Abnormal data detection early warning method of life support system

Abstract

The invention discloses an abnormal data detection early warning method of a life support system. The method comprises the steps of determining an abnormal detection item according to current disease information and historical disease information of a to-be-detected user to obtain detection item demand information; setting alarm thresholds for different types of detection items according to equipment specifications and historical disease information to obtain a set threshold data set; based on the detection item demand information, detecting the physical state of the to-be-detected user by adopting a life support system to obtain a user physiological monitoring data set, and creating a body digital twinborn body of the to-be-detected user; detecting whether each device in the life support system runs abnormally or not, if not, determining that the device runs normally, if yes, adjusting the abnormal device in the life support system, adjusting the abnormal device to be in a normal state, then outputting a device running normal signal, and re-detecting the physical state of the to-be-detected user. And updating the user physiological monitoring data set. The method has the effect of improving the abnormal data detection early warning accuracy of the life support system.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to a method for detecting and warning of abnormal data in a life support system. Background Technology

[0002] Currently, life support systems are a set of technologies, equipment, and procedures that provide and maintain the essential conditions for survival of living organisms (primarily humans). Its core objective is to create a safe and controllable artificial environment to cope with extreme or dangerous environments where natural resources are unavailable. Life support systems play a crucial role in the medical field, temporarily or permanently replacing key physiological functions through technological means, buying time for critically ill patients, maintaining vital signs, and creating conditions for disease recovery. When life support systems malfunction, they can cause varying degrees of adverse effects on users; therefore, detecting and issuing early warnings for abnormal data from life support systems is essential.

[0003] Existing methods for detecting and warning of abnormal data in life support systems involve monitoring various vital signs of users and issuing alarms when abnormal data is detected. However, these methods do not consider situations where equipment malfunctions, resulting in no alarm or false alarms. This wastes significant human and material resources and can seriously endanger user safety. Furthermore, the accuracy of these methods for detecting and warning of abnormal data in life support systems is relatively low and needs improvement. Summary of the Invention

[0004] To improve the accuracy of abnormal data detection and early warning in life support systems, this application provides a method for abnormal data detection and early warning in life support systems.

[0005] This application provides a method for detecting and warning of abnormal data in a life support system, which adopts the following technical solution: A method for detecting and warning of abnormal data in a life support system includes the following steps: Step S1: Based on the current and historical disease information of the user to be tested, determine the abnormal detection items of the user to be tested and obtain the detection item requirement information. Step S2: Based on the physical characteristics of the user to be tested, the device specifications, and historical medical information, set alarm thresholds for different types of test items to obtain a set threshold dataset; Step S3: Based on the detection project requirements information, the life support system is used to detect the physical status of the user to be tested to obtain the user's physiological monitoring dataset, and a digital twin of the user's body is created. Step S4: Detect whether there are any abnormalities in the operation of each device in the life support system. If there are no abnormalities, the device is operating normally. If there are abnormalities, adjust the abnormal device in the life support system. After adjusting it to a normal state, output a normal device operation signal, re-detect the physical state of the user under test, and update the user's physiological monitoring dataset. Step S5: Based on the set threshold dataset and the user physiological monitoring dataset, perform multimodal analysis of the body physiological state of the user to be tested to obtain the user's body characteristic judgment result, and record the user's abnormal characteristic information; Step S6: After receiving the user's physical characteristics determination result, determine whether the life support system has a false alarm for the user to be tested. If there is no false alarm, output the cross-validation success result. Step S7: Upon receiving the successful verification result, the system will issue a graded warning and respond to the user under test based on the abnormal user characteristic information, record the response and rescue information, and upload it to the blockchain system.

[0006] Preferably, basic detection item information is obtained by acquiring real-time relevant detection items of basic vital signs when using a life support system based on big data technology; Based on the patient's historical medical history, the additional testing items required for the patient are determined, resulting in the first category of additional testing item information; Based on the current illness of the user being tested, the additional tests required for the user are determined, resulting in the second category of additional test information; The basic testing item information, the first type of additional testing item information, and the second type of additional testing item information are combined to form preliminary testing item determination information. It is determined whether there are redundant testing items in the preliminary testing item determination information. If there are, the redundancy is deleted to obtain the testing item requirement information.

