Accompanying data abnormity analysis system based on AI analysis

Abstract

The invention relates to the technical field of data processing, in particular to an accompanying data anomaly analysis system based on AI analysis, which is characterized in that accompanying data of a target time length before a current time point is acquired, a nursing entity is extracted from the accompanying data, and values of environmental parameters and sign data of a target object after the nursing entity is executed for a certain time length are acquired; and accompanying feature data are constructed. And then performing anomaly prediction based on the accompanying feature data by using an anomaly prediction model reflecting a sign change rule after nursing to obtain an anomaly prediction result. And finally, performing comprehensive analysis on the physical sign data in the target time period based on the abnormal prediction result to obtain a prediction analysis result. The abnormal physical sign data of the patient in the accompanying process is found according to the physical sign changes which the patient should respond to after multiple nursing operations are executed on the patient in the accompanying process, and therefore the abnormity of the patient can be found timely and accurately.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and specifically to an abnormal analysis system for escort data based on AI analysis. Background Art

[0002] Escort data refers to various information collected during medical care, especially for patients who need special care (such as the elderly, chronic disease patients, postoperative recovery patients, etc.). For example, personal information, vital sign data, nursing data, etc.

[0003] Abnormal analysis of escort data refers to the process of identifying data points that do not conform to the normal pattern or expectation when collecting and processing various data related to patient escort. This kind of analysis is of great significance. By analyzing the abnormal situations in the escort data, problems or potential risks existing in the nursing process can be found, which helps to discover risk factors that may lead to medical accidents in advance, such as medication errors, operation mistakes, etc. However, the existing abnormal analysis of escort data relies on the personal experience of medical staff, and it is easy to have situations of analysis omission and analysis error. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide an abnormal analysis system for escort data based on AI analysis to solve the problems in the background art.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] An abnormal analysis system for escort data based on AI analysis of the present invention includes the following steps:

[0007] An acquisition module, configured to acquire the escort data of a target object in a target time period, where the escort data includes the vital sign data, nursing records, and environmental parameter values at multiple time points, and the target time period is a time period with a target duration before the current time point;

[0008] A feature extraction module, configured to extract the vital sign change data and nursing entities at multiple time points from the escort data, and construct escort feature data based on the vital sign change data at the first time point, the environmental parameter value at the first time point, the nursing entity at the second time point, where the first time point is later than the second time point by a unit duration;

[0009] An AI prediction module, configured to input the escort feature data into a pre-constructed abnormal prediction model to obtain an abnormal prediction result, where the abnormal analysis model represents the change situation of vital sign data after different nursing entities are implemented on a patient under various environmental conditions;

[0010] Anomaly analysis module, configured to perform anomaly analysis on the vital sign data of the target object in the target time period based on the anomaly prediction result.

[0011] In an embodiment of the present application, extracting vital sign change data at multiple time points from the accompanying care data, including:

[0012] Extracting the values of multiple vital sign parameters at multiple time points from the vital sign data of the accompanying care data;

[0013] Performing normalization processing on the values of each vital sign parameter at multiple time points to obtain multiple normalized sequences;

[0014] Calculating the slope of each time point in the multiple normalized sequences based on central difference to obtain the change characteristics of multiple vital sign parameters; and constructing vital sign change data at multiple time points based on the change characteristics of multiple vital sign parameters.

[0015] In an embodiment of the present application, extracting nursing entities at multiple time points from the accompanying care data, including:

[0016] Performing word segmentation on the nursing records in the accompanying care data to obtain multiple words;

[0017] Matching the multiple words with a pre-constructed entity library to obtain nursing entities that match the entities in the pre-constructed entity library, where the nursing entities include medication record entities, nursing operation entities, and user behavior entities.

