Method and system for post-hemorrhagic stroke edema analysis and treatment planning

Abstract

The present invention discloses a method and system for analyzing edema after hemorrhagic stroke and for treatment planning. The method involves collecting and preprocessing edema volume data for each patient at different time points. Then, using a Gaussian fitting algorithm to construct a curve of the edema volume evolution over time and calculating the residual from the actual value, dividing the patients into subgroups using a deep learning clustering algorithm, fitting the edema volume for each subgroup at different time points and calculating the residual from the actual value. The change in edema volume is calculated, trends of change are identified, and the corresponding treatment plans are combined with the corresponding edema evolution trend and analyzed using an association rule algorithm to identify the treatment plan with the highest correlation. The present invention utilizes advanced algorithms to accurately analyze changes in edema volume after hemorrhagic stroke and improve the accuracy of patient prognosis prediction. Detailed analysis of different patient subgroups supports more personalized treatment plans and contributes to improved treatment efficacy.

Description

[Technical Field]

[0001] The present invention relates to the technical field of clinical intelligence, and in particular to a method and system for analyzing edema after hemorrhagic stroke and treatment planning. [Background technology]

[0002] Hemorrhagic stroke is a severe cerebral hemorrhage caused by non-traumatic intracerebral vascular rupture. Its acute onset and rapid progression often result in high mortality rates and severe neurological dysfunction. This disease not only poses significant health risks to patients, but also places a heavy economic burden on society and families. Especially in the later stages of hemorrhagic stroke, hematoma expansion is considered an important risk factor for poor prognosis, causing increased intracranial pressure, worsening neurological function, and even threatening life.

[0003] In recent years, edema around the hematoma has attracted increasing clinical attention as an important marker of secondary injury after hemorrhagic stroke. Such edema can cause compression of brain tissue, affect neuronal function, cause further brain tissue damage, and exacerbate neurological dysfunction. Therefore, dynamic monitoring and early prediction of edema around the hematoma are crucial for improving patient survival and quality of life.

[0004] With the continuous advancement of medical imaging technology, non-invasive dynamic monitoring of brain damage after hemorrhagic stroke has become possible. Furthermore, the rapid development of artificial intelligence technology has revolutionized data processing and analysis in the medical field. By integrating multi-source data, including imaging features, clinical information, and treatment plans, accurate prediction of patient outcomes after hemorrhagic stroke and personalized evaluation of treatment effects can be achieved. Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention proposes a comprehensive analysis method based on a Gaussian fitting algorithm, a deep learning clustering algorithm, and an association rule algorithm. The purpose of this method is to accurately predict and evaluate the change trends of edema volume in patients after hemorrhagic stroke and the correlation between these changes and different treatment methods. Using the Gaussian fitting algorithm, the present invention can construct a curve of edema volume change over time and calculate the residual between the actual observed value and the fitted curve. Using the deep learning clustering algorithm, the present invention can further distinguish different patient subgroups and perform a dedicated edema volume trend analysis for each subgroup. Finally, using the association rule algorithm, the present invention can identify the treatment method most closely related to the progression trend of edema, providing important support for clinical treatment decision-making. [Means for solving the problem]

[0006] In view of the above-mentioned existing problems, the present invention is proposed.

[0007] Therefore, the technical problem that the present invention aims to solve is as follows: to effectively monitor and predict the progression of hematoma expansion and surrounding edema after hemorrhagic stroke, and accurately evaluate the relationship between treatment methods and the patient's edema volume change tendency, thereby creating a personalized treatment plan.