[0007] Preferably, a project standard range dataset is obtained based on big data technology, wherein the project standard range dataset includes the standard numerical range information of each detection item in the detection project requirement information; Based on the historical medical information of the users to be tested, the threshold range of each test item is initially adjusted to obtain the preliminary threshold adjustment information for the test items. The equipment specifications of each device in the life support system are obtained to obtain the system equipment model information. Based on big data technology, the degree of result bias of each model of equipment in the historical testing process is obtained to obtain the test result bias information of each device. Based on the bias information of the detection results of each device, the preliminary adjustment information of the project threshold for the corresponding detection item is adjusted again to obtain the project threshold setting information; When the life support system performs tests on the user under test, the threshold setting information of each test item in the test item requirement information is combined to form a set threshold dataset.

[0008] Preferably, based on the detection item requirement information, the user physiological monitoring dataset is obtained by using the physical state characteristics of the user to be tested through various devices in the life support system. The user physiological monitoring dataset includes the detection value information of each detection item in the detection item requirement information. The first type of user's image information is obtained by taking real-time photos of the user's body surface. The second type of user image information was obtained by scanning the whole body of the user under test using CT scanning technology; By combining image information taken by the first type of user and image information taken by the second type of user, a digital twin of the body of the user to be tested is created.

[0009] Preferably, the image information is obtained by real-time shooting of the user under test using a camera; Based on the captured image information, determine whether the placement of the detection contact part between the life support system and the user under test is reasonable. If it is unreasonable, output the result of abnormal device placement. Based on the captured image information, determine the degree of abnormality of the placement position to obtain the device placement abnormality information. If it is reasonable, output the result of normal device placement. The system checks whether the power supply / battery status of each device in the life support system is abnormal. If it is abnormal, it outputs the result of abnormal power supply and records the abnormal power supply information. If it is not abnormal, it outputs the result of normal power supply. Real-time monitoring of the operating status of each device in the life support system to determine whether a mechanical failure has occurred. If a mechanical failure exists, the system outputs the mechanical failure result and records the mechanical failure information. If no mechanical failure exists, the system outputs the result indicating that the device is mechanically normal. Based on the captured image information, determine the usage status of consumables of each device in the life support system and determine whether the consumables are exhausted. If the consumables are about to run out, output the device consumables exhaustion result and record the device consumables exhaustion information. If the consumables are sufficient, output the device consumables sufficient result. The abnormal equipment placement information, abnormal equipment power supply information, abnormal equipment mechanical fault information, and information on depleted equipment consumables are combined to form abnormal equipment operation information.

[0010] Preferably, when receiving results of abnormal equipment placement, abnormal equipment power supply, equipment mechanical failure, or depletion of equipment consumables, the system determines that there is an abnormal equipment operation and outputs an abnormal equipment operation result. When receiving results of normal equipment placement, normal equipment power supply, normal equipment mechanics, or sufficient equipment consumables, the system determines that there is no abnormal equipment operation and outputs a normal equipment operation result. Upon receiving a result indicating abnormal equipment operation, adjustments are made to each device in the life support system based on the abnormal equipment operation information until a result indicating normal equipment operation is received. Upon receiving a result indicating abnormal device operation, the system re-detects the physical condition characteristics of the user under test using various devices within the life support system, based on the detection item requirements, and updates the detection values ​​of each item in the user's physiological monitoring dataset.

[0011] Preferably, the detection value information of each detection item in the user physiological monitoring dataset is compared with the item threshold setting information of the corresponding detection item in the set threshold dataset to obtain the item parameter comparison result; Based on the comparison results of project parameters, it is determined whether the detection value information of each detection item is normal. If the detection value information of a detection item is within the range of the project threshold setting information of the corresponding detection item, the detection value information of the detection item is normal and the detection node outputs a normal result; otherwise, the detection value information of the detection item is in an abnormal state and the detection node outputs an abnormal result. Based on the detection value information of each detection item in the user physiological monitoring dataset, a time series diagram of the detection value of each detection item is drawn. Based on the time series diagram of the detection value of each detection item, it is determined whether there is any abnormality in the change of the detection value information of each detection item. If there is an abnormal change in the detection value information, the abnormal result of the detection change is output. If there is no abnormal change in the detection value information, the normal result of the detection change is output. When a normal result is received from the detection node and a normal result is received from the detection change, a normal result for the user feature is output; otherwise, an abnormal result for the user feature is output. When an abnormal result for the user feature is received, the abnormal information for the user feature is recorded. The results of normal user characteristics and abnormal user characteristics are used to determine the user's physical characteristics.

