A method and system for analyzing intensive care for stroke based on prediction model

Abstract

The present invention discloses a method and system for analyzing intensive care of stroke based on a prediction model, and belongs to the field of medical care. The present invention substitutes acquired brain CT image data and neurological function score data into a severity prediction model to predict severity, imports acquired physical characteristic data of stroke patients into a physical constitution prediction model to predict physical constitution influence, imports acquired severity prediction results and physical constitution influence prediction results into a monitoring level analysis model to output monitoring level analysis, accurately analyzes the severity of intensive care of stroke through brain CT image data and neurological function score data, and simultaneously accurately analyzes the patient's monitoring level by integrating the patient's physical characteristic data, thereby improving the accuracy of monitoring analysis and the monitoring efficiency.

Description

Technical Field

[0001] The invention belongs to the field of medical care, and specifically relates to a method and system for analyzing intensive care for stroke based on a prediction model. Background Art

[0002] Stroke intensive care is a special care model for acute stroke patients, especially those with severe conditions, life-threatening conditions or high-risk complications. Stroke intensive care refers to the process of close monitoring, treatment and care of stroke patients in the hospital intensive care unit (ICU). Since stroke may cause severe neurological dysfunction and life-threatening conditions, intensive care is crucial to improving patient survival and prognosis.

[0003] In the process of intensive care for stroke, the existing technology cannot perform evolution analysis on brain CT image data and neurological function score data, which leads to the inability to analyze the severity of stroke according to the evolution process of brain CT image data and neurological function score data. At the same time, it is impossible to comprehensively analyze the patient's physical characteristics data to accurately analyze the patient's monitoring level, resulting in low accuracy and monitoring efficiency of monitoring analysis. Most of the existing technologies have the above problems;

[0004] For example, a Chinese patent with application publication number CN107707606A discloses a stroke emergency screening and supervision medical networking system and its application method, which includes a stroke screening Internet of Things system, a stroke emergency command and dispatch management system, a stroke emergency Internet of Things system, an intensive care system, and a stroke emergency screening and supervision information system. This invention focuses on the screening, prevention, treatment and health management of stroke. By building a medical and health cooperation platform, using the medical Internet of Things as the main means to check underlying diseases, and under the guidance of medical experts, a new model of precision medicine and health management is constructed to actively screen stroke patients. Hospitals that receive emergency requests respond to emergency needs in a timely manner, follow up on subsequent treatment measures and guide subsequent treatment, thereby rationally responding to major public health threats brought about by stroke, improving the overall treatment efficiency and level of patients, and reducing the incidence, recurrence, disability and mortality of stroke;

[0005] The above patents all have the problems raised by this background technology. In order to solve the problems raised by this background technology, this application designs a stroke intensive care analysis method and system based on a prediction model. Summary of the invention

[0006] In view of the deficiencies of the prior art, the present invention proposes a method and system for analyzing intensive care for stroke based on a prediction model.

[0007] To achieve the above object, the present invention provides the following technical solution: a method for analyzing intensive care for stroke based on a prediction model, which comprises the following specific steps:

[0008] S1. Obtain physical characteristics data, brain CT image data and neurological function score data of stroke patients;

[0009] S2, judging whether stroke monitoring is needed by obtaining the brain CT image data and the neurological function score data, if the judgment is yes, proceed to S3, if the judgment is no, end;

[0010] S3, substituting the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction;

[0011] S4, importing the acquired physical characteristics data of stroke patients into the physical constitution prediction model to predict the impact of physical constitution;

[0012] S5. Import the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis output.

[0013] It should be noted here that as a preferred technical solution of a stroke intensive care analysis method and system based on a prediction model, the specific steps of obtaining the physical characteristic data, brain CT image data and neurological function score data of stroke patients are:

[0014] S11. Collecting physical characteristic data of stroke patients within a set period through a physical data collection terminal, wherein the physical characteristic data of the patients include body temperature data, blood pressure data and heart rate data, and storing them in a storage module, wherein the set period is the duration of this hospitalization;

[0015] S12, acquiring all brain CT images of stroke patients within a set period through an image data acquisition terminal, and storing them in a storage module;

[0016] S13. Collect all neurological function score data of stroke patients within a set period and store them in a storage module, wherein the neurological function score data is score data corresponding to the neurological function score table NIHSS.

