Personalized health management method and system based on endometriosis

Abstract

The invention discloses a personalized health management method and system based on endometriosis. The method comprises the following steps: acquiring physical state intervention data of a patient based on a personalized health management scheme; determining a comprehensive quantitative evaluation score corresponding to the body state intervention data through a comprehensive quantitative evaluation formula, and predicting a disease change trend through a linear regression algorithm; acquiring a microbiome sample analysis result of the patient; performing deep analysis on the comprehensive quantitative evaluation score, the illness state change trend and the microbiome sample analysis result through a preset scheme adjustment strategy analysis model; and optimizing the personalized health management scheme based on the health management adjustment strategy, and performing rehabilitation intervention on the living environment of the patient in linkage with the smart home. According to the method, the personalized health management scheme is adjusted and the intelligent home is linked to intervene by comprehensively quantifying the evaluation score, the illness state change trend and the microbiome sample analysis result, so that a reasonable health management mode is provided for the patient in time.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent healthcare, and particularly relates to a personalized health management method and system based on endometriosis. Background Art

[0002] With the rapid development of medical technology and the increasing emphasis on health management by people, personalized intelligent health management solutions for various chronic diseases have gradually become a research hotspot.

[0003] Endometriosis, as a common chronic gynecological disease, brings many physical discomforts and life troubles to patients, seriously affecting the quality of life of patients. Traditional management of endometriosis mainly relies on patients going to the hospital for regular check-ups, and doctors give relatively broad treatment suggestions and rehabilitation guidance based on limited outpatient examination data, such as basic physiological indicators and lifestyle habits understood through simple inquiries.

[0004] Regarding the traditional related technologies: First, incomplete data collection leads to a superficial understanding of the patient's condition, and it is impossible to accurately grasp the root cause and development trend of the disease. For example, based on only occasional blood indicators and a rough understanding of lifestyle habits, it is difficult to insight into the potential impact of the patient's daily environmental factors and subtle changes in the daily rhythm on the condition, and key pathogenic factors are easily missed; Second, the static and phased management mode makes the health management plan lack timeliness and pertinence.

[0005] The above content is only used to assist in understanding the technical solution of the present invention, and does not represent an admission that the above content is prior art. Summary of the Invention

[0006] The main purpose of the present invention is to provide a personalized health management method and system based on endometriosis, aiming to solve the technical problem of how to provide a reasonable health management plan for patients in a timely manner.

[0007] To achieve the above purpose, the present invention provides a personalized health management method based on endometriosis, and the personalized health management method based on endometriosis includes: Determine a personalized health management plan corresponding to the condition of endometriosis according to the daily data of the patient's physical state, and collect the physical state intervention data of the patient based on the personalized health management plan; Determine the comprehensive quantitative evaluation score corresponding to the physical state intervention data through a comprehensive quantitative evaluation formula, and predict the disease change trend through a linear regression algorithm according to the comprehensive quantitative evaluation score; Obtain the microbiome sample of the patient, and analyze the microbiome sample through a preset bioinformatics analysis method to obtain the analysis result of the microbiome sample; Performing an in-depth analysis of the comprehensive quantitative assessment score, the disease change trend, and the microbiome sample analysis results through a preset scheme adjustment strategy analysis model to obtain a health management adjustment strategy; The personalized health management program is optimized based on the health management adjustment strategy, so that the optimized personalized health management program can be linked with the smart home to perform rehabilitation intervention on the patient's living environment.

[0008] The present invention first determines the personalized health management plan corresponding to the endometriosis condition according to the daily data of the patient's physical state, and collects the patient's physical state intervention data based on the personalized health management plan, determines the comprehensive quantitative evaluation score corresponding to the physical state intervention data through a comprehensive quantitative evaluation formula, and predicts the trend of the disease change through a linear regression algorithm based on the comprehensive quantitative evaluation score, and then obtains the patient's microbiome sample, analyzes the microbiome sample through a preset bioinformatics analysis method to obtain the microbiome sample analysis result, and then performs a deep analysis of the comprehensive quantitative evaluation score, the disease change trend and the microbiome sample analysis result through a preset scheme adjustment strategy analysis model to obtain a health management adjustment strategy, and finally optimizes the personalized health management plan based on the health management adjustment strategy, so that the optimized personalized health management plan links the smart home to perform rehabilitation intervention on the patient's living environment. The present invention customizes and optimizes personalized health management plans for endometriosis patients by collecting multi-source data, processing, modeling and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a structural schematic diagram of a personalized health management method and device based on endometriosis in a hardware operating environment involved in an embodiment of the present invention; Figure 2 It is a flow chart of the first embodiment of the personalized health management method based on endometriosis of the present invention; Figure 3 This is a structural block diagram of the first embodiment of the personalized health management system based on endometriosis of the present invention.

[0010] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0012] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of a personalized health management method for endometriosis in the hardware operating environment of an embodiment of the present invention.

[0013] As shown in Figure 1 , the personalized health management method and device based on endometriosis may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. Optionally, the memory 1005 may also be a storage system independent of the aforementioned processor 1001.

[0014] Those skilled in the art can understand that Figure 1 the structure shown in

[0015] As shown in Figure 1 , in the memory 1005 as a storage medium, there may be included an operating system, a network communication module, a user interface module, and a personalized health management program based on endometriosis.

[0016] In Figure 1 the personalized health management device based on endometriosis shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the personalized health management device of the present invention may be arranged in the personalized health management device based on endometriosis. The personalized health management device based on endometriosis calls the personalized health management program stored in the memory 1005 through the processor 1001, and executes the personalized health management method provided by the embodiments of the present invention.

[0017] The embodiments of the present invention provide a personalized health management method based on endometriosis. Referring toFigure 2 , Figure 2 This is a flow chart of the first embodiment of the personalized health management method based on endometriosis of the present invention.

[0018] In this embodiment, the personalized health management method based on endometriosis includes the following steps: Step S10: Determine a personalized health management plan corresponding to the endometriosis condition according to the patient's daily physical condition data, and collect the patient's physical condition intervention data based on the personalized health management plan.

[0019] It is easy to understand that the execution subject of this embodiment can be a personalized health management system based on endometriosis with functions such as data processing, network communication and program running, or it can be other computer devices with similar functions, etc., and this embodiment is not limited.

[0020] It should be noted that daily physical condition data includes physiological indicators, lifestyle, sleep status, dietary intake, environmental information, and activity data in home and hospital scenarios.

[0021] Physiological indicators: quantitative parameters that reflect the patient's physical function status, such as blood pressure, heart rate, hormone levels, etc. Blood pressure can be measured by an electronic sphygmomanometer, heart rate can be monitored with a smart bracelet or electrocardiogram equipment, and hormone levels generally require blood samples to be collected and tested by professional laboratories. Indicators such as estrogen and progesterone are crucial to the diagnosis of endometriosis. Abnormal hormone fluctuations may induce or aggravate the condition.

[0022] Lifestyle: covers the patient's daily behavioral habits, including exercise habits, smoking and drinking, etc. Exercise habits can be recorded through mobile phone sports software. The type, frequency and duration of exercise. If the patient insists on moderate aerobic exercise for more than 30 minutes three times a week, such as jogging and yoga, it will be beneficial to recovery; smoking and drinking conditions are reported by the patient himself or supervised by his family. Long-term smoking and excessive drinking will interfere with the body's endocrine system and are not conducive to recovery.

[0023] Sleep status: includes information such as sleep duration and sleep quality. The sleep duration can be monitored by the smart bracelet, and the sleep quality is comprehensively judged based on the patient's subjective feelings (such as whether they are easily awakened and have many dreams) combined with the proportion of deep sleep and light sleep monitored by the bracelet. Ensuring 7-8 hours of high-quality sleep every night is conducive to physical recovery.

[0024] Dietary intake: refers to the amount of food and nutrients consumed by the patient every day. With the help of a diet record app, patients can record the details of their three meals a day and snacks to understand their intake of protein, dietary fiber, fat, etc. For example, a long-term high-fat, low-fiber diet may cause inflammation in the body and affect the condition.

[0025] Environmental information: It involves relevant elements of the living and medical environment where the patient is located, such as indoor temperature, humidity, and the pollution level around the living area. Indoor temperature and humidity are measured by temperature and humidity sensors, and the pollution level refers to the local air quality report. Appropriate indoor temperature and humidity and a clean air environment contribute to the physical and mental comfort of the patient and are beneficial to recovery.

[0026] Activity data in home and hospital scenarios: In the home scenario, the activity trajectory and sedentary time of the patient are monitored through smart home devices. For example, cameras combined with intelligent analysis software are used to determine the activity range and duration of the patient in the living room and bedroom. In the hospital scenario, the hospital information system is relied on to record the processes and time nodes of the patient's medical treatment, examinations, and treatments, which is convenient for medical staff to understand the progress of the condition.

[0027] Furthermore, the processing method for determining the personalized health management plan corresponding to the endometriosis condition based on the patient's daily physical condition data is to construct a feature dataset related to the endometriosis condition through a feature selection algorithm according to the patient's daily physical condition data; input the feature dataset into a preset endometriosis condition classification model, obtain different levels of condition-related features through multiple neuron nodes in the hidden layer, and output the condition classification result through the preset endometriosis condition classification model according to the condition-related features; determine the personalized health management plan corresponding to the endometriosis condition based on the daily physical condition data and the condition classification result.

[0028] In specific implementation, it is also necessary to perform data preprocessing on the patient's daily physical condition data, and the data preprocessing includes data cleaning, data normalization, and data filling. Then, according to the preprocessed daily physical condition data, relevant feature data related to the endometriosis condition is screened out through a feature selection algorithm, and the screened feature data related to the endometriosis condition is used to construct a feature dataset related to the endometriosis condition.

[0029] Data cleaning: Since the collected data comes from diverse sources, there may be some incorrect or invalid data. At this time, it is necessary to identify and eliminate these outliers by setting reasonable data ranges, logical checks, etc., to ensure the accuracy of the data. Taking blood pressure data as an example, if the recorded systolic blood pressure is higher than 200 mmHg and the diastolic blood pressure is higher than 120 mmHg, which significantly exceeds the normal physiological range, it is necessary to verify or discard this data.

[0030] Data normalization: The value ranges of different types of health data vary greatly. Therefore, various types of data need to be unified to a similar value range, generally using methods such as linear transformation to make them comparable.

[0031] Data filling: Some data may be missing due to equipment failures, patients' forgetting to record, etc. For example, if a smart bracelet fails to record sleep data on a certain day, it is necessary to fill in the data according to the patient's past sleep data pattern, using methods such as mean, median, or time series-based prediction methods.

[0032] It should be noted that the feature selection algorithms include the Relief algorithm, information gain algorithm, etc. Taking the Relief algorithm as an example, it repeatedly samples and compares the differences in various features between similar samples (data of patients with similar degrees of endometriosis) and dissimilar samples (data of healthy people or patients with other diseases), and assigns a weight to each feature. The higher the weight, the more the feature can distinguish different categories of samples, that is, the stronger the correlation with the endometriosis condition. After algorithm screening, those features that contribute little to the condition judgment and are redundant are removed, leaving the most representative feature subset (i.e., the feature dataset related to the endometriosis condition).

