Exosome-based muscular tissue injury hemostasis repair capability data analysis system and method

Abstract

The invention relates to the technical field of data analysis, and discloses an exosome-based muscular tissue injury hemostasis repair capability data analysis system and method. According to the method, experimental animals in the same batch are collected, different control groups are set, meanwhile, a fluorescence microscope is used for monitoring the set experimental animals in each group, a sample set is constructed, and meanwhile image data in the constructed sample set are preprocessed through an image preprocessing method; selecting each organization index in the preprocessed image data as an analysis standard through an index selection method, clustering the selected organization indexes in the preprocessed image data through a hierarchical clustering method, and calculating the molar content of each selected organization index after clustering; and meanwhile, the tissue injury hemostasis repair capability effectiveness of each tissue index is determined through comparison, so that the reliability of hemostasis repair capability data analysis is ensured.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and specifically to a system and method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes. Background Art

[0002] Continuously exploring new filling materials is the best choice for clinical treatment. At present, medical absorbable gelatin sponge is widely used due to its advantages such as good hemostatic effect, easy processing and economical price. However, due to the certain irritation and toxicity of medical absorbable gelatin sponge, it is limited, making exosome materials a research hotspot.

[0003] The existing Chinese patent application CN108399225A, this method removes redundant information in the biometric information by comparing and repetitively screening the process, thus avoiding the low-density value attribute of big data. At the same time, the biometric information that is conducive to identification is concentrated in the database by taking photos, and the system's autonomous identification function is realized through repeated operations. The repeated operations of big data are used to improve the accuracy of system identification. The verification process avoids the possibility of misjudgment, but it has certain limitations. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the shortcomings of the prior art, the present invention provides a data analysis system and method for the hemostatic repair ability of muscle layer tissue damage based on exosomes, which has the advantages of high efficiency, accuracy, and reliability, and solves the problem that medical absorbable gelatin sponge has certain irritation and toxicity that causes limitations.

[0006] (II) Technical solution

[0007] In order to solve the technical problem that the above-mentioned medical absorbable gelatin sponge has certain irritation toxicity and is limited, the present invention provides the following technical solution:

[0008] This embodiment discloses a method for analyzing the hemostatic and repairing ability of muscular tissue damage based on exosomes, which specifically includes the following steps:

[0009] S1. Collect the same batch of experimental animals, and set up an exosome group, a normal saline group, and a blank control group. At the same time, monitor the experimental animals in each group using a fluorescence microscope, and collect the image data after monitoring in real time to construct a sample set;

[0010] S2, preprocessing the image data in the constructed sample set by an image preprocessing method to obtain preprocessed image data;

[0011] S3. Since different tissues will present different color states after the sample is stained, various tissue indicators in the pre-processed image data are selected through the indicator selection method and used as analysis standards;

[0012] S4, clustering the selected tissue indicators in the preprocessed image data by a hierarchical clustering method, and calculating the molar content of each selected tissue indicator after clustering;

[0013] S5. Summarize the molar content of each selected tissue index in the exosome group, normal saline group, and blank control group, and compare them to determine the tissue injury hemostasis and repair ability of each tissue index.

[0014] The present invention collects experimental animals of the same batch and sets up different control groups, monitors the experimental animals of each group by using a fluorescence microscope, and constructs a sample set, preprocesses the image data in the constructed sample set by an image preprocessing method, selects various tissue indicators in the preprocessed image data as analysis standards by an indicator selection method, clusters the selected tissue indicators in the preprocessed image data by a hierarchical clustering method, calculates the molar content of each selected tissue indicator after clustering, and compares and determines the tissue damage hemostasis and repair ability utility of each tissue indicator, thereby ensuring the reliability of the hemostasis and repair ability data analysis.

[0015] Preferably, the method of collecting the same batch of experimental animals and setting up an exosome group, a saline group, and a blank control group, and monitoring the experimental animals in each group by using a fluorescence microscope, and collecting the monitored image data in real time to construct a sample set includes the following steps:

[0016] S11, set up exosome group, normal saline group and blank control group;

[0017] After cutting open the erector spinae muscles of the experimental animals, the exosome group was set up: absorbable gelatin sponge blocks soaked in exosome extract were implanted into the back wound;

[0018] Normal saline group: absorbable gelatin sponge soaked in 0.9% normal saline was implanted into the back wound;

[0019] A blank control group was set up: only the model surgery was performed, and the muscular tissue, subcutaneous tissue and skin were sutured layer by layer;

[0020] S12, setting a monitoring period for each group of experimental animals, and collecting wound healing samples from each group of experimental animals based on the set monitoring period and staining them;

[0021] S13. Extract image data from wound healing samples of experimental animals after staining using a fluorescence microscope and construct a sample set.

