A method for processing follow-up data of patients with gynecological endometriosis

Abstract

The present invention relates to the field of medical data processing technology, and in particular to a method for processing review and follow-up data of patients with gynecological endometriosis, comprising: obtaining basic information, condition information, and follow-up content in the review and follow-up data of the patient; analyzing the degree of change of the condition information and follow-up content in the review and follow-up data of the patient, and determining the normality of the review and follow-up data of the patient; obtaining the correlation between the review and follow-up data of any two patients according to the similarity of the review and follow-up data of the two patients and the difference in the normality of the review and follow-up data; and classifying different patients into several categories according to the correlation of the review and follow-up data between the patients. The present invention improves the accuracy of the description of the correlation of the review and follow-up data between different patients, and improves the accuracy of the classification of the review and follow-up data of the patients.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to a method for processing review and follow-up data of gynecological endometriosis patients. Background Art

[0002] Endometriosis is a common gynecological disease. Its main pathological feature is that the endometrium (including glands and stroma) grows abnormally outside the uterine cavity. This lesion can cause symptoms such as dysmenorrhea, abnormal menstruation, and infertility. With the development of modern medicine, regular follow-up has become a crucial part of the patient's treatment process. Therefore, systematic processing and analysis of the follow-up data of gynecological endometriosis patients can better understand the dynamic changes of the disease and provide patients with more accurate and personalized treatment plans, thereby achieving better treatment effects.

[0003] When processing the patient's follow-up data, the follow-up data can be classified according to differences in the patient's treatment plan and symptom manifestations; however, endometriosis is a disease that is prone to recurrence. With the patient's regular follow-up, new symptom characteristics may appear at different review stages. Therefore, the patient's condition will fluctuate to a certain extent, resulting in dynamic changes in the follow-up data, which in turn reduces the accuracy of the classification results of the follow-up data. Summary of the invention

[0004] The invention provides a method for processing reexamination and follow-up data of patients with gynecological endometriosis, so as to solve the existing problems.

[0005] The present invention provides a method for processing follow-up data of patients with gynecological endometriosis using the following technical solution:

[0006] An embodiment of the present invention provides a method for processing review and follow-up data of patients with gynecological endometriosis, the method comprising the following steps:

[0007] Acquire the patient's review and follow-up data, wherein the review and follow-up data at least includes the patient's condition information and follow-up content; perform word segmentation processing on the condition information and the follow-up content respectively to obtain a plurality of word segmentation sequences;

[0008] Analyze the degree of change of the word segmentation sequence of the condition information in the patient's reexamination and follow-up data to determine the normal coefficient of the patient's reexamination and follow-up data; obtain the degree of recurrence of the patient's condition according to the degree of change of the word segmentation sequence of the follow-up content in the patient's reexamination and follow-up data; and determine the normal degree of the patient's reexamination and follow-up data based on the normal coefficient of the patient's reexamination and follow-up data and the degree of recurrence of the condition;

[0009] Obtaining the correlation between the review and follow-up data of any two patients according to the similarity between the review and follow-up data of the two patients and the difference in the normality of the review and follow-up data;

[0010] All patients with endometriosis were classified according to the relevance of their follow-up data.

[0011] Preferably, the specific method for obtaining the word segmentation sequence is:

[0012] The patient's condition information includes the patient's symptoms and medication information of the medication taken, and the follow-up content includes the patient's condition recovery;

[0013] Use JIEBA word segmentation tool to segment the disease information and follow-up content in the review and follow-up data;

[0014] In the result of word segmentation processing of medical condition information: the sequence composed of symptom segmentations is recorded as segmentation sequence 1, and the sequence composed of drug information segmentations is recorded as segmentation sequence 2; in the result of word segmentation processing of follow-up content: the sequence composed of medical condition recovery status segmentations is recorded as segmentation sequence 3.