[0008] In order to solve the above technical problems, the present invention proposes the following technical solution: A method for analyzing edema after hemorrhagic stroke and for treatment planning, comprising: Collect and preprocess edema volume data for each patient at different time points; fitting the pre-processed data using a Gaussian fitting algorithm to construct a curve of the evolution of edema volume over time for the whole patient and calculating the residual between said curve and the actual values; Dividing patients into multiple subgroups using a deep learning clustering algorithm, fitting a Gaussian function to the edema amounts of patients in each subgroup at different time points, and calculating the residual between the fitted curves of each subclass and the actual values ​​of each subclass; Calculating the amount of change in edema volume, identifying the change trend, combining with the corresponding treatment plan, analyzing the degree of association between the treatment method and the edema progression trend using an association rule algorithm, and identifying the treatment plan that is most closely related to the edema progression trend.

[0009] In a preferred embodiment of the method for analyzing edema after hemorrhagic stroke and treatment planning according to the present invention, the pre-processing includes cleansing the edema volume data at different time points for each patient, correcting errors and outliers, performing mean value filling and linear interpolation for missing edema volume data, time-aligning the edema volume data based on the patient's examination time and onset time, and performing standardization processing for the edema volume data.

[0010] In a preferred embodiment of the method for analyzing edema after hemorrhagic stroke and for planning treatment according to the present invention, fitting the pre-processed data using the Gaussian fitting algorithm comprises: sorting the pre-processed edema volume data by patient and time point to form a structured dataset, with each row representing the edema volume for each patient at a particular time point; The formula for selecting the Gaussian function is: JPEG2026502318000002.jpg1231(1) where v represents the time from onset to examination, a, b, and c are the parameters of the Gaussian function curve of the patient's edema versus time after fitting, a represents the peak value of the curve, b represents the mean value of the curve, and c represents the width of the curve; For each patient's data, estimate the Gaussian model parameters a, b, and c using least squares and perform iterative parameter optimization using gradient descent; and applying a Gaussian function to the time series data for each patient to generate a fitting curve of the change in edema volume over time.

[0011] In a preferred embodiment of the method for analyzing edema after hemorrhagic stroke and for planning treatment according to the present invention, calculating the residual between the curve and the actual value comprises: Residual error is defined as the difference between the predicted value of the Gaussian fitting curve and the actual observed value, and the calculation formula is as follows: JPEG2026502318000003.jpg1127(2) However, R i represents the fitting residual at time point i, and Y i represents the actual value of the edema volume at time point i, JPEG2026502318000004.jpg85 represents the predicted value of the Gaussian fitting curve at the same time point, and calculating the residuals for the edema volume data for each patient at each time point with the corresponding points on the Gaussian fitting curve.

[0012] In a preferred embodiment of the method for analyzing edema after hemorrhagic stroke and for formulating a treatment plan according to the present invention, dividing patients into a plurality of subgroups using the deep learning clustering algorithm comprises: For the personal information features of different patients, feature extraction is performed on the patient data using an autoencoder. In the encoding process, the original feature data x is converted into a low-dimensional latent variable h by the encoder, and the calculation formula is as follows: JPEG2026502318000005.jpg831(3) where W1 and b1 are the weight and bias of the encoder, respectively, σ is the activation function, and m is the feature value in the feature encoder. In the decoding process, the latent variable h after the encoding process is decoded to the original data dimension, and the reconstructed data JPEG2026502318000006.jpg76 is obtained, and the calculation formula is as follows: JPEG2026502318000007.jpg1042(4) where W2 represents the weight of the feature decoder, b2 represents the bias of the feature decoder, and The parameters of the autoencoder are iteratively optimized by a loss function, which is a function of the original data x and the reconstructed data It is defined to minimize the difference between JPEG2026502318000008.jpg76, and its formula is as follows: JPEG2026502318000009.jpg841(5) where G represents the original feature information dataset for each patient, and G R represents the original feature information dataset for each patient reconstructed by the encoder; The low-dimensional representation h obtained by training the autoencoder is used for K-Means clustering. Based on the preset number of clusters k, the algorithm divides patients into k clusters, each consisting of patients with similar characteristics (C1, C2…C k ) and the minimized squared error E is expressed as: JPEG2026502318000010.jpg1031(6) however, TIFF2026502318000011.tif45 represents the vector norm, l represents the low-dimensional representation of patient information, and μ i Cluster C i and representing the mean vector of

[0013] In a preferred embodiment of the method for analyzing edema after hemorrhagic stroke and for planning treatment according to the present invention, calculating the residual between the fitted curve of each subclass and the actual value of each subclass includes the following steps: This involves performing Gaussian fitting on the patient data in each subgroup separated by the deep learning clustering algorithm, and calculating the residual between the Gaussian fitting curve for each subgroup and the actual edema volume data for each patient in the subgroup.