[0022] The present invention provides an exosome group, a physiological saline group, and a blank control group, and sets a monitoring period for each group of experimental animals. Based on the set monitoring period, wound healing samples of each group of experimental animals are collected and stained. At the same time, data is collected on the stained samples and a sample set is constructed, thereby improving the validity of the data.

[0023] Preferably, the step of preprocessing the image data in the constructed sample set by using an image preprocessing method to obtain the preprocessed image data comprises the following steps:

[0024] S21, setting a window, wherein the window includes the image pixels in the sample set to be processed and the surrounding neighborhood of the image pixels in the sample set;

[0025] S22, for each pixel in the window, calculating the product of the spatial domain weight and the gray value weight as the weight of the pixel;

[0026] S23, calculating the weight of each pixel respectively;

[0027] S24, multiplying the weight of each pixel by its corresponding pixel value, and summing the weighted values ​​of all pixels;

[0028] S25, dividing the weighted sum by the weighted sum to obtain preprocessed image data.

[0029] Preferably, for each pixel in the window, calculating the product of the spatial domain weight and the gray value weight as the weight of the pixel comprises the following steps:

[0030] The pixel weight calculation formula is as follows:

[0031] w=w1×w2;

[0032]

[0033] Among them, w1 represents the spatial domain weight, w2 represents the gray value weight, w represents the pixel weight, and e represents the natural logarithm. f(α ,η) represents the pixel value, α,η represents the pixel coordinate, f(β ,γ) represents the neighborhood pixels, β,γ represent the neighborhood coordinates, and σ represents the Gaussian function parameters.

[0034] The present invention sets a window and for each pixel in the window, obtains the weight of each pixel by calculating the product of the spatial domain weight and the gray value weight, and standardizes the image data based on the calculated weight of each pixel, thereby ensuring the stability of the image data.

[0035] Preferably, the method of selecting various tissue indicators in the preprocessed image data by the indicator selection method and using them as analysis criteria includes the following steps:

[0036] S31, collecting scoring data of various tissue indicators in the preprocessed image data from different experts;

[0037] Assuming there are n existing organizational indicators and m preprocessed image data, the original scoring matrix A is established as:

[0038]

[0039] Among them, a mn represents the score of the mth preprocessed image data with respect to the nth tissue index;

[0040] S32, processing the original matrix A and updating the score in the original matrix A;

[0041] S33. Determine the final score of each organizational indicator through the entropy weight method;

[0042] S34. Arrange the final scores of each organizational indicator in descending order, set the number k of selected organizational indicators, and select the first k organizational indicators as analysis criteria.

[0043] Preferably, the processing of the original matrix A and updating the scores in the original matrix A comprises the following steps:

[0044] According to the positive organization index calculation formula and the negative organization index calculation formula, the data in the original matrix A is processed and the score in the original matrix A is updated;

[0045] Positive organizational indicator calculation formula:

[0046]

[0047] in, represents the score of the mth preprocessed image data with respect to the nth positive tissue index;

[0048] Negative organizational indicator calculation formula:

[0049]

[0050] in, Represents the score of the mth preprocessed image data regarding the nth negative tissue indicator.

[0051] Preferably, determining the final score of each organizational indicator by the entropy weight method comprises the following steps:

[0052] S331. The formula for determining the weight of organizational indicators using the entropy weight method is:

[0053]

[0054] Among them, P ij is the weight value of the jth tissue index in the i-th preprocessed image data; Y ij is the set value of the weight of the j-th tissue index in the i-th preprocessed image data;

[0055] Calculate the entropy value u of the jth organizational indicator j , the formula is:

[0056]

[0057] S332. Calculate the utility value g of the j-th organizational indicator information j , the formula is: g j =1-u j ;

[0058] S333. Give weights to organizational indicators and obtain the weights of each organizational indicator and determine the final score for each organizational indicator

[0059] Among them, W j is the organizational indicator weight of the jth organizational indicator, V j is the final score of the jth organizational indicator.

[0060] The present invention is adopted.