[0033] It also should be noted that the preset endometriosis condition classification model is a classification model constructed and pre-trained based on machine learning or deep learning algorithms. Common ones include support vector machine (SVM), neural network model, etc. Taking the neural network model as an example, it simulates the structure of human brain neurons and includes an input layer, a hidden layer, and an output layer. The input layer receives the feature dataset related to the endometriosis condition, such as key feature values like estrogen level, dietary fiber intake, sleep quality score, etc.; the hidden layer consists of multiple neuron nodes, and through complex weight assignment and activation function operations, it performs deep feature extraction and transformation on the input data. Different hidden layers can learn different levels of condition-related features. For example, the first hidden layer may focus on the interactive effects of hormones and diet, and the second layer may explore the condition clues under the combined action of physiological indicators and lifestyle; the output layer gives the condition classification result based on the processing results of the hidden layer.

[0034] It should be understood that the condition classification results include mild, moderate, or severe.

[0035] Further, the method for determining a personalized health management plan corresponding to the endometriosis condition based on the daily physical condition data and the disease classification result is as follows: detect whether there is heart rate variability data and galvanic skin response data in the daily physical condition data; if not, determine the personalized health management plan corresponding to the endometriosis condition according to the mapping relationship between the disease classification result and the personalized health management plan; if so, input the heart rate variability data and the galvanic skin response data into a preset mental stress quantification model to output the patient's mental stress level; according to the mental stress level and the disease classification result, the personalized health management plan corresponding to the endometriosis condition can be determined according to the mapping relationship between the mental stress level, the disease classification result and the personalized health management plan.

[0036] Heart rate variability (HRV) data: It reflects the subtle changes in the cardiac cycle and is obtained by measuring the differences in the time intervals between consecutive heartbeats. These differences are regulated by the autonomic nervous system. Among them, the low-frequency (LF) component is often related to sympathetic nerve activity, the high-frequency (HF) component is associated with parasympathetic nerve activity, and indicators such as the LF / HF ratio can reflect the balance state of the sympathetic-parasympathetic nerves, thereby indicating the mental stress situation. Heart rate variability (HRV) data is usually obtained by means of professional wearable devices, such as medical-grade smart bracelets or chest strap heart rate monitors. These devices are equipped with high-precision sensors that can continuously and real-time record the heartbeats interval time, calculate various HRV indicators according to the established algorithms through the supporting software, and upload the data to the health management system via wireless transmission methods such as Bluetooth or Wi-Fi.

[0037] Galvanic skin response (GSR) data: Also known as electrodermal activity, it is collected based on the principle that the skin sweat glands change their electrical conductivity under the influence of emotions, psychology and other factors. When a person is under mental stress, the sympathetic nerve is excited, the sweat glands secrete more, which makes the skin conductance increase. By detecting the fluctuation parameters of the skin conductance, such as sweating frequency and conductance change rate, the mental stress changes can be detected. Obtaining galvanic skin response data: Special skin sensors need to be equipped, which are mostly integrated in wearable devices or special hand and wrist monitoring devices. The sensor contacts the skin surface, real-time senses the skin conductance changes, and also sends the data to the background via wireless transmission and converts it into GSR data such as sweating frequency and conductance change rate for analysis.

[0038] Patient mental stress quantification model: This is a model constructed based on machine learning algorithms and is trained using a large number of samples with different mental stress levels and corresponding heart rate variability and galvanic skin response data. The model learns the complex relationship between the two types of data features and the stress level, so that it can accurately output the stress level based on the input real-time HRV and GSR data.

[0039] Based on the mapping relationship between the mental stress level of the patient, the classification result of the patient's condition, and the personalized health management plan, analyze and determine the patient's personalized health management plan.

[0040] The patient's personalized health management plan includes psychological counseling strategies and dietary planning guidelines.

[0041] Regarding the psychological counseling strategy, for example, patient F's condition is moderate and the mental stress is mild.

[0042] For patients like patient F with moderate condition and mild mental stress, the system automatically pushes online psychological counseling appointment notices to the patient's mobile device twice a week. The counseling time slots are intelligently selected by the system based on big data analysis of the patient's daily active time, and each session lasts 40 minutes. The psychological counseling interface is built with an intelligent guidance program that can initially help the patient sort out the cognitive confusion about the disease and relieve anxiety. At the same time, a mobile application with relaxation training levels is pushed, setting daily automatic reminders to guide the patient to complete 20 minutes of interesting relaxation exercises.

[0043] Regarding the dietary planning guidelines, for patient F with moderate condition and mild mental stress, the system automatically pushes personalized breakfast suggestions to the patient's mobile device every morning based on the nutrition database and big data of the patient's past dietary preferences, such as a cup of yogurt rich in probiotics paired with a vegetable sandwich. The intelligent recipe generator is used to ensure a balanced nutrition for lunch and dinner, and the food ingredients are selected considering the local seasonal supply. For example, beef stir-fried with broccoli and brown rice is recommended for lunch, and vegetable tofu soup and sweet potatoes are pushed for dinner.

[0044] It should also be noted that the mapping relationship between the condition classification result and the personalized health management plan: This is a pre-established scientific correspondence system, which is summarized based on a large amount of clinical experience, medical research, and past case data. The mental stress level is divided into mild, moderate, and severe stress. The preset mental stress quantification model is the mental stress quantification model for patients who have completed training.

[0045] For example, for patients diagnosed with mild endometriosis, the corresponding personalized health management plan focuses on fine-tuning of lifestyle. This may include at least 150 minutes of moderate-intensity aerobic exercise per week, such as jogging and swimming, to promote blood circulation in the body and relieve pelvic congestion symptoms; in terms of diet, increase the intake of foods rich in dietary fiber, such as whole wheat bread and fresh fruits and vegetables, and ensure that the daily dietary fiber intake reaches 25 - 30 grams, which helps regulate hormone metabolism and reduce inflammatory reactions; at the same time, ensure 7 - 8 hours of high-quality sleep every night to maintain stable body endocrine and improve the body's self-healing ability through regular work and rest.

[0046] Treatment plan for moderate patients: When the condition classification result of moderate patients is input, the corresponding plan will upgrade the intervention measures. In addition to continuing with appropriate exercise, diet, and sleep requirements, drug-assisted treatment needs to be introduced. The drug selection is based on individual patient differences and may involve non-steroidal anti-inflammatory drugs such as ibuprofen to relieve pain symptoms, or oral contraceptives to regulate hormone levels and inhibit endometrial growth. Also, patients are required to have regular follow-up visits every month, and the condition changes are closely monitored through means such as ultrasound examinations and blood hormone tests to adjust the treatment strategy in a timely manner.

[0047] Treatment plan for severe patients: For severe patients, the mapping relationship points to a more comprehensive and powerful intervention combination. On the one hand, surgical treatment becomes an important option, such as laparoscopic surgery to remove ectopic lesions and reduce the body burden; after the surgery, drugs are used to consolidate the curative effect, such as gonadotropin-releasing hormone agonists to reduce estrogen levels and prevent recurrence. At the level of rehabilitation care, professional rehabilitation therapists are arranged to formulate personalized rehabilitation training plans to strengthen pelvic muscle strength and improve the pelvic environment; psychological counseling is synchronized. Since the psychological pressure brought about by long-term illness will affect the rehabilitation process, professional psychological counselors help patients establish a positive attitude towards the disease through regular communication.

[0048] It should also be noted that the mapping relationship between the mental stress level, the condition classification result, and the personalized health management plan also belongs to a pre-established scientific correspondence system, which is summarized based on a large amount of clinical experience, medical research, and past case data.

[0049] Furthermore, interventions are implemented for patients based on the personalized health management plan, and the physical state intervention data of patients during the implementation of the personalized health management plan is monitored and obtained.

[0050] In this embodiment, for mild patients, if the plan requires 150 minutes of moderate-intensity aerobic exercise per week, a sports APP can be used to remind patients of their exercise schedule and record exercise data to ensure that the exercise is carried out; in terms of diet, a dietitian can communicate online and, based on the patient's daily diet check-in, provide real-time guidance on food combination and intake adjustment; for sleep, a smart sleep monitoring device continuously tracks. If it is found that the patient has insufficient sleep for several consecutive days, sleep aid tips will be pushed in a timely manner.

[0051] When moderate patients receive drug intervention, medical staff use a medication reminder APP to accurately set the medication time and dose reminders. After the patient takes the medicine, they give feedback on the APP, which is convenient for medical staff to follow up on compliance; during regular follow-up visits, the hospital information system makes advance appointments for examination items to optimize the medical treatment process and ensure the continuity of condition monitoring.

[0052] Before and after the surgery of severe patients, the medical team makes careful preparations. During the surgery, the procedures are strictly followed. After the surgery, the rehabilitation therapist guides the rehabilitation training at home or online according to the plan, and the psychological counselor conducts regular video communication to ensure the implementation of multi-faceted interventions.

[0053] Monitoring and obtaining body status intervention data: Wearable devices such as smart bracelets and patch-type physiological index monitors can be used to collect data such as heart rate, blood pressure, and sleep rhythm in real time and transmit them to the cloud via Bluetooth. Smart home devices continuously capture the activity trajectories of the home scene and the environmental temperature and humidity, and analyze the patient's living patterns and comfort. Hospital examination data, such as the healing status of postoperative wounds and hormone indicators after drug treatment, are summarized and integrated through the hospital information system to form a complete data chain, providing a basis for subsequent plan adjustment and dynamically tracking the effectiveness of personalized health management plans.

[0054] Step S20: Determine the comprehensive quantitative evaluation score corresponding to the body status intervention data through a comprehensive quantitative evaluation formula, and predict the trend of disease change based on the comprehensive quantitative evaluation score through a linear regression algorithm.

[0055] It should be noted that the comprehensive quantitative evaluation method is a multi-dimensional and comprehensive evaluation system that integrates medical professional knowledge and data analysis technology. It covers various factors such as physiological indicators, lifestyle compliance, symptom relief, and psychological state. For example, in the dimension of physiological indicators, score ranges are set based on key physiological data. For example, if blood pressure is stable within the normal range, a certain basic score can be obtained, and if there are fluctuations, points will be deducted according to the degree; the same is true for heart rate. A normal sinus rhythm and no abnormal fluctuations will add points, otherwise points will be deducted. For lifestyle compliance, if mild patients complete the weekly exercise duration, reasonable diet, and ensure sleep as required, they can obtain a higher score in this section. Conversely, there are corresponding point deduction rules for each unqualified item.

[0056] Degree of symptom relief: Focus on the reduction of symptoms related to endometriosis in patients, such as the frequency and intensity of pain attacks. If the pain originally occurred 3 times a week and after a period of implementation of the management plan, the number of attacks decreased to 1 time, corresponding points can be obtained according to the preset scoring rules; the same is true for the improvement of symptoms such as pelvic distension and menstrual disorders. Psychological state assessment uses professional psychological scales. For example, if a patient was initially more anxious and depressed, and through psychological counseling and the establishment of rehabilitation confidence, the mentality gradually becomes peaceful, and the score in the psychological section will also increase.