[0014] In a preferred embodiment of the method for analyzing edema after hemorrhagic stroke and for formulating a treatment plan according to the present invention, analyzing the degree of association between a treatment method and a tendency for edema to progress using the association rule algorithm includes: extracting edema volume change and corresponding treatment planning information from the pre-processed dataset; Defining item sets X and Y, where X represents a specific treatment plan and Y represents a specific edema volume change trend; Support calculation is performed, and the support Support(X) of data itemset X is the ratio of the number of transactions containing X in D to the total number of transactions in D, and the formula is shown as follows: JPEG2026502318000012.jpg1033(7) where D represents the progression of edema, x represents the specific treatment plan selected, count(x) represents the number of specific treatment plans selected, and count(D) represents the number of cases in which the tendency of edema progression is suppressed; The support of the association rule X=>Y is The support of JPEG2026502318000013.jpg410 is equal to the formula shown below: JPEG2026502318000014.jpg1171(8) where X represents a particular treatment plan selected and Y represents another particular treatment plan selected; JPEG2026502318000015.jpg617 represents the number of the two selected treatment plans; Calculate the reliability of X=>Y using the following formula: JPEG2026502318000016.jpg1254(9) Evaluate rules for the likelihood that other treatment plans will appear when a particular treatment plan is included, provided that: JPEG2026502318000017.jpg721 represents the support for both treatment methods, and Support(x) represents the support for both treatment methods. Iteratively generating frequent itemsets and corresponding association rules using the Apriori algorithm, and identifying the rules with the highest support and confidence from the generated association rules.

[0015] 1. A system for post-hemorrhagic stroke edema analysis and treatment planning, comprising: The method includes a data pre-processing module, a Gaussian fitting module, a patient subgroup classification module, and a treatment plan related analysis module; The data preprocessing module preprocesses the edema volume data of each patient at different time points; the Gaussian fitting module fits the pre-processed data using a Gaussian fitting algorithm to construct a curve of the evolution of the edema volume over time for the whole patient, and calculates the residual between the curve and the actual value; The patient subgroup classification module divides patients into multiple subgroups using a deep learning clustering algorithm, performs Gaussian function fitting on the edema amounts of patients in each subgroup at different time points, and calculates the residual between the fitted curves of each subclass and the actual values ​​of each subclass; The treatment plan association analysis module calculates the change in edema volume, identifies the change trend, and combines it with the corresponding treatment plan, and uses an association rule algorithm to analyze the association between the treatment method and the edema progression trend, and identifies the treatment plan that is most closely related to the edema progression trend.

[0016] A computer device includes a memory and a processor, wherein a computer program is stored in the memory, and the steps of the above method are realized when the processor executes the computer program.

[0017] a computer-readable storage medium having a computer program stored therein; When the computer program is executed by a processor, the steps of the above method are implemented. [Effects of the Invention]

[0018] The beneficial effects of the present invention are as follows: the use of advanced algorithms to accurately analyze edema volume changes after hemorrhagic stroke and improve the accuracy of predicting patient prognosis; by analyzing different patient subgroups in detail, the present invention supports more personalized treatment methods and helps improve treatment efficacy; and by identifying and accurately predicting the progression of hematoma and edema early, effective measures can be taken in a timely manner, helping to improve patient prognosis and quality of life. [Brief explanation of the drawings]

