Intelligent search system and method for medical service

Abstract

The invention is suitable for the technical field of medical information, and provides an intelligent search system and method for medical services, and the system comprises a user data collection module, a hidden demand analysis module, a medical service recommendation module, a physiological data real-time monitoring module and a dynamic adjustment module. Wherein the user data acquisition module is used for acquiring health multi-modal data of a user; the recessive demand analysis module is used for mining recessive health demands which are not clearly expressed by the user; the medical service recommendation module is used for generating a medical service recommendation list; the physiological data real-time monitoring module is used for triggering reanalysis of hidden demands; and the dynamic adjustment module is used for dynamically adjusting the content of the medical service recommendation list. According to the method, personalized and dynamic medical service recommendation is realized through multi-modal data acquisition, implicit demand analysis, real-time monitoring and dynamic adjustment, the recommendation accuracy and response speed are remarkably improved, and the medical service experience of the user is improved.

Description

Technical Field

[0001] The present invention belongs to the field of medical information technology, and particularly relates to an intelligent search system and method for medical services. Background Art

[0002] With the rapid development of medical informatization and intelligent technologies, more and more users obtain medical services through the Internet or mobile devices. However, traditional medical service search systems mainly rely on users to input clear health needs or disease information, and the provided services are often based on static keyword matching, lacking in-depth analysis of users' implicit health needs. At the same time, most existing medical service recommendation systems do not take into account the real-time health status of users and are difficult to dynamically adjust the recommended content, thus unable to meet users' personalized and timely medical needs. In addition, various information such as users' behavioral data, health history data, and wearable device data has not been effectively integrated, resulting in poor accuracy and practicality of medical service recommendations.

[0003] The problems existing in the prior art mainly focus on the following aspects: First, existing medical service search systems cannot effectively identify and analyze users' implicit health needs that are not clearly expressed, and it is difficult to provide truly personalized medical service recommendations; Second, the system lacks a real-time monitoring and dynamic adjustment mechanism for users' health data and is difficult to provide corresponding medical service recommendations in a timely manner when users' health conditions change; Finally, in medical service recommendations, how to optimize by combining users' geographical locations, the availability of medical resources, and time arrangements remains an unsolved problem. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent search method for medical services, aiming to solve the technical problems existing in the prior art determined in the background art.

[0005] The present invention is implemented as follows. An intelligent search system for medical services, the method includes: A user data collection module, configured to collect multi-modal data of users' health questionnaires, historical medical records, physical examination reports, physiological data and behavioral data provided by wearable devices, and perform cleaning and formatting processing on the multi-modal data to ensure the integrity and consistency of the data; An implicit demand analysis module, configured to, based on the data provided by the user data collection module, use a behavior pattern analysis algorithm to mine users' implicit health needs that are not clearly expressed, and sort the priorities of the implicit health needs; A medical service recommendation module, which is used to match corresponding medical services, relevant departments, doctors, and examination items according to the latent health needs analyzed by the latent demand analysis module, in combination with a medical knowledge graph and medical big data, and make real-time recommendations based on the user's geographical location and the time of medical service providers, so as to generate a medical service recommendation list; A physiological data real-time monitoring module, which is used to continuously monitor the user's health status. When an abnormality is detected, it will update the multimodal data in the user data acquisition module and trigger a re-analysis of latent demands; A dynamic adjustment module, which is used to dynamically adjust the content of the medical service recommendation list according to the health abnormality information monitored by the physiological data real-time monitoring module and the results of the re-analysis by the latent demand analysis module.

[0006] As a further solution of the present invention, the user data acquisition module includes: A health questionnaire acquisition unit: which is used to receive and store questionnaire data related to the user's health status filled in by the user. The questionnaire content includes health information such as the user's basic information, past medical history, current symptoms, and medication situation; A historical diagnosis and treatment record acquisition unit: which is used to obtain the user's historical diagnosis and treatment records from medical institutions or authorized medical data platforms, including information such as medical records, diagnostic reports, prescriptions, and surgical records; A physical examination report acquisition unit: which is used to collect and store the user's past physical examination reports, including various physical examination results such as blood routine, urine routine, imaging examinations, and electrocardiograms; A wearable device data acquisition unit: which is used to collect real-time physiological data generated by the intelligent health devices worn by the user, including heart rate, blood pressure, sleep status, and exercise data; A behavior data acquisition unit: which is used to monitor and record the user's daily behavior data, including steps, activity level, and daily work and rest habits; A data cleaning unit: which is used to clean the collected multimodal data, including removing redundant data, correcting error data, and eliminating outliers, to ensure the integrity and accuracy of the collected data; A data formatting processing unit: which is used to format the cleaned multimodal data to unify the data storage format.