[0057] Calculate the comprehensive quantitative evaluation score: Weighted sum the scores of each dimension according to the preset weights. For example, the weight of physiological indicators accounts for 40%, the lifestyle compliance accounts for 30%, the degree of symptom remission accounts for 20%, and the mental state accounts for 10%. Suppose a patient scores 70 points in physiological indicators, 80 points in lifestyle compliance, 60 points in symptom remission, and 80 points in mental state. Then the comprehensive quantitative evaluation score = 70×0.4 + 80×0.3 + 60×0.2 + 80×0.1 = 72 points. Through such a precise quantification method, intuitively present the overall state of the patient after the implementation of the personalized health management plan, provide a key quantitative basis for subsequent plan adjustment, and accurately control the rehabilitation process.

[0058] In specific implementation, according to the mapping relationship between the interval range where the comprehensive quantitative evaluation score falls and the adjustment strategy of the personalized health management plan, analyze and determine the adjustment strategy of the personalized health management plan, and adjust the personalized health management plan according to the adjustment strategy of the personalized health management plan; the personalized health management plan can be adjusted according to the comprehensive quantitative evaluation score and the trend of disease change to obtain the health management adjustment strategy; the personalized health management plan can also be adjusted according to the comprehensive quantitative evaluation score, the trend of disease change and the analysis results of the microbiome sample to obtain the health management adjustment strategy; if there is the latest endometrial image of the patient, the personalized health management plan can be adjusted according to the comprehensive quantitative evaluation score, the trend of disease change, the endometrial image and the analysis results of the microbiome sample to obtain the health management adjustment strategy.

[0059] In the embodiment of analyzing and determining the adjustment strategy of the personalized health management plan according to the mapping relationship between the interval range where the comprehensive quantitative evaluation score falls and the adjustment strategy of the personalized health management plan, the mapping relationship between the comprehensive quantitative evaluation score interval and the adjustment strategy: This is a key guideline constructed based on a large number of clinical practices and data analyses. For example, if the comprehensive quantitative evaluation score is in the interval of 80 - 100 points, it indicates that the patient's rehabilitation state is good, and the corresponding adjustment strategy may be to maintain the existing plan and appropriately relax some of the stricter restrictions in the lifestyle. For example, if a mild patient originally exercised 150 minutes per week, it can be moderately reduced by 10 - 20 minutes, and occasionally a small amount of desserts are allowed in the diet to improve life satisfaction, while continuously monitoring to ensure the stability of the condition.

[0060] It should also be noted that in the embodiment of adjusting the personalized health management plan according to the comprehensive quantitative evaluation score and the trend of disease change to obtain the health management adjustment strategy, a linear regression method needs to be used to fit the curve of the comprehensive quantitative evaluation score changing with time within the preset time period, and the result information of the change trend is determined according to the slope of the curve.

[0061] The linear regression method is a statistical method widely used in the field of data analysis, aiming to find the linear relationship between variables. In this step, time is used as the independent variable (x), and the comprehensive quantitative evaluation scores corresponding to different time points within the preset time period are used as the dependent variable (y). Through a series of mathematical calculations, a straight line that best fits the distribution of these data points is found, minimizing the sum of the squares of the distances from each data point to this line, so as to approximately describe the changing pattern of the comprehensive quantitative evaluation scores over time with this line.

[0062] Change curve: That is, the straight line obtained through linear regression, which reflects how the comprehensive quantitative evaluation scores change over time. It intuitively shows the dynamic trend of the patient's health management effectiveness during the entire preset time period. The abscissa is time, and the ordinate is the comprehensive quantitative evaluation score. Each point on the curve represents the patient's comprehensive quantitative evaluation status at a certain moment.

[0063] Curve slope: The slope is an important feature of a straight line, indicating the degree of inclination of the line. In this change curve, the slope reflects the change rate of the comprehensive quantitative evaluation scores. A positive slope means the scores increase over time, indicating that the patient's health condition is improving; a slope of zero means the scores are stable, and the condition is in a stable period; a negative slope indicates that the scores are decreasing, suggesting that the health management plan may need to be adjusted, and there is a risk of the patient's condition deteriorating.

[0064] Obtain the health management adjustment strategy and adjust the personalized health management plan according to the comprehensive quantitative evaluation scores and the trend of the disease condition change. The specific operation process is as follows: First, collect the comprehensive quantitative evaluation scores of this patient within several consecutive preset time periods in the past (such as the past 3 three-week periods, a total of 9 weeks), as well as the corresponding time marks, to ensure the accuracy and continuity of the data. These data are usually extracted from the cloud storage system, which has completed the calculation and storage of the comprehensive quantitative evaluation scores within each period.

[0065] Secondly, import the prepared data into professional statistical analysis software or a module with linear regression function, start the linear regression algorithm, and let the software automatically find the optimal fitting straight line according to the built-in mathematical model. According to the fitted straight line equation, directly read the slope value. For example, in the above example, the slope is 0.5. Since the slope is greater than 0, it can be judged that the comprehensive quantitative evaluation scores of this patient show an upward trend during this period, and the health condition is gradually improving; if the slope is -0.3, it indicates that the scores are decreasing, and the condition has a tendency to worsen.

[0066] Analyze and determine the personalized health management plan adjustment strategy according to the mapping relationship between the interval range where the current comprehensive quantitative evaluation score falls, the result information of the change trend, and the personalized health management plan adjustment strategy.

[0067] Comprehensive Quantitative Assessment Score Range: Based on a large amount of clinical practice and data analysis, the comprehensive quantitative assessment scores are divided into different ranges, and each range represents a different level of recovery status. For example, scores from 85 to 100 can be set as excellent, indicating that the patient's physical functions have recovered well, symptoms have significantly alleviated, and the quality of life is high; scores from 60 to 84 are medium, meaning that the patient has made some progress, but there is still room for improvement, and some symptoms or indicators need to be improved; scores below 60 are poor, indicating that the current health management plan is not effective and the condition has not been effectively controlled.

[0068] Trend Result Information: The trend of the comprehensive quantitative assessment score over time judged by the curve slope is divided into three types: rising, stable, and falling. A rising trend indicates that the patient is developing towards recovery; stable means that the plan maintains the stability of the condition; falling warns of the risk of deterioration of the condition and requires timely adjustment of strategies.

[0069] Mapping Relationship between the Range of the Current Comprehensive Quantitative Assessment Score, the Result Information of the Trend, and the Adjustment Strategy of the Personalized Health Management Plan: This is a pre-constructed scientific correlation system. Combining different score ranges and trends, corresponding adjustment strategies are matched. The adjustment strategies mainly involve aspects of exercise, diet, and rehabilitation monitoring.

[0070] Taking the example of a patient in the "excellent" range (85 - 100 points) with a rising score, the following settings are as follows: Exercise: Smart exercise devices (such as smart bracelets or exercise APPs) automatically identify the patient's daily exercise habits. When it detects that the exercise items are single for a consecutive week, it will push exercise adjustment suggestions, suggesting that 1 - 2 outdoor cycling or yoga activities can be arranged within this week, each lasting 30 - 45 minutes, and provide the surrounding cycling routes and online yoga course links. During the exercise process, the device real-time monitors the heart rate, exercise duration, etc., to ensure that the exercise intensity is appropriate. Once abnormalities are found, it will give a voice reminder in a timely manner.

[0071] Diet: The diet record APP, based on the patient's recent diet data, pushes messages on weekends, informing that 1 - 2 small servings of low-sugar desserts or sugar-free functional beverages can be purchased this week, and at the same time attaching the nutritional component description and recommendations of nearby merchants for purchase, helping the patient maintain a balanced diet while enjoying appropriate food.

[0072] Rehabilitation Monitoring: The health management system automatically adjusts the follow-up visit reminder, changing the original monthly in-person follow-up visit reminder to once every two months, and through the APP pop-up window at a fixed time every month, guiding the patient to connect the smart wearable device and upload recent body data (such as indicators like heart rate, blood pressure, sleep, etc.) with one key for remote evaluation by the doctor. The system presets the abnormal data threshold, and once it finds that the data fluctuation exceeds the normal range, it immediately pushes an emergency follow-up visit notice.

[0073] Step S30: Obtain the microbiome sample of the patient, and analyze the microbiome sample through a preset bioinformatics analysis method to obtain the microbiome sample analysis result.

[0074] Step S40: Deeply analyze the comprehensive quantitative evaluation score, the disease condition change trend, and the microbiome sample analysis result through a preset plan adjustment strategy analysis model to obtain a health management adjustment strategy.

[0075] Furthermore, the method of analyzing the microbiome sample through a preset bioinformatics analysis method to obtain the microbiome sample analysis result is to perform microbiome analysis on the microbiome sample through high-throughput sequencing technology to obtain microbiome sequencing data; analyze the composition information, diversity information, and functional information of the microbial community according to the microbiome sequencing data through a preset bioinformatics analysis method; determine the microbiome sample analysis result according to the composition information, diversity information, and functional information of the microbial community.

[0076] It should be noted that the microbiome sample includes a fecal sample or a vaginal secretion sample.

[0077] Samples for microbiome analysis: For patients with endometriosis, fecal samples and vaginal secretion samples contain rich microbial information, which can reflect the composition, diversity, and functional status of the microbial community in the body.

[0078] Fecal sample collection: The patient is required to collect samples during a specific time period following a standard procedure to ensure sample quality and representativeness. Vaginal secretion sample collection: The vaginal secretion sample is collected by professional medical staff or under their guidance using a sterile vaginal swab.

[0079] Perform microbiome analysis on the sample using high-throughput sequencing technology to obtain microbiome sequencing data.

[0080] High-throughput sequencing technology can simultaneously sequence a large number of DNA fragments, enabling rapid and low-cost acquisition of the sequence information of the microbiome in the sample. It labels nucleotides through special chemical methods and records the base sequence during DNA synthesis, and is suitable for the analysis of complex microbial communities.

[0081] The microbiome sequencing data is the output result after the operation of high-throughput sequencing technology, presented in the form of sequence information.

[0082] The sample operation process is as follows: 1. Extract and purify the microbial genomic DNA from fecal or vaginal secretion samples to ensure purity and integrity. 2. Fragment the DNA, add adapters to construct a library, and after passing the detection, dilute and load it onto the sequencer. 3. Set parameters for sequencing to generate raw data (FASTQ format).

[0083] The data is processed as follows: 1. Use FastQC to evaluate the quality of the raw data and filter out low-quality reads. 2. Use Cutadapt to remove adapters, trim low-quality bases, and convert the format to prepare for subsequent analysis. Finally, high-quality sequencing data is obtained for microbiome analysis.

[0084] Based on the obtained microbiome sequencing data, the composition, diversity, and functional information of the microbial community are analyzed and determined using preset bioinformatics analysis methods as the results of the microbiome analysis.

[0085] Community composition analysis: Use alignment tools (such as BLAST) to align the sequencing reads with the microbial genome database to determine the species or taxonomic units they belong to and count the abundances. At the same time, a classification model is constructed with the help of machine learning algorithms (such as the Naive Bayes classifier) to assist in classification to clarify the composition structure of the microbial community.