[0019] In order to more clearly describe the technical solutions of the embodiments of the present invention, the following will briefly describe the drawings necessary for describing the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative efforts. [Figure 1] 1 is an overall flowchart of a method for analyzing edema after hemorrhagic stroke and formulating a treatment plan according to a first embodiment of the present invention; [Figure 2] 1 is a flowchart of the association rule Apriori algorithm of the analysis method for post-hemorrhagic stroke edema analysis and treatment planning provided by a first embodiment of the present invention; [Figure 3] FIG. 10 is a fitting diagram of the edema volume vs. time of the whole patient in the analysis method for edema analysis and treatment planning after hemorrhagic stroke provided by the second embodiment of the present invention. [Figure 4] FIG. 10 is a curve diagram showing the change in edema volume over time in patients of subclass 0 of the method for analyzing edema after hemorrhagic stroke and planning treatment provided by the second embodiment of the present invention. [Figure 5] FIG. 10 is a curve diagram showing the change in edema volume over time in patients of subclass 1 of the method for analyzing edema after hemorrhagic stroke and formulating a treatment plan provided by the second embodiment of the present invention. [Figure 6] FIG. 10 is a curve diagram showing the change in edema volume over time in patients of subclass 2 of the method for analyzing edema after hemorrhagic stroke and formulating a treatment plan provided by the second embodiment of the present invention. [Figure 7] FIG. 10 is a curve diagram showing the change in edema volume over time in patients of subclass 3 of the method for analyzing edema after hemorrhagic stroke and formulating a treatment plan provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention clearer and easier to understand, the following describes in detail specific embodiments of the present invention with reference to the drawings in the specification, and the clearly described embodiments are only some of the embodiments of the present invention, and not all of the embodiments. All other embodiments that can be obtained by those skilled in the art based on the embodiments of the present invention without requiring creative efforts should fall within the protection scope of the present invention.

[0021] Example 1 Referring to FIGS. 1 and 2, there is provided a method for analyzing edema after hemorrhagic stroke and treatment planning, which is one embodiment of the present invention, and includes: In S1, preprocessing is performed on the edema volume data for each patient at different time points.

[0022] The edema volume data for each patient at different time points is cleansed, errors and outliers are corrected, mean value filling and linear interpolation are performed for missing edema volume data, time alignment is performed for the edema volume data based on the patient's examination time and onset time, and standardization processing is performed for the edema volume data.

[0023] In S2, the pre-processed data is fitted using a Gaussian fitting algorithm to construct a curve of the evolution of the edema volume over time for the whole patient, and the residual between the curve and the actual values ​​is calculated.

[0024] The pre-processed edema volume data was sorted by patient and time point to form a structured dataset, with each row representing the edema volume for each patient at a particular time point. The formula for selecting the Gaussian function is: JPEG2026502318000018.jpg1231(1) where x represents the time from onset to examination, and a, b, and c are parameters of the Gaussian function, representing the peak value of the curve, the mean value, and the width of the descriptive curve, respectively.

[0025] For each patient's data, the Gaussian model parameters a, b, and c are estimated using least squares. A Gaussian function is applied to the time series data for each patient to generate a fitting curve for the change in edema volume over time.

[0026] Residual error is defined as the difference between the predicted value of the Gaussian fitting curve and the actual observed value, and the calculation formula is as follows: JPEG2026502318000019.jpg1127(2) However, R i represents the residual for the i-th data point, and Y i represents the actual observation, JPEG2026502318000020.jpg75 represents the predicted value of the Gaussian fitting curve at the same time point. For each patient's edema volume data at each time point, the residuals from the corresponding points on the Gaussian fitting curve are calculated.

[0027] In S3, a deep learning clustering algorithm is used to divide the patients into multiple subgroups, and a Gaussian function is fitted to the edema volume of each subgroup at different time points, and the residual between the fitted curve for each subclass and the actual value for each subclass is calculated.