[0015] Preferably, the analysis of the degree of change of the word segmentation sequence of the condition information in the patient's reexamination and follow-up data to determine the normal coefficient of the patient's reexamination and follow-up data includes the following specific methods:

[0016] Obtain the coefficient of symptom change at each follow-up examination for any patient;

[0017] Obtaining the similarity of the segmentation sequence 2 in the segmentation processing results of each review and follow-up data of the patient;

[0018] Combining the similarity of the segmentation sequence 2 with the symptom change degree coefficient to obtain the normal coefficient of the patient's follow-up data;

[0019] The coefficient of symptom change degree of the patient at each follow-up examination is inversely proportional to the normal coefficient of the follow-up data; the similarity of the segmentation sequence 2 in the segmentation processing result of each follow-up data is directly proportional to the normal coefficient of the follow-up data.

[0020] Preferably, the specific method for obtaining the symptom change degree coefficient of any patient at each follow-up examination is:

[0021] The participles that did not appear in the nth review follow-up data but appeared in the n+1th review follow-up data were recorded as newly added participles;

[0022] The specific calculation method of the symptom change coefficient at any follow-up visit of the patient is:

[0023] ;

[0024] In the formula, represents the coefficient of symptom change of the mth patient at the n+1th follow-up visit, It indicates the number of new segmentations in segmentation sequence 1 during the n+1th follow-up of the mth patient; represents the number of segmentations in segmentation sequence 1 for the mth patient at the nth follow-up visit; It indicates the number of identical segmentations in segmentation sequence 1 in the follow-up data of the mth patient, the n+1th and the previous review data.

[0025] Preferably, the specific method for obtaining the similarity of the segmentation sequence 2 in the segmentation processing result of each review and follow-up data of the patient is:

[0026] Calculate the difference between the word sequence 2 of each follow-up data of the patient and the word sequence 2 of the previous follow-up data. Correlation coefficient, and the The correlation coefficient is recorded as the similarity of the segmentation sequence 2 in the segmentation processing results of each follow-up data of the patient.

[0027] Preferably, the method of obtaining the degree of recurrence of the patient's condition according to the degree of change of the word segmentation sequence of the follow-up content in the patient's reexamination follow-up data includes the following specific methods:

[0028] Obtain the sentiment score of word sequence 3 in the follow-up data of any patient since treatment, and calculate the mean of the sentiment score of the patient;

[0029] Construct a sentiment score scatter plot according to the review follow-up time; wherein the horizontal axis is the review follow-up time; the vertical axis is the sentiment score value in the word segmentation sequence 3 corresponding to the review follow-up time; use the least squares method to fit the data points in the sentiment score scatter plot to obtain a sentiment score fluctuation curve, and calculate the average slope on the sentiment score fluctuation curve;

[0030] The degree of recurrence of the patient's condition is obtained by combining the patient's emotional score mean and the average slope of the emotional score fluctuation curve;

[0031] The mean value of the patient's emotion score is inversely proportional to the degree of recurrence of the patient's condition, and the average slope of the patient's emotion score fluctuation curve is directly proportional to the degree of recurrence of the patient's condition.

[0032] Preferably, the normal coefficient of the patient's follow-up data and the degree of disease recurrence are comprehensively considered to determine the normal degree of the patient's follow-up data, including the specific method of:

[0033] The specific calculation method for the normal degree of the patient's follow-up data is:

[0034] ;

[0035] In the formula, Indicates the normality of the follow-up data of the mth patient; represents the normal coefficient of the follow-up data of the mth patient; represents the recurrence degree of the mth endometriosis patient; represents the normalization function; Represents an exponential function with a natural constant as base.

[0036] Preferably, the obtaining of the correlation between the review and follow-up data of any two patients based on the similarity between the review and follow-up data of the two patients and the difference in the normality of the review and follow-up data comprises the following specific methods:

[0037] Obtaining the similarity of the segmentation sequences in the review and follow-up data of any two patients, wherein the similarity quantifies the similarity of the review and follow-up data of any two patients;

[0038] The specific calculation method for the correlation between the follow-up data of any two patients is:

[0039] ;

[0040] In the formula, represents the correlation between the follow-up data of the pth and qth patients; Indicates the normality of the follow-up data of the pth patient; Indicates the normality of the follow-up data of the qth patient; represents the similarity of the segmentation sequence k in the follow-up data of the p-th and q-th patients, ; Indicates the absolute value symbol; represents an exponential function with a natural constant as base; Represents the normalization function.