[0086] Diversity analysis: Calculate the α-diversity index (such as Shannon and Simpson indices) to evaluate the diversity within a single sample, and its value reflects species richness and evenness. Use methods based on distance matrices (such as Bray-Curtis distance), such as principal coordinate analysis (PCoA), to analyze β-diversity and compare the differences in community composition between samples.

[0087] Functional information analysis: Align and annotate the reads with functional gene databases (such as KEGG and COG) to determine functional genes, count the abundances, and map them to metabolic pathways to analyze the functional potential. It is also possible to perform functional prediction and annotation on the sequences after metagenomic assembly to discover new genes and functions.

[0088] In the embodiment of adjusting the personalized health management plan according to the comprehensive quantitative evaluation score, the trend of the disease condition change, and the results of the microbiome sample analysis to obtain the health management adjustment strategy, the results of the microbiome analysis, the range of the interval where the current comprehensive quantitative evaluation score falls, and the result information of the change trend are input into the pre-trained personalized health management plan adjustment strategy analysis model, and the matching personalized health management plan adjustment strategy is output.

[0089] The personalized health management plan adjustment strategy analysis model is an intelligent model built relying on cutting-edge machine learning or deep learning algorithms. During the training stage of the model, it fully learns different microbial community characteristics, various change patterns of the comprehensive quantitative evaluation score, and the corresponding precise health management plan adjustment strategies, so as to have the ability to quickly give scientific and reasonable adjustment strategies based on the input information.

[0090] Input of microbiome analysis results: that is, inputting detailed information obtained through in-depth bioinformatics analysis of microbiome sequencing data into the model. This information includes extremely detailed species composition of the microbial community, accurate to the relative abundance of each key microorganism in the sample, such as the proportion of a certain butyrate-producing bacterium in fecal samples is only 0.5%, but there are abnormal fluctuations; it also covers the unique functional characteristics of microorganisms, such as the specific pathways and efficiency changes of certain microorganisms involved in estrogen metabolism. These microscopic data provide key basis for the model to gain insight into the causes of diseases from a microbial perspective and formulate targeted strategies.

[0091] Comprehensive quantitative assessment score range and change trend input: The current comprehensive quantitative assessment score range directly reflects the level of the patient's overall health management effectiveness at this stage. For example, 60-80 points means that the patient is in a medium state that needs to be improved; and the change trend information is like a dynamic weather vane, showing whether the condition is moving in an upward direction of accelerated recovery, maintaining a stable state of the status quo, or a downward channel tending to worsen. The combination of the two inputs allows the model to grasp the trend of the patient's condition from a macro level, complementing the micro data of the microbiome, so that the generated adjustment strategy is more in line with the actual needs of the patient.

[0092] Strategy generation and output: After receiving relevant data, the model quickly conducts in-depth analysis of the input information based on the internal complex neural network architecture or decision tree and other algorithmic logic. It quickly searches and matches the most appropriate health management plan adjustment strategy in the massive knowledge reserve. For example, when it is identified that the diversity of the vaginal microbial community is significantly reduced and the comprehensive quantitative evaluation score is in a low segment and shows a downward trend, the model will quickly lock in a set of highly targeted adjustment strategies including the use of vaginal microecological regulators, enhanced nutritional support, increased frequency of psychological counseling, and encrypted follow-up arrangements. Finally, the generated matching strategy is output in the form of structured text, charts, or interactive interfaces to facilitate intuitive understanding and subsequent execution by medical staff and patients.

[0093] An example of adjusting a personalized health management plan based on a comprehensive quantitative assessment score, a trend of disease changes, endometrial imaging, and microbiome sample analysis results to obtain a health management adjustment strategy: Obtain the patient's latest imaging diagnostic data, including accurate presentation of lesion location (such as the specific location of endometriosis lesions in the pelvic cavity), type (cystic, solid or mixed, etc.), and severity (based on lesion size, adhesion to surrounding tissues, and other indicators).

[0094] The patient's latest imaging diagnostic data, microbiome analysis results, comprehensive quantitative assessment scores and their changing trends are input into the microbiome-imaging-health data association model to generate an optimized health management plan.

[0095] This model is built on cutting-edge deep learning algorithms, such as deep convolutional neural networks (DCNN), recurrent neural networks (RNN) and their variants, and its internal architecture has been carefully designed and repeatedly optimized. During the model training phase, a large amount of case data from patients with endometriosis from different regions, different age groups, and different disease severity was invested, covering a rich variety of imaging features, microbiome information, and corresponding clinically proven effective health management strategies.

[0096] Step S50: Optimizing the personalized health management program based on the health management adjustment strategy, so that the optimized personalized health management program can be linked to the smart home to perform rehabilitation intervention on the patient's living environment.

[0097] Furthermore, the processing method for optimizing the personalized health management plan based on the health management adjustment strategy is to determine whether the health management adjustment strategy meets the smart home linkage intervention conditions; if so, target monitoring of key areas in the patient's living space, and determine the dynamics of specific functional microbial communities related to endometriosis and the spatiotemporal characteristic information of the microbial communities based on the key areas in the living space, the microbial community dynamics including species, quantity and activity; analyze the dynamics of specific functional microbial communities and the spatiotemporal characteristic information of the microbial communities to obtain a smart home collaborative processing plan; optimize the personalized health management plan according to the health management adjustment strategy and the smart home collaborative processing plan.

[0098] It should be understood that key areas within the living space generally include places that patients frequently come into contact with on a daily basis and are closely related to their health, such as bedrooms.

[0099] Specific functional microbial communities: refers specifically to those microbial communities that are known or suspected to be functionally associated with the pathogenesis, progression, and recovery of endometriosis. On the one hand, there are microorganisms involved in human immune regulation, such as certain lactic acid bacteria, which can stimulate intestinal-associated lymphoid tissue, regulate systemic immune responses, and affect the local immune microenvironment of the endometrium; on the other hand, there are microorganisms involved in hormone metabolism, such as some intestinal bacteria that have the ability to metabolize estrogen, and their metabolites may interfere with the normal level of estrogen in the body.

[0100] Species: refers to the species diversity of microorganisms in key areas of living space. Different species of microorganisms have different biological characteristics and functions, such as Escherichia coli and Staphylococcus aureus among bacteria, and Candida albicans among fungi.

[0101] Quantity: Accurately reflects the individual abundance of each microbial species. Fluctuations in quantity are often closely related to environmental factors and human activities. For example, increased humidity in the bathroom may lead to a surge in the number of molds, and frequent human contact may introduce new species and change the balance of the original microbial population, which is associated with the increase or decrease in susceptibility to endometriosis.

[0102] Activity: Measures the vigor of microbial metabolism, reproduction and other life activities. Active microbial communities usually mean stronger environmental adaptability and potential greater influence. By testing indicators such as microbial secretion enzyme activity and gas production rate, we can judge their activity and understand their immediate effects on the microecology of living spaces and human health.

[0103] The spatiotemporal analysis features include time features and space features. Time features: track the changing patterns of microbial communities from a temporal dimension. Spatial features: analyze the distribution differences of microbial communities in different key areas, and compare bedrooms, bathrooms, and kitchens horizontally. Due to different functions, temperature, humidity, light and other conditions, the composition of microbial communities in these areas is very different.

[0104] In the specific implementation, the dynamics of specific functional microbial communities related to endometriosis and the spatiotemporal characteristics of microbial communities corresponding to key areas in the living space can be obtained by the following methods: Deployment of detection device: In the bedroom, air microbial sensors are placed above the bed and near the pillow. This type of sensor uses the principle of laser scattering. When microbial particles in the air pass through the detection area, the laser is scattered. The sensor identifies the approximate type of microorganism based on the intensity and angle of the scattered light, and estimates its number through a built-in algorithm. At the same time, fiber optical sensors are embedded inside the bedding. The substances produced by microbial metabolism will change the refractive index of the optical fiber, monitor the activity of the microorganisms, and transmit the data to the data acquisition terminal in real time.

[0105] Data collection and transmission: Each detection device is equipped with a low-power Bluetooth or Wi-Fi module, which packages and sends the collected microbial community data to the home gateway or repeater at set time intervals, such as once an hour.

[0106] After receiving the data, the cloud server first cleans the data to remove outliers caused by sensor failure, electromagnetic interference and other factors. It then uses a data fusion algorithm to integrate the microbial species, quantity and activity data from different sensors and different regions into a unified data format.

[0107] Professional statistical software is used to fit curves to data on changes in the microbial community in the same area over time. For example, the fluctuation trend of the number of a certain beneficial bacteria in the bedroom over a few months is observed, and the periodic change patterns are explored to correlate with the onset period of endometriosis.

[0108] Spatial distribution visualization uses Geographic Information System (GIS) or similar mapping software to present the characteristics of microbial communities in different regions in the form of maps, heat maps, etc., visually showing spatial differences.

[0109] Furthermore, the way to analyze the dynamics of specific functional microbial communities and the spatio-temporal characteristic information of microbial communities to obtain the intelligent home collaborative processing solution is to determine whether there are visitors entering and leaving within a preset time range; if so, obtain the entry and exit details information of the visitors, and the entry and exit details information includes the number of visitors, the staying area, and the staying duration corresponding to the staying area; when the staying area is the key rehabilitation area of the patient, determine whether the staying duration corresponding to the staying area is greater than the preset duration; if so, determine the concentration of harmful microorganisms in the key rehabilitation area according to the dynamics of specific functional microbial communities and the spatio-temporal characteristic information of microbial communities; generate an intelligent home control instruction according to the concentration of harmful microorganisms and the preset harmful microorganism concentration threshold range; determine the intelligent home collaborative processing solution according to the intelligent home control instruction.

[0110] It should be understood that special disease data of endometriosis and body status monitoring data are obtained; the special disease data of endometriosis and body status monitoring data are analyzed by a rehabilitation sensitivity scoring method to determine the comprehensive rehabilitation sensitivity score; according to the comprehensive rehabilitation sensitivity score, the preset harmful microorganism concentration threshold range is determined through the mapping relationship of different categories of harmful microorganism threshold concentration ranges.

[0111] In specific implementation, the special disease data of endometriosis of the patient and the all-round physiological and living status monitoring data of the patient are obtained, standardized, and the comprehensive rehabilitation sensitivity score is analyzed and determined by using a rehabilitation sensitivity scoring method.

[0112] Collect two aspects of key data: one is the special disease data of endometriosis exclusive to the patient, such as information on the severity of the disease, the location and size of the lesion, etc.; the other is the all-round physiological and living status monitoring data covering various physical indicators, daily diet, exercise, sleep, etc. After collection, standardize these complex data, unify the data format and magnitude, and make them comparable. Then, use a professional rehabilitation sensitivity scoring method to deeply analyze and integrate the above data, and accurately determine the comprehensive rehabilitation sensitivity score that can reflect the patient's rehabilitation potential and trend, providing a key reference basis for subsequent diagnosis and treatment.