[0017] It should be noted that, as a preferred technical solution of a stroke intensive care analysis method and system based on a prediction model, the step of determining whether stroke monitoring is required by obtaining the brain CT image data and neurological function score data in S2 includes the following specific steps:

[0018] S21, obtaining all brain CT image data of the patient within a set period, obtaining the last brain CT image data of the patient within the set period, and importing the obtained last brain CT image data into the first image abnormality value calculation formula to calculate the first image abnormality value, wherein the first image abnormality value calculation formula of the last brain CT image is: , where m is the number of pixels in the specified part of the brain CT image, xi is the pixel value of the i-th pixel in the brain CT image, and xiz is the pixel value of the i-th pixel in the brain CT image when no stroke occurs;

[0019] S22, obtaining all brain CT image data obtained within a set period, importing all of them into a first image abnormality value calculation formula to calculate first image abnormality values ​​of all brain CT images obtained within the set period, obtaining a phase difference value of the first image abnormality values ​​of two adjacent brain CT images, importing the first image abnormality value of the last brain CT image and the phase difference value of the first image abnormality values ​​of all two adjacent brain CT images into a second image abnormality value calculation formula to calculate a second image abnormality value, wherein the second image abnormality value calculation formula is: , where a is the first proportion coefficient, N is the total number of all brain CT images taken within the set period, Xt(j+1) is the first image abnormality value of the j+1th brain CT image within the set period, Xtj is the first image abnormality value of the jth brain CT image within the set period, and t(j+1) is the time from the acquisition time of the j+1th brain CT image to the end of the set period;

[0020] S23, obtaining all neurological function scoring data obtained within a set period, calculating the average value of all neurological function scoring data, and obtaining the average neurological function scoring value within the set period;

[0021] S24. Compare the obtained average neurological function score with the set average neurological function score threshold, and compare the calculated second image abnormality value with the set second image abnormality threshold. If the average neurological function score is greater than or equal to the set average neurological function score threshold and / or the second image abnormality value is greater than or equal to the set second image abnormality threshold, it is determined that stroke monitoring is required; if the average neurological function score is less than the set average neurological function score threshold and the second image abnormality value is less than the set second image abnormality threshold, it is determined that stroke monitoring is not required.

[0022] It should be noted that, as a preferred technical solution of a stroke intensive care analysis method and system based on a prediction model, the step of substituting the acquired brain CT image data and neurological function score data into a severity prediction model for severity prediction includes the following specific steps:

[0023] S31, obtaining the calculated second image abnormality value and neurological function score average value of the stroke monitoring patient;

[0024] S32, substituting the acquired abnormal value of the second image of the stroke monitoring patient and the average value of the neurological function score into the severity calculation formula to calculate the severity of the stroke monitoring patient, wherein the severity calculation formula is: , where c is the image abnormality ratio, is the average value of neurological function score, and W is the standard value of neurological function score.

[0025] It should be noted that, as a preferred technical solution of a stroke intensive care analysis method and system based on a prediction model, the method of importing the acquired physical characteristics data of stroke patients into a physical constitution prediction model to predict the impact of physical constitution includes the following specific contents:

[0026] The physical characteristic data of the stroke monitoring patient within the set period are obtained, and the physical characteristic data of the stroke monitoring patient within the set period are imported into the physical constitution influence value calculation formula to calculate the physical constitution influence value, wherein the physical constitution influence value calculation formula is: , where G is the number of types of physical characteristic data, T is the set cycle duration, sg is the proportion coefficient of the g-th physical characteristic data, pgt is the specific value of the g-th physical characteristic data at time t, and pgtm is the median value of the safety range of the g-th physical characteristic data;

[0027] It should be noted that as a preferred technical solution of a stroke intensive care analysis method and system based on a prediction model, the method of importing the obtained severity prediction results and physical constitution influence prediction results into the monitoring level analysis model to perform monitoring level analysis output includes the following specific contents:

[0028] S51. Obtain the calculated severity and the calculated physical constitution impact value, and substitute the calculated severity and the calculated physical constitution impact value into the monitoring value calculation formula to calculate the monitoring value, wherein the monitoring value calculation formula is: , where exp() is the power of the natural constant e;

[0029] S52. Divide the calculated monitoring value by the set monitoring threshold to obtain the monitoring abnormality value. If the monitoring abnormality value is less than or equal to 1, perform level 3 monitoring; if the monitoring abnormality value is greater than 1 and less than or equal to 3, perform level 2 monitoring; if the monitoring abnormality value is greater than 3, perform level 1 monitoring.

