Data processing system for lactation reexamination and follow-up visit of endometriosis patient

Abstract

The invention discloses an endometriosis patient lactation review follow-up visit data processing system, and relates to the technical field of medical data processing, the system collects patient hospital information, follow-up visit records, autonomous filling data and equipment monitoring data, and adopts a BERT medical NLP model to carry out vectorization processing on text illness state records; standardized storage is realized in combination with NER, and a patient data set is sorted according to a time sequence; scoring abnormal data by using a local abnormal factor and an isolated forest algorithm, and providing an abnormal data processing strategy; a trend change coefficient BQS is calculated based on time sequence analysis and an EWMA method, and the change trend of the illness state is monitored and early warning is carried out; the method comprises the following steps: analyzing patient condition tracks through Euclidean distance, cosine similarity and a clustering algorithm, calculating a similarity score Sxs, judging condition similarity, and providing a corresponding strategy for high-similarity patients.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and particularly to a data processing system for the follow-up review of endometriosis patients during lactation. 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. The occurrence and development of endometriosis are closely related to lactation and menstrual patterns. Due to changes in lactation and menstrual patterns during lactation, the disease may improve or progress. However, there is currently a lack of management guidelines for this disease during lactation both at home and abroad, resulting in the lack of an effective management model during this special physiological stage of female lactation. Close follow-up during lactation has become a crucial link in the treatment process of patients. When processing the follow-up review data of patients, the data can be classified according to multiple aspects such as the patient's menstrual pattern during lactation, lactation situation, differences in treatment plans, and symptom manifestations.

[0003] However, endometriosis is a disease that is prone to recurrence and has dynamic changes. As patients undergo regular follow-up reviews, new symptom characteristics may appear at different review stages. Therefore, the patient's condition has a certain degree of volatility during lactation, resulting in dynamic changes in the follow-up review data. This change makes it difficult for traditional data classification methods to accurately reflect the patient's condition and reduces the accuracy of data classification results. If data anomalies cannot be effectively detected, the trend of disease changes cannot be analyzed, and the similarity between the patient's disease trajectory and that of a certain type of historical case cannot be monitored, many problems may occur, such as data anomalies not being discovered in a timely manner, the lack of effective disease trend analysis, and the lack of similarity analysis between the disease trajectory and historical cases. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides a data processing system for the follow-up review of endometriosis patients during lactation to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: A data processing system for the follow-up review of endometriosis patients during lactation, including a data collection module, a data structured classification module, a data anomaly detection module, a disease trend analysis module, and a disease similarity monitoring module;

[0006] The data collection module is used to collect the patient's hospital information data, follow-up information data, self-reported information data, and device monitoring data;

[0007] The data structured classification module is used to perform vectorization processing on text - type medical records, traditional Chinese medicine data, and lactation - related health data through the BERT medical NLP model, extract semantic features, and use part - of - speech tagging and named entity recognition NER to convert disease descriptions, drug names, and symptoms into standard terms. The disease names are standardized using ICD - 10 coding, and the drug names are unified using the ATC classification system; the processed text data is stored in a structured database format, and a patient dataset is established; all timestamps are converted to a standard format, the disease data is arranged in the order of the patient's follow - up time. For data related to lactation, time markers related to the lactation cycle are added, including the colostrum stage, lactation stage, and involution stage, and the follow - up records and physiological data are matched;

[0008] The data anomaly detection module is used to adopt the local outlier factor and isolation forest algorithms to score abnormal data points, detect abnormal data in real - time, combine with the data in the patient dataset, calculate the comprehensive data score Stotal, and compare and analyze it with the first threshold Q1 to judge whether the data conforms to the distribution and give strategies for abnormal data;

[0009] The disease trend analysis module is used to adopt time - series analysis, sliding window technology, and exponentially weighted moving average EWMA method to monitor the dynamic changes of the patient's condition in real - time, calculate the disease trend change coefficient BQS, and compare and analyze it with the second threshold Q2 to judge whether the patient's disease change trend is normal and give strategies for abnormal trends;

[0010] The disease similarity monitoring module is used to use multiple similarity measurement methods including Euclidean distance and cosine similarity to compare the disease change trends between different patients and group the patients; and monitor the conditions of the grouped patient populations and similar historical cases in real - time, calculate the similarity score coefficient Sxs, and compare and analyze it with the third threshold Q3 to judge whether the disease trajectory of the patient is similar to that of a certain type of historical case and give strategies for patients with similar disease trajectories.