[0041] Preferably, the specific method for obtaining the similarity of the segmentation sequences in the review and follow-up data of any two patients is:

[0042] The specific calculation method of the segmentation similarity of the segmentation sequences in the review and follow-up data of any two patients is:

[0043] ;

[0044] In the formula, represents the maximum number of follow-up visits between the pth and qth patients; The pth and qth patients had the segmentation sequence k in the jth follow-up. Correlation coefficient.

[0045] Preferably, the method of classifying all patients with endometriosis according to the correlation of the patients' follow-up data includes:

[0046] The silhouette coefficient method was used to determine the number of clusters;

[0047] According to the correlation of the follow-up data between patients, the k-means clustering algorithm was used to cluster all patients with endometriosis in the hospital information system, and the classification results of several patients were obtained.

[0048] The beneficial effects of the technical solution of the present invention are: by analyzing the degree of change of the disease information in the follow-up data of the patient, the normal coefficient of the follow-up data of each patient is determined, indicating the similarity of the treatment plan during each follow-up of the patient; according to the degree of change of the follow-up content in the follow-up data, the degree of recurrence of the patient's disease is obtained, and the degree of recurrence of the disease effectively describes the patient's recognition of the treatment effect; the normal coefficient of the follow-up data of the patient and the degree of recurrence of the disease are combined to determine the normal degree of the follow-up data of the patient, thereby reflecting the effect of the treatment plan; according to the similarity of the follow-up data of any two patients and the difference in the normal degree of the follow-up data, the correlation between the follow-up data of the two patients is obtained; finally, the patients are classified according to the correlation of the follow-up data between the patients, and the treatment effect of each type of patients is monitored. The present invention analyzes the symptom changes, treatment medication and treatment effect in all the follow-up data of the patients during the treatment process, obtains the correlation between the follow-up data of any two patients, classifies all patients according to the correlation, and improves the accuracy of the classification of the follow-up data of the patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 The present invention is a flowchart of the steps of a method for processing review and follow-up data of patients with gynecological endometriosis. DETAILED DESCRIPTION

[0051] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method for processing follow-up data of a gynecological endometriosis patient according to the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0052] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0053] The following is a detailed description of a method for processing review and follow-up data of gynecological endometriosis patients provided by the present invention in conjunction with the accompanying drawings.

[0054] See also Figure 1 , which shows a flowchart of a method for processing review and follow-up data of gynecological endometriosis patients provided by an embodiment of the present invention, the method comprising the following steps:

[0055] Step S001: Obtain the basic information, condition information and follow-up content of the patient's review and follow-up data.

[0056] It should be noted that endometriosis is a common gynecological disease. By analyzing the follow-up data of patients during treatment, patients with endometriosis in the hospital (all referred to as patients) are classified to effectively monitor the disease progression and treatment effect of each type of patient.

[0057] Specifically, in order to implement the review and follow-up data processing method for gynecological endometriosis patients proposed in this embodiment, it is first necessary to obtain the review and follow-up data of all patients with endometriosis in the hospital information system. The specific process is as follows:

[0058] First, the basic information, medical condition information and follow-up content of the patient are obtained in the hospital information system. The basic information includes the patient's name, age, etc.; the medical condition information includes the patient's symptoms and medication information; and the follow-up content includes the patient's description of the recovery of the disease.

[0059] Then, the JIEBA word segmentation tool was used to segment the obtained review and follow-up data; the sequence of symptom segmentations in the result of the word segmentation of the disease information was recorded as segmentation sequence 1, and the sequence of drug information segmentations was recorded as segmentation sequence 2; the sequence of disease recovery segmentations in the result of the word segmentation of the follow-up content was recorded as segmentation sequence 3. For example, the segmentation sequence 1 of a patient is ["dysmenorrhea", "constipation"], the segmentation sequence 2 is ["raloxifene", "progesterone", "leuprorelin acetate"], and the segmentation sequence 3 is ["dysmenorrhea", "improved", "constipation", "not improved"].

[0060] It should be noted that the obtained review and follow-up data is text data, so it is necessary to use the JIEBA word segmentation tool to perform word segmentation preprocessing on the text data; the JIEBA word segmentation tool is a Chinese word segmentation library based on Python, which will not be repeated in this embodiment.