[0061] Preferably, clustering the selected tissue indicators in the preprocessed image data by a hierarchical clustering method and calculating the molar content of each selected tissue indicator after clustering comprises the following steps:

[0062] S41, setting each selected tissue indicator as a cluster, and the pixel value corresponding to the color of the tissue indicator after dyeing is the cluster center of the cluster;

[0063] S42, clustering each pixel point of the preprocessed image data by pixel value comparison and image binarization to obtain clustered image data;

[0064] When the pixel value of a certain pixel in the preprocessed image data is consistent with the cluster center, the grayscale value of the pixel is set to 1;

[0065] When the pixel value of a certain pixel in the preprocessed image data is inconsistent with the cluster center, the gray value of the pixel is set to 0;

[0066] When the grayscale value is set to 1, the pixel is black, and when the grayscale value is set to 0, the pixel is white.

[0067] The black image data obtained after clustering each selected tissue index is summarized, and the obtained black image data is set as the image data after clustering.

[0068] The present invention clusters each pixel point of the preprocessed image data by pixel value comparison and image binarization, completes image separation and clustering by setting different gray values, and improves the accuracy of calculating the molar content of each selected tissue index after clustering.

[0069] Preferably, the summarizing the molar content of each selected tissue index in the exosome group, the saline group, and the blank control group, and comparing them to determine the tissue injury hemostasis and repair ability of each tissue index comprises the following steps:

[0070] According to the actual results of each sample in the sample set, the accuracy of the hemostatic and repair ability of the sample tissue damage in the sample set after comparing the molar content of each selected tissue index is determined;

[0071] Set the actual result threshold of each sample, set the positive sample as the sample whose actual result is greater than or equal to the actual result threshold, and the negative sample as the sample whose actual result is less than the actual result threshold;

[0072]

[0073] Among them, ACC represents the utility accuracy of the hemostatic repair ability of sample tissue damage, TP represents the number of positive samples determined by the tissue indicator, TN represents the number of negative samples determined by the tissue indicator, FP represents the number of positive samples wrongly judged by the tissue indicator, and FN represents the number of negative samples wrongly judged by the tissue indicator;

[0074] Set the accuracy threshold of the hemostatic repair ability of sample tissue damage. When the calculated accuracy is greater than or equal to the set accuracy threshold, reduce the number of selected tissue indicators until convergence;

[0075] When the calculated accuracy is less than the set accuracy threshold, the number of selected tissue indicators is increased until convergence.

[0076] The present invention determines the accuracy of the hemostatic repair ability utility of tissue damage in the sample set after comparing the molar content of each selected tissue indicator based on the actual results of each sample in the sample set, and visualizes the accuracy of the hemostatic repair ability utility by calculating the analysis results of the samples, thereby improving the reliability of the accuracy of the hemostatic repair ability utility.

[0077] This embodiment also discloses a data analysis system for hemostasis and repair ability of muscular tissue damage based on exosomes, including: a data acquisition module, a data processing module, a tissue index selection module, a molar content determination module, and a repair ability analysis module;

[0078] The data acquisition module is used to collect image data in the sample in real time and transmit it to the data processing module;

[0079] The data processing module is used to filter the received image data;

[0080] The tissue index selection module is used to select various tissue indexes in the image data according to the expert scores;

[0081] The molar content determination module is used to calculate the molar content of each tissue index in the selected image data;

[0082] The repair capacity analysis module is used to determine the relationship between the repair capacity and various tissue indicators.

[0083] (III) Beneficial effects

[0084] Compared with the prior art, the present invention provides a data analysis system and method for hemostasis and repair ability of muscular tissue damage based on exosomes, which has the following beneficial effects:

[0085] 1. The present invention collects experimental animals from the same batch and sets up different control groups. It monitors the experimental animals in each group by using a fluorescence microscope and constructs a sample set. It preprocesses the image data in the constructed sample set by an image preprocessing method, and selects various tissue indicators in the preprocessed image data as analysis standards by an indicator selection method. It clusters the selected tissue indicators in the preprocessed image data by a hierarchical clustering method, calculates the molar content of each selected tissue indicator after clustering, and compares and determines the tissue damage hemostasis and repair ability utility of each tissue indicator, thereby ensuring the reliability of the hemostasis and repair ability data analysis.

[0086] 2. The invention sets up an exosome group, a saline group, and a blank control group, and sets a monitoring period for each group of experimental animals. Based on the set monitoring period, wound healing samples of each group of experimental animals are collected and stained. At the same time, data is collected on the stained samples and a sample set is constructed, thereby improving the validity of the data.