[0030] A stroke intensive care analysis system based on a prediction model, which is implemented based on the above-mentioned stroke intensive care analysis method based on a prediction model, and specifically includes a data acquisition module, a monitoring judgment module, a severity prediction module, a physical condition influence prediction module and a monitoring level analysis output module, wherein the data acquisition module is used to obtain physical characteristic data, brain CT image data and neurological function score data of stroke patients;

[0031] The monitoring judgment module is used to determine whether stroke monitoring is needed by acquiring the brain CT image data and the neurological function score data;

[0032] The severity prediction module is used to substitute the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction;

[0033] The physical constitution influence prediction module is used to import the acquired physical constitution characteristic data of the stroke patient into the physical constitution prediction model to perform physical constitution influence prediction;

[0034] The monitoring level analysis output module is used to import the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis and output.

[0035] An electronic device comprises: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0036] The processor executes the above-mentioned stroke intensive care analysis method based on the prediction model by calling the computer program stored in the memory.

[0037] A computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned stroke intensive care analysis method based on a prediction model.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention obtains physical characteristic data, brain CT image data and neurological function score data of a stroke patient, determines whether stroke monitoring is needed by obtaining the obtained brain CT image data and neurological function score data, substitutes the obtained brain CT image data and neurological function score data into a severity prediction model to perform severity prediction, imports the obtained physical characteristic data of the stroke patient into a physical constitution prediction model to perform physical constitution influence prediction, imports the obtained severity prediction result and physical constitution influence prediction result into a monitoring level analysis model to perform monitoring level analysis output, performs evolution analysis on the brain CT image data and the neurological function score data, and then accurately analyzes the severity of severe stroke through the evolution process of the brain CT image data and the neurological function score data, and simultaneously comprehensively analyzes the patient's physical characteristic data to accurately analyze the patient's monitoring level, thereby improving the accuracy of monitoring analysis and the monitoring efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a schematic diagram of the overall process of a stroke intensive care analysis method based on a prediction model of the present invention;

[0041] Figure 2 This is a schematic flow chart of step S2 of a stroke intensive care analysis method based on a prediction model of the present invention;

[0042] Figure 3 It is a schematic diagram of the overall framework of a stroke intensive care analysis system based on a prediction model of the present invention;

[0043] Figure 4 It is a schematic diagram of the connection between the information collection terminal and the medical display terminal of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present application and its application or use.

[0045] In order to solve the technical problems raised in the background technology, the present invention provides a preferred embodiment: Figure 4 As shown, the implementation scenario of this embodiment is demonstrated, and the implementation scenario is: the information collection terminal collects stroke patient information and transmits the patient information to the server, and the server analyzes and calculates the patient's patient monitoring level and displays it on the medical display terminal;

[0046] The specific contents of this embodiment are:

[0047] like Figure 1-Figure 2 As shown, a method and system for analyzing intensive care for stroke based on a prediction model include the following specific steps:

[0048] S1. Obtain physical characteristics data, brain CT image data and neurological function score data of stroke patients;

[0049] In this embodiment, the specific steps of obtaining the physical characteristic data, brain CT image data and neurological function score data of stroke patients are as follows:

[0050] S11. Collecting physical characteristic data of stroke patients within a set period through a physical data collection terminal, wherein the physical characteristic data of the patients include body temperature data, blood pressure data and heart rate data, and storing them in a storage module, wherein the set period is the duration of this hospitalization;

[0051] S12, acquiring all brain CT images of stroke patients within a set period through an image data acquisition terminal, and storing them in a storage module;

[0052] S13. Collect all neurological function score data of stroke patients within a set period and store them in a storage module, wherein the neurological function score data is the score data corresponding to the neurological function score table NIHSS, wherein the neurological function score table NIHSS is a scale widely used to evaluate the degree of neurological function impairment in stroke patients. The following is some key information about the NIHSS score table:

[0053] definition

[0054] NIHSS is a standardized, quantitative assessment tool used to quickly evaluate the patient's neurological status after a stroke.