[0113] Rehabilitation Sensitivity Scoring Method: This is a quantitative evaluation system constructed based on medical big data, clinical experience, and cutting-edge research results. The specific scoring calculation is as follows: Call the rehabilitation sensitivity scoring model and input the preprocessed standardized data. Inside the model, according to the established weight distribution and algorithm logic, first conduct in-depth analysis on the special disease condition data of endometriosis, score different disease condition characteristics, such as giving 0 - 5 points according to the size of the ectopic lesion, 0 - 3 points for the pain degree, etc.; then score the physiological and living status data, such as getting 1 point for meeting the daily vegetable and fruit intake standard, 2 points for exercising 3 times or more per week, etc. Finally, comprehensively combine the scores of each part, and through calculation methods such as weighted summation, obtain the comprehensive rehabilitation sensitivity score of the patient. This score accurately reflects the difficulty of the patient's rehabilitation in the current state, providing a key basis for determining the threshold concentration range of harmful microorganisms in the follow-up.

[0114] For example, for a patient with endometriosis, the ultrasound shows that the lesions involve both ovaries, the CA125 value is 80 U / mL, has dysmenorrhea for 5 days per month, and the pain score is 7 points; in terms of living status, the diet is greasy, exercises once a week, has poor sleep quality, and has mild anxiety in the psychological assessment. After collecting, integrating, and preprocessing the data and inputting it into the rehabilitation sensitivity scoring model, the calculated comprehensive rehabilitation sensitivity score is 45 points (out of 100). Based on this, the corresponding threshold concentration range of harmful microorganisms can be further analyzed, providing precise guidance for the intelligent home environment regulation and assisting the patient's rehabilitation.

[0115] According to the mapping relationship between the interval range where the comprehensive rehabilitation sensitivity score falls and the threshold concentration range of different categories of harmful microorganisms, analyze and determine the threshold concentration range of different categories of harmful microorganisms.

[0116] Threshold Concentration Range of Different Categories of Harmful Microorganisms: This covers the maximum safe content limits per unit volume of air and per unit area of the surface of objects in different areas (key rehabilitation areas such as bedrooms, bathrooms, kitchens, etc.) of the patient's living space for various harmful microorganisms such as bacteria (such as Escherichia coli, Staphylococcus aureus), fungi (such as Candida albicans), viruses (such as common easily infective types like influenza virus, etc.). These threshold concentration ranges are not fixed but are dynamically adjusted according to the patient's comprehensive rehabilitation sensitivity score. For example, for a patient with low rehabilitation sensitivity, the threshold concentration of Staphylococcus aureus in the bedroom air may be set to a relatively loose 150 CFU / m³ (CFU: Colony Forming Unit), but for a highly sensitive patient, to minimize the infection risk, the threshold concentration will be tightened to 80 CFU / m³ to protect the patient from excessive invasion of harmful microorganisms during the corresponding rehabilitation stage.

[0117] Mapping relationship: This is a precise correspondence logic constructed based on medical expertise, microbiological research, and long-term clinical practice. It defines the one-to-one correspondence rules between different comprehensive rehabilitation sensitivity score ranges and the threshold concentration ranges of various harmful microorganisms. For example, when the patient's comprehensive rehabilitation sensitivity score falls within the moderately sensitive range of 31 - 60 points, the threshold concentration range of Escherichia coli on the bathroom floor is determined to be 40 - 60 CFU / 100 cm², and the concentration threshold of Candida albicans in the air is 80 - 120 CFU / m³; when the score is in the highly sensitive range of 61 - 100 points, the corresponding threshold concentrations are correspondingly reduced, with Escherichia coli being 20 - 40 CFU / 100 cm² and Candida albicans being 50 - 80 CFU / m³. Through this precise mapping, it ensures that the harmful microorganism prevention and control measures are closely adapted to the individual rehabilitation needs of the patient.

[0118] In the specific implementation, the smart home collaborative processing solution includes a series of environmental adjustment instructions and life assistance suggestions implemented through smart home devices. For example, the smart air conditioner automatically regulates the bedroom temperature to 22 - 24°C based on factors such as indoor and outdoor temperature and humidity, and the activity level of microorganisms, and turns on the fresh air function regularly; the smart lighting system turns on ultraviolet disinfection during the period when there is no one in the bathroom to purify the air and inhibit the growth of harmful microorganisms; the smart appliances in the kitchen automatically execute cleaning and disinfection procedures based on the usage time period and microbiological monitoring data, aiming to create a healthy living space for the patient and assist in the recovery of endometriosis from the environmental aspect.

[0119] Preset time range: This is a time period preset according to the patient's daily life pattern, rehabilitation needs, and the monitoring and control capabilities of the smart home system. Visitor entry and exit: It refers to the behavior of other people entering the patient's living space except for the patient and long-term resident family members.

[0120] The operation process is as follows: Deployment of the smart access control and sensing system: Install a smart access control system at the entrance door of the house. This system integrates multiple identification technologies such as face recognition, fingerprint recognition, or Bluetooth sensing. It can not only accurately identify authorized personnel (patients and family members) and visitors but also precisely record the entry and exit times of visitors with a time accuracy of up to the second level. When an unauthorized person is identified, it is immediately marked as a visitor in the smart home central control system, and the first identification time is recorded as the entry time. At the same time, deploy passive infrared sensors or micro-motion sensors at key channels and room entrances inside the house. These sensors form a sensing network that can real-time sense the movement trajectories of people. Once it detects that a person enters from the direction of the entrance door and moves into the house, and the smart access control system has marked the person as a visitor, it further confirms the movement route of the visitor inside the house to provide data support for subsequent analysis of the staying area.

[0121] The intelligent access control system and the in-house sensor network are connected to the smart home control system via wired (such as Ethernet cables, ensuring stable and high-speed data transmission, suitable for newly built smart homes) or wireless (such as the ZigBee protocol, with low power consumption and strong self-organizing network capabilities, facilitating deployment in old houses for renovation) means.

[0122] The control system is built-in with intelligent analysis algorithms, which perform real-time screening on visitor entry and exit data based on the preset time range parameters.

[0123] If no visitor entry or exit is detected, the original settings are maintained.

[0124] Original settings: Refers to a set of parameter configurations that are pre-set and run by each smart home device according to the patient's daily habits, rehabilitation needs, and previous environmental monitoring data when the smart home collaborative processing solution does not detect special situations (such as visitor entry and exit in the current step).

[0125] If visitor entry and exit are detected, the detailed information of the visitor entry and exit is obtained. The detailed information includes the number of visitors, the staying areas, and the duration.

[0126] Number of visitors: That is, the total number of external personnel entering the patient's living space. Staying area: Refers to the specific rooms or spatial ranges that the visitor has set foot in the patient's home. Staying duration: Records the time consumed by the visitor from entering to leaving each staying area.

[0127] The operation process is as follows: When the smart home control system determines that there are visitors entering and exiting within the preset time range, it immediately activates the in-house all-round intelligent sensor network. In addition to the intelligent access control system at the entrance door continuously recording the information of the entering and exiting personnel, the passive infrared sensors and micro-vibration sensors arranged in each room and key passages work together to accurately track the movement path of the visitors. These sensors sense based on the infrared rays emitted by the human body or the vibration signals generated by subtle movements. Once a visitor enters a certain area, they immediately feedback to the control system in real time. Through multi-sensor data fusion, a complete activity trajectory of the visitor in the house is constructed.

[0128] Based on the sensor feedback data, the control system uses the built-in data analysis algorithm to accurately count the number of visitors. Each time a new visitor face information is recognized or a new human body signal is sensed, the visitor number counter increases accordingly to ensure accuracy. At the same time, according to the movement trajectory of the visitors and combined with the room layout information (pre-entered into the control system), the staying areas of each visitor are accurately located and marked by room, such as the living room, master bedroom, secondary bedroom, bathroom, etc. For the staying duration, the control system calculates the time difference by recording the timestamps of the visitor entering and leaving each area, accurate to the second, so as to comprehensively master the detailed activity time of the visitors in different areas.

[0129] The detailed information such as the number of visitors, the area they stayed in, and the length of time they stayed is integrated to form a structured data record. Arranged in chronological order, each record contains fields such as the visitor's entry time, the number of visitors, the area they stayed in, and the length of time they stayed. These fields are stored in the local database of the smart home control system, making it easy to retrieve and query in subsequent steps, providing a detailed data basis for further analyzing the impact of visitors on the patient's rehabilitation environment and formulating response strategies.

[0130] Based on the visitor entry and exit details, analyze whether there are visitors who have stayed in key rehabilitation areas for longer than the preset time.

[0131] Key recovery area: This is a specially designated living space that has a key impact on the recovery of endometriosis patients based on their recovery needs. Generally speaking, the bedroom is the core place for patients to rest and recuperate.

[0132] Preset duration: This is a time threshold that is pre-set based on multiple factors such as the growth and reproduction characteristics of microorganisms in different environments, the rate of transmission and diffusion, and the potential degree of harm to patient health, after scientific research and verification with a large amount of practical data.

[0133] The operation process is as follows: Data retrieval and screening: After obtaining the visitor entry and exit details, the smart home central control system immediately retrieves relevant data from the local database. Focusing on the number of visitors, the corresponding stay area and stay duration of each visitor, the data screening tool is used to quickly identify the visitor records entering the key recovery areas (bedroom, bathroom, kitchen), and extract these key data separately to prepare for the next step of risk analysis.

[0134] Time comparison analysis: For the selected visitor records in key rehabilitation areas, the central control system compares the length of stay of each visitor in the corresponding area with the preset length of stay. Through the built-in time comparison algorithm, the difference in the length of stay is accurately calculated to determine whether it has exceeded the time limit.

[0135] Risk determination output: After analyzing the length of stay of visitors in all key rehabilitation areas one by one, the central control system generates the final risk determination result based on whether there is any overtime situation.

[0136] If yes, the target rehabilitation area is determined, and it is analyzed whether the concentration of harmful microorganisms in the target rehabilitation area exceeds the preset harmful microorganism threshold concentration range.

[0137] Target recovery area: refers to the key recovery area where visitors are judged to have stayed for longer than the preset time. Since visitors stay for a long time, there is an increased risk of introducing external microorganisms and causing them to grow and multiply in the area. Therefore, these areas become the focus of subsequent attention and prevention and control.

[0138] Concentration of harmful microorganisms: This covers the content indicators of various microbial populations that may have a negative impact on the health of patients, especially those that are not conducive to the recovery of endometriosis. It includes bacteria (such as Staphylococcus aureus, which is likely to cause infections and interfere with the body's recovery; Escherichia coli, which may contaminate the living environment and affect the patient's quality of life), fungi (such as Candida albicans, which thrives in humid environments and is likely to cause local infections, exacerbating the patient's physical discomfort), viruses (such as influenza virus, etc., which can weaken the patient's immunity and hinder the recovery process), etc. The quantity per unit volume of air and per unit area of the surface of objects in the target recovery area. Precise monitoring of its concentration changes can promptly identify the degree of microbial contamination and provide a key basis for taking targeted measures.

[0139] Preset threshold concentration range of harmful microorganisms: This is a concentration limit scientifically set based on medical research, microbiological experiments, and the accumulation of a large number of clinical cases, taking into account the vulnerable immune status and recovery needs of endometriosis patients, and comprehensively considering factors such as the pathogenic risks and growth characteristics of different harmful microorganism species. For example, for the bedroom environment, through research, it has been found that when the concentration of Staphylococcus aureus in the air exceeds 100 CFU / m³ (CFU: colony-forming unit) and the concentration of Escherichia coli on the surface of objects is higher than 50 CFU / 100 cm², it will significantly increase the patient's infection probability and hinder recovery. These numerical ranges constitute the preset threshold concentration range of the corresponding microorganisms in the bedroom area. Once the detected value exceeds this range, it means that the environmental risk has increased and immediate intervention is required.