[0018] In an embodiment of the present application, constructing accompanying care feature data based on the vital sign change data at the first time point, the value of the environmental parameter at the first time point, and the nursing entity at the second time point, including:

[0019] Extracting the second time point of each nursing record from the accompanying care data;

[0020] Determining the first time point based on the second time point and the unit duration, and extracting the vital sign change data and the value of the environmental parameter at the first time point from the accompanying care data;

[0021] Performing vectorization processing on multiple groups of vital sign change data at the first time point, the value of the environmental parameter at the first time point, and the nursing entity at the second time point to obtain multiple groups of accompanying care feature data.

[0022] In an embodiment of the present application, the method for constructing the anomaly analysis model includes:

[0023] Obtaining multiple accompanying care sample data, where the accompanying care sample data includes a set of vital sign change sample data, sample values of environmental parameters, and nursing entities;

[0024] Vectorize and annotate the accompanying care sample data to obtain the label of each accompanying care sample data, where the label is the normal probability and the abnormal probability;

[0025] Construct training data based on the vectorized accompanying care sample data and the labels to obtain a training data set;

[0026] Train an artificial neural network based on the training data set to obtain an anomaly analysis model.

[0027] In an embodiment of the present application, input the accompanying care feature data into a pre-constructed anomaly analysis model to obtain an anomaly analysis result, including:

[0028] Input each group of accompanying care feature data into the pre-constructed anomaly analysis model respectively to obtain multiple groups of anomaly analysis results, where each group of anomaly analysis results includes a normal probability and an abnormal probability.

[0029] In an embodiment of the present application, perform anomaly analysis on the vital sign data of the target object in the target time period based on the anomaly prediction result, including:

[0030] Calculate the average normal probability and the average abnormal probability in multiple groups of anomaly analysis results, and when the average abnormal probability is greater than a preset probability threshold, determine that the vital sign data of the target object in the target time period is abnormal, otherwise, determine that the vital sign data of the target object in the target time period is normal.

[0031] In an embodiment of the present application, when the vital sign data of the target object in the target time period is abnormal, send the anomaly conclusion and the accompanying care data of the target object in the target time period to the target medical staff.

[0032] In an embodiment of the present application, the vital sign data includes heart rate, blood pressure, body temperature, and respiratory rate.

[0033] In an embodiment of the present application, the environmental parameters include temperature, humidity, air quality, noise level, and air flow velocity.

[0034] The beneficial effects of the present invention are as follows: An abnormal analysis system for escort data based on AI analysis according to the present invention extracts the escort data for a target duration before the current time point, extracts the nursing entities from the escort data, and the values of the environmental parameters and the physical sign data of the target object after the nursing entity has been executed for a certain duration, and constructs escort feature data. Then, an abnormal prediction model that reflects the physical sign change law after nursing is used to perform abnormal prediction based on the escort feature data to obtain an abnormal prediction result. Finally, a comprehensive analysis of the physical sign data for the target time period is performed based on the abnormal prediction result to obtain a prediction analysis result. This application aims to find the abnormal physical sign data of the patient during the escort process by analyzing the physical sign changes that the patient should exhibit after performing various nursing operations on the patient during the escort process, so as to be able to find the abnormalities of the patient in a timely and accurate manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention will be further described below in conjunction with the drawings and embodiments:

[0036] Figure 1 is a structural diagram of an abnormal analysis system for escort data based on AI analysis shown in an embodiment of the present application;

[0037] Figure 2 is a flowchart for extracting physical sign change data in the present application;

[0038] Figure 3 is a flowchart for extracting escort entities in an embodiment of the present application;

[0039] Figure 4 is a schematic diagram of the input data construction process in an embodiment of the present application;

[0040] Figure 5 is a schematic diagram of the construction process of an abnormal prediction model in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] The following specific examples illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0042] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the layers related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape, and size of the layers during actual implementation. The type, quantity, and ratio of each layer during actual implementation can be arbitrarily changed, and the layer layout type may also be more complex.

[0043] In the following description, a large number of details are explored to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details.