[0012] Preferably, after receiving the user's body feature determination result, the system determines whether the user under test has abnormal physical signs based on the captured image information. If abnormal physical signs are found, the system determines whether they match the user's abnormal feature information. If they match, the system outputs a successful cross-validation result. Upon receiving a successful cross-validation result, the abnormal user characteristic information is marked and displayed on the digital twin of the user under test.

[0013] Preferably, after receiving the successful cross-validation result, the warning level information of the user detection warning is obtained by judging the warning level of the life support system when there is a warning requirement for the detection of the user under test based on the abnormal information of user characteristics. Based on user detection and early warning level information, abnormal user characteristic information is displayed and marked in a hierarchical manner on the digital twin of the user's body; The system performs graded early warnings based on user detection and warning level information, and sends the user detection and warning level information to the background monitoring system via wireless communication module. Upon receiving the user detection and warning level information, the system responds and conducts targeted rescue. Record the response and rescue information of the users under test and upload it to the blockchain system for storage.

[0014] In summary, this application includes at least one of the following beneficial technical effects: 1. By conducting personalized analysis of the current and historical medical information of the patients to be tested, the required testing items for the users are determined, and the testing item demand information is obtained. Based on the equipment specifications of each device in the life support system and the historical medical information of the users to be tested, alarm thresholds for different testing items are preset to obtain a set threshold dataset. The alarm threshold settings for the users to be tested improve the accuracy of abnormal data detection and early warning of the life support system. Based on the life support system, the users to be tested are tested to obtain a user physiological monitoring dataset. The operating status of the equipment is detected to determine whether the inaccurate detection is caused by the operating status of the equipment. If the operating status of the equipment is abnormal, it is adjusted to the normal state and the detection is repeated, and the user physiological monitoring dataset is updated, which reduces false alarms caused by equipment failure and further improves the accuracy of abnormal data detection and early warning of the life support system. Based on the set threshold dataset and the user physiological monitoring dataset, multimodal analysis of the body physiological status of the users to be tested is performed to obtain the user's body characteristic judgment results and record the user characteristic abnormal information. The status of the users to be tested is detected and cross-validated, which further improves the accuracy of abnormal data detection and early warning of the life support system. 2. By using a camera to capture images of the user's body surface and employing CT scanning technology, a full-body scan of the user using X-rays can be performed, clearly displaying organ structures and allowing observation of arteries / veins. Combining the user's body surface and internal organ structures and blood vessels, a digital twin of the user's body is created. This more accurate digital twin facilitates subsequent detection and early warning of abnormal data in the life support system, further improving the accuracy of abnormal data detection and early warning in the life support system. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the abnormal data detection and early warning method of the life support system, which is the main feature of this embodiment. Detailed Implementation

[0016] The present application will be further described in detail below with reference to the accompanying drawings.

[0017] This application discloses a method for detecting and warning of abnormal data in a life support system.

[0018] A method for detecting and warning of abnormal data in a life support system includes the following steps: Reference Figure 1 Step S1 involves determining the abnormal testing items for the user based on their current and historical illness information, thus obtaining the required testing items. Step S1 specifically includes the following sub-steps: Step S11: Based on big data technology, obtain the basic detection item information by acquiring the real-time relevant detection items of basic vital signs when using the life support system.

[0019] Specifically, the basic testing information includes respiratory system testing, circulatory system testing, body temperature monitoring, intracranial pressure monitoring, and electroencephalogram (EEG) testing. Respiratory system testing includes tidal volume, respiratory rate, airway pressure, oxygen concentration, end-tidal carbon dioxide concentration, and blood oxygen saturation. Circulatory system testing includes heart rate, blood pressure, central venous pressure, pulmonary artery pressure, cardiac output, and mixed venous oxygen saturation.