[0028] Feature extraction is performed on the patient data using an autoencoder. In the encoding process, the original data x is converted into a low-dimensional latent variable h by the encoder. The calculation formula is as follows: JPEG2026502318000021.jpg841(3) where W1 and b1 are the weight and bias of the encoder, respectively, and σ is the activation function. During the decoding process, the latent variable h is decoded back to the original data dimension, and the reconstructed data JPEG2026502318000022.jpg76 is obtained, and the calculation formula is as follows: JPEG2026502318000023.jpg740(4) where W2 and b2 are the weights and biases of the decoder, which are obtained by optimizing the parameters of the autoencoder to differentiate the original data x and the reconstructed data To minimize the difference between JPEG2026502318000024.jpg76, the formula is as follows: JPEG2026502318000025.jpg841(5) The low-dimensional representation h obtained by training the autoencoder is used for K-Means clustering. Based on the preset number of clusters k, the algorithm divides patients into k clusters, each consisting of patients with similar characteristics (C1, C2…C k ) and the minimized square error E is expressed as follows: JPEG2026502318000026.jpg1031(6) However, μ i is the cluster c i is the mean value vector of JPEG2026502318000027.jpg1353(10) Gaussian fitting is performed on the patient data in each subgroup separated by the deep learning clustering algorithm, and the residual between the Gaussian fitting curve for each subgroup and the actual edema volume data for each patient in the subgroup is calculated.

[0029] The deep learning clustering algorithm (Autoencoder K-Means Clustering) combines the feature learning of autoencoders and the clustering capabilities of K-Means. It first uses an autoencoder to train the data and obtain a low-dimensional encoded representation H. Then, it uses the K-Means clustering algorithm to cluster H into K clusters, and improves the clustering performance of K-Means through feature learning of autoencoders.

[0030] In S4, the change in edema volume is calculated, the change trend is identified, and combined with the corresponding treatment plan, an association rule algorithm is used to analyze the correlation between the treatment method and the edema progression trend, and the treatment plan with the highest correlation with the edema progression trend is identified.

[0031] Extracting edema volume change and corresponding treatment plan information from the pre-processed dataset, and defining item sets X and Y, where X represents a specific treatment plan and Y represents a specific edema volume change trend; Support calculation is performed. The support of a data item set X, Support(X), is the ratio of the number of transactions in D that include X to the total number of transactions in D, and the formula is shown as follows: JPEG2026502318000028.jpg1033(7) where D represents the progression of edema, x represents the specific treatment plan selected, count(x) represents the number of specific treatment plans selected, and count(D) represents the number of suppressed progression trends of edema. The support of the association rule X=>Y is calculated by the item set It is equal to the support of JPEG2026502318000029.jpg410, as shown in the following formula: JPEG2026502318000030.jpg1171(8) where X represents a particular selected treatment regimen and Y represents another particular selected treatment regimen; JPEG2026502318000031.jpg617 represents the number of the two selected treatment plans, and calculates the reliability of X=>Y using the following formula: JPEG2026502318000032.jpg1254(9) Evaluate rules for the likelihood that other treatment plans will appear when a particular treatment plan is included, provided that: JPEG2026502318000033.jpg721 represents the support level for both of the two treatment methods, and Support(X) represents the support level for all of the treatment methods.

[0032] The Apriori algorithm is used to iteratively generate frequent itemsets and corresponding association rules, and from the generated association rules, the rules with the highest support and confidence are identified.

[0033] As shown in Figure 2, the Apriori algorithm is an iterative method. k Connect using the (k+1)th candidate set C k+1 Then, perform pruning to generate the (k+1)th most frequent itemset L k+1 Then, the database is scanned multiple times and the above operation is repeated until no frequent itemsets can be found. After the frequent itemset mining is completed, one k-th frequent itemset L k For L, we can generate 2(2k-1) meaningful rules (with non-blank terms), but none of them satisfy the condition. k Generate an association rule "s → (1-s)" that satisfies the requirements only if the confidence of the non-blank subset S of is greater than the minimum confidence.