[0007] As a further solution of the present invention, the latent demand analysis module includes: A behavior pattern analysis unit, which is used to deeply mine the user's health data and behavior data based on the multi-modal data provided by the user data collection module, and by analyzing the user's daily activity patterns, sleep patterns, eating habits, exercise frequency and the correlation with the historical health status, identify the implicit health needs not explicitly expressed by the user, including sub-healthy states, potential chronic disease risks or long-term unhealthy living habits; A health risk assessment unit, which is used to, after the behavior pattern analysis unit mines out the implicit health needs, quantify the risks of the implicit needs according to a predetermined risk assessment model, and calculate the potential threat degree of each implicit health need to the user's health; the risk assessment model will comprehensively evaluate by combining the user's age, gender, past medical history and current health data, generate a risk score, and determine the severity of the implicit health needs according to the size of the score; A priority ranking unit, which is used to rank the implicit health needs based on the risk quantification results.

[0008] As a further solution of the present invention, the health risk assessment unit includes: Based on the identified implicit health needs, combined with the user's multi-modal health data, determine the potential health risks, analyze the user's age, gender, past medical history, current health status and living habits, and identify the risk factors related to the implicit health needs; Quantify each implicit health need according to a predetermined risk assessment model, and the risk level of each health need Is defined as a weighted combination of several risk factors, and the calculation formula is: ; Among them, Represents the overall risk value of any implicit health need, Represents the weight of the th risk factor, Represents the th risk factor's risk value, Is the total number of risk factors related to the implicit health need corresponding to the risk level ; Calculate the priority of each implicit health need according to the predetermined rules in the risk assessment model: ; Among them, Represents the priority of the implicit health need, Represents the overall risk tolerance of the user's health.

[0009] As a further solution of the present invention, the medical service recommendation module includes: A knowledge graph matching unit, which is used to perform matching through a medical knowledge graph based on implicit health needs. Each medical resource in the knowledge graph is represented by a node, and the relationship between nodes is represented by an edge; A big data analysis unit, which is used to utilize medical big data for retrieval and analysis, and recommend medical services and doctors that match the implicit health needs; A geographical location optimization unit, which is used to, based on the user's current geographical location, analyze the physical distances of medical institutions and doctors, and screen medical resources according to the priority recommendation rule of proximity, and generate a medical resource recommendation list; A time dynamic adjustment unit, which is used to, by docking with the internal system of medical institutions, obtain the idle time of doctors and equipment in real time, and dynamically adjust the medical resource recommendation list according to the appointment time window of medical service providers and the available status of equipment.

[0010] Optionally, the geographical location optimization unit specifically includes: Location acquisition: Based on the user's current geographical location, use the global positioning system in the mobile device or the address information manually input by the user to determine the user's current location, providing a basis for subsequent distance calculation; Distance calculation: Calculate the physical distances between the user's current location and the locations of each medical institution and doctor; Priority recommendation: After calculating the physical distances between all medical institutions and the user, sort the medical institutions or doctors according to the distance size, and give priority to recommending medical resources that are closer.

[0011] As a further solution of the present invention, the physiological data real-time monitoring module includes: A data acquisition unit, which is used to continuously acquire the user's physiological data through wearable devices, including heart rate, blood pressure, respiratory rate, and blood oxygen saturation; An anomaly detection unit, which is used to perform real-time analysis on the acquired physiological data according to preset health thresholds to detect whether there are abnormal conditions; when any physiological index exceeds the health threshold, add a potential health risk mark; A data update and analysis trigger unit, which is used to automatically update the multi-modal health data when there is a potential health risk mark.

[0012] As a further solution of the present invention, the dynamic adjustment module includes: An abnormal data input unit, which is used to receive the updated multi-modal health data, extract the user's latest physiological data, and input it into the implicit demand analysis module to re-evaluate the user's health status; A demand update analysis unit, which is used to re-analyze the user's health status and re-evaluate the risk level of the user's implicit health needs. The priority update formula for implicit health needs is: ; wherein, represents the updated priority of latent health needs, is the newly detected abnormal risk value, and are the weight coefficients for priority adjustment; The recommended list dynamic adjustment unit is used to dynamically adjust the medical resource recommended list according to the updated priority of latent health needs provided by the demand update analysis unit.