[0011] Preferably, the data acquisition module includes a hospital information acquisition unit, a follow - up acquisition unit, a patient self - reporting unit, a device acquisition unit, and a traditional Chinese medicine data acquisition unit;

[0012] The hospital information acquisition unit is used to extract the patient's disease records, examination reports, and medication data from the hospital's electronic medical record EMR;

[0013] The follow - up acquisition unit is used to collect the recovery situation, symptom changes, patient feedback data, ultrasound examination data, prolactin test data, milk production, menstrual pattern, lactation - related disease descriptions, medication conditions, and breastfeeding impact data recorded by doctors during the patient's follow - up;

[0014] The patient self-reporting unit is used for patients to submit daily symptoms, perceived changes in the condition, medication information, menstrual patterns, and lactation status through a mobile application and an online questionnaire;

[0015] The device acquisition unit is used to collect patients' physiological indicators including heart rate, blood pressure, body temperature, and respiratory rate using physiological monitoring devices;

[0016] The traditional Chinese medicine data acquisition unit is used to collect the four diagnostic information of traditional Chinese medicine, including inspection data, interrogation data, and palpation data, and collect data on traditional Chinese medicine prescriptions, acupuncture, and massage therapies.

[0017] Preferably, the data structured classification module includes a data processing unit and a time series processing unit;

[0018] The data processing unit is used to vectorize the text-based medical records and follow-up contents of patients using a BERT pre-trained medical NLP model, extract semantic features, and perform structured storage; combine part-of-speech tagging and named entity recognition NER to convert disease descriptions, drug names, and symptom contents into standard terms to reduce text ambiguity; perform standardized mapping on symptoms and disease names using the ICD-10 international disease classification code to make all disease descriptions consistent, and use the ATC anatomical therapeutic chemical classification system code for drug names and treatment plans to unify drug names from different sources; store the processed text data in a structured database format and establish a patient dataset; ·

[0019] The time series processing unit is used to unify the time format of the data, convert all timestamps into a standardized format, and arrange the patients' condition data in sequence according to the patients' follow-up time points to make the time order of the condition data correct and continuous.

[0020] Preferably, the data anomaly detection module includes a first calculation unit and a first analysis unit;

[0021] The first calculation unit is used to score abnormal data points using the local outlier factor and isolation forest algorithms, identify possible unreasonable data, detect abnormal data in real time, and calculate and obtain the comprehensive data score Stotal after dimensionless processing in combination with the data in the patient dataset. The formula is as follows:

[0022]

[0023] In the formula, n represents the total number of data indicators, Av i represents the i-th data indicator, f(Av i ) represents the abnormal scoring function corresponding to the data indicator, a1 represents the weight coefficient of the indicator score, which is obtained through model training, m represents the total number of features used to calculate the data features, D jDenote the j-th data feature score, a2 represents the weight coefficient of the feature score, which is obtained through model training, w1 represents the weight coefficient for adjusting the influence of the data index score, and w2 represents the weight coefficient for adjusting the influence of the data feature score.

[0024] Preferably, the first analysis unit is used to preset a first threshold Q1 in advance, and compare and analyze the data anomaly coefficient SJY with the first threshold Q1 to obtain the first evaluation result, including:

[0025] When the data comprehensive score Stotal ≤ the first threshold Q1, it indicates that the data conforms to the distribution, there is no anomaly, and continuous monitoring is carried out;

[0026] When the data comprehensive score Stotal > the first threshold Q1, it indicates that the data does not conform to the distribution, there are data errors, and there is a risk of disease progression and detection errors, triggering the first warning instruction and generating the first strategy: initiate manual review of the medical history and clinical examination.

[0027] Preferably, the disease trend analysis module includes a second calculation unit and a second analysis unit;

[0028] The second calculation unit is used to adopt time series analysis, sliding window technology, and exponentially weighted moving average EWMA method to monitor the dynamic changes of the patient's condition in real time. Combining the patient dataset data, after dimensionless processing, calculate and obtain the disease trend change coefficient BQS, and the formula is as follows:

[0029]

[0030] In the formula, M represents the total number of disease characteristics, t represents the reexamination time point, T represents the length of the time window period, d is the tiny change in the differential time, Bv k (t) represents the value of the k-th disease index at time t, γ represents the time decay factor, α k represents the weight coefficient of the k-th disease index.

[0031] Preferably, the second analysis unit is used to preset a second threshold Q2 in advance, and compare and analyze the disease trend change coefficient BQS with the second threshold Q2 to obtain the second evaluation result, including:

[0032] When the disease trend change coefficient BQS < the second threshold Q2, it indicates that the disease change trend is normal and there is no risk of disease deterioration, and continuous monitoring is carried out;

[0033] When the disease trend change coefficient BQS ≥ the second threshold Q2, it indicates that the disease change trend is abnormal and there is a risk of disease deterioration, triggering the second warning instruction and generating the second strategy: give suggestions for adjusting the treatment plan.