[0055] Purpose

[0056] To assess the initial neurological deficit in stroke patients;

[0057] Monitor changes in neurological function during stroke treatment;

[0058] Predict the patient's clinical prognosis;

[0059] The NIHSS rating scale covers 11 assessment items, including:

[0060] 1. Neurological dysfunction (including motor and sensory function).

[0061] 2. Level of consciousness (including eye movements and speech ability).

[0062] 3. Language function.

[0063] 4. Cranial nerve function (such as facial paralysis, visual field loss).

[0064] 5. Regulatory abilities (e.g., gait, balance).

[0065] Rating Range

[0066] Scores range from 0 to 42, with higher scores indicating more severe neurological deficits;

[0067] A score of 0 indicates no neurological deficit;

[0068] The higher the score, the more severe the neurological damage and the worse the prognosis;

[0069] In conclusion, the NIHSS score sheet is a simple and effective tool that is of great significance for assessing the degree of neurological deficit and prognosis of stroke patients;

[0070] S2, judging whether stroke monitoring is needed by obtaining the brain CT image data and the neurological function score data, if the judgment is yes, proceed to S3, if the judgment is no, end;

[0071] In this embodiment, determining whether stroke monitoring is needed by acquiring the brain CT image data and the neurological function score data in S2 includes the following specific steps:

[0072] S21, obtaining all brain CT image data of the patient within a set period, obtaining the last brain CT image data of the patient within the set period, and importing the obtained last brain CT image data into the first image abnormality value calculation formula to calculate the first image abnormality value, wherein the first image abnormality value calculation formula of the last brain CT image is: , where m is the number of pixels in the specified part of the brain CT image, xi is the pixel value of the i-th pixel in the brain CT image, and xiz is the pixel value of the i-th pixel in the brain CT image when no stroke occurs;

[0073] S22, obtaining all brain CT image data obtained within a set period, importing all of them into a first image abnormality value calculation formula to calculate first image abnormality values ​​of all brain CT images obtained within the set period, obtaining a phase difference value of the first image abnormality values ​​of two adjacent brain CT images, importing the first image abnormality value of the last brain CT image and the phase difference value of the first image abnormality values ​​of all two adjacent brain CT images into a second image abnormality value calculation formula to calculate a second image abnormality value, wherein the second image abnormality value calculation formula is: , where a is the first proportion coefficient, N is the total number of all brain CT images taken within the set period, Xt(j+1) is the first image abnormality value of the j+1th brain CT image within the set period, Xtj is the first image abnormality value of the jth brain CT image within the set period, and t(j+1) is the time from the acquisition time of the j+1th brain CT image to the end of the set period;

[0074] S23, obtaining all neurological function scoring data obtained within a set period, calculating the average value of all neurological function scoring data, and obtaining the average neurological function scoring value within the set period;

[0075] S24, comparing the obtained average neurological function score with the set average neurological function score threshold, and comparing the calculated second image abnormality value with the set second image abnormality threshold; if the average neurological function score is greater than or equal to the set average neurological function score threshold and / or the second image abnormality value is greater than or equal to the set second image abnormality threshold, it is determined that stroke monitoring is required; if the average neurological function score is less than the set average neurological function score threshold and the second image abnormality value is less than the set second image abnormality threshold, it is determined that stroke monitoring is not required;

[0076] S3, substituting the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction;

[0077] In this embodiment, substituting the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction includes the following specific steps:

[0078] S31, obtaining the calculated second image abnormality value and neurological function score average value of the stroke monitoring patient;

[0079] S32, substituting the acquired abnormal value of the second image of the stroke monitoring patient and the average value of the neurological function score into the severity calculation formula to calculate the severity of the stroke monitoring patient, wherein the severity calculation formula is: , where c is the image abnormality ratio, is the mean value of neurological function score, and W is the standard value of neurological function score;