[0020] Step S12: Based on the historical medical information of the user to be tested, determine the additional testing items that the user needs to be tested to obtain the information on the first type of additional testing items.

[0021] Here are some examples: The patient's historical medical information includes the type, severity, and duration of the illness. If the patient's historical medical information includes diabetes, then the first category of additional tests includes fasting blood glucose, two-hour postprandial blood glucose, glycated hemoglobin, and pancreatic function. If the patient's historical medical information includes cerebral infarction, then the first category of additional tests includes cerebral artery CT and cerebral artery MRI.

[0022] Step S13: Based on the current illness of the user to be tested, determine the additional testing items that the user needs to be tested and obtain the second type of additional testing item information.

[0023] Here are some examples: If the user's current illness includes liver disease, the second category of additional testing information will include liver function tests, hepatitis B five markers, four infection markers, and four liver fibrosis markers. If the user's current illness includes acute brain injury, the second category of additional testing information will include intracranial pressure testing.

[0024] Step S14: The basic test item information, the first type of additional test item information, and the second type of additional test item information are combined to form the preliminary test item judgment information. It is determined whether there are redundant test items in the preliminary test item judgment information. If there are, the redundancy is deleted to obtain the test item requirement information.

[0025] Reference Figure 1Step S2 involves setting alarm thresholds for different types of testing items based on the user's physical characteristics, device specifications, and historical medical information, thus obtaining a set threshold dataset. Step S2 specifically includes the following sub-steps: Step S21: Obtain the project standard range dataset based on big data technology. The project standard range dataset includes the standard numerical range information of each test item in the test project requirement information.

[0026] Here are some examples: For the fasting blood glucose test in the test requirements information: the internationally accepted normal value for fasting blood glucose is 3.9~6.1mmol / L. For the two-hour postprandial blood glucose test in the test requirements information: the internationally accepted normal value for two-hour postprandial blood glucose is <7.8mmol / L.

[0027] Step S22: Based on the historical disease information of the user to be tested, the threshold range of each test item is initially adjusted to obtain the preliminary adjustment information of the item threshold.

[0028] Here is an example: If the user's historical disease type information includes diabetes, then the fasting blood glucose threshold range in the preliminary adjustment information of the project threshold is 4.4-6.1 mmol / L, and the postprandial two-hour blood glucose range in the preliminary adjustment information of the project threshold is 4.4-8.0 mmol / L.

[0029] Step S23: Obtain the equipment specifications of each device in the life support system to obtain the system equipment model information. Based on big data technology, obtain the result bias information of each device model in the historical testing process to obtain the test result bias information of each device.

[0030] In practical applications, if a particular model of thermometer consistently measures values ​​that are 0.05 degrees Celsius higher than normal, then the bias information for the corresponding temperature measurement result of that thermometer is +0.05 degrees Celsius. Similarly, if a particular model of blood glucose meter consistently measures values ​​that are 0.1 mmol / L higher than normal during historical use, then the bias information for the result of that model of blood glucose meter is +0.1 mmol / L.

[0031] Step S24: Based on the bias information of the detection results of each device, the preliminary adjustment information of the project threshold for the corresponding detection item is adjusted again to obtain the project threshold setting information.

[0032] For example, if the bias information of the blood glucose meter in the life support system is +0.1 mmol / L, and the initial adjustment information of the project thresholds indicates that the fasting blood glucose threshold range is 4.4-6.1 mmol / L and the postprandial two-hour blood glucose range is 4.4-8.0 mmol / L, then the initial adjustment information of the project thresholds for the corresponding test items, namely the fasting blood glucose test and the postprandial two-hour blood glucose test, is adjusted to obtain the project threshold setting information. In the project threshold setting information, the fasting blood glucose threshold is set to 4.3-6.0 mmol / L, and the postprandial two-hour blood glucose threshold is set to 4.3-7.9 mmol / L.

[0033] Step S25: When the life support system performs detection on the user to be tested, the threshold setting information of each detection item in the detection item requirement information is combined to form a set threshold dataset.