[0061] So far, the word segmentation sequence of the reexamination and follow-up data of all endometriosis patients is obtained through the above method.

[0062] Step S002: Analyze the degree of change of the disease condition information in the patient's reexamination and follow-up data to determine the normal coefficient of each patient's reexamination and follow-up data; obtain the degree of recurrence of the patient's disease condition according to the degree of change of the follow-up content in the reexamination and follow-up data; and determine the normal degree of the patient's reexamination and follow-up data by combining the normal coefficient of the patient's reexamination and follow-up data and the degree of recurrence of the disease condition.

[0063] It should be noted that each patient will undergo multiple follow-up visits during the treatment process, and each follow-up visit will have corresponding follow-up data. The normality of the follow-up data can be determined by analyzing the changes in the follow-up data of the patients. If the changes in the follow-up data of the multiple follow-up visits are small, it means that the similarity of the follow-up data of each follow-up visit is high, which indicates that the follow-up results of the patients tend to be normal.

[0064] S1: Analyze the degree of change in the patient's follow-up data and determine the normal coefficient of each patient's follow-up data.

[0065] It should be noted that the degree of change of the disease condition information in all the reexamination and follow-up data of any patient during the entire treatment process is analyzed, and the normal coefficient of the reexamination and follow-up data of the patient is determined according to the degree of change.

[0066] Specifically, for any patient, obtain the segmentation sequence 1 and segmentation sequence 2 in the follow-up data of the patient's treatment so far;

[0067] Then the calculation expression of the normal coefficient of the follow-up data of any patient is:

[0068] ;

[0069] In the formula, represents the normal coefficient of the follow-up data of the mth patient; N represents the number of follow-up visits of the patient; It indicates the number of segmentations added to segmentation sequence 1 (i.e., not appearing in the nth follow-up) during the n+1th follow-up of the mth patient; represents the number of segmentations in segmentation sequence 1 for the mth patient at the nth follow-up visit; It indicates the number of identical segmentations in segmentation sequence 1 in the follow-up data of the mth patient's n+1th and previous reexaminations; Represents the segmentation sequence 2 in the follow-up data of the mth patient, the nth and n+1th review Correlation coefficient; Represents an exponential function with a natural constant as base.

[0070] It should be noted that Represents a sequence of two word segments Correlation coefficient; In addition, get the two word segmentation sequences The correlation coefficient is a well-known technology and will not be described in detail in this embodiment.

[0071] It is further necessary to explain that It is used to describe the coefficient of symptom change at each follow-up visit of the patient, where The smaller the value, the fewer new symptoms the patient has in the n+1th follow-up examination; The smaller the value of , the less symptoms the patient experienced during the nth follow-up examination occurred during the n+1th follow-up examination; therefore, The smaller the value, the less symptoms there are in the n+1th follow-up compared to the nth time, and the fewer new symptoms there are, and the greater the normal coefficient of the patient's follow-up data.

[0072] It reflects the similarity of the segmentation sequence 2 in the nth and n+1th follow-up visits. The larger the value, the higher the similarity of the treatment drugs in the nth and n+1th follow-up visits, the more similar the corresponding treatment plans are, and the larger the normal coefficient of the patient's follow-up data is. By analyzing the difference in the number of segmentations in the segmentation sequence 1 and the correlation between the segmentation sequence 2 in the two adjacent follow-up data, we can get and , characterizing the degree of change in the word segmentation sequence of the patient's follow-up data.

[0073] At this point, the normal coefficient of the patient's follow-up data was obtained through the above method.

[0074] S2: Obtain the degree of recurrence of the patient's condition based on the degree of change in the follow-up content in the reexamination follow-up data.

[0075] It should be noted that the word segmentation sequence 3 in the review and follow-up data is a personal subjective descriptive term with a certain emotional color. Therefore, by calculating the emotional score in the word segmentation sequence 3 in the current review and follow-up data, the patient's recognition of the treatment effect can be understood. When the emotional color is positive and the emotional score is high, the patient's recognition of the treatment effect is high; when the emotional color is negative and the emotional score is low, it means that the patient's recognition of the treatment effect is low. Therefore, the changes in the emotional score value in the word segmentation sequence 3 in the review and follow-up data of the patient since treatment are analyzed to determine the consistency of the patient's review and follow-up data.