[0080] S4, importing the acquired physical characteristics data of stroke patients into the physical constitution prediction model to predict the impact of physical constitution;

[0081] In this embodiment, the acquired physical characteristics data of stroke patients are imported into the physical constitution prediction model to predict the physical constitution influence, including the following specific contents:

[0082] The physical characteristic data of the stroke monitoring patient within the set period are obtained, and the physical characteristic data of the stroke monitoring patient within the set period are imported into the physical constitution influence value calculation formula to calculate the physical constitution influence value, wherein the physical constitution influence value calculation formula is: , where G is the number of types of physical characteristic data, T is the set cycle duration, sg is the proportion coefficient of the g-th physical characteristic data, pgt is the specific value of the g-th physical characteristic data at time t, and pgtm is the median value of the safety range of the g-th physical characteristic data;

[0083] S5. Importing the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis output;

[0084] In this embodiment, the obtained severity prediction results and physical constitution impact prediction results are introduced into the monitoring level analysis model to analyze and output the monitoring level, including the following specific contents:

[0085] S51. Obtain the calculated severity and the calculated physical constitution impact value, and substitute the calculated severity and the calculated physical constitution impact value into the monitoring value calculation formula to calculate the monitoring value, wherein the monitoring value calculation formula is: , where exp() is the power of the natural constant e;

[0086] S52, dividing the calculated monitoring value by the set monitoring threshold value to obtain the monitoring abnormality value. If the monitoring abnormality value is less than or equal to 1, perform level 3 monitoring; if the monitoring abnormality value is greater than 1 and less than or equal to 3, perform level 2 monitoring; if the monitoring abnormality value is greater than 3, perform level 1 monitoring;

[0087] Finally, implement the appropriate level of supervision:

[0088] Level 3 stroke monitoring: Applicable to patients with low abnormal monitoring values, requiring basic monitoring measures, such as general vital signs monitoring and basic nursing care;

[0089] Level II stroke monitoring: for patients with moderate abnormal monitoring values ​​who require more monitoring and treatment, such as more frequent vital sign monitoring, medication adjustments, and early rehabilitation measures;

[0090] Level 1 stroke monitoring: for patients with higher abnormal monitoring values, requiring the most stringent monitoring and treatment, which may include advanced life support measures, emergency medical interventions, and ongoing therapeutic management;

[0091] It should be specifically explained here that the first proportion coefficient, the second image abnormality threshold, the average threshold of the neurological function score, the image abnormality proportion, the proportion coefficient of the g-th physical characteristic data and the monitoring threshold in this embodiment are determined as follows: at least 5000 groups of physical characteristic data, brain CT image data and neurological function score data of stroke patients are obtained, experts are hired to classify the risk level of stroke patients, the physical characteristic data, brain CT image data and neurological function score data of stroke patients are substituted into the calculation of the monitoring level, and the monitoring level and classification results are imported into the fitting software to output the first proportion coefficient, the second image abnormality threshold, the average threshold of the neurological function score, the image abnormality proportion, the proportion coefficient of the g-th physical characteristic data and the monitoring threshold that meet the highest classification judgment accuracy.

[0092] It should be noted in this embodiment that the advantages of this embodiment over the prior art are: obtaining physical characteristic data, brain CT image data and neurological function score data of stroke patients, judging whether stroke monitoring is needed by obtaining the obtained brain CT image data and neurological function score data, substituting the obtained brain CT image data and neurological function score data into the severity prediction model for severity prediction, importing the obtained physical characteristic data of stroke patients into the physical constitution prediction model for physical constitution influence prediction, importing the obtained severity prediction results and physical constitution influence prediction results into the monitoring level analysis model for monitoring level analysis output, performing evolution analysis on brain CT image data and neurological function score data, and then accurately analyzing the severity of stroke through the evolution process of brain CT image data and neurological function score data, and at the same time comprehensively analyzing the patient's physical characteristic data to accurately analyze the patient's monitoring level, thereby improving the accuracy of monitoring analysis and the monitoring efficiency.