[0034] Reference Figure 1 Step S3 involves using a life support system to monitor the physical condition of the user under test in real time, based on the test project requirements, to obtain a user physiological monitoring dataset and create a digital twin of the user's body. Step S3 specifically includes the following sub-steps: Step S31: Based on the detection item requirement information, the physical condition characteristics of the user to be tested are detected using each device in the life support system to obtain the user physiological monitoring dataset. The user physiological monitoring dataset includes the detection value information of each detection item in the detection item requirement information.

[0035] Step S32: Based on the camera, the first type of user image information is obtained by real-time detection of the body surface of the user to be tested.

[0036] Step S33: Based on CT scanning technology, X-rays are used to perform a full-body scan of the user to be tested to obtain image information of the second type of user.

[0037] Step S34: Combine the image information captured by the first type of user and the image information captured by the second type of user to create a digital twin of the user's body.

[0038] In practical applications, cameras can only capture images of the user's body surface, while CT scans, using X-rays, can clearly display organ structures and observe arteries and veins. By combining the user's body surface, internal organ structures, and blood vessels, a digital twin of the user's body can be created. This digital twin is more accurate and facilitates the detection and early warning of abnormal data in the life support system.

[0039] Reference Figure 1Step S4 involves detecting any operational abnormalities in the life support system during the user's testing process. If no abnormalities are found, the system is considered to be operating normally. If abnormalities are found, the malfunctioning devices in the life support system are adjusted and restored to normal operation. A normal operation signal is then output, the user's physical condition is re-detected, and the user's physiological monitoring dataset is updated. Step S4 specifically includes the following sub-steps: Step S41: Obtain image information by taking real-time photos of the user under test using a camera.

[0040] Step S42: Based on the captured image information, determine whether the placement of the detection contact part between the life support system and the user under test is reasonable. If it is unreasonable, output the device placement abnormality result. Based on the captured image information, determine the degree of abnormality of the placement position to obtain the device placement abnormality information. If it is reasonable, output the device placement normal result.

[0041] Step S43: Check whether the power / battery status of each device in the life support system is abnormal. If it is abnormal, output the device power supply abnormality result and record the device power supply abnormality information. If there is no abnormality, output the device power supply normal result.

[0042] Step S44: Real-time monitoring of the operating status of each device in the life support system to determine whether a mechanical fault has occurred. If a mechanical fault exists, output the mechanical fault result and record the mechanical fault information. If no mechanical fault exists, output the normal mechanical result.

[0043] Step S45: Based on the captured image information, determine the usage status of consumables of each device in the life support system and determine whether the consumables are exhausted. If the consumables are about to run out, output the device consumables exhaustion result and record the device consumables exhaustion information. If the consumables are sufficient, output the device consumables sufficient result.

[0044] Step S46: Abnormal equipment placement information, abnormal equipment power supply information, abnormal equipment mechanical fault information, and information on depleted equipment consumables are combined to form abnormal equipment operation information.

[0045] Step S47: When receiving results of abnormal equipment placement, abnormal equipment power supply, mechanical equipment failure, or depletion of equipment consumables, determine that the life support system has an abnormal equipment operation and output the abnormal equipment operation result. When receiving results of normal equipment placement, normal equipment power supply, normal equipment mechanics, or sufficient equipment consumables, determine that the life support system does not have an abnormal equipment operation and output the normal equipment operation result.

[0046] Step S48: After receiving the abnormal equipment operation result, adjust each device in the life support system based on the abnormal equipment operation information until the normal equipment operation result is received.

[0047] Step S49: Upon receiving the abnormal device operation result, based on the detection item requirement information, the physical status characteristics of the user to be tested are re-detected using each device in the life support system, and the detection value information of each detection item in the user physiological monitoring dataset is updated.

[0048] Reference Figure 1 Step S5 involves performing multimodal analysis on the body physiological state of the user under test based on the set threshold dataset and the user's physiological monitoring dataset to obtain the user's body characteristic determination results and record abnormal user characteristic information. Step S5 specifically includes the following sub-steps: Step S51: Compare the detection value information of each detection item in the user physiological monitoring dataset with the item threshold setting information of the corresponding detection item in the set threshold dataset to obtain the item parameter comparison result.