[0140] The operation process is as follows: Area locking and marking: When the smart home central control system detects that a visitor has been in the critical recovery area for more than the preset duration, based on the previously recorded information on the areas where the visitor stayed, it quickly locks the corresponding target recovery area and makes a clear mark in the system. For example, mark the bedroom as the "high-risk bedroom area" and the bathroom as the "potentially contaminated bathroom area", etc. At the same time, push the relevant information (location, area, function, etc.) of this area to the subsequent monitoring and prevention and control modules to ensure precise focus on processing.

[0141] Microbial sampling and detection startup: The central control system immediately instructs the microbial monitoring equipment deployed in the target recovery area to start working. These devices include air microbial samplers, which draw a certain volume of air through a suction pump so that the microorganisms in the air adhere to a special culture medium or sampling filter membrane; and surface sampling swabs for objects, which are used to wipe and collect microbial samples from the surfaces of bedding, sanitary ware, countertops, etc. The sampling frequency is dynamically adjusted according to the risk level. It can be increased to once every 15 minutes during high-risk periods (such as in the short term after a visitor leaves) and once every hour during normal periods to ensure timely capture of changes in the concentration of harmful microorganisms.

[0142] Concentration Analysis and Judgment: The collected microbial samples are quickly sent to a small laboratory equipped with the smart home or an analysis module with rapid detection capabilities (such as a portable detector based on microfluidic chips and immunofluorescence technology). Using advanced detection techniques, harmful microorganisms in the samples are accurately identified and quantitatively analyzed, and the obtained concentration data is transmitted back to the central control system in real time. The central control system is built-in with a comparison algorithm, which compares the concentrations of various harmful microorganisms detected with the preset threshold concentration ranges of harmful microorganisms one by one to determine whether it exceeds the standard. For example, if the concentration of Staphylococcus aureus detected in the bedroom air reaches 120 CFU / m³, exceeding the set threshold, the central control system determines that the concentration of harmful microorganisms in the target rehabilitation area exceeds the standard, triggering subsequent enhanced prevention and control steps.

[0143] If all test results are within the threshold range, it is determined that it does not exceed the standard, and it enters a relatively mild routine maintenance process to avoid overreaction affecting the normal life of the patient. If so, it is marked that there is a risk in the target rehabilitation area, and the local air purifier located in the target rehabilitation area is preferentially instructed to enter the enhanced mode, and the fan speed is increased to the maximum speed to maximize the air circulation volume.

[0144] Marking that there is a risk in the target rehabilitation area: This is a prominent warning sign for a specific area by the smart home central control system based on the judgment of factors such as the concentration of harmful microorganisms.

[0145] The enhanced mode of the air purifier is a preset high-performance operating state designed to cope with high-pollution risk scenarios. When this mode is activated, the device will increase the fan speed to the maximum designed speed to achieve the maximum air circulation volume. On the one hand, the high-speed rotating fan can strongly suck the surrounding air, so that more air containing harmful microorganisms quickly passes through the filter inside the purifier. The high-efficiency filter material in the filter (such as a HEPA filter, with a retention efficiency of more than 99.97% for fine particles and microorganisms) can effectively capture harmful substances such as bacteria, fungal spores, and viruses; on the other hand, the powerful air circulation can quickly dilute the concentration of harmful microorganisms in the target rehabilitation area and reduce the degree of microbial contamination in a short time, providing a relatively safe local environment for the patient.

[0146] In addition, before preferentially instructing the local air purifier located in the target rehabilitation area to enter the enhanced mode, there are the following steps: Analyze whether to retrieve the latest imaging diagnosis data of the patient.

[0147] If the answer is no, then continue with the subsequent steps.

[0148] If the answer is yes, then conduct in-depth analysis to obtain the key features of the lesion, including not only basic information such as the location, size, and type of the lesion, but also refining to the organizational structure changes inside the lesion.

[0149] The system quickly and accurately retrieves the latest imaging diagnosis data of the patient and conducts a comprehensive in-depth analysis using professional imaging analysis software with high-resolution analysis capabilities. During this process, it fully explores the key features of the lesion. In addition to accurately locking the position coordinates of the lesion in the body, precisely measuring its three-dimensional size, and clarifying its basic types such as cystic, solid, or mixed, it further focuses on the fine changes in the internal tissue structure of the lesion. Through image texture analysis technology, the degree of uniformity of the lesion tissue is interpreted. Once it is found that the texture is rough and the gray value fluctuates frequently and significantly in local areas, it is very likely to indicate that the inflammation inside the lesion is in a highly active stage, cell proliferation is disordered, and tissue stability is damaged; closely observe the sharpness of the lesion edge. An unclear edge is often a visual signal of the invasion and spread of diseased cells outward, indicating an increased risk of disease progression; use professional measurement tools to accurately quantify the edema range and severity around the lesion. The continuous expansion of the edema area is like a red light warning, indicating that the local microenvironment deterioration is intensifying. These key features obtained from the in-depth analysis will serve as the core basis for subsequent precise regulation of the air purifier.

[0150] The key features of the lesion are input into a pre-constructed correlation model for adjusting the function parameters of the air purifier, and the adjusted function parameters of the air purifier are output, and the parameter adjustment is made according to the adjusted function parameters of the air purifier.

[0151] This model is based on the accumulation of a large amount of historical case data, covering the purification requirements of the rehabilitation environment of patients in different disease stages and with different imaging features; at the same time, it integrates the principles of cutting-edge air purification technologies, including the filtration efficiency of various filter materials for pollutants of different particle sizes, the inactivation effect of different sterilization technologies on microorganisms, and the ventilation performance of the fan under different working conditions and other professional knowledge. After receiving the lesion feature data, the model quickly outputs the optimized and adjusted function parameters of the air purifier for the current situation through complex internal algorithm operations.

[0152] According to the adjusted function parameters of the air purifier output by the correlation model, the on-site intelligent control system quickly performs precise parameter adjustment operations on the local air purifier. Whether it is the change of the fan speed, the switching of the filter working mode, or the activation and strengthening of the sterilization function, it is strictly executed according to the instructions to ensure that the purifier can provide the most suitable air purification service based on the current disease imaging features of the patient.

[0153] Furthermore, it is also possible to obtain endometriosis images of the patient; analyze the endometriosis images through image texture analysis technology to determine the lesion characteristic information; adjust the function parameters of the air purifier according to the lesion characteristic information so that the air purifier with adjusted function parameters generates an air purification instruction; adjust the smart home control instruction according to the air purification instruction so that the adjusted smart home control instruction determines the smart home collaborative processing plan.

[0154] The smart home control instruction is a direct operation command sent by the smart home central control system to specific devices, presented in a specific code format or signal form, enabling the devices to accurately execute corresponding actions.

[0155] Deeply analyze the key characteristics of the lesion, including not only basic information such as the location, size, and type of the lesion, but also refine to the organizational structure changes inside the lesion.

[0156] The system quickly and accurately retrieves the latest image diagnosis data of the patient and conducts a full - range in - depth analysis with professional image analysis software with high - resolution analysis capabilities. During this process, fully explore the key characteristics of the lesion. In addition to accurately locking the position coordinates of the lesion in the body, precisely measuring its three - dimensional size, and clarifying its basic types such as cystic, solid, or mixed, further focus on the fine changes in the organizational structure inside the lesion. Through image texture analysis technology, interpret the degree of uniformity of the lesion tissue. Once it is found that the texture is rough and the gray - scale value fluctuates frequently and significantly in local areas, it is very likely to indicate that the inflammation inside the lesion is in a highly active stage, cell proliferation is disordered, and tissue stability is damaged; closely observe the sharpness of the lesion edge. An unclear edge is often a visual signal of the invasion and spread of diseased cells, indicating an increased risk of disease progression; use professional measurement tools to accurately quantify the edema range and severity around the lesion. The continuous expansion of the edema area is like a red light, warning of the exacerbation of the local micro - environment deterioration. These key characteristics obtained through in - depth analysis will serve as the core basis for subsequent precise regulation of the air purifier.

[0157] Input the key characteristics of the lesion into the pre - constructed correlation model for adjusting the function parameters of the air purifier, output the adjusted function parameters of the air purifier, and make parameter adjustments according to the adjusted function parameters of the air purifier.

[0158] This model is based on the accumulation of a vast amount of historical case data, covering the purification requirements of the rehabilitation environment of patients in different disease stages and with different imaging characteristics. At the same time, it integrates the principles of cutting-edge air purification technologies, including the filtration efficiency of various filter materials for pollutants of different particle sizes, the inactivation effect of different sterilization technologies on microorganisms, and the ventilation performance of fans under different operating conditions, etc. After receiving the lesion feature data, the model quickly outputs the optimized and adjusted function parameters of the air purifier for the current situation through complex internal algorithm operations.

[0159] According to the adjusted function parameters of the air purifier output by the associated model, the on-site intelligent control system quickly performs precise parameter adjustment operations on the local air purifier. Whether it is the change of the fan speed, the switching of the filter working mode, or the activation and strengthening of the sterilization function, they are all strictly executed according to the instructions to ensure that the purifier can provide the most suitable air purification service based on the current disease imaging characteristics of the patient.

[0160] It should also be noted that frequently used electrical appliances mean more energy consumption, potential heat radiation release, and frequent disturbances to the indoor temperature and humidity. These factors may indirectly affect the health status of patients. Especially for patients with endometriosis, a stable and suitable living environment is crucial.

[0161] Deploy intelligent sensors at key positions in each room of the patient's home. For the characteristics related to the use of electrical appliances, install an electricity consumption monitoring socket at the electrical appliance plug, which can accurately record the power-on start time and power-off shutdown time of each electrical appliance, so as to count the usage frequency and duration. At the same time, using indoor positioning technology (such as positioning based on Bluetooth beacons or Wi-Fi signal strength), combined with the pre-entered room layout and electrical appliance position information, when the electrical appliance is running, the distance between it and the human body can be estimated in real time. For example, set positioning reference points around the sofa in the living room. When the TV is turned on, the system dynamically updates the distance value according to the relative position between the patient's mobile phone signal sitting on the sofa and the TV.

[0162] Regarding the measurement of electromagnetic field intensity, use a portable electromagnetic field intensity tester to conduct regular (such as once an hour) patrol measurements in the areas where the patient often moves (bedroom, living room, kitchen, etc.), and record the electromagnetic field intensity values at different positions and different time periods. The tester has the ability to detect multiple frequency bands and can capture the changes of power frequency and radio frequency electromagnetic fields at the same time to ensure a comprehensive reflection of the electromagnetic environment.

[0163] Data integration: The data from the power consumption monitoring socket, indoor positioning system, and electromagnetic field strength tester are aggregated through the smart home central control system. Indexed by the unique patient identifier, various electrical appliance usage data and electromagnetic field strength data at the same time and in the same room are associated to construct a multi-dimensional original dataset. For example, the data record of the living room at a certain moment includes the usage frequency, duration, distance from the human body of the TV, and the electromagnetic field strength value at this position in the living room, facilitating subsequent unified processing.