[0093] This embodiment also provides a stroke intensive care analysis system based on a prediction model, which is implemented based on the above-mentioned stroke intensive care analysis method based on a prediction model. Figure 3 As shown, it specifically includes a data acquisition module, a monitoring judgment module, a severity prediction module, a physical influence prediction module and a monitoring level analysis output module, wherein the data acquisition module is used to obtain physical characteristic data, brain CT image data and neurological function score data of stroke patients;

[0094] A monitoring judgment module is used to determine whether stroke monitoring is needed based on the acquired brain CT image data and neurological function score data;

[0095] A severity prediction module is used to substitute the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction;

[0096] A physical constitution influence prediction module is used to import the acquired physical constitution characteristic data of stroke patients into a physical constitution prediction model to predict physical constitution influence;

[0097] The monitoring level analysis output module is used to import the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis and output.

[0098] This embodiment also provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0099] The processor executes the above-mentioned stroke intensive care analysis method based on the prediction model by calling the computer program stored in the memory.

[0100] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors and one or more memories, wherein the memory stores at least one computer program, and the computer program is loaded and executed by the processor to implement a stroke intensive care analysis method based on a prediction model provided in the above method embodiment. The electronic device may also include other components for realizing the functions of the device. For example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here;

[0101] This embodiment also provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0102] When the computer program runs on a computer device, the computer device is enabled to execute the above-mentioned stroke intensive care analysis method based on a prediction model.

[0103] For example, the computer readable storage medium can be a read-only memory, a random access memory, a read-only CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0104] The above embodiments can be implemented in whole or in part by software, hardware, 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 a computer instruction or computer program is loaded or executed on a computer, a process or function according to an 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 device. 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, computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or / and a wireless network. 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.

Claims

1. A method for analyzing intensive care for stroke based on a prediction model, characterized in that: It includes the following specific steps: S1. Obtain physical characteristics data, brain CT image data and neurological function score data of stroke patients; S2, judging whether stroke monitoring is needed by obtaining the brain CT image data and the neurological function score data, if the judgment is yes, proceed to S3, if the judgment is no, end; S3, substituting the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction; The specific steps include: S31, obtaining the calculated second image abnormality value and neurological function score average value of the stroke monitoring patient; S32, substituting the acquired abnormal value of the second image of the stroke monitoring patient and the average value of the neurological function score into the severity calculation formula to calculate the severity of the stroke monitoring patient, wherein the severity calculation formula is: , where c is the image abnormality ratio, is the average value of the neurological function score, W is the standard value of the neurological function score, a is the first proportion coefficient, N is the total number of all brain CT images taken within the set period, Xt(j+1) is the first image abnormality value of the j+1th brain CT image within the set period, Xtj is the first image abnormality value of the jth brain CT image within the set period, t(j+1) is the time from the acquisition time of the j+1th brain CT image to the end of the set period, and Xtm is the first image abnormality value of the last brain CT image; S4, importing the acquired physical characteristics data of stroke patients into the physical constitution prediction model to predict the impact of physical constitution; The specific contents include: The physical characteristic data of the stroke monitoring patient within the set period of the last brain care are obtained, and the physical characteristic data of the stroke monitoring patient within the set period are imported into the physical constitution influence value calculation formula to calculate the physical constitution influence value, wherein the physical constitution influence value calculation formula is: , where G is the number of types of physical characteristic data, T is the set cycle duration, sg is the proportion coefficient of the g-th physical characteristic data, pgt is the specific value of the g-th physical characteristic data at time t, and pgtm is the median value of the safety range of the g-th physical characteristic data; S5. Import the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis output.