[0013] Another object of the present invention is to provide an intelligent search method for medical services, and the method includes: S1. Collect and preprocess the user's multimodal health data, including health questionnaires, diagnosis and treatment records, physiological data and behavior data of wearable devices; S2. Through behavior pattern analysis and risk assessment models, mine and quantify the user's latent health needs, and generate a list of latent needs with priority ranking; S3. Combine the medical knowledge graph and big data analysis to match medical services, and dynamically generate a medical resource recommended list according to the geographical location and resource availability; S4. Continuously monitor the user's physiological indicators, trigger data update and re-analysis of latent needs; S5. Based on the user's physiological indicators and the list of latent needs after the trigger data update, adjust the content of the medical resource recommended list in real time.

[0014] The beneficial effects of the present invention are: Through behavior pattern analysis, risk quantification and priority ranking, the system can accurately match medical services, departments, doctors and examination items related to the user's health status, so as to provide personalized medical service recommendations. At the same time, by combining the user's geographical location and the real-time availability of medical service providers, the system ensures that the recommended medical services are both time-sensitive and can effectively meet the user's health needs.

[0015] Through real-time monitoring of physiological data and dynamic adjustment mechanisms, the automatic update of the recommended list according to changes in the user's health status is realized, ensuring that corresponding medical service recommendations can be provided in a timely manner when the user's health status is abnormal. This innovative dynamic adjustment and real-time monitoring function significantly improves the accuracy and response speed of medical service recommendations, enhances the user's medical service experience, and solves the problem that the recommended content cannot be dynamically adjusted in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic diagram of the intelligent search system for medical services according to an embodiment of the present invention; Figure 2Schematic diagram of the intelligent search method for medical services according to an embodiment of the present invention. Detailed implementation manners

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. 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.

[0018] As Figure 1 shown, an intelligent search system for medical services, the system includes: A user data acquisition module, configured to collect and preprocess multi-modal health data of a user, including health questionnaires, medical records, physiological data and behavior data of wearable devices; The user data acquisition module includes: A health questionnaire acquisition unit: configured to receive and store questionnaire data related to the health status filled in by the user, and the questionnaire content includes health information such as the basic information, past medical history, current symptoms, and medication situation of the user; A historical medical record acquisition unit: configured to obtain the historical medical records of the user from a medical institution or an authorized medical data platform, including information such as medical records, diagnosis reports, prescriptions, and surgical records; A physical examination report acquisition unit: configured to collect and store the past physical examination reports of the user, including various physical examination results such as blood routine, urine routine, imaging examinations, and electrocardiograms; A wearable device data acquisition unit: configured to collect real-time physiological data generated by the intelligent health device worn by the user, including heart rate, blood pressure, sleep status, and exercise data; A behavior data acquisition unit: configured to monitor and record the daily behavior data of the user, including the number of steps, activity level, and daily work and rest habits; A data cleaning unit: configured to perform data cleaning on the collected multi-modal data, including removing redundant data, correcting error data, and eliminating outliers, to ensure the integrity and accuracy of the collected data, so as to avoid interference of invalid or incorrect data on subsequent analysis; A data formatting processing unit: configured to perform formatting processing on the cleaned multi-modal data to unify the data storage format; through the collaborative work of the above units, the user data acquisition module can effectively collect and process user data from multiple sources, ensure the integrity, consistency and availability of the data, provide high-quality basic data for implicit demand analysis, and then improve the overall analysis and recommendation accuracy of the system.

[0019] An implicit demand analysis module, configured to mine and quantify the implicit health needs of the user through behavior pattern analysis and a risk assessment model, and generate a list of implicit needs with priority ranking; The implicit demand analysis module includes: A behavior pattern analysis unit for analyzing the user's multimodal health data, deeply mining the user's health data and behavior data using behavior pattern analysis algorithms, and conducting activity pattern analysis, sleep pattern analysis, eating habit analysis, exercise frequency analysis, and health correlation analysis on the user to identify the user's latent health needs, including sub-healthy states, potential chronic disease risks, or long-term unhealthy living habits; A health risk assessment unit for quantitatively assessing the risks of latent health needs through a risk assessment model, calculating the potential threat level of each latent health need to the user's health. The risk assessment model will comprehensively evaluate by combining the user's age, gender, medical history, and current health data, generate a risk score, and determine the severity of the latent health need based on the score size; A priority ranking unit for ranking the latent health needs based on the risk quantification results.

[0020] The priority ranking unit can ensure that latent health needs with higher risks are processed first in the medical service recommendation module to improve the system's response speed to the user's health risks; through the collaborative work of the above units, the latent need analysis module can deeply mine the user's unexpressed latent health needs, accurately identify them by combining the user's behavior patterns and health data, and scientifically rank the latent needs through a quantified health risk assessment model. This innovative latent health need analysis mechanism not only improves the system's sensitivity to the user's health status but also ensures that high-risk latent needs can be promptly responded to through priority ranking.