[0049] Step S52: Based on the comparison results of project parameters, determine whether the detection value information of each detection project is normal. If the detection value information of a detection project is within the range of the project threshold setting information of the corresponding detection project, the detection value information of the detection project is normal and the detection node outputs a normal result. Otherwise, the detection value information of the detection project is in an abnormal state and the detection node outputs an abnormal result.

[0050] Here is an example to illustrate: If the detection value of a detection item is outside the threshold setting range of the corresponding detection item, the more the detection value exceeds the threshold setting range of the corresponding detection item, the higher the degree of anomaly of the detection node.

[0051] Step S53: Based on the detection value information of each detection item in the user physiological monitoring dataset, draw the time series diagram of the detection value of each detection item. Based on the time series diagram of the detection value of each detection item, determine whether there is any abnormality in the change of the detection value information of each detection item of the user under test. If there is an abnormal change in the detection value information, output the abnormal change result. If there is no abnormal change in the detection value information, output the normal change result.

[0052] Here is an example to illustrate: If the change in the detection value of a test item exceeds the preset threshold for the change in the detection value, it is determined that the change in the detection value of that test item is abnormal. The more the change in the detection value exceeds the preset threshold for the change in the detection value, the higher the degree of abnormality of the change in the detection value.

[0053] Step S54: When a normal result of the detection node is received and a normal result of the detection change is received, output a normal result of the user feature; otherwise, output an abnormal result of the user feature. When an abnormal result of the user feature is received, record the abnormal information of the user feature.

[0054] The user feature anomaly information includes user feature anomaly type information, user feature anomaly location information, and user feature anomaly degree information. The user feature anomaly degree information is jointly determined by the anomaly degree of the detection node and the anomaly degree of the change in the detected value. The higher the anomaly degree of the detection node, the higher the user feature anomaly degree information; the higher the anomaly degree of the change in the detected value, the higher the user feature anomaly degree information.

[0055] Step S55: The combination of normal user characteristic results and abnormal user characteristic results forms the user's physical characteristic determination result.

[0056] Reference Figure 1 Step S6: After receiving the user's physical characteristic assessment result, determine whether the life support system has any false alarms for the user being tested. If there are no false alarms, output a successful cross-validation result. Step S6 specifically includes the following sub-steps: Step S61: After receiving the user's body feature determination result, determine whether the user under test has abnormal physical signs based on the captured image information. If there are abnormal physical signs, determine whether they match the user's abnormal feature information. If they match, output the cross-validation success result.

[0057] Here's an example: If the image information shows that the user's face is cyanotic, and the abnormal user feature information shows that the abnormal user feature type is abnormal blood oxygen saturation, then facial cyanosis is a sign that appears when blood oxygen saturation is abnormal. In other words, the abnormal signs that the user shows at this time are consistent with the abnormal user feature information, and the cross-validation result is output as successful.

[0058] Step S62: After receiving the successful cross-validation result, mark and display the user feature anomaly type information and user feature anomaly location information in the user feature anomaly information on the digital twin of the user's body.

[0059] Reference Figure 1 Step S7: Upon receiving a successful verification result, based on the user's abnormal characteristic information, a tiered early warning system is implemented for the user under test, and a rescue response is initiated. The rescue response information is recorded and uploaded to the blockchain system. Step S7 specifically includes the following sub-steps: Step S71: After receiving the successful cross-validation result, the warning level of the life support system for detecting the user under test is determined based on the abnormal user characteristic information, and the user detection warning level information is obtained.

[0060] Step S72: Based on the user detection warning level information, the abnormal user feature information is displayed and marked in a hierarchical manner on the digital twin of the user's body.

[0061] Step S73: Based on the user detection warning level information, a graded warning is issued. The user detection warning level information is sent to the background monitoring system via the wireless communication module. Upon receiving the user detection warning level information, a response is made and a targeted rescue is carried out based on the user characteristic anomaly type information and the user characteristic anomaly location information.

[0062] Step S74: Record the response and rescue information for the user under test and upload the response and rescue information to the blockchain system for storage. The response and rescue information includes the warning time, warning content, rescue time, rescue content, and rescue personnel information for the user under test.