[0164] Preprocessing execution: Use data cleaning algorithms to identify and remove error values and missing values in the original dataset. Such as the incorrect power-on records caused by the power outage and restart of the power consumption monitoring socket, and the distance missing values occasionally caused by the signal loss of the indoor positioning system. Ensure data integrity by means of data interpolation or directly deleting abnormal records. Then, according to the pre-set standardization rules, transform the cleaned data. As mentioned before, normalize and calibrate the electrical appliance usage frequency, duration, distance, and electromagnetic field strength respectively, and finally generate a standardized dataset for subsequent analysis, providing data support for accurately understanding the impact of electrical appliances and the electromagnetic environment on the patient's health.

[0165] Combine the usage-related characteristics of various electrical appliances in the patient's home with the electromagnetic field strength to construct interaction terms.

[0166] Interaction terms: In the field of data analysis, interaction terms are used to measure the impact of the synergistic effect between two or more variables on the result variable. In this context, the interaction terms are constructed by combining the usage-related characteristics (usage frequency, usage duration, distance from the human body) of various electrical appliances in the patient's home with the electromagnetic field strength.

[0167] The generation of interaction terms involves the following process: Use data processing algorithms to construct interaction terms according to the pre-set combination rules. Taking the table lamp in a certain room as an example, multiply the usage frequency, usage duration, and distance from the human body of the table lamp by the electromagnetic field strength in the room at that time respectively to generate three interaction terms: table lamp usage frequency × electromagnetic field strength, table lamp usage duration × electromagnetic field strength, table lamp distance from the human body × electromagnetic field strength. The purpose of this is to capture the co-variation relationship between variables in different dimensions, because considering only the electrical appliance usage characteristics or the electromagnetic field strength alone cannot fully reflect the overall picture of their potential impact on the patient's health, and the interaction terms can uncover the deep-seated associations hidden behind the data.

[0168] For multiple rooms and multiple types of electrical appliances in the home, repeat the above operations one by one to construct a series of interaction terms covering various combinations of electrical appliances and electromagnetic field strength.

[0169] Conduct statistical feature analysis of the usage-related characteristics of various electrical appliances in the patient's home by time period to obtain data with refined time series characteristics.

[0170] Statistical feature analysis by time period: This is a data analysis method that divides the continuous time axis into multiple time periods according to specific rules, such as morning, noon, and evening in a day, or weekdays and weekends in a week, and describes the data statistically for each time period.

[0171] Data with refined time series characteristics: The data set obtained after the above statistical characteristics analysis divided by time periods is more detailed and has a sense of time hierarchy than the original appliance usage records. It includes statistics such as the average usage frequency of a certain type of appliance in each time period, the maximum and minimum usage time, and the standard deviation of the average distance between the human body and the appliance in different time periods.

[0172] The operation process is as follows: Time period setting: According to the patient's daily life patterns and research purposes, reasonably set the time period division plan.

[0173] Data grouping and statistics: From the pre-processed standard data set, the usage-related characteristic data of various types of electrical appliances are grouped according to the set time period.

[0174] Data integration and output: Integrate the statistical feature data of each time period and each type of electrical appliance to construct a complete and detailed time series feature data set. To ensure the integrity and consistency of the data, each data record is clearly marked with the time period, electrical appliance category, and the corresponding statistical feature values, so that subsequent steps can be directly called for analysis. For example, a data record is "After get off work on weekdays, the smart rice cooker is used an average of 0.6 times, the maximum usage time is 1.5 hours, the minimum usage time is 0.5 hours, and the standard deviation of the distance between the human body and the rice cooker is 0.2 meters." And so on. After all the data are integrated, they are output as a key part of the subsequent model input, helping to accurately understand the impact of electrical appliance use on the health of endometriosis patients and promote the formulation of personalized health management plan adjustment strategies.

[0175] The data of interaction terms and refined time series features are input into the trained multivariate linear regression model, and the prediction results are output, where the prediction results include the influence coefficient of each factor on the comprehensive quantitative evaluation score.

[0176] Completed Multivariate Linear Regression Model: This is a mathematical model constructed and fully trained and optimized using the multivariate linear regression algorithm based on a large amount of historical data. Multivariate linear regression aims to explore the linear relationship between multiple independent variables (in this scenario, the interaction terms generated, numerous variables involved in refining time series feature data, such as interaction variables of electrical appliance usage frequency, duration, distance, and electromagnetic field intensity, as well as statistical characteristic values of various electrical appliances at different time periods) and a dependent variable (the comprehensive quantitative assessment score of the patient, comprehensively reflecting various aspects of the physical state and rehabilitation progress of endometriosis patients).

[0177] Influence Coefficients in Prediction Results: The influence coefficients output by the model represent the strength and direction of the influence of each independent variable on the dependent variable (comprehensive quantitative assessment score).

[0178] The operation process is as follows: Data Preparation: Extract the interaction term dataset (including the numerical values of numerous interaction variables generated by the combination of various electrical appliances and electromagnetic field intensity) and the refined time series feature dataset (statistics such as the average usage frequency and maximum and minimum usage durations of each electrical appliance at different time periods) from the data results collected and processed in the previous steps, ensuring that the data is complete, accurate, and in a standardized format, matching the input requirements of the multivariate linear regression model.

[0179] Model Loading: Retrieve the completed multivariate linear regression model from the model repository. This model has been optimized for parameters using a large amount of data from a similar patient population during the previous training stage and has good generalization ability.

[0180] Prediction Execution: Use the prepared interaction term data and refined time series feature data as input variables and batch or sample-by-sample input them into the loaded multivariate linear regression model. The model performs complex mathematical operations on the input data according to the internally constructed linear relationship expression. For example, for each piece of input data containing multiple interaction terms and time series feature values, the model calculates a corresponding predicted value according to the established calculation rules of multiplying and summing with the coefficient matrix and variables. This predicted value is the estimated result of the patient's comprehensive quantitative assessment score. During the calculation process, the model also outputs the influence coefficients corresponding to each input variable (i.e., each factor), and these coefficients intuitively reflect the relative importance and direction of the role of each factor in influencing the comprehensive quantitative assessment score.

[0181] Result Arrangement: Collect the prediction results output by the model, including the predicted values of the comprehensive quantitative assessment scores corresponding to each sample and the influence coefficients of each factor.

[0182] Analyze and determine the adjustment strategy for the personalized health management plan based on the mapping relationship between the prediction results, the range of intervals where the current comprehensive quantitative assessment score falls, the result information of the change trend, and the adjustment strategy of the personalized health management plan.

[0183] The operation process is as follows: Data integration: Collect the predicted result data output by step S83D, including the influence coefficients of various factors and the predicted values of the comprehensive quantitative evaluation scores. At the same time, obtain the patient's current comprehensive quantitative evaluation score and the range of the interval it falls into, as well as the information on the recent change trend of this score (calculated by continuously monitoring the historical evaluation scores). Integrate these data into an analysis dataset to ensure the integrity and relevance of the data, providing a basis for subsequent accurate analysis. For example, a record in the integrated dataset contains: the influence coefficient of the interaction term between the usage duration of a certain bedroom electrical appliance and the electromagnetic field intensity is -0.3, the influence coefficient of the average usage frequency of the intelligent rice cooker after work on weekdays is 0.2, the current comprehensive quantitative evaluation score is 45 points (falling into the interval of 31 - 60 points), and the score has shown a stable trend in the past month.

[0184] Strategy matching: Adjust the mapping relationship of the strategy according to the pre-set personalized health management plan, and compare the integrated dataset with the mapping table. Use the comprehensive quantitative evaluation score interval, change trend, and influence coefficients of key factors in the dataset as indexes to search for the corresponding adjustment strategies in the mapping table. For example, for the above example data, if the mapping relationship indicates that when the score is in the interval of 31 - 60 points, the score is stable, and there are negative factors of electrical appliance electromagnetic fields in the bedroom, the strategies to be taken include: suggesting that the patient move the high-electromagnetic-field-intensity electrical appliances in the bedroom (such as the wireless router) more than 2 meters away from the bed, adjust the usage time of bedroom electrical appliances, and avoid using them for a long time at night; encouraging the appropriate increase in the use of the intelligent rice cooker after work on weekdays and following the nutritional recipe to make healthy meals to promote recovery.

[0185] Strategy generation and optimization: Refine and optimize according to the initially matched adjustment strategies in combination with the patient's special circumstances (such as living habits, family environment layout, etc.). If the space in the patient's home is limited and the wireless router cannot be moved 2 meters away, consider using a shielding cover to reduce electromagnetic radiation; if the patient has special dietary preferences, adjust the recipe recommendations of the intelligent rice cooker under the premise of following nutritional principles to make it more in line with the taste and improve the patient's compliance. Finally, generate a complete and practical set of personalized health management plan adjustment strategies, covering multiple aspects such as electrical appliance use, environmental management, and lifestyle, and customize a rehabilitation path for patients with endometriosis to help improve health.

[0186] Furthermore, it is also possible to identify and determine the scenarios where the patient is located, and the scenarios include sleep scenarios, work scenarios, and leisure scenarios.

[0187] The operation process is as follows: Data collection: Use various sensors in the smart home system to collect information. The time sensor provides real-time time data as the basis for initially judging the time of the scene, such as determining whether it is within the probable range of the night sleep period. The human body sensors are distributed at key positions in each room (bed in the bedroom, sofa in the living room, chair in the study, etc.), and continuously feedback the activity status of people by sensing the movement and position changes of the human body. For example, detecting that someone is lying still in bed in the bedroom at night for a long time may imply a sleep scene. The ambient light sensor measures the indoor light intensity. If the light is dim and the bedroom curtains are closed, it conforms to the characteristics of the prelude to sleep; in the living room, moderate light accompanied by the TV being on may indicate a leisure scene. The device status sensor monitors the start and stop of electrical appliances. The working of the computer and printer means a working scene, and the smart speaker playing music helps to judge the leisure scene. The multi-source data lays a solid foundation for accurately identifying the scene.

[0188] Scene inference: The smart home central control system is built-in with intelligent algorithms and conducts comprehensive analysis based on various types of data collected. A judgment matrix is constructed with time information as the horizontal axis and information such as human activities, device status, and ambient light as the vertical axis. For example, at 10:30 pm, the human body sensor in the bedroom shows that someone is lying flat in bed for more than 30 minutes, the lights are off, and the ambient light is extremely dim. The central control system determines it as a sleep scene based on this; if at 3 pm, there are signs of activity in the human body sensor in the study, the computer has been on for more than 1 hour, and the printer is working intermittently, it is determined as a working scene; at 7 pm in the evening, the lights in the living room are softly turned on, the TV is playing a program, and the human body sensor in the sofa area frequently senses human activities, and it is inferred as a leisure scene. The accuracy of scene recognition is ensured through multi-dimensional cross-verification.