[0076] Specifically, first, the sentiment score value of the word segmentation sequence 3 in the follow-up data of any patient treated so far is obtained.

[0077] Then, a sentiment score scatter plot is constructed according to the review and follow-up time, wherein the horizontal axis is the review and follow-up time, and the vertical axis is the sentiment score value in the word segmentation sequence 3 corresponding to the review and follow-up time.

[0078] Finally, the least squares method is used to fit the data points in the sentiment score scatter plot to obtain the sentiment score fluctuation curve.

[0079] The calculation expression for the degree of recurrence of any patient's condition is:

[0080] ;

[0081] In the formula, represents the degree of recurrence of the mth patient’s condition, represents the number of follow-up visits for the mth patient; represents the sentiment score of word sequence 3 in the nth follow-up data of the mth patient; represents the average slope of the emotion score fluctuation curve of the mth patient; Represents an exponential function with a natural constant as base.

[0082] It should be noted that the degree of change in the follow-up content in the patient's review and follow-up data is described by the mean sentiment score of the segmentation sequence 3 in the patient's review and follow-up data and the slope of the patient's sentiment score fluctuation curve; the degree of fluctuation of the sentiment score fluctuation curve is characterized according to the average slope of the patient's sentiment score fluctuation curve, and then the degree of recognition of the patient's treatment effect is determined; when the sentiment score fluctuation degree is smaller, the patient's recognition of the effects of multiple treatments is more consistent, indicating that the probability of the patient's condition recurring is smaller, and the effect of the treatment plan is better. When the mean sentiment score of the segmentation sequence 3 in the patient's review and follow-up data since the treatment is completed, it means that the patient's recognition of the treatment effect is higher, the degree of recurrence of the patient's condition is smaller, and the effect of the treatment plan is better.

[0083] It should be further explained that the NTUSD sentiment dictionary is used to calculate the score value of the word sequence 3 in the patient review follow-up data; the NTUSD sentiment dictionary is a public Chinese word database based on a text sentiment binary division method, which will not be described in detail in this embodiment.

[0084] So far, the degree of recurrence of the patient's disease is obtained through the above method.

[0085] S3: Determine the normality of the patient's reexamination and follow-up data by comprehensively considering the normality coefficient of the patient's reexamination and follow-up data and the degree of disease recurrence.

[0086] It should be noted that the normality of the patient's reexamination and follow-up data is comprehensively judged based on the similarity of the patient's reexamination and follow-up data and the degree of disease recurrence.

[0087] Specifically, the calculation expression for the normal degree of the patient's follow-up data is:

[0088] ;

[0089] In the formula, Indicates the normality of the follow-up data of the mth patient; represents the similarity of the follow-up data of the mth patient; Indicates the degree of recurrence of the mth patient's condition; represents an exponential function with a natural constant as base; Represents the normalization function.

[0090] It should be further explained that, when the patient's disease recurrence is smaller and the similarity of the reexamination and follow-up data is higher, the normality of the patient's reexamination and follow-up data is higher, which means that the patient's treatment plan is more effective.

[0091] So far, the normality of the patient's follow-up data has been obtained through the above method.

[0092] Step S003: according to the similarity of the review and follow-up data of any two patients and the difference in the normality of the review and follow-up data, obtain the correlation between the review and follow-up data of the two patients.

[0093] It should be noted that the similarity of different segmentation sequences in the review and follow-up data of any two patients and the difference between the normal levels of the review and follow-up data are analyzed to obtain the correlation of the review and follow-up data of the two patients.

[0094] Specifically, the similarity of the word segmentation sequences in the review and follow-up data of any two patients and the normality of the review and follow-up data are obtained.

[0095] In the follow-up data of any two patients, the calculation expression of the similarity of the segmentation of any segmentation sequence is:

[0096] ;

[0097] In the formula, , represents the similarity of the segmentation sequence k in the follow-up data of the p-th and q-th patients; represents the maximum number of follow-up visits between the pth and qth patients; Indicates that the p-th and q-th patients in the j-th follow-up examination, the segmentation sequence k Correlation coefficient.