[0063] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for detecting and warning of abnormal data in a life support system, characterized in that, include: Step S1: Based on the current and historical disease information of the user to be tested, determine the abnormal detection items of the user to be tested and obtain the detection item requirement information. Step S2: Based on the equipment specifications of each device in the life support system and the historical medical information of the user to be tested, set alarm thresholds for different types of test items to obtain a set threshold dataset. Step S3: Based on the detection project requirements information, the life support system is used to detect the physical status of the user to be tested to obtain the user's physiological monitoring dataset, and a digital twin of the user's body is created. Step S4: Detect whether there are any abnormalities in the operation of each device in the life support system. If there are no abnormalities, the device is operating normally. If there are abnormalities, adjust the abnormal device in the life support system. After adjusting it to a normal state, output a normal device operation signal, re-detect the physical state of the user under test, and update the user's physiological monitoring dataset. Step S5: Based on the set threshold dataset and the user physiological monitoring dataset, perform multimodal analysis of the body physiological state of the user to be tested to obtain the user's body characteristic judgment result, and record the user's abnormal characteristic information; Step S6: After receiving the user's physical characteristics determination result, determine whether the life support system has a false alarm for the user to be tested. If there is no false alarm, output the cross-validation success result. Step S7: Upon receiving the successful verification result, the system will issue a graded warning and respond to the user under test based on the abnormal user characteristic information, record the response and rescue information, and upload it to the blockchain system.

2. The abnormal data detection and early warning method for a life support system according to claim 1, characterized in that, Step S1 specifically includes: Based on big data technology, real-time relevant detection items of basic vital signs are obtained when using life support systems to obtain basic detection item information; Based on the patient's historical medical history, the additional testing items required for the patient are determined, resulting in the first category of additional testing item information; Based on the current illness of the user being tested, the additional tests required for the user are determined, resulting in the second category of additional test information; The basic testing item information, the first type of additional testing item information, and the second type of additional testing item information are combined to form preliminary testing item determination information. It is determined whether there are redundant testing items in the preliminary testing item determination information. If there are, the redundancy is deleted to obtain the testing item requirement information.

3. The abnormal data detection and early warning method for a life support system according to claim 2, characterized in that, Step S2 specifically includes: The project standard range dataset is obtained based on big data technology. The project standard range dataset includes the standard numerical range information of each detection item in the detection project requirement information. Based on the historical medical information of the users to be tested, the threshold range of each test item is initially adjusted to obtain the preliminary threshold adjustment information for the test items. The equipment specifications of each device in the life support system are obtained to obtain the system equipment model information. Based on big data technology, the degree of result bias of each model of equipment in the historical testing process is obtained to obtain the test result bias information of each device. Based on the bias information of the detection results of each device, the preliminary adjustment information of the project threshold for the corresponding detection item is adjusted again to obtain the project threshold setting information; When the life support system performs tests on the user under test, the threshold setting information of each test item in the test item requirement information is combined to form a set threshold dataset.

4. The abnormal data detection and early warning method for a life support system according to claim 3, characterized in that, Step S3 specifically includes: Based on the detection project requirement information, the physical condition characteristics of the user to be tested are detected by each device in the life support system to obtain the user physiological monitoring dataset, which includes the detection value information of each detection project in the detection project requirement information. The first type of user's image information is obtained by taking real-time photos of the user's body surface. The second type of user image information was obtained by scanning the whole body of the user under test using CT scanning technology; By combining image information taken by the first type of user and image information taken by the second type of user, a digital twin of the body of the user to be tested is created.