2. A stroke intensive care analysis method based on a prediction model as claimed in claim 1, characterized in that: Determining whether stroke monitoring is required by acquiring the brain CT image data and the neurological function score data in S2 includes the following specific steps: S21, obtaining all brain CT image data of the patient within a set period, obtaining the last brain CT image data of the patient within the set period, and importing the obtained last brain CT image data into the first image abnormality value calculation formula to calculate the first image abnormality value, wherein the first image abnormality value calculation formula of the last brain CT image is: , where m is the number of pixels in the specified part of the brain CT image, xi is the pixel value of the i-th pixel in the brain CT image, and xiz is the pixel value of the i-th pixel in the brain CT image when no stroke occurs; S22, obtaining all brain CT image data obtained within a set period, importing all of them into a first image abnormality value calculation formula to calculate first image abnormality values ​​of all brain CT images obtained within the set period, obtaining a phase difference value of the first image abnormality values ​​of two adjacent brain CT images, importing the first image abnormality value of the last brain CT image and the phase difference value of the first image abnormality values ​​of all two adjacent brain CT images into a second image abnormality value calculation formula to calculate a second image abnormality value, wherein the second image abnormality value calculation formula is: , where a is the first proportion coefficient, N is the total number of all brain CT images taken within the set period, Xt(j+1) is the first image abnormality value of the j+1th brain CT image within the set period, Xtj is the first image abnormality value of the jth brain CT image within the set period, and t(j+1) is the time from the acquisition time of the j+1th brain CT image to the end of the set period; S23, obtaining all neurological function scoring data obtained within a set period, calculating the average value of all neurological function scoring data, and obtaining the average neurological function scoring value within the set period; S24. Compare the obtained average neurological function score with the set average neurological function score threshold, and compare the calculated second image abnormality value with the set second image abnormality threshold. If the average neurological function score is greater than or equal to the set average neurological function score threshold and / or the second image abnormality value is greater than or equal to the set second image abnormality threshold, it is determined that stroke monitoring is required; if the average neurological function score is less than the set average neurological function score threshold and the second image abnormality value is less than the set second image abnormality threshold, it is determined that stroke monitoring is not required.

3. A method for analyzing intensive care for stroke based on a prediction model as claimed in claim 2, characterized in that: The step of importing the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis output includes the following specific contents: S51. Obtain the calculated severity and the calculated physical constitution impact value, and substitute the calculated severity and the calculated physical constitution impact value into the monitoring value calculation formula to calculate the monitoring value, wherein the monitoring value calculation formula is: , where exp() is the power of the natural constant e; S52. Divide the calculated monitoring value by the set monitoring threshold to obtain the monitoring abnormality value. If the monitoring abnormality value is less than or equal to 1, perform level 3 monitoring; if the monitoring abnormality value is greater than 1 and less than or equal to 3, perform level 2 monitoring; if the monitoring abnormality value is greater than 3, perform level 1 monitoring.

4. A method for analyzing intensive care for stroke based on a prediction model as claimed in claim 3, characterized in that: The specific steps of obtaining the physical characteristic data, brain CT image data and neurological function score data of the stroke patient are as follows: S11, collecting physical characteristic data of stroke patients within a set period through a physical data collection terminal, wherein the physical characteristic data of the patients include body temperature data, blood pressure data and heart rate data, and storing them in a storage module; S12, acquiring all brain CT images of stroke patients within a set period through an image data acquisition terminal, and storing them in a storage module; S13. Collect all neurological function score data of stroke patients within a set period and store them in a storage module, wherein the neurological function score data is score data corresponding to the neurological function score table NIHSS.

5. A stroke intensive care analysis system based on a prediction model, which is implemented based on the stroke intensive care analysis system based on a prediction model according to any one of claims 1 to 3, characterized in that: It specifically includes a data acquisition module, a monitoring judgment module, a severity prediction module, a physical influence prediction module and a monitoring level analysis output module, wherein the data acquisition module is used to obtain physical characteristic data, brain CT image data and neurological function score data of stroke patients; The monitoring judgment module is used to determine whether stroke monitoring is needed by acquiring the brain CT image data and the neurological function score data; The severity prediction module is used to substitute the acquired brain CT image data and neurological function score data into the severity prediction model to perform severity prediction; The physical constitution influence prediction module is used to import the acquired physical constitution characteristic data of the stroke patient into the physical constitution prediction model to perform physical constitution influence prediction; The monitoring level analysis output module is used to import the obtained severity prediction results and physical constitution impact prediction results into the monitoring level analysis model to perform monitoring level analysis and output.

6. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; It is characterized in that the processor executes the stroke intensive care analysis method based on the prediction model as described in any one of claims 1-3 by calling the computer program stored in the memory.

7. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute a stroke intensive care analysis method based on a prediction model as described in any one of claims 1 to 3.

Images