[0034] Preferably, the disease similarity monitoring module includes a grouping unit, a third calculation unit, and a third analysis unit;

[0035] The grouping unit is used to compare the disease change trends among different patients using multiple similarity measurement methods including Euclidean distance and cosine similarity. Based on the similarity between disease characteristics, the nonlinear clustering algorithms K-means and DBSCAN are used to group the patients, and a group of patients with similar disease trajectories is established.

[0036] Preferably, the third calculation unit is used to monitor the disease conditions of the patient group and similar historical cases in real time. Combining the data in the patient data set, after dimensionless processing, the similarity score coefficient Sxs is calculated, and the formula is as follows:

[0037]

[0038] In the formula, M represents the total number of disease characteristics, Bv k (t) represents the value of the k-th disease index at time t, represents the disease index value of the historical case at the same time point t, α k represents the weight coefficient of the k-th disease index, represents the maximum value of the k-th disease index, represents the minimum value of the k-th disease index.

[0039] Preferably, the third analysis unit is used to preset a third threshold Q3 in advance, and compare and analyze the similarity score coefficient Sxs with the third threshold Q3 to obtain the third evaluation result, including:

[0040] When the similarity score coefficient Sxs < the third threshold Q3, it means that the disease trajectory of the patient does not resemble the trajectory of a certain type of historical case, and continuous monitoring is required;

[0041] When the similarity score coefficient Sxs ≥ the third threshold Q3, it means that the disease trajectory of the patient resembles the trajectory of a certain type of historical case, triggering a third warning instruction and generating a third strategy: prompting that the disease change of the patient requires special attention, and recommending corresponding treatment measures or intervention plans.

[0042] The present invention provides a data processing system for the follow-up review during lactation of endometriosis patients. It has the following beneficial effects:

[0043] (1) The data processing system for the follow-up review during lactation of endometriosis patients can effectively reduce the ambiguity in text data and ensure the accurate classification of follow-up review data by using the BERT medical NLP model to structurally process the patient's condition records, and using the ICD-10 coding to standardize disease descriptions and the ATC classification system to unify drug names. This improvement makes the classification results of the data more reliable and avoids the accuracy problems of traditional classification methods when dealing with the dynamically changing conditions of patients.

[0044] (2) The data processing system for the follow-up review during lactation of endometriosis patients can detect anomalies in patient data through the local outlier factor and isolation forest algorithms, and combine comprehensive scoring methods to timely discover data anomalies. Through comparative analysis with the set threshold, it can trigger an alarm when data anomalies occur, effectively avoiding the failure to timely detect anomalies in the condition data, thereby improving the accuracy and response speed of condition monitoring.

[0045] (3) The data processing system for the follow-up review during lactation of endometriosis patients can, through time series analysis, sliding window technology, and the exponentially weighted moving average EWMA method, real-time track the dynamic changes of the patient's condition and calculate the condition trend change coefficient BQS. When the condition change trend is abnormal, the system can automatically trigger an alarm and give suggestions for adjusting the treatment plan, thereby ensuring the personalization and timeliness of the patient's treatment process.

[0046] (4) The data processing system for the follow-up review during lactation of endometriosis patients can, through similarity measurement methods such as Euclidean distance and cosine similarity, compare the condition change trends between different patients and group patients with similar condition trajectories. By real-time monitoring the condition similarity, the system can identify patients similar to historical cases and provide them with special attention and corresponding treatment intervention plans, thereby optimizing the treatment effect of patients. Brief Description of the Drawings

[0047] Figure 1 It is a block diagram and flowchart of the data processing system for the follow-up review during lactation of endometriosis patients according to the present invention. Detailed Embodiments

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] Embodiment 1

[0050] Please refer to Figure 1 , the present invention provides a data processing system for the follow-up review of endometriosis patients during lactation, including a data collection module, a data structured classification module, a data anomaly detection module, a disease trend analysis module, and a disease similarity monitoring module;

[0051] The data collection module is used to collect the hospital information data, follow-up information data, self-reported information data, and device monitoring data of patients;

[0052] The data structured classification module is used to perform vectorization processing on text-based disease records, traditional Chinese medicine data, and lactation-related health data through the BERT medical NLP model, extract semantic features, and use part-of-speech tagging and named entity recognition NER to convert disease descriptions, drug names, and symptoms into standard terms. The disease names are standardized using the ICD-10 coding, and the drug names are unified using the ATC classification system; the processed text data is stored in a structured database format, and a patient dataset is established; all timestamps are converted into a standard format, the disease data is arranged in the order of the patient's review time, and for data related to lactation, time markers related to the lactation cycle are added, including the colostrum period, lactation period, and involution period, and the follow-up records and physiological data are matched;