[0021] The behavior pattern analysis unit specifically includes: Activity pattern analysis: Based on the user's behavior data, analyze the user's daily activity pattern, calculate the user's total daily activity volume and activity intensity, identify whether the user has a long-term behavior pattern of insufficient or excessive activity by cumulatively calculating the daily activity situation and comparing it with historical activity data, so as to prompt possible sub-healthy states or exercise-related health risks; Sleep pattern analysis: By analyzing the proportion of the user's sleep duration and deep sleep duration, evaluate the user's overall sleep quality, compare the user's actual sleep quality with the standard sleep reference value, and determine whether there is a long-term lack of sleep or low sleep quality, thus prompting possible sleep disorders or related health problems; Eating habit analysis: By analyzing the proportion of nutritional components (such as protein, fat, and carbohydrates) ingested by the user daily, evaluate whether the diet structure is balanced, calculate the deviation between the nutritional components ingested by the user and the recommended standard intake. If the diet structure is unbalanced for a long time, it will prompt the risk of metabolic diseases; Exercise frequency analysis: By analyzing the number of times a user exercises per week and the duration of each exercise, it evaluates whether the user's exercise frequency meets health recommendations. If the user's exercise frequency is too low or too high, the system will prompt potential risks of cardiovascular diseases or sports injuries; Health relevance analysis: Based on the user's historical medical records and the results of behavior pattern analysis, it evaluates the degree of association between the user's behavior and past health problems. By counting the frequency of behavior occurrence and combining historical health data, it determines the correlation between the predefined behavior and health problems. If the degree of association is high, it will prompt the user of the risk of recurrence of health problems.

[0022] The functions and algorithm processes of each step in the pattern analysis unit are as follows: Activity pattern analysis step: Based on the user's behavior data, it analyzes the user's daily activity pattern, uses time series analysis methods to identify the time distribution and intensity changes of the user's daily activities, and calculates the user's total daily activity volume and activity intensity , the formula is: ; Among them, represents the total daily activity volume, represents the intensity of the th activity, represents the duration of this activity, is the number of activity items; By comparing with historical activity data, it identifies whether the user has long-term insufficient or excessive activity behavior patterns, and prompts possible sub-healthy states or sports-related risks; Sleep pattern analysis step: By collecting the user's sleep duration and the proportion of deep sleep time , it uses the sleep pattern analysis algorithm to evaluate the user's sleep quality. The user's sleep quality is calculated by the formula: ; Among them, represents the user's total daily sleep duration, represents the proportion of deep sleep time in the total sleep duration. By comparing this value with the standard reference sleep quality, it determines whether the user has a long-term low sleep quality situation, thus prompting potential sleep disorders or health problems; Diet habit analysis step: By analyzing the user's daily diet data, it calculates the balance of the diet structure. The diet balance index is defined as the proportion deviation of the key nutrients (protein, fat, carbohydrates) ingested by the user, and is calculated by the following formula: ; wherein, is the amount of protein ingested by the user per day, is the amount of fat ingested per day, is the amount of carbohydrates ingested per day; , , are the recommended intakes of protein, fat and carbohydrates respectively; a larger value indicates an unbalanced diet structure of the user, which may suggest the risk of metabolic diseases; Exercise frequency analysis step: Based on the user's exercise data, analyze whether the exercise frequency and intensity meet the health standards. The exercise frequency index is calculated by the following formula: ; wherein, is the number of times the user exercises per week, is the average duration of each exercise. This value is compared with the standard recommended in the health guidelines. If it deviates from the standard for a long time, it indicates that the user may have insufficient exercise or excessive exercise, which may lead to potential cardiovascular diseases or exercise injury risks; Health relevance analysis step: Based on the user's historical diagnosis and treatment records and the analysis results of behavior patterns, use the association rule mining algorithm to calculate the correlation degree between the user's behavior pattern and health problems. This correlation degree is calculated by the following formula: ; wherein, represents the correlation degree between the user's behavior pattern and health problems, represents the number of times the behavior related to health problems occurs, represents the total number of behaviors. A higher correlation degree indicates a strong correlation between this behavior pattern and the user's health status, suggesting that the user may face related health risks; Through the above steps, the behavior pattern analysis unit can comprehensively analyze the user's daily activities, sleep, diet, exercise frequency and their correlation with the historical health status, systematically mine the implicit health needs that the user has not clearly expressed, combine multi-dimensional behavior analysis and health correlation assessment, not only can identify the sub-healthy state, but also can effectively prompt the risk of chronic diseases or bad living habits, enhancing the personalization and accuracy of medical service recommendations.