[0034] In the above embodiment, the system further includes an analysis system for post-hemorrhagic stroke edema analysis and treatment planning, specifically, The data pre-processing module pre-processes the edema volume data for each patient at different time points. The Gaussian fitting module fits the pre-processed data using a Gaussian fitting algorithm to construct a curve of the evolution of the edema volume over time for the whole patient, and calculates the residual between the curve and the actual values. The patient subgroup classification module divides patients into multiple subgroups using a deep learning clustering algorithm, performs Gaussian function fitting on the edema amounts of patients in each subgroup at different time points, and calculates the residual between the fitted curve for each subclass and the actual value for each subclass. The treatment plan association analysis module calculates the change in edema volume, identifies the change trend, and combines it with the corresponding treatment plan, and analyzes the association between the treatment method and the edema progression trend using an association rule algorithm to identify the treatment plan that is most closely related to the edema progression trend.

[0035] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O interface are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. An operating system, a computer program, and a database are stored in the non-volatile storage medium. The internal memory provides an environment for the execution of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of a power monitoring system. The I / O interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for analyzing edema after hemorrhagic stroke and formulating a treatment plan is realized.

[0036] As will be understood by those skilled in the art, all or part of the processes in the above-described methods can be achieved by instructing relevant hardware using a computer program. The above-described computer program can be stored in a non-volatile computer-readable storage medium, and when the computer program is executed, it can include the processes of the above-described method embodiments. Any reference to memory, database, or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM {registered trademark}), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, the RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database according to the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a distributed database based on a blockchain. The processor according to the embodiments provided herein may be, but is not limited to, a general-purpose processor, a central processor, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc.

[0037] Example 2 3 to 7 show an embodiment of the present invention, which provides a method for analyzing edema after hemorrhagic stroke and treatment planning, and scientifically demonstrates the beneficial effects of the present invention through calculation of economic effects and simulation / comparative experiments.

[0038] Figure 3 shows a curve of the change in edema over time for all patients measured using Gaussian function fitting according to the present invention, and a curve of the change in edema over time measured using Gaussian function fitting for each of 100 patients divided into four subclasses using a deep learning clustering algorithm.

[0039] In a real environment, the time from the onset of symptoms to the first test after admission to hospital is random, independent, and countable. The number of subsequent tests and the time of the test for the patient also show strong independence and randomness in the statistical data. Therefore, for a large amount of patient data, data with zero tests in the subsequent data are cleared. There are a total of 450 data points for all patients, and the results of fitting the curve using a Gaussian fitting algorithm are shown in Figure 3 below.

[0040] As can be seen from the analysis of Figure 3, most of the time points where the patient's edema volume was large were concentrated within xx hours after the onset of the disease. The data showed a tendency to decrease over time, with clear peaks. The fitted data (curve) was determined to have remarkable accuracy. Fitting the curve with a Gaussian fitting algorithm clearly maintained the distribution characteristics of the data and the superior smoothness of the fitted curve. By calculating the residual between the actual value and the corresponding time point of the fitted curve, the serial number of the time point corresponding to the residual was found, and the residual values ​​corresponding to all serial numbers for each patient were matched. Then, a mean square error calculation was performed on the residual values ​​of all serial numbers corresponding to the patient, and finally the residual value between the patient's actual value and the fitted curve of the change in edema over time for the entire patient was obtained.

[0041] Similarly, to further explore the individual differences in the progression pattern of edema volume over time, we added relevant information about the patient's personal history, medical history, and onset. Using a deep learning algorithm, clustering was performed, and the 100 patients were clustered into four subclasses (subclass 0, subclass 1, subclass 2, and subclass 3). The curves were fitted using a Gaussian fitting algorithm, as shown in Figure 4.