[0053] The data anomaly detection module is used to adopt the local outlier factor and isolation forest algorithms to score the abnormal data points, detect abnormal data in real time, combine the data in the patient dataset, calculate the comprehensive data score Stotal, and compare and analyze it with the first threshold Q1 to determine whether the data conforms to the distribution and give strategies for the abnormal data;

[0054] The disease trend analysis module is used to adopt time series analysis, sliding window technology, and exponential weighted moving average EWMA method to monitor the dynamic changes of the patient's disease in real time, calculate the disease trend change coefficient BQS, and compare and analyze it with the second threshold Q2 to determine whether the patient's disease change trend is normal and give strategies for the abnormal trend;

[0055] The disease similarity monitoring module is used to use various similarity measurement methods including Euclidean distance and cosine similarity to compare the disease change trends between different patients, group the patients; and monitor the disease conditions of the grouped patient populations and similar historical cases in real time, calculate the similarity score coefficient Sxs, and compare and analyze it with the third threshold Q3 to determine whether the disease trajectory of the patient is similar to that of a certain type of historical case and give strategies for the patients with similar disease trajectories.

[0056] In this embodiment, through the comprehensive application of the data acquisition module, data structured classification module, data anomaly detection module, disease trend analysis module, and disease similarity monitoring module, this system can accurately collect and process the disease data of patients, detect abnormal conditions in real time, and analyze the changing trends of the disease. At the same time, through the similarity analysis of the disease trajectories of different patients, patients similar to historical cases can be identified, and then the personalized treatment plan can be optimized. The intelligent data processing and predictive analysis of this system can effectively improve the accuracy and timeliness of disease monitoring, ensuring the best personalized adjustment and intervention during the patient's treatment process.

[0057] Embodiment 2

[0058] This embodiment is an explanatory description based on Embodiment 1. Specifically, the data acquisition module includes a hospital information acquisition unit, a follow-up acquisition unit, a patient self-reporting unit, a device acquisition unit, and a traditional Chinese medicine data acquisition unit;

[0059] The hospital information acquisition unit is used to extract the patient's disease records, examination reports, and medication data from the hospital's electronic medical record (EMR);

[0060] The follow-up acquisition unit is used to collect the recovery situation, symptom changes, patient feedback data, ultrasound examination data, prolactin test data, milk production volume, menstrual pattern, descriptions of lactation-related diseases, medication situation, and data on the impact of breastfeeding recorded by doctors during the patient follow-up process;

[0061] The patient self-reporting unit is used for patients to submit daily symptoms, perceived disease changes, medication situation information, menstrual pattern, and lactation situation through the mobile application and online questionnaire;

[0062] The device acquisition unit is used to use physiological monitoring devices to collect the patient's physiological indicators, including heart rate, blood pressure, body temperature, and respiratory rate;

[0063] The traditional Chinese medicine data acquisition unit is used to collect the four diagnostic information of traditional Chinese medicine, including inspection data, interrogation data, and palpation data, and collect data on traditional Chinese medicine prescriptions, acupuncture, and massage therapies.

[0064] In this embodiment, by integrating the hospital information acquisition unit, the follow-up acquisition unit, the patient self-reporting unit, the device acquisition unit, and the traditional Chinese medicine data acquisition unit, this system can comprehensively and real-time collect the patient's disease data, follow-up records, symptom changes, and physiological indicators. The feedback information provided by patients through the mobile application and online questionnaire is combined with the medical device monitoring data, ensuring the accurate monitoring and dynamic tracking of the patient's health status. This all-round data acquisition method not only improves the coverage and real-time nature of the data but also better supports disease analysis and personalized treatment decisions, promoting the refinement and scientificization of patient health management.

[0065] Example 3

[0066] This example is an explanatory note based on Example 2. Specifically, the data structured classification module includes a data processing unit and a time series processing unit;

[0067] The data processing unit is used to vectorize the text-based medical records and follow-up content of patients using the BERT pre-trained medical NLP model, extract semantic features, and perform structured storage; combined with part-of-speech tagging and named entity recognition NER, convert disease descriptions, drug names, and symptom content into standard terms to reduce text ambiguity; for symptoms and disease names, use the ICD-10 international disease classification code for standard mapping to make all disease descriptions consistent, and for drug names and treatment plans, use the ATC anatomical therapeutic chemical classification system code to unify drug names from different sources; store the processed text data in a structured database format and establish a patient dataset;·

[0068] The time series processing unit is used to unify the time format of the data, convert all timestamps into a standard format, and arrange the patient's condition data in sequence according to the patient's reexamination time points, so that the time sequence of the condition data is correct and continuous.