[0044] Figure 1 It is a structural diagram of an abnormal analysis system for escort data based on AI analysis shown in an embodiment of the present application. As Figure 1 shown: An abnormal analysis system for escort data based on AI analysis in this embodiment includes:

[0045] An acquisition module 110, configured to acquire the escort data of a target object during a target time period, where the escort data includes the physical sign data, nursing records, and environmental parameter values at multiple time points, and the target time period is a time period with a target duration before the current time point;

[0046] Escort data is the data generated during the inpatient escort period of a patient. Generally, medical staff will regularly perform physical sign detections, nursing operations such as dressing changes / turning over on the patient. In addition, environmental sensors in the ward will also monitor the environment. Thus, escort data is formed.

[0047] In the present application, the environmental parameters include temperature, humidity, air quality, noise level, and air flow velocity. The physical sign data includes heart rate, blood pressure, body temperature, and respiratory rate.

[0048] A feature extraction module 120, configured to extract the physical sign change data and nursing entities at multiple time points from the escort data, and construct escort feature data based on the physical sign change data at the first time point, the value of the environmental parameter at the first time point, and the nursing entity at the second time point, where the first time point is later than the second time point by a unit duration;

[0049] Among them, the nursing entity is the nursing operation actually performed by medical staff on the patient, such as feeding medicine, dressing change, turning over, etc. These operations will produce certain effects in the subsequent time, thus causing certain changes in the physical signs of the patient, such as a decrease in body temperature, a decrease in blood pressure, and a stable heart rate. Therefore, it is also necessary to extract the physical sign change data. In addition, the value of the environmental parameter in the ward will also affect the physical signs of the patient. For example, when the temperature is high or the air quality is poor, the respiratory rate, heart rate, etc. of the patient will increase significantly.

[0050] Figure 2 It is a flowchart for extracting the physical sign change data in the present application. As Figure 2 shown, in an embodiment of the present application, extracting the physical sign change data at multiple time points from the escort data includes:

[0051] S210, extract the values of multiple physical sign parameters at multiple time points from the physical sign data of the accompanying care data;

[0052] Extract the values of multiple physical sign parameters at multiple time points from the physical sign data in the accompanying care data. This step involves collecting and organizing physical sign information at different time points, such as heart rate, blood pressure, etc., to prepare for subsequent analysis.

[0053] S220, perform normalization processing on the values of each physical sign parameter at multiple time points to obtain multiple normalized sequences;

[0054] Perform normalization processing on the values of each physical sign parameter at different time points. The purpose is to eliminate the influence of the dimension between different indicators so that they can be compared and analyzed on the same scale. The result after normalization is a series of normalized sequences, and each sequence corresponds to the change of a physical sign parameter over time.

[0055] S230, calculate the slope of each time point in the multiple normalized sequences based on central difference to obtain the change characteristics of multiple physical sign parameters; and construct the physical sign change data of multiple time points based on the change characteristics of multiple physical sign parameters.

[0056] Calculate the slope of each time point in these normalized sequences based on the central difference method to capture the change characteristics of the physical sign parameters over time. Central difference is a numerical differentiation method, which is suitable for estimating the derivative of a function at a certain point. The rate of change obtained by this method can reflect how quickly or slowly the physical sign parameters change over time. Finally, construct a data set about the physical sign changes of multiple time points based on these change characteristics, and this data set can provide in-depth insights into the dynamic changes of physical sign parameters.

[0057] Figure 3 This is the extraction flow chart of the accompanying care entity in an embodiment of the present application. As Figure 3 shown, in an embodiment of the present application, extracting the care entities at multiple time points from the accompanying care data includes:

[0058] S310, perform word segmentation processing on the care records in the accompanying care data to obtain multiple words;

[0059] First, perform word segmentation processing on the care records in the accompanying care data. Word segmentation refers to the process of splitting continuous text into lexical units (words or phrases) with certain semantics. In Chinese processing, since there is no obvious word boundary, specific word segmentation tools or algorithms need to be used to achieve this process. For example, the bidirectional maximum matching method.