[0087] 3. The invention sets a window and for each pixel in the window, obtains the weight of each pixel by calculating the product of the spatial domain weight and the gray value weight, and standardizes the image data based on the calculated weight of each pixel, thereby ensuring the stability of the image data.

[0088] 4. The invention collects the scoring data of various tissue indicators in the preprocessed image data from different experts, determines the final score of each tissue indicator by the entropy weight method, arranges the final scores of each tissue indicator in descending order, sets the number k of selected tissue indicators, and selects the first k tissue indicators as the analysis criteria, thereby ensuring the reliability and accuracy of the selected tissue indicators.

[0089] 5. The present invention clusters each pixel point of the preprocessed image data by pixel value comparison and image binarization, completes image separation and clustering by setting different gray values, and improves the accuracy of calculating the molar content of each selected tissue indicator after clustering.

[0090] 6. The present invention determines the accuracy of the hemostatic repair ability utility of tissue damage in the sample set after comparing the molar content of each selected tissue indicator based on the actual results of each sample in the sample set, and visualizes the accuracy of the hemostatic repair ability utility by calculating the analysis results of the samples, thereby improving the reliability of the accuracy of the hemostatic repair ability utility. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 This is a schematic diagram of the structure for analyzing the hemostatic and repairing ability of the exosomes of the present invention for muscular tissue damage. DETAILED DESCRIPTION

[0092] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0093] Example 1

[0094] See also Figure 1 This embodiment discloses a method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes, which specifically includes the following steps:

[0095] S1. Collect the same batch of experimental animals, and set up an exosome group, a normal saline group, and a blank control group. At the same time, monitor the experimental animals in each group using a fluorescence microscope, and collect the image data after monitoring in real time to construct a sample set;

[0096] The same batch of experimental animals were collected, and an exosome group, a saline group, and a blank control group were set up. At the same time, the experimental animals in each group were monitored by using a fluorescence microscope, and the image data after monitoring was collected in real time to construct a sample set, including the following steps:

[0097] S11, set up exosome group, normal saline group and blank control group;

[0098] After cutting open the erector spinae muscles of the experimental animals, the exosome group was set up: absorbable gelatin sponge blocks soaked in exosome extract were implanted into the back wound;

[0099] Normal saline group: absorbable gelatin sponge soaked in 0.9% normal saline was implanted into the back wound;

[0100] A blank control group was set up: only the model surgery was performed, and the muscular tissue, subcutaneous tissue and skin were sutured layer by layer;

[0101] S12, setting a monitoring period for each group of experimental animals, and collecting wound healing samples from each group of experimental animals based on the set monitoring period and staining them;

[0102] S13, extracting image data from wound healing samples of experimental animals after staining through a fluorescence microscope, and constructing a sample set;

[0103] S2, preprocessing the image data in the constructed sample set by an image preprocessing method to obtain preprocessed image data;

[0104] Preprocessing the image data in the constructed sample set by an image preprocessing method to obtain the preprocessed image data includes the following steps:

[0105] S21, setting a window, wherein the window includes the image pixels in the sample set to be processed and the surrounding neighborhood of the image pixels in the sample set;

[0106] S22, for each pixel in the window, calculating the product of the spatial domain weight and the gray value weight as the weight of the pixel;

[0107] The pixel weight calculation formula is as follows:

[0108] w=w1×w2;

[0109]

[0110] Among them, w1 represents the spatial domain weight, w2 represents the gray value weight, w represents the pixel weight, and e represents the natural logarithm. f(α ,η) represents the pixel value, α,η represents the pixel coordinate, f(β ,γ) represents the neighborhood pixels, β,γ represent the neighborhood coordinates, and σ represents the Gaussian function parameters;

[0111] S23, calculating the weight of each pixel respectively;

[0112] S24, multiplying the weight of each pixel by its corresponding pixel value, and summing the weighted values ​​of all pixels;

[0113] S25, dividing the weighted sum by the weighted sum to obtain preprocessed image data;

[0114] S3. Since different tissues will present different color states after the sample is stained, various tissue indicators in the pre-processed image data are selected through the indicator selection method and used as analysis standards;

[0115] The various tissue indicators in the preprocessed image data are selected by the indicator selection method and used as analysis criteria, including the following steps:

[0116] S31, collecting scoring data of various tissue indicators in the preprocessed image data from different experts;

[0117] Assuming there are n existing organizational indicators and m preprocessed image data, the original scoring matrix A is established as:

[0118]

[0119] Among them, a mn represents the score of the mth preprocessed image data with respect to the nth tissue index;

[0120] S32, processing the original matrix A and updating the score in the original matrix A;

[0121] According to the positive organization index calculation formula and the negative organization index calculation formula, the data in the original matrix A is processed and the score in the original matrix A is updated;

[0122] Positive organizational indicator calculation formula:

[0123]

[0124] in, represents the score of the mth preprocessed image data with respect to the nth positive tissue index;

[0125] Negative organizational indicator calculation formula:

[0126]

[0127] in, represents the score of the mth preprocessed image data with respect to the nth negative tissue index;

[0128] S33. Determine the final score of each organizational indicator through the entropy weight method;

[0129] S331. The formula for determining the weight of organizational indicators using the entropy weight method is:

[0130]

[0131] Among them, P ij is the weight value of the jth tissue index in the i-th preprocessed image data; Y ij is the set value of the weight of the j-th tissue index in the i-th preprocessed image data;

[0132] Calculate the entropy value u of the jth organizational indicator j , the formula is:

[0133]

[0134] S332. Calculate the utility value g of the j-th organizational indicator information j , the formula is: g j =1-u j ;

[0135] S333. Give weights to organizational indicators and obtain the weights of each organizational indicator and determine the final score for each organizational indicator

[0136] Among them, W j is the organizational indicator weight of the jth organizational indicator, V j is the final score of the jth organizational indicator;

[0137] S34, arranging the final scores of each organizational indicator in descending order, setting the number k of selected organizational indicators, and selecting the first k organizational indicators as analysis criteria;

[0138] S4, clustering the selected tissue indicators in the preprocessed image data by a hierarchical clustering method, and calculating the molar content of each selected tissue indicator after clustering;

[0139] Clustering the selected tissue indicators in the preprocessed image data by a hierarchical clustering method and calculating the molar content of each selected tissue indicator after clustering includes the following steps:

[0140] S41, setting each selected tissue indicator as a cluster, and the pixel value corresponding to the color of the tissue indicator after dyeing is the cluster center of the cluster;

[0141] S42, clustering each pixel point of the preprocessed image data by pixel value comparison and image binarization to obtain clustered image data;

[0142] When the pixel value of a certain pixel in the preprocessed image data is consistent with the cluster center, the grayscale value of the pixel is set to 1;

[0143] When the pixel value of a certain pixel in the preprocessed image data is inconsistent with the cluster center, the gray value of the pixel is set to 0;

[0144] When the grayscale value is set to 1, the pixel is black, and when the grayscale value is set to 0, the pixel is white.

[0145] Summarize the black image data obtained after clustering each selected tissue indicator, and set the obtained black image data as the image data after clustering;

[0146] S43, determining the molar content of each selected tissue index by an area calculation formula based on the obtained clustered image data;

[0147] The area of ​​the image data after clustering of each tissue index = the number of pixels × the size of each pixel;

[0148] Set the molar content represented by each pixel;

[0149] The molar content of each selected tissue index = the area of ​​the image data after clustering of each tissue index × the molar content represented by each pixel;

[0150] S5. Summarize the molar content of each selected tissue index in the exosome group, normal saline group, and blank control group, and compare them to determine the tissue injury hemostasis and repair ability of each tissue index;

[0151] Summarizing the molar content of each selected tissue index in the exosome group, saline group, and blank control group, and comparing them, and determining the tissue injury hemostasis and repair ability of each tissue index includes the following steps:

[0152] According to the actual results of each sample in the sample set, the accuracy of the hemostatic and repair ability of the sample tissue damage in the sample set after comparing the molar content of each selected tissue index is determined;

[0153] Set the actual result threshold of each sample, set the positive sample as the sample whose actual result is greater than or equal to the actual result threshold, and the negative sample as the sample whose actual result is less than the actual result threshold;

[0154]

[0155] Among them, ACC represents the utility accuracy of the hemostatic repair ability of sample tissue damage, TP represents the number of positive samples determined by the tissue indicator, TN represents the number of negative samples determined by the tissue indicator, FP represents the number of positive samples wrongly judged by the tissue indicator, and FN represents the number of negative samples wrongly judged by the tissue indicator;