[0042] As can be seen from Figures 4 to 7, the edema volume of patients in different subclasses shows significant changes within 1,000 hours after onset, and then stabilizes after 1,000 hours. By calculating the residuals between the actual values ​​of each subclass and their fitted curves and analyzing the data, it was found that the residuals between the actual data of each patient in each subclass obtained using the deep learning clustering algorithm and the curve fitted with a Gaussian function, as well as the residuals between the fitted curve of edema volume over time for all patients, were significantly reduced. As can be seen, the subclass groups obtained using the deep learning clustering algorithm, and the curve fitted with a Gaussian function, conform to the change patterns of the actual data and are closer to the actual situation, which has important implications for analyzing changes in patient condition in the field of regenerative medicine.

[0043] It should be noted that the above embodiments are only for illustrating the technical solutions of the present invention, and are not intended to be limiting. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be included in the scope of the claims of the present invention.

Claims

1. 1. A method for analyzing edema after hemorrhagic stroke and for treatment planning, comprising: Collect and preprocess edema volume data for each patient at different time points; fitting the pre-processed data using a Gaussian fitting algorithm to construct a curve of the evolution of edema volume over time for the whole patient and calculating the residual between said curve and the actual values; Dividing patients into multiple subgroups using a deep learning clustering algorithm, fitting a Gaussian function to the edema amounts of patients in each subgroup at different time points, and calculating the residual between the fitted curves of each subclass and the actual values ​​of each subclass; 1. A method for analyzing edema after hemorrhagic stroke and treatment plans, comprising: calculating the amount of change in edema volume, identifying a change trend, combining with a corresponding treatment plan, analyzing the degree of association between the treatment method and the edema progression trend using an association rule algorithm, and identifying the treatment plan that is most closely related to the edema progression trend.

2. 2. The method for analyzing edema after hemorrhagic stroke and treatment planning according to claim 1, wherein the preprocessing includes: cleansing the edema volume data at different time points for each patient, correcting errors and outliers, performing mean value filling and linear interpolation for missing edema volume data, time-aligning the edema volume data based on the patient's examination time and onset time, and performing standardization processing for the edema volume data.

3. Fitting the pre-processed data using the Gaussian fitting algorithm comprises: sorting the pre-processed edema volume data by patient and time point to form a structured dataset, with each row representing the edema volume for each patient at a particular time point; The formula for selecting the Gaussian function is: （１） where v represents the time from onset to examination, a, b, and c are the parameters of the Gaussian function curve of the patient's edema versus time after fitting, a represents the peak value of the curve, b represents the mean value of the curve, and c represents the width of the curve; For each patient's data, estimate the Gaussian model parameters a, b, and c using least squares fitting and perform iterative parameter optimization using gradient descent; and applying a Gaussian function to the time-series data of each patient to generate a fitting curve of the change in edema volume over time.

4. Calculating the residual between the curve and the actual value comprises: Residual error is defined as the difference between the predicted value of the Gaussian fitting curve and the actual observed value, and the calculation formula is as follows: （２） However, R ｉ represents the fitting residual at time point i, and Y ｉ represents the actual value of the edema volume at time point i, represents the predicted value at the same time point of the Gaussian fitting curve, and and calculating residuals between the edema volume data of each patient at each time point and the corresponding points on the Gaussian fitting curve.