[0069] In this example, through the BERT pre-trained medical NLP model to vectorize the patient's condition records and follow-up content, combined with part-of-speech tagging and named entity recognition technology NER, this system can effectively extract the semantic features of disease descriptions, drug names, and symptom content, automatically convert text data into standard terms, and use the ICD-10 and ATC coding systems to standardize the condition and treatment information. Combining the time series processing unit to standardize and sort the timestamps of the reexamination data to ensure the accuracy and continuity of the condition data. This process not only improves the readability and consistency of the condition data but also provides high-quality structured data support for subsequent data analysis and decision-making.

[0070] Example 4

[0071] This example is an explanatory note based on Example 3. Specifically, the data anomaly detection module includes a first calculation unit and a first analysis unit;

[0072] The first calculation unit is used to use the local outlier factor and isolation forest algorithm to score the abnormal data points, identify possible unreasonable data, and detect abnormal data in real time. Combining the data in the patient dataset, after dimensionless processing, calculate and obtain the comprehensive data score Stotal, and the formula is as follows:

[0073]

[0074] In the formula, n represents the total number of data indicators, Av i represents the i-th data indicator, f(Av i ) represents the abnormal score function corresponding to the data indicator, a1 represents the weight coefficient of the indicator score, which is obtained through model training, m represents the total number of features used to calculate data features, D j represents the score of the j-th data feature, a2 represents the weight coefficient of the feature score, which is obtained through model training, w1 represents the weight coefficient for adjusting the influence of the data indicator score, and w2 represents the weight coefficient for adjusting the influence of the data feature score.

[0075] In this embodiment, through the local outlier factor and the isolation forest algorithm, the data anomaly detection module can perform real-time anomaly scoring on the patient's condition data and identify possible unreasonable data. Combining the patient data set, the comprehensive score Stotal is calculated after dimensionless processing to accurately analyze whether the data conforms to the normal distribution and compare it with the preset threshold to timely detect potential abnormal data. This function effectively improves the sensitivity and accuracy of data monitoring, can trigger an alarm when the data is abnormal, supports medical staff to quickly take countermeasures, and reduces misdiagnosis or treatment delays caused by data problems.

[0076] Embodiment 5,

[0077] This embodiment is an explanatory description based on Embodiment 4. Specifically, the first analysis unit is used to preset a first threshold Q1 in advance and compare and analyze the data anomaly coefficient SJY with the first threshold Q1 to obtain the first evaluation result, including:

[0078] When the data comprehensive score Stotal ≤ the first threshold Q1, it indicates that the data conforms to the distribution, there is no anomaly, and continuous monitoring is carried out;

[0079] When the data comprehensive score Stotal > the first threshold Q1, it indicates that the data does not conform to the distribution, there are data errors, and there is a risk of disease progression and detection errors, triggering a first warning instruction and generating a first strategy: initiate manual review of the medical history and clinical examination.

[0080] In this embodiment, through the set first threshold Q1, real-time comparative analysis of the data comprehensive score Stotal can accurately judge whether the patient data conforms to the normal distribution. When the score exceeds the threshold, the system will timely identify potential data errors, disease progression or detection error risks and trigger a warning instruction. This mechanism effectively reduces the misdiagnosis risk caused by the failure to timely detect data problems, and at the same time initiates manual review of the medical history and clinical examination to ensure the accurate adjustment of the treatment plan and improve the safety and reliability of medical decision-making.

[0081] Embodiment 6,

[0082] This embodiment is an explanatory description carried out in Embodiment 3. Specifically, the disease trend analysis module includes a second calculation unit and a second analysis unit;

[0083] The second calculation unit is used to adopt time series analysis, sliding window technology and exponentially weighted moving average (EWMA) method to monitor the dynamic changes of the patient's condition in real time. Combining the data in the patient dataset, after dimensionless processing, the disease trend change coefficient BQS is calculated and obtained. The formula is as follows:

[0084]

[0085] In the formula, M represents the total number of disease characteristics, t represents the reexamination time point, T represents the length of the time window period, d is the tiny change amount in the differential time, and Bv k (t) represents the value of the kth disease index at time t, γ represents the time decay factor, and α k represents the weight coefficient of the kth disease index.

[0086] In this embodiment, the disease trend analysis module can monitor the dynamic changes of the patient's condition in real time by adopting time series analysis, sliding window technology and exponentially weighted moving average (EWMA) method. By calculating the disease trend change coefficient BQS, the system can timely identify the tiny fluctuations of the disease changes, so as to predict the development trend of the patient's condition. This function helps doctors to more accurately master the real-time changes of the condition and provide personalized treatment plans for patients, ensuring early intervention and improving the treatment effect and the prognosis of patients.