[0156] The accuracy threshold of the hemostatic repair ability of sample tissue damage is set. When the calculated accuracy is greater than or equal to the set accuracy threshold, the number of selected tissue indicators is reduced until convergence;

[0157] When the calculated accuracy is less than the set accuracy threshold, the number of selected tissue indicators is increased until convergence;

[0158] Example 2

[0159] This embodiment also discloses a data analysis system for hemostasis and repair ability of muscular tissue damage based on exosomes, including: a data acquisition module, a data processing module, a tissue index selection module, a molar content determination module, and a repair ability analysis module;

[0160] The data acquisition module is used to collect image data in the sample in real time and transmit it to the data processing module;

[0161] The data processing module is used to filter the received image data;

[0162] The tissue index selection module is used to select various tissue indexes in the image data according to the expert scores;

[0163] The molar content determination module is used to calculate the molar content of each tissue index in the selected image data;

[0164] The repair capacity analysis module is used to determine the relationship between the repair capacity and various tissue indicators.

[0165] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing the hemostatic and repairing ability of muscular tissue damage based on exosomes, characterized in that: The following steps are involved: S1. Collect the same batch of experimental animals, and set up an exosome group, a normal saline group, and a blank control group. At the same time, monitor the experimental animals in each group using a fluorescence microscope, and collect the image data after monitoring in real time to construct a sample set; S2, preprocessing the image data in the constructed sample set by an image preprocessing method to obtain preprocessed image data; S3. Since different tissues will present different color states after the sample is stained, various tissue indicators in the pre-processed image data are selected through the indicator selection method and used as analysis standards; S4, clustering the selected tissue indicators in the preprocessed image data by a hierarchical clustering method, and calculating the molar content of each selected tissue indicator after clustering; S5. Summarize the molar content of each selected tissue index in the exosome group, normal saline group, and blank control group, and compare them to determine the tissue injury hemostasis and repair ability of each tissue index.

2. According to claim 1, a method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes, characterized in that: The method of collecting the same batch of experimental animals and setting up an exosome group, a saline group, and a blank control group, and monitoring the experimental animals in each group by using a fluorescence microscope, and collecting the monitored image data in real time to construct a sample set includes the following steps: S11, set up exosome group, normal saline group and blank control group; After cutting open the erector spinae muscles of the experimental animals, the exosome group was set up: absorbable gelatin sponge blocks soaked in exosome extract were implanted into the back wound; Normal saline group: absorbable gelatin sponge soaked in 0.9% normal saline was implanted into the back wound; A blank control group was set up: only the model surgery was performed, and the muscular tissue, subcutaneous tissue and skin were sutured layer by layer; S12, setting a monitoring period for each group of experimental animals, and collecting wound healing samples from each group of experimental animals based on the set monitoring period and staining them; S13. Extract image data from wound healing samples of experimental animals after staining using a fluorescence microscope and construct a sample set.

3. According to claim 1, a method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes, characterized in that: The method of preprocessing the image data in the constructed sample set by the image preprocessing method to obtain the preprocessed image data comprises the following steps: S21, setting a window, wherein the window includes the image pixels in the sample set to be processed and the surrounding neighborhood of the image pixels in the sample set; S22, for each pixel in the window, calculating the product of the spatial domain weight and the gray value weight as the weight of the pixel; S23, calculating the weight of each pixel respectively; S24, multiplying the weight of each pixel by its corresponding pixel value, and summing the weighted values ​​of all pixels; S25, dividing the weighted sum by the weighted sum to obtain preprocessed image data.

4. According to claim 3, a method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes, characterized in that: For each pixel in the window, calculating the product of the spatial domain weight and the gray value weight as the weight of the pixel includes the following steps: The pixel weight calculation formula is as follows: w=w1×w2; Among them, w1 represents the spatial domain weight, w2 represents the gray value weight, w represents the pixel weight, and e represents the natural logarithm. f(α ,η) represents the pixel value, α,η represents the pixel coordinate, f(β ,γ) represents the neighborhood pixels, β,γ represent the neighborhood coordinates, and σ represents the Gaussian function parameters.