[0060] S320, match the multiple words with a pre-built entity library to obtain care entities that match the entities in the pre-built entity library, where the care entities include medication record entities, care operation entities, and user behavior entities.

[0061] Match the multiple words obtained in the previous step with a pre-built entity library, aiming to identify key information in the care record and classify it into specific care entities. These care entities include, but are not limited to:

[0062] Medication record entity: Involves information such as drug name, dosage, administration time, etc.

[0063] Care operation entity: For example, care activities such as wound dressing change, measuring vital signs, etc.

[0064] User behavior entity: Refers to the behavior records of patients, such as diet situation, exercise amount, etc.

[0065] By matching with a pre-built entity library, structured key information can be extracted from a large amount of text information, which is convenient for subsequent data analysis and utilization. This method helps to extract valuable information from unstructured care records, improve the utilization efficiency of information, support more accurate medical decisions and the design of personalized care plans. In addition, it can also help medical staff quickly obtain important information and reduce the workload caused by manually reading and understanding a large amount of text.

[0066] In addition, Figure 4 For the input data construction process schematic diagram in an embodiment of the present application, as Figure 4 shown, the present application needs to extract data at different time points to construct input data, including:

[0067] S410, extract the second time point of each care record from the accompanying care data;

[0068] The second time point here refers to the exact time point when the care behavior occurs, such as the administration time or the time when a certain care operation is performed.

[0069] S420, determine the first time point based on the second time point and the unit duration, and extract the vital sign change data and the values of environmental parameters at the first time point from the accompanying care data;

[0070] Based on the second time point and the set unit duration (e.g., half an hour before and after the nursing behavior), one or more "first time points" are determined. These first time points are set to capture the patient's state before or after a specific nursing behavior. Then, the sign change data (such as the change trend of heart rate, blood pressure, etc.) and the values of environmental parameters (such as ward temperature, humidity, etc.) at these first time points are extracted from the accompanying care data. This step aims to collect data that can reflect the impact of nursing behavior on the patient's signs.

[0071] S430, vectorize the sign change data of multiple groups of first time points, the values of environmental parameters at the first time points, and the nursing entities at the second time point to obtain multiple groups of accompanying care feature data.

[0072] Vectorize the sign change data, the values of environmental parameters, and the corresponding nursing entities at the second time point obtained from multiple groups of first time points. This means converting these different types of data into a unified mathematical expression form (such as a vector) for subsequent analysis and model training. The final result is to generate multiple groups of accompanying care feature data, which can be used for various applications, such as predicting the effect of nursing behavior, optimizing the nursing plan, or identifying potential health risks, etc.

[0073] This process comprehensively considers the time dimension of nursing behavior, the changes in the patient's physiological state, and the surrounding environmental factors, providing strong support for in-depth understanding of the nursing effect and its influencing factors.

[0074] The AI prediction module 130 is used to input the accompanying care feature data into a pre-constructed anomaly prediction model to obtain an anomaly prediction result, where the anomaly analysis model characterizes the changes in sign data after implementing different nursing entities on the patient under various environmental conditions;

[0075] In this application, an AI model is used to characterize the sign change characteristics of the patient after various nursing operations under various environmental conditions. Figure 5 It is a schematic diagram of the construction process of the anomaly prediction model in an embodiment of this application, as Figure 5 shown, the construction process of the anomaly prediction model includes:

[0076] S510, obtain multiple accompanying care sample data, where the accompanying care sample data includes a set of sign change sample data, sample values of environmental parameters, and nursing entities;

[0077] S520, vectorize and label the accompanying care sample data to obtain the label of each accompanying care sample data, where the label is the normal probability and the abnormal probability;

[0078] Convert the physical sign changes, environmental parameters, and care entities in the companion sample data into a unified mathematical representation (such as a vector). This step is to enable the data to be processed by machine learning algorithms or artificial neural networks. Assign a label to each companion sample data, where the label represents the normal probability and abnormal probability corresponding to the sample data. For example: If a certain sample data reflects that the patient's physical signs and environment are within the normal range, its label may be "Normal probability: 60%, Abnormal probability 40%".