[0189] Scene confirmation: After the central control system initially infers the scene, it further verifies and confirms. For example, for a suspected sleep scene, check again whether the air conditioner temperature in the bedroom is adjusted to 24 - 26 °C suitable for sleep, and whether the smart curtains are completely light-blocking. If the conditions match, it is finally confirmed; in the working scene, verify whether the lighting brightness in the office area meets the working standards, and whether the height of the desks and chairs is set according to ergonomics. If it is correct, it is recognized; when confirming the leisure scene, check whether the air purifier in the leisure area is turned on to maintain fresh air, and whether the music of the smart speaker is playing smoothly. The accurate scene recognition after secondary verification provides a reliable premise for the implementation of subsequent smart home linkage intervention strategies.

[0190] An intelligent collaborative action plan generated based on the adjustment strategy of the patient's personalized health management plan, comprehensively considering the patient's condition (such as the rehabilitation needs of endometriosis), the scene where the patient is located (sleep, work, leisure scene, etc.), and the functional characteristics of smart home devices. For example, in the sleep scene, to help the patient fall asleep, the linkage strategy may include a series of coordinated operations such as dimming the lights, reducing the air conditioner wind speed, turning on the air purifier and setting it to silent mode, aiming to create a sleep-friendly environment in all aspects and promote the patient's physical recovery.

[0191] The direct operation commands sent by the smart home central control system to specific devices are presented in a specific code format or signal form, allowing the device to accurately perform the corresponding actions. These instructions vary according to the type and function of different devices. For example, for smart lighting, the control instructions may include brightness adjustment values ​​(expressed as a percentage, such as 30% brightness), color temperature switching instructions (such as switching to 2700K warm yellow light); for smart air conditioners, there are temperature setting values ​​(such as 25℃), wind speed gear instructions (such as low wind speed), etc., to ensure that the equipment operates as required and achieve the goals of the linkage intervention strategy.

[0192] The central control system uses the pre-stored smart home device layout to quickly locate the target device in the patient's scene. In the leisure scene, the smart lighting equipment, air purifier and smart speaker in the living room are found by positioning, and the generated control instructions are sent to the corresponding devices using wireless communication technology (such as Wi-Fi, ZigBee, etc.). After receiving the instructions, the device immediately performs the actions required, such as changing the brightness and color temperature of the light, starting the air purifier, and playing music on the speaker, thereby creating a comfortable leisure environment for the patient, realizing the implementation of the smart home linkage intervention strategy, and helping the patient improve his quality of life during rehabilitation.

[0193] In this embodiment, first, a personalized health management plan corresponding to the endometriosis condition is determined according to the daily data of the patient's physical state, and the patient's physical state intervention data is collected based on the personalized health management plan, and the comprehensive quantitative evaluation score corresponding to the physical state intervention data is determined by a comprehensive quantitative evaluation formula, and the trend of the disease change is predicted by a linear regression algorithm based on the comprehensive quantitative evaluation score, and then the patient's microbiome sample is obtained, and the microbiome sample is analyzed by a preset bioinformatics analysis method to obtain the microbiome sample analysis result, and then the comprehensive quantitative evaluation score, the disease change trend and the microbiome sample analysis result are deeply analyzed by a preset solution adjustment strategy analysis model to obtain a health management adjustment strategy, and finally the personalized health management plan is optimized based on the health management adjustment strategy, so that the optimized personalized health management plan links the smart home to perform rehabilitation intervention on the patient's living environment. This embodiment customizes and optimizes personalized health management plans for endometriosis patients by collecting multi-source data, processing, modeling and analysis.

[0194] Reference Figure 3 , Figure 3 This is a structural block diagram of the first embodiment of the personalized health management system based on endometriosis of the present invention.

[0195] like Figure 3 As shown, the personalized health management system based on endometriosis proposed in the embodiment of the present invention includes: A generating module 3001 is used to determine a personalized health management plan corresponding to the condition of endometriosis according to the daily physical condition data of the patient, and collect the physical condition intervention data of the patient based on the personalized health management plan; Prediction module 3002, used to determine the comprehensive quantitative evaluation score corresponding to the physical condition intervention data through a comprehensive quantitative evaluation formula, and predict the disease condition change trend through a linear regression algorithm according to the comprehensive quantitative evaluation score; An analysis module 3003 is used to obtain a microbiome sample of the patient, and analyze the microbiome sample using a preset bioinformatics analysis method to obtain a microbiome sample analysis result; The analysis module 3003 is further used to perform in-depth analysis on the comprehensive quantitative evaluation score, the disease change trend and the microbiome sample analysis result through a preset scheme adjustment strategy analysis model to obtain a health management adjustment strategy; The optimization module 3004 is used to optimize the personalized health management plan based on the health management adjustment strategy, so that the optimized personalized health management plan can be linked to the smart home to perform rehabilitation intervention on the patient's living environment.

[0196] Other embodiments or specific implementations of the personalized health management system based on endometriosis of the present invention can refer to the above-mentioned method embodiments, which will not be described in detail here.

[0197] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0198] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0199] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware data platform. Of course, they can also be implemented through hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as a read-only memory / random access memory, magnetic disk, optical disk), and includes several instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0200] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A personalized health management method based on endometriosis, characterized in that: The personalized health management method based on endometriosis comprises the following steps: Determine a personalized health management plan corresponding to the condition of endometriosis according to the patient's daily physical condition data, and collect the patient's physical condition intervention data based on the personalized health management plan; Determine the comprehensive quantitative evaluation score corresponding to the physical condition intervention data through a comprehensive quantitative evaluation formula, and predict the trend of disease change through a linear regression algorithm based on the comprehensive quantitative evaluation score; Obtaining a microbiome sample from the patient, and analyzing the microbiome sample using a preset bioinformatics analysis method to obtain a microbiome sample analysis result; Performing an in-depth analysis of the comprehensive quantitative assessment score, the disease change trend, and the microbiome sample analysis results through a preset scheme adjustment strategy analysis model to obtain a health management adjustment strategy; The personalized health management program is optimized based on the health management adjustment strategy, so that the optimized personalized health management program can be linked with the smart home to perform rehabilitation intervention on the patient's living environment.

2. The method according to claim 1, characterized in that The step of determining a personalized health management plan corresponding to the endometriosis condition based on the patient's daily physical condition data includes: Based on the daily data of the patient's physical condition, a feature data set related to endometriosis is constructed through a feature selection algorithm; Inputting the feature data set into a preset endometriosis disease classification model, obtaining disease-related features at different levels through multiple neuron nodes in a hidden layer, and outputting disease classification results through the preset endometriosis disease classification model according to the disease-related features; A personalized health management plan corresponding to the endometriosis condition is determined according to the daily physical condition data and the condition classification result.

3. The method according to claim 2, characterized in that The step of determining a personalized health management plan corresponding to the endometriosis condition according to the daily physical condition data and the condition classification result comprises: Detecting whether the daily physical status data contains heart rate variability data and galvanic skin response data; If so, inputting the heart rate variability data and the skin electrical response data into a preset mental stress quantification model, and outputting the mental stress level of the patient; A personalized health management plan corresponding to the endometriosis condition is determined according to the mental stress level and the condition classification result.

4. The method according to claim 3, characterized in that The step of analyzing the microbiome sample by a preset bioinformatics analysis method to obtain a microbiome sample analysis result comprises: Performing microbiome analysis on the microbiome sample by high-throughput sequencing technology to obtain microbiome sequencing data; Analyzing the composition information, diversity information and functional information of the microbial community according to the microbiome sequencing data by a preset bioinformatics analysis method; The microbiome sample analysis results are determined based on the composition information, diversity information and functional information of the microbial community.

5. The method according to claim 4, characterized in that The step of optimizing the personalized health management program based on the health management adjustment strategy includes: Determining whether the health management adjustment strategy meets the smart home linkage intervention conditions; If so, the key areas in the patient's living space are targeted for monitoring, and the dynamics of the specific functional microbial community related to the endometriosis condition and the spatiotemporal characteristic information of the microbial community are determined according to the key areas in the living space, wherein the dynamics of the microbial community include species, quantity and activity; Analyze the dynamics of the specific functional microbial community and the spatiotemporal characteristic information of the microbial community to obtain a smart home collaborative processing solution; The personalized health management plan is optimized according to the health management adjustment strategy and the smart home collaborative processing plan.

6. The method according to claim 5, characterized in that The step of analyzing the dynamics of the specific functional microbial community and the spatiotemporal characteristic information of the microbial community to obtain a smart home collaborative processing solution includes: Determine whether there are visitors entering or leaving within the preset time range; If yes, obtaining the visitor's entry and exit details information, the entry and exit details information including the number of visitors, the stay area and the stay time corresponding to the stay area; When the stay area is a key rehabilitation area for the patient, determining whether the stay time corresponding to the stay area is greater than a preset time; If so, determining the concentration of harmful microorganisms in the key rehabilitation area according to the dynamics of the specific functional microbial community and the spatiotemporal characteristic information of the microbial community; Generate a smart home control instruction according to the harmful microorganism concentration and a preset harmful microorganism concentration threshold range; A smart home collaborative processing solution is determined according to the smart home control instruction.

7. The method according to claim 6, characterized in that Before the step of generating a smart home control instruction according to the harmful microorganism concentration and a preset harmful microorganism concentration threshold range, the method further includes: Obtain endometriosis-specific disease data and physical condition monitoring data; Analyzing the endometriosis-specific disease data and the physical condition monitoring data by a rehabilitation sensitivity scoring method to determine a comprehensive rehabilitation sensitivity score; According to the comprehensive recovery sensitivity score, the preset threshold range of harmful microorganism concentration is determined by mapping the threshold concentration ranges of different categories of harmful microorganisms.

8. The method according to claim 7, characterized in that The step of determining the smart home collaborative processing solution according to the smart home control instruction includes: obtaining endometriosis imaging of the patient; Analyzing the endometriosis image by image texture analysis technology to determine lesion feature information; adjusting the function parameters of the air purifier according to the lesion characteristic information, so that the air purifier after the function parameters are adjusted generates an air purification instruction; The smart home control instruction is adjusted according to the air purification instruction so that the adjusted smart home control instruction determines a smart home collaborative processing solution.

9. A personalized health management system based on endometriosis, characterized in that: The personalized health management system based on endometriosis includes: A generation module, configured to determine a personalized health management plan corresponding to the condition of endometriosis according to the daily physical condition data of the patient, and to collect the physical condition intervention data of the patient based on the personalized health management plan; A prediction module, used to determine the comprehensive quantitative evaluation score corresponding to the physical condition intervention data through a comprehensive quantitative evaluation formula, and predict the disease condition change trend through a linear regression algorithm based on the comprehensive quantitative evaluation score; An analysis module, used to obtain a microbiome sample of the patient, and analyze the microbiome sample using a preset bioinformatics analysis method to obtain a microbiome sample analysis result; The analysis module is further used to perform in-depth analysis on the comprehensive quantitative evaluation score, the disease change trend and the microbiome sample analysis results through a preset scheme adjustment strategy analysis model to obtain a health management adjustment strategy; The optimization module is used to optimize the personalized health management plan based on the health management adjustment strategy, so that the optimized personalized health management plan can be linked to the smart home to perform rehabilitation intervention on the patient's living environment.

Images