5. The method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes according to claim 1, characterized in that: The method of selecting various tissue indicators in the preprocessed image data as analysis criteria includes the following steps: S31, collecting scoring data of various tissue indicators in the preprocessed image data from different experts; Assuming there are n existing organizational indicators and m preprocessed image data, the original scoring matrix A is established as: Among them, a mn represents the score of the mth preprocessed image data with respect to the nth tissue index; S32, processing the original matrix A and updating the score in the original matrix A; S33. Determine the final score of each organizational indicator through the entropy weight method; S34. Arrange the final scores of each organizational indicator in descending order, set the number k of selected organizational indicators, and select the first k organizational indicators as analysis criteria.

6. A method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes according to claim 5, characterized in that: The processing of the original matrix A and updating the scores in the original matrix A comprises the following steps: According to the positive organization index calculation formula and the negative organization index calculation formula, the data in the original matrix A is processed and the score in the original matrix A is updated; Positive organizational indicator calculation formula: in, represents the score of the mth preprocessed image data with respect to the nth positive tissue index; Negative organizational indicator calculation formula: in, Represents the score of the mth preprocessed image data regarding the nth negative tissue indicator.

7. The method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes according to claim 5, characterized in that: Determining the final score of each organizational indicator by the entropy weight method includes the following steps: S331. The formula for determining the weight of organizational indicators using the entropy weight method is: Among them, P ij is the weight value of the jth tissue index in the i-th preprocessed image data; Y ij is the set value of the weight of the j-th tissue index in the i-th preprocessed image data; Calculate the entropy value u of the jth organizational indicator j , the formula is: S332. Calculate the utility value g of the j-th organizational indicator information j , the formula is: g j =1-u j ; S333. Give weights to organizational indicators and obtain the weights of each organizational indicator and determine the final score for each organizational indicator Among them, W j is the organizational indicator weight of the jth organizational indicator, V j is the final score of the jth organizational indicator.

8. The method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes according to claim 1, characterized in that: The method of clustering the selected tissue indicators in the preprocessed image data by a hierarchical clustering method and calculating the molar content of each selected tissue indicator after clustering comprises the following steps: S41, setting each selected tissue indicator as a cluster, and the pixel value corresponding to the color of the tissue indicator after dyeing is the cluster center of the cluster; S42, clustering each pixel point of the preprocessed image data by pixel value comparison and image binarization to obtain clustered image data; When the pixel value of a certain pixel in the preprocessed image data is consistent with the cluster center, the grayscale value of the pixel is set to 1; When the pixel value of a certain pixel in the preprocessed image data is inconsistent with the cluster center, the gray value of the pixel is set to 0; When the grayscale value is set to 1, the pixel is black, and when the grayscale value is set to 0, the pixel is white. The black image data obtained after clustering each selected tissue index is summarized, and the obtained black image data is set as the image data after clustering.

9. The method for analyzing the hemostatic and repairing ability of muscle layer tissue damage based on exosomes according to claim 1, characterized in that: The method of summarizing the molar content of each selected tissue index in the exosome group, the saline group, and the blank control group, and comparing them to determine the tissue injury hemostasis and repair ability of each tissue index comprises the following steps: According to the actual results of each sample in the sample set, the accuracy of the hemostatic and repair ability of the sample tissue damage in the sample set after comparing the molar content of each selected tissue index is determined; Set the actual result threshold of each sample, set the positive sample as the sample whose actual result is greater than or equal to the actual result threshold, and the negative sample as the sample whose actual result is less than the actual result threshold; Among them, ACC represents the utility accuracy of the hemostatic repair ability of sample tissue damage, TP represents the number of positive samples determined by the tissue indicator, TN represents the number of negative samples determined by the tissue indicator, FP represents the number of positive samples wrongly judged by the tissue indicator, and FN represents the number of negative samples wrongly judged by the tissue indicator; Set the accuracy threshold of the hemostatic repair ability of sample tissue damage. When the calculated accuracy is greater than or equal to the set accuracy threshold, reduce the number of selected tissue indicators until convergence; When the calculated accuracy is less than the set accuracy threshold, the number of selected tissue indicators is increased until convergence.

10. A system for implementing the exosome-based muscular tissue damage hemostasis and repair ability data analysis method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, data processing module, tissue index selection module, molar content determination module and repair capacity analysis module; The data acquisition module is used to collect image data in the sample in real time and transmit it to the data processing module; The data processing module is used to filter the received image data; The tissue index selection module is used to select various tissue indexes in the image data according to the expert scores; The molar content determination module is used to calculate the molar content of each tissue index in the selected image data; The repair capacity analysis module is used to determine the relationship between the repair capacity and various tissue indicators.

Images