[0098] The calculation expression for the correlation of the review and follow-up data of any two patients is:

[0099] ;

[0100] In the formula, represents the correlation between the follow-up data of the pth and qth patients; Indicates the normality of the follow-up data of the pth patient; Indicates the normality of the follow-up data of the qth patient; represents the similarity of the segmentation sequence k in the follow-up data of the p-th and q-th patients; Indicates the absolute value symbol; represents an exponential function with a natural constant as base; Represents the normalization function.

[0101] It should be noted that the smaller the difference in the normal degree of the review and follow-up data of any two patients, the greater the correlation between the review and follow-up data of the two patients; the greater the mean similarity of all word segmentation sequences in the review and follow-up data of the two patients, the greater the correlation between the review and follow-up data of the two patients.

[0102] So far, the correlation between the reexamination and follow-up data of any two patients is obtained through the above method.

[0103] Step S004: Classify the patients according to the correlation between the review and follow-up data of the patients.

[0104] It should be noted that, based on the correlation of reexamination and follow-up data between patients, all patients with endometriosis in the hospital information system were clustered, and the symptoms of patients in each cluster and the corresponding treatment effects were monitored and analyzed.

[0105] Specifically, first, the number of clusters is determined according to the silhouette coefficient method;

[0106] Then, according to the correlation of the follow-up data between patients, the k-means clustering algorithm was used to cluster all the patients with endometriosis in the hospital information system, and the classification results of several patients were obtained.

[0107] At this point, this embodiment is completed.

[0108] It should be noted that the The model is only used to represent negative correlation and constrain the output of the model to be in In the specific implementation, it can be replaced by other models with the same purpose. This embodiment is only based on The model is described as an example without any specific limitation. is the input to the model.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for processing review and follow-up data of patients with gynecological endometriosis, characterized in that: The method comprises the following steps: Acquire the patient's review and follow-up data, wherein the review and follow-up data at least includes the patient's condition information and follow-up content; perform word segmentation processing on the condition information and the follow-up content respectively to obtain a plurality of word segmentation sequences; Analyze the degree of change of the word segmentation sequence of the condition information in the patient's reexamination and follow-up data to determine the normal coefficient of the patient's reexamination and follow-up data; obtain the degree of recurrence of the patient's condition according to the degree of change of the word segmentation sequence of the follow-up content in the patient's reexamination and follow-up data; and determine the normal degree of the patient's reexamination and follow-up data based on the normal coefficient of the patient's reexamination and follow-up data and the degree of recurrence of the condition; Obtaining the correlation between the review and follow-up data of any two patients according to the similarity between the review and follow-up data of the two patients and the difference in the normality of the review and follow-up data; All patients with endometriosis were classified according to the relevance of their follow-up data; The method of comprehensively considering the normal coefficient of the patient's follow-up data and the degree of disease recurrence to determine the degree of normality of the patient's follow-up data includes the following specific methods: The specific calculation method for the normal degree of the patient's follow-up data is: ； In the formula, Indicates the normality of the follow-up data of the mth patient; represents the normal coefficient of the follow-up data of the mth patient; represents the degree of recurrence of the mth endometriosis patient; represents the normalization function; represents an exponential function with a natural constant as base; The method of obtaining the correlation between the review and follow-up data of any two patients based on the similarity between the review and follow-up data of the two patients and the difference in the normality of the review and follow-up data includes the following specific methods: Obtaining the similarity of the segmentation sequences in the review and follow-up data of any two patients, wherein the similarity quantifies the similarity of the review and follow-up data of any two patients; The specific calculation method for the correlation between the follow-up data of any two patients is: ； In the formula, represents the correlation between the follow-up data of the pth and qth patients; Indicates the normality of the follow-up data of the pth patient; Indicates the normality of the follow-up data of the qth patient; represents the similarity of the segmentation sequence k in the follow-up data of the p-th and q-th patients, ; Indicates the absolute value symbol; represents an exponential function with a natural constant as base; represents the normalization function; The specific calculation method of the segmentation similarity of the segmentation sequences in the review and follow-up data of any two patients is: ； In the formula, represents the maximum number of follow-up visits between the pth and qth patients; The pth and qth patients had the segmentation sequence k in the jth follow-up. Correlation coefficient; The method of classifying all patients with endometriosis according to the correlation of the patients' follow-up data includes: The silhouette coefficient method was used to determine the number of clusters; According to the correlation of the follow-up data between patients, the k-means clustering algorithm was used to cluster all patients with endometriosis in the hospital information system, and the classification results of several patients were obtained.

