediction model assesses whether the prone ventilation patient triggers ECMO intervention conditions according to the oxygenation index data: If an evaluation result that triggers ECMO intervention conditions is received, an activation command is issued to the ECMO device; otherwise, monitoring continues; A message transmission module, used for transmitting the activation instruction to the ECMO device; HIS system, used to record the HIS medical records of patients undergoing prone ventilation; A monitoring module is used to monitor the HIS system and determine whether the HIS medical record file of the patient with prone ventilation in the HIS system has new oxygenation index data written into it: if so, read the oxygenation index data written into the HIS medical record file of the patient with prone ventilation in real time, and import the preset prone adaptation prediction model; otherwise, continue to monitor; Backend database, used for backend database storage; An ECMO device, configured to respond to the activation instruction and intervene in ventilation treatment of the prone ventilation patient; The model management module, data analysis module, ECMO ventilation adaptation assessment module, message transmission module, HIS system, monitoring module and background database are all deployed on the background server; The respiratory monitoring system and the ECMO device are respectively connected to the delivery background server for communication.

7. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 5.

Images