5. Dividing patients into a plurality of subgroups using the deep learning clustering algorithm includes: For the personal information features of different patients, feature extraction is performed on the patient data using an autoencoder. In the encoding process, the original feature data x is converted into a low-dimensional latent variable h by the encoder, and the calculation formula is as follows: （３） However, W １ and b １ are the weights and biases of the encoder, respectively, σ is the activation function, and m is the feature value in the feature encoder; In the decoding process, the latent variable h after the encoding process is decoded to the original data dimension, and the reconstructed data The calculation formula is as follows: （４） However, W ２ represents the weights of the feature decoder, and b ２ represents the bias of the feature decoder; The parameters of the autoencoder are iteratively optimized by a loss function, which is a function of the original data x and the reconstructed data and its formula is as follows: （５） where G represents the original feature information dataset for each patient, and G Ｒ represents the original feature information dataset for each patient reconstructed by the encoder; The low-dimensional representation h obtained by training the autoencoder is used for K-Means clustering. Based on the preset number of clusters k, the algorithm divides patients into k clusters, each consisting of patients with similar characteristics (C １ , C ２ …C ｋ ) and the minimized squared error E is expressed as: （６） however, represents the norm of the vector, l represents the low-dimensional representation of the patient information, and μ ｉ Cluster C ｉ and expressing the mean value vector of:

6. Calculating the residual between the fitted curve of each subclass and the actual value of each subclass is The method for analyzing edema after hemorrhagic stroke and planning treatment as described in claim 5, characterized in that it includes performing Gaussian fitting on the patient data in each subgroup divided by the deep learning clustering algorithm, and calculating the residual between the Gaussian fitting curve of each subgroup and the actual edema volume data of each patient in the subgroup.

7. Analyzing the degree of association between a treatment method and a tendency for progression of edema using the association rule algorithm includes: extracting edema volume change and corresponding treatment planning information from the pre-processed dataset; Defining item sets X and Y, where X represents a particular treatment plan and Y represents a particular edema volume change trend; Support calculation is performed, and the support Support(X) of an item set X in the data is the ratio of the number of transactions containing X in D to the total number of transactions in D, and the formula is shown as follows: （７） where D represents the progression state of edema, x represents the selected specific treatment plan, count(x) represents the number of the selected specific treatment plans, and count(D) represents the number of cases in which the tendency of edema progression is suppressed; The support of the association rule X => Y is The formula is shown as follows: （８） where X represents a particular selected treatment plan and Y represents another particular selected treatment plan; represents the number of two treatment plans selected; Calculate the reliability of X => Y using the following formula: （９） Evaluate rules for the likelihood that other treatment plans will appear when a particular treatment plan is included, provided that: represents the support that both of the two treatment methods exist, and Support(x) represents the support that both of the former treatment methods exist; 7. The method for analyzing edema after hemorrhagic stroke and treatment planning according to claim 6, further comprising: iteratively generating frequent itemsets and corresponding association rules using the Apriori algorithm; and identifying the rule with the highest support and confidence from the generated association rules.

8. 10. A system for analyzing edema after hemorrhagic stroke and treatment planning using the method of claim 1, comprising: The method includes a data pre-processing module, a Gaussian fitting module, a patient subgroup classification module, and a treatment plan related analysis module; The data preprocessing module preprocesses the edema volume data of each patient at different time points; the Gaussian fitting module fits the pre-processed data using a Gaussian fitting algorithm to construct a curve of the evolution of the edema volume over time for the whole patient, and calculates the residual between the curve and the actual value; The patient subgroup classification module divides patients into multiple subgroups using a deep learning clustering algorithm, performs Gaussian function fitting on the edema amounts of patients in each subgroup at different time points, and calculates the residual between the fitted curves of each subclass and the actual values ​​of each subclass; The treatment plan association analysis module calculates the change in edema volume, identifies the change trend, and combines it with the corresponding treatment plan. The system for analyzing edema after hemorrhagic stroke and treatment plan is characterized in that the module uses an association rule algorithm to analyze the association between the treatment method and the edema progression trend, and identifies the treatment plan that is most closely related to the edema progression trend.

9. 1. A computer device comprising: a memory and a processor, a computer program stored in the memory; 10. A computer apparatus, wherein the steps of the method of claim 1 are implemented when the processor executes the computer program.

10. 1. A computer-readable storage medium, comprising: a computer program stored in the computer-readable storage medium; 10. A computer-readable storage medium, comprising: a computer program that, when executed by a processor, performs the steps of the method of claim 1.

Images