[0079] If a certain sample data reflects abnormal conditions in the patient's physical signs or environment (such as too high heart rate or too low environmental temperature), its label may be "Normal probability 30%, Abnormal probability 70%".

[0080] S530, construct training data based on the vectorized companion sample data and labels to obtain a training data set;

[0081] S540, train an artificial neural network based on the training data set to obtain an anomaly analysis model.

[0082] During the training process, divide the training data set into a training set and a validation set. Use the training set to iteratively train the neural network and optimize the network parameters to minimize the prediction error. Use the validation set to evaluate the model performance and avoid overfitting.

[0083] Specifically, input each group of companion feature data into a pre-constructed anomaly analysis model respectively to obtain multiple groups of anomaly analysis results, where each group of anomaly analysis results includes a normal probability and an abnormal probability. For example, Normal probability: 35%, Abnormal probability 65%.

[0084] The anomaly analysis module 140 is used to perform anomaly analysis on the physical sign data of the target object in the target time period based on the anomaly prediction results.

[0085] For example: Calculate the average normal probability and average abnormal probability in multiple groups of anomaly analysis results, and when the average abnormal probability is greater than a preset probability threshold, determine that the physical sign data of the target object in the target time period is abnormal, otherwise, determine that the physical sign data of the target object in the target time period is normal.

[0086] When the physical sign data of the target object in the target time period is abnormal, send the anomaly conclusion and the companion data of the target object in the target time period to the target medical staff.

[0087] An escort data anomaly analysis system based on AI analysis according to the present invention extracts, from the escort data of a target duration before the current time point, the care entities, the values of the environmental parameters after the care entities have been executed for a certain duration, and the physical sign data of the target object, and constructs escort feature data. Then, an anomaly prediction model that reflects the physical sign change pattern after care is used to perform anomaly prediction based on the escort feature data to obtain an anomaly prediction result. Finally, a comprehensive analysis of the physical sign data for the target time period is performed based on the anomaly prediction result to obtain a prediction analysis result. This application aims to find the abnormal physical sign data of a patient during the escort process by analyzing the physical sign changes that the patient should exhibit after various care operations are performed on the patient during the escort process, so as to be able to find the abnormalities of the patient in a timely and accurate manner.

[0088] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements any one of the methods in this embodiment, where the method is the execution logic of this system.

[0089] This embodiment also provides an electronic terminal, including: a processor and a memory;

[0090] The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the terminal executes any one of the methods in this embodiment.

[0091] For the computer-readable storage medium in this embodiment, those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to a computer program. The foregoing computer program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disk, or optical disc that can store program codes.

[0092] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store a computer program, the communication interface is used for communication, and the processor and the transceiver are used to run the computer program so that the electronic terminal executes each step of the above method.

[0093] In this embodiment, the memory may include a random access memory (Random Access Memory, abbreviated as RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0094] The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0095] In the above embodiments, although the present invention has been described in conjunction with specific embodiments of the present invention, many substitutions, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art based on the foregoing description. The embodiments of the present invention are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims.

[0096] The above embodiments are only illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those of ordinary skill in the art in the technical field without departing from the spirit and technical idea disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. An abnormal analysis system for escort data based on AI analysis, characterized in that, Including: An acquisition module, configured to acquire the accompanying care data of a target object in a target time period, where the accompanying care data includes physical sign data, nursing records, and environmental parameter values at multiple time points, and the target time period is a time period of a target duration before the current time point; A feature extraction module, configured to extract physical sign change data and nursing entities at multiple time points from the accompanying care data, and construct accompanying care feature data based on the physical sign change data at a first time point, the environmental parameter value at the first time point, the nursing entity at a second time point, where the first time point is later than the second time point by a unit duration; An AI prediction module, configured to input the accompanying care feature data into a pre-constructed anomaly prediction model to obtain an anomaly prediction result, where the anomaly analysis model characterizes the change of physical sign data after different nursing entities are implemented on a patient under various environmental conditions; An anomaly analysis module, configured to perform anomaly analysis on the physical sign data of the target object in the target time period based on the anomaly prediction result.