2. A method for processing review and follow-up data of gynecological endometriosis patients according to claim 1, characterized in that: The specific method for obtaining the word segmentation sequence is: The patient's condition information includes the patient's symptoms and medication information of the medication taken, and the follow-up content includes the patient's condition recovery; Use JIEBA word segmentation tool to segment the disease information and follow-up content in the review and follow-up data; In the result of word segmentation processing of medical condition information: the sequence composed of symptom segmentations is recorded as segmentation sequence 1, and the sequence composed of drug information segmentations is recorded as segmentation sequence 2; in the result of word segmentation processing of follow-up content: the sequence composed of medical condition recovery status segmentations is recorded as segmentation sequence 3.

3. A method for processing review and follow-up data of gynecological endometriosis patients according to claim 2, characterized in that: The specific method of analyzing the degree of change of the word segmentation sequence of the condition information in the patient's reexamination and follow-up data to determine the normal coefficient of the patient's reexamination and follow-up data includes: Obtain the coefficient of symptom change at each follow-up examination for any patient; Obtaining the similarity of the segmentation sequence 2 in the segmentation processing results of each review and follow-up data of the patient; Combining the similarity of the segmentation sequence 2 with the symptom change degree coefficient to obtain the normal coefficient of the patient's follow-up data; The coefficient of symptom change degree of the patient at each follow-up examination is inversely proportional to the normal coefficient of the follow-up data; The similarity of the segmentation sequence 2 in the segmentation processing results of each review and follow-up data is proportional to the normal coefficient of the review and follow-up data.

4. A method for processing review and follow-up data of gynecological endometriosis patients according to claim 3, characterized in that: The specific method for obtaining the symptom change degree coefficient of any patient at each follow-up examination is: The participles that did not appear in the nth review follow-up data but appeared in the n+1th review follow-up data were recorded as newly added participles; The specific calculation method of the symptom change coefficient at any follow-up visit of the patient is: ； In the formula, represents the coefficient of symptom change of the mth patient at the n+1th follow-up visit, It indicates the number of new segmentations in segmentation sequence 1 during the n+1th follow-up of the mth patient; represents the number of segmentations in segmentation sequence 1 for the mth patient at the nth follow-up visit; It indicates the number of identical segmentations in segmentation sequence 1 in the follow-up data of the mth patient, the n+1th and the previous review data.

5. The method for processing review and follow-up data of gynecological endometriosis patients according to claim 3, characterized in that: The specific method for obtaining the similarity of the segmentation sequence 2 in the segmentation processing results of each follow-up data of the patient is: Calculate the difference between the word sequence 2 of each follow-up data of the patient and the word sequence 2 of the previous follow-up data. Correlation coefficient, and the The correlation coefficient is recorded as the similarity of the segmentation sequence 2 in the segmentation processing results of each follow-up data of the patient.

6. The method for processing review and follow-up data of gynecological endometriosis patients according to claim 2, characterized in that: The specific method of obtaining the degree of recurrence of the patient's condition according to the degree of change of the word segmentation sequence of the follow-up content in the patient's reexamination follow-up data includes: Obtain the sentiment score of word sequence 3 in the follow-up data of any patient since treatment, and calculate the mean of the sentiment score of the patient; Construct a scatter plot of emotional scores according to the review and follow-up time; the horizontal axis is the review and follow-up time; The ordinate is the sentiment score value in the word segmentation sequence 3 corresponding to the review follow-up time; the least squares method is used to fit the data points in the sentiment score scatter plot to obtain the sentiment score fluctuation curve, and the average slope on the sentiment score fluctuation curve is calculated; The degree of recurrence of the patient's condition is obtained by combining the patient's emotional score mean and the average slope of the emotional score fluctuation curve; The mean value of the patient's emotion score is inversely proportional to the degree of recurrence of the patient's condition, and the average slope of the patient's emotion score fluctuation curve is directly proportional to the degree of recurrence of the patient's condition.

Images