[0087] Embodiment 7

[0088] This embodiment is an explanatory description carried out in Embodiment 6. Specifically, the second analysis unit is used to preset a second threshold Q2 in advance, and compare and analyze the disease trend change coefficient BQS with the second threshold Q2 to obtain the second evaluation result, including:

[0089] When the disease trend change coefficient BQS < the second threshold Q2, it indicates that the disease change trend is normal and there is no risk of disease deterioration, and continuous monitoring is carried out;

[0090] When the disease trend change coefficient BQS ≥ the second threshold Q2, it indicates that the disease change trend is abnormal and there is a risk of disease deterioration, triggering a second warning instruction and generating a second strategy: giving suggestions for adjusting the treatment plan.

[0091] In this embodiment, through the combination of the disease trend analysis module and the second analysis unit, the system can automatically monitor the disease trend change coefficient BQS and compare it with the preset second threshold Q2. When the disease change trend is abnormal, the system will trigger an alarm and automatically generate suggestions for adjusting the treatment plan. This function helps doctors identify and take intervention measures in the early stage of disease deterioration, thereby minimizing the risk of patients and ensuring the timely optimization of the treatment plan, as shown in the following table:

[0092]

[0093] Table 1.

[0094] Example 8

[0095] This embodiment is an explanatory description based on Embodiment 7. Specifically, the disease similarity monitoring module includes a grouping unit, a third calculation unit, and a third analysis unit;

[0096] The grouping unit is used to compare the disease change trends between different patients using multiple similarity measurement methods including Euclidean distance and cosine similarity. Through the similarity between disease characteristics, the patients are grouped using the non - linear clustering algorithms K - means and DBSCAN, and a group of patients with similar disease trajectories is established.

[0097] In this embodiment, through the grouping unit of the disease similarity monitoring module, the system can compare the disease change trends of patients using multiple similarity measurement methods such as Euclidean distance and cosine similarity, and group the patients with similar disease trajectories through the non - linear clustering algorithms K - means and DBSCAN. This function enables the system to accurately identify groups of patients with similar disease developments, providing a basis for formulating personalized treatment plans.

[0098] Example 9

[0099] This embodiment is an explanatory description based on Embodiment 8. Specifically, the third calculation unit is used to monitor the disease conditions of the patient group and similar historical cases in real - time. After dimensionless processing in combination with the patient dataset data, the similarity score coefficient Sxs is calculated, and the formula is as follows:

[0100]

[0101] In the formula, M represents the total number of disease characteristics, Bv k (t) represents the value of the k - th disease index at time t, represents the disease index value of the historical case at the same time point t, α k represents the weight coefficient of the k - th disease index, represents the maximum value of the k - th disease index, Represents the minimum value of the k-th disease index.

[0102] In this embodiment, the third calculation unit monitors the changes in the conditions of the patient group in real time and compares them with similar historical cases. The system can calculate and obtain the similarity scoring coefficient Sxs. This scoring coefficient can reflect the similarity between the patient's current condition and historical cases, thereby helping doctors identify whether the patient's condition conforms to certain typical disease course patterns and providing real-time data support for personalized treatment interventions. This monitoring and scoring mechanism helps to detect potential disease risks in advance and optimize treatment decisions.

[0103] Embodiment 10

[0104] This embodiment is an explanatory description based on Embodiment 9. Specifically, the third analysis unit is used to preset the third threshold Q3 in advance and compare the similarity scoring coefficient Sxs with the third threshold Q3 for comparative analysis. The obtained third evaluation results include:

[0105] When the similarity scoring coefficient Sxs < the third threshold Q3, it indicates that the disease trajectory of the patient does not resemble the trajectory of a certain type of historical case, and continuous monitoring is required;

[0106] When the similarity scoring coefficient Sxs ≥ the third threshold Q3, it indicates that the disease trajectory of the patient resembles the trajectory of a certain type of historical case, triggering the third warning instruction and generating the third strategy: prompting that the patient's condition change requires special attention and recommending corresponding treatment measures or intervention plans.

[0107] In this embodiment, by setting the third threshold Q3 by the third analysis unit and comparing the similarity scoring coefficient Sxs, the system can determine whether the disease trajectory of the patient resembles the trajectory of a certain type of historical case. When the disease trajectory resembles that of a historical case, the system will trigger an alarm and recommend corresponding treatment or intervention plans, thereby ensuring that the changes in the patient's condition are promptly noticed and dealt with. This mechanism can significantly improve the accuracy of disease management and reduce the risk of disease deterioration, as shown in the following table:

[0108]

[0109] Table 2.