2. The abnormal analysis system for escort data based on AI analysis according to claim 1, wherein Extracting physical sign change data at multiple time points from the accompanying care data includes: Extracting values of multiple physical sign parameters at multiple time points from the physical sign data of the accompanying care data; Performing normalization processing on the values of each physical sign parameter at multiple time points to obtain multiple normalized sequences; Calculating the slope of each time point in the multiple normalized sequences based on central difference to obtain the change characteristics of multiple physical sign parameters; and constructing physical sign change data at multiple time points based on the change characteristics of multiple physical sign parameters.

3. The abnormal analysis system for escort data based on AI analysis according to claim 1, characterized in that Extracting nursing entities at multiple time points from the accompanying care data includes: Performing word segmentation on the nursing records in the accompanying care data to obtain multiple words; Matching the multiple words with a pre-constructed entity library to obtain nursing entities that match the entities in the pre-constructed entity library, where the nursing entities include medication record entities, nursing operation entities, and user behavior entities.

4. An abnormal analysis system for escort data based on AI analysis according to claim 1, characterized in that, Constructing accompanying care feature data based on the physical sign change data at the first time point, the environmental parameter value at the first time point, and the nursing entity at the second time point includes: Extracting the second time point of each nursing record from the accompanying care data; Determining the first time point based on the second time point and the unit duration, and extracting the physical sign change data and environmental parameter value at the first time point from the accompanying care data; Performing vectorization processing on multiple groups of physical sign change data at the first time point, environmental parameter values at the first time point, and nursing entities at the second time point to obtain multiple groups of accompanying care feature data.

5. The abnormal analysis system for escort data based on AI analysis according to claim 1, characterized in that, The construction method of the anomaly analysis model includes: Obtaining multiple accompanying care sample data, where the accompanying care sample data includes a set of physical sign change sample data, sample values of environmental parameters, and nursing entities; Performing vectorization and annotation on the accompanying care sample data to obtain the label of each accompanying care sample data, where the label is a normal probability and an abnormal probability; Constructing training data based on the vectorized accompanying care sample data and the label to obtain a training data set; Training an artificial neural network based on the training data set to obtain an anomaly analysis model.

6. The abnormal analysis system for escort data based on AI analysis according to claim 4, characterized in that, Input the accompanying care feature data into a pre-constructed anomaly analysis model to obtain an anomaly analysis result, including: Input each group of accompanying care feature data into a pre-constructed anomaly analysis model respectively to obtain multiple groups of anomaly analysis results, where each group of anomaly analysis results includes a normal probability and an abnormal probability.

7. An abnormal analysis system for escort data based on AI analysis according to claim 6, characterized in that, Based on the anomaly prediction result, conduct an anomaly analysis on the physical sign data of the target object during the target time period, including: Calculate the average normal probability and the average abnormal probability in multiple groups of anomaly analysis results, and when the average abnormal probability is greater than a preset probability threshold, determine that the physical sign data of the target object during the target time period is abnormal; otherwise, determine that the physical sign data of the target object during the target time period is normal.

8. An abnormal analysis system for escort data based on AI analysis according to claim 7, characterized in that, When the physical sign data of the target object during the target time period is abnormal, send the anomaly conclusion and the accompanying care data of the target object during the target time period to the target medical staff.

9. An abnormal analysis system for escort data based on AI analysis according to claim 1, characterized in that, The physical sign data includes heart rate, blood pressure, body temperature, and respiratory rate.

10. The abnormal analysis system for escort data based on AI analysis according to claim 1, wherein, The environmental parameters include temperature, humidity, air quality, noise level, and air flow velocity.

Images