5. The abnormal data detection and early warning method for a life support system according to claim 4, characterized in that, Step S4 specifically includes: The image information is obtained by capturing real-time images of the user under test using a camera. Based on the captured image information, determine whether the placement of the detection contact part between the life support system and the user under test is reasonable. If it is unreasonable, output the result of abnormal device placement. Based on the captured image information, determine the degree of abnormality of the placement position to obtain the device placement abnormality information. If it is reasonable, output the result of normal device placement. The system checks whether the power supply / battery status of each device in the life support system is abnormal. If it is abnormal, it outputs the result of abnormal power supply and records the abnormal power supply information. If it is not abnormal, it outputs the result of normal power supply. Real-time monitoring of the operating status of each device in the life support system to determine whether a mechanical failure has occurred. If a mechanical failure exists, the system outputs the mechanical failure result and records the mechanical failure information. If no mechanical failure exists, the system outputs the result indicating that the device is mechanically normal. Based on the captured image information, determine the usage status of consumables of each device in the life support system and determine whether the consumables are exhausted. If the consumables are about to run out, output the device consumables exhaustion result and record the device consumables exhaustion information. If the consumables are sufficient, output the device consumables sufficient result. The abnormal equipment placement information, abnormal equipment power supply information, abnormal equipment mechanical fault information, and information on depleted equipment consumables are combined to form abnormal equipment operation information.

6. The abnormal data detection and early warning method for a life support system according to claim 5, characterized in that, Step S4 also includes: When receiving results of abnormal equipment placement, abnormal equipment power supply, equipment mechanical failure, or depletion of equipment consumables, the life support system is determined to have an abnormal equipment operation and outputs an abnormal equipment operation result. When receiving results of normal equipment placement, normal equipment power supply, normal equipment mechanics, or sufficient equipment consumables, the life support system is determined to have no abnormal equipment operation and outputs a normal equipment operation result. Upon receiving a result indicating abnormal equipment operation, adjustments are made to each device in the life support system based on the abnormal equipment operation information until a result indicating normal equipment operation is received. Upon receiving a result indicating abnormal device operation, the system re-detects the physical condition characteristics of the user under test using various devices within the life support system, based on the detection item requirements, and updates the detection values ​​of each item in the user's physiological monitoring dataset.

7. The abnormal data detection and early warning method for a life support system according to claim 6, characterized in that, Step S5 specifically includes: The detection values ​​of each detection item in the user physiological monitoring dataset are compared with the threshold setting information of the corresponding detection item in the set threshold dataset to obtain the comparison results of the item parameters. Based on the comparison results of project parameters, it is determined whether the detection value information of each detection item is normal. If the detection value information of a detection item is within the range of the project threshold setting information of the corresponding detection item, the detection value information of the detection item is normal and the detection node outputs a normal result; otherwise, the detection value information of the detection item is in an abnormal state and the detection node outputs an abnormal result. Based on the detection value information of each detection item in the user physiological monitoring dataset, a time series diagram of the detection value of each detection item is drawn. Based on the time series diagram of the detection value of each detection item, it is determined whether there is any abnormality in the change of the detection value information of each detection item. If there is an abnormal change in the detection value information, the abnormal result of the detection change is output. If there is no abnormal change in the detection value information, the normal result of the detection change is output. When a normal result is received from the detection node and a normal result is received from the detection change, a normal result for the user feature is output; otherwise, an abnormal result for the user feature is output. When an abnormal result for the user feature is received, the abnormal information for the user feature is recorded. The results of normal user characteristics and abnormal user characteristics are used to determine the user's physical characteristics.

8. The abnormal data detection and early warning method for a life support system according to claim 7, characterized in that, Step S6 specifically includes: After receiving the user's physical characteristics assessment result, the system determines whether the user has any abnormal physical signs based on the captured image information. If abnormal physical signs are found, the system determines whether they match the user's abnormal characteristics information. If they match, the system outputs a successful cross-validation result. Upon receiving a successful cross-validation result, the abnormal user characteristic information is marked and displayed on the digital twin of the user under test.

9. The abnormal data detection and early warning method for a life support system according to claim 8, characterized in that, Step S7 specifically includes: After receiving the successful cross-validation result, the warning level information of the user detection warning level is obtained by determining the warning level when the life support system needs to detect the user under test based on the abnormal user characteristic information. Based on user detection and early warning level information, abnormal user characteristic information is displayed and marked in a hierarchical manner on the digital twin of the user's body; The system performs graded early warnings based on user detection and warning level information, and sends the user detection and warning level information to the background monitoring system via wireless communication module. Upon receiving the user detection and warning level information, the system responds and conducts targeted rescue. Record the response and rescue information of the users under test and upload it to the blockchain system for storage.

Images