[0110] The setting of the size of the threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0111] The above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value. The coefficients in the formula are set by those skilled in the art according to the actual situation. As described above, this is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, shall be covered by the protection scope of the present invention.

Claims

1. A data processing system for follow-up review during lactation of patients with endometriosis, characterized in that, It includes a data collection module, a data structured classification module, a data anomaly detection module, a disease trend analysis module, and a disease similarity monitoring module; The data collection module is used to collect the hospital information data, follow-up information data, self-reported information data, and device monitoring data of patients; The data structured classification module is used to perform vectorization processing on text-based disease records, traditional Chinese medicine data, and lactation-related health data through the BERT medical NLP model, extract semantic features, and use part-of-speech tagging and named entity recognition NER to convert disease descriptions, drug names, and symptoms into standard terms. The disease names are standardized using the ICD-10 coding, and the drug names are unified using the ATC classification system; the processed text data is stored in a structured database format, and a patient dataset is established; all timestamps are converted into a standard format, and the disease data is arranged in the order of the patient's follow-up time. For data related to lactation, time markers related to the lactation cycle are added, including the colostrum period, lactation period, and involution period, and the follow-up records and physiological data are matched; The data anomaly detection module is used to adopt the local outlier factor and isolation forest algorithms to score abnormal data points, detect abnormal data in real time, combine the data in the patient dataset, calculate the comprehensive data score Stotal, and compare and analyze it with the first threshold Q1 to judge whether the data conforms to the distribution, and give strategies for abnormal data; The disease trend analysis module is used to adopt time series analysis, sliding window technology, and exponentially weighted moving average EWMA method to monitor the dynamic changes of the patient's disease in real time, calculate the disease trend change coefficient BQS, and compare and analyze it with the second threshold Q2 to judge whether the patient's disease change trend is normal, and give strategies for abnormal trends; The disease similarity monitoring module is used to use various similarity measurement methods including Euclidean distance and cosine similarity to compare the disease change trends between different patients and group the patients; and monitor the disease conditions of the grouped patient populations and similar historical cases in real time, calculate the similarity score coefficient Sxs, and compare and analyze it with the third threshold Q3 to judge whether the disease trajectory of the patient is similar to that of a certain type of historical case, and give strategies for patients with similar disease trajectories.

2. The follow-up data processing system for lactating patients with endometriosis according to claim 1, wherein The data collection module includes a hospital information collection unit, a follow-up collection unit, a patient self-reporting unit, a device collection unit, and a traditional Chinese medicine data collection unit; The hospital information collection unit is used to extract the disease records, examination reports, and medication data of patients from the hospital electronic medical record EMR; The follow-up collection unit is used to collect the recovery situation, symptom changes, patient feedback data, ultrasound examination data, prolactin test data, milk secretion volume, menstrual pattern, lactation-related disease descriptions, medication conditions, and breastfeeding impact data recorded by doctors during the patient follow-up process; The patient self-reporting unit is used for patients to submit daily symptoms, perceived disease changes, medication information, menstrual pattern, and lactation situation through the mobile application and online questionnaire; The device acquisition unit is used to collect patients' physiological indicators including heart rate, blood pressure, body temperature, and respiratory rate using a physiological monitoring device; The traditional Chinese medicine data acquisition unit is used to collect the four diagnostic information of traditional Chinese medicine, including inspection data, inquiry data, and palpation data, and collect data on traditional Chinese medicine prescriptions, acupuncture, and massage therapies.

3. The data processing system for follow-up review during lactation of endometriosis patients according to claim 2, characterized in that The data structured classification module includes a data processing unit and a time series processing unit; The data processing unit is used to perform vectorization processing on the text-based medical records and follow-up content of patients using a BERT pre-trained medical NLP model, extract semantic features, and perform structured storage; combine part-of-speech tagging and named entity recognition NER to convert disease descriptions, drug names, and symptom content into standard terms to reduce text ambiguity; perform standardized mapping on symptoms and disease names using the ICD-10 international disease classification code to make all disease descriptions consistent, and use the ATC anatomical therapeutic chemical classification system code for drug names and treatment plans to unify drug names from different sources; Store the processed text data in a structured database format and establish a patient dataset; · The time series processing unit is used to unify the time format of the data, convert all timestamps into a standardized format, and arrange the patients' disease data in sequence according to the patients' follow-up time points to ensure that the time sequence of the disease data is correct and continuous.

4. The data processing system for follow-up review during lactation of endometriosis patients according to claim 3, wherein The data anomaly detection module includes a first calculation unit and a first analysis unit; The first calculation unit is used to use the local outlier factor and isolation forest algorithms to score outlier data points, identify possible unreasonable data, and detect abnormal data in real time. Combine the data in the patient dataset, and after dimensionless processing, calculate and obtain the comprehensive data score Stotal, and the formula is as follows: where n represents the total number of data indicators, Av i represents the i-th data indicator, f(Av i ) represents the anomaly scoring function corresponding to the data indicator, a1 represents the weight coefficient of the indicator score, obtained by model training, m represents the total number of features used to calculate data features, D j represents the score of the j-th data feature, a2 represents the weight coefficient of the feature score, obtained by model training, w1 represents the weight coefficient for adjusting the influence of the data indicator score, w2 represents the weight coefficient for adjusting the influence of the data feature score, 0 < w1 < 1, 0 < w2 < 1, and w1 + w2 = 1.

5. The follow-up data processing system for lactating patients with endometriosis according to claim 4, wherein, The first analysis unit is used to preset a first threshold Q1 in advance, and compare and analyze the data anomaly coefficient SJY with the first threshold Q1 to obtain the first evaluation result including: When the comprehensive data score Stotal ≤ the first threshold Q1, it means that the data conforms to the distribution, there is no anomaly, and continuous monitoring is carried out; When the comprehensive data score Stotal > the first threshold Q1, it means that the data does not conform to the distribution, there are data errors, and there is a risk of disease progression and detection errors, triggering a first warning instruction and generating a first strategy: initiate manual review of the medical history and clinical examination.

6. The data processing system for follow-up review during lactation of endometriosis patients according to claim 3, wherein The disease trend analysis module includes a second calculation unit and a second analysis unit; The second calculation unit is used to use time series analysis, sliding window technology, and exponentially weighted moving average EWMA method to monitor the dynamic changes of the patients' diseases in real time. Combine the data in the patient dataset, and after dimensionless processing, calculate and obtain the disease trend change coefficient BQS, and the formula is as follows: Where M represents the total number of disease characteristics, t represents the reexamination time point, T represents the length of the time window period, d is the tiny change amount in the differential time, and Bv k (t) represents the value of the k-th disease index at time t, γ represents the time decay factor, and α k represents the weight coefficient of the k-th disease index.

7. A data processing system for follow-up review during lactation of endometriosis patients according to claim 6, characterized in that The second analysis unit is used to preset a second threshold Q2 in advance, and compare and analyze the disease trend change coefficient BQS with the second threshold Q2 to obtain the second evaluation result including: When the disease trend change coefficient BQS < the second threshold Q2, it means that the disease change trend is normal and there is no risk of disease deterioration, and continuous monitoring is carried out; When the disease condition trend change coefficient BQS ≥ the second threshold Q2, it indicates that the disease condition change trend is abnormal and there is a risk of disease deterioration, triggering a second warning instruction and generating a second strategy: giving suggestions for adjusting the treatment plan.

8. A data processing system for follow-up review during lactation of endometriosis patients according to claim 7, characterized in that, The disease condition similarity monitoring module includes a grouping unit, a third calculation unit, and a third analysis unit; The grouping unit is used to compare the disease condition change trends between different patients using a variety of similarity measurement methods including Euclidean distance and cosine similarity. Through the similarity between disease condition characteristics, the K-means and DBSCAN non-linear clustering algorithms are used to group patients and establish a group of patients with similar disease condition trajectories.

9. A data processing system for follow-up review during lactation of patients with endometriosis according to claim 8, characterized in that The third calculation unit is used to monitor the disease conditions of the patient group and similar historical cases in real time. After dimensionless processing in combination with the data in the patient dataset, the similarity score coefficient Sxs is calculated, and the formula is as follows: Where M represents the total number of disease characteristics, Bv k (t) represents the value of the k-th disease index at time t, represents the disease index value of historical cases at the same time point t, α k represents the weight coefficient of the k-th disease index, represents the maximum value of the k-th disease index, represents the minimum value of the k-th disease index.

10. A data processing system for follow-up review during lactation of endometriosis patients according to claim 9, characterized in that, The third analysis unit is used to preset a third threshold Q3 in advance and compare and analyze the similarity score coefficient Sxs with the third threshold Q3 to obtain the third evaluation result, including: When the similarity score coefficient Sxs < the third threshold Q3, it indicates that the disease condition trajectory of the patient is not similar to the trajectory of a certain type of historical case, and continuous monitoring is carried out; When the similarity score coefficient Sxs ≥ the third threshold Q3, it indicates that the disease condition trajectory of the patient is similar to the trajectory of a certain type of historical case, triggering a third warning instruction and generating a third strategy: prompting that the patient's disease condition change requires special attention and recommending corresponding treatment measures or intervention plans.

Images