[0023] The health risk assessment unit includes: Based on the identified latent health needs, combined with the user's multimodal health data, potential health risks are determined. Analyze the user's age, gender, past medical history, current health status, and lifestyle habits to identify risk factors related to the latent health needs. Each latent health need will be associated with corresponding health risk factors, providing a basis for subsequent risk assessment; After determining the risk factors related to the latent needs, each latent health need is quantified according to a predetermined risk assessment model, and the risk level of each health need is defined as a weighted combination of several risk factors, and the calculation formula is: ; where, represents the overall risk value of any latent health need, represents the weight of the th risk factor, the risk value of the th risk factor, is the total number of risk factors related to the latent health need corresponding to the risk level By calculating the weighted values of each risk factor, the threat level of each latent health need to the user's health is quantified; ; where, represents the priority of the latent health need, represents the overall risk tolerance of the user's health.

[0024] If has a high value, it indicates that this latent need should be processed first. The system sorts each latent need according to the calculated priority to ensure that high-risk needs can receive medical service recommendations in a timely manner; Through the above steps, the health risk assessment unit can scientifically quantify each latent health need and sort its priority according to the potential threat level to the user's health. This assessment mechanism not only combines the comprehensive assessment of multiple health risk factors but also reasonably allocates the priority of needs through a predetermined risk assessment model to ensure that the system can promptly identify and handle high-risk latent needs, thereby improving the pertinence and effectiveness of medical service recommendations.

[0025] The medical service recommendation module is used to match medical services by combining a medical knowledge graph and big data analysis, and dynamically generate a list of medical resource recommendations based on geographical location and resource availability; The medical service recommendation module includes: A knowledge graph matching unit, which is used to perform matching through a medical knowledge graph based on implicit health needs. Each medical resource in the knowledge graph is represented by a node, and the relationship between nodes is represented by an edge; Assume that the medical concepts of the user's health needs are represented as nodes , and the nodes of medical services, departments, and doctors are respectively , , . By calculating the association degree between the user demand node and these medical resource nodes , the formula is as follows: ; Among them, is the weight between node and node in the knowledge graph, representing the association strength between medical concepts and medical resources, represents the direct association relationship between nodes, is the distance between nodes, is the number of optional medical resource nodes. By calculating the value, the system can determine the matching degree between medical resources and user needs, so as to recommend the most relevant medical services; A data analysis unit, which is used to perform retrieval and analysis using medical big data and recommend medical services and doctors that match implicit health needs; Assume that the number of historical cases included in the medical big data is , and the similarity between each case and the user's implicit needs is , and the influence weight of the user's historical data on the recommendation result is . The big data analysis unit calculates the weighted average similarity of all relevant cases , and its formula is as follows: ; Among them, represents the similarity between the th historical case and the user's implicit needs, represents the influence weight of the user's health history on the recommendation. Based on the calculated average similarity , the system selects the medical services and doctors that best match the user's health needs for recommendation to ensure that the recommended services have high effectiveness and adaptability; A geographical location optimization unit, which is used to, based on the user's current geographical location, screen medical resources by analyzing the physical distances between medical institutions and doctors, and generate a medical resource recommendation list according to the rule of giving priority to the closer distance; a time dynamic adjustment unit, which is used to, by docking with the internal system of medical institutions, obtain the idle time of doctors and equipment in real time, and dynamically adjust the medical resource recommendation list according to the appointment time window of medical service providers and the available status of equipment, so as to ensure that users can obtain timely medical services; through the collaborative effect of each unit, the medical service recommendation module can match relevant services in the medical knowledge graph and big data according to the latent health needs, and perform real-time optimization and dynamic adjustment in combination with the user's geographical location and the time arrangement of medical service providers, so as to ensure that the recommended medical services are accurate, efficient and meet the actual needs of users.

[0026] The geographical location optimization unit specifically includes: Location acquisition: Based on the user's current geographical location, use the Global Positioning System (GPS) in the mobile device or the address information manually input by the user to determine the user's current location, providing a basis for subsequent distance calculation; Distance calculation: Calculate the physical distance between the user's current location and the locations of each medical institution and doctor. This physical distance is calculated by the straight-line distance between the location information of the medical institution or doctor and the user's location, and specifically uses the spherical cosine theorem for distance estimation. The formula is: ; Among them, is the distance between the user and the medical institution, is the radius of the earth, , is the longitude and latitude coordinates of the user, , is the longitude and latitude coordinates of the medical institution or doctor. Through this formula, the system can accurately calculate the distance between the user and the medical resources; Priority recommendation: After calculating the physical distances between all medical institutions and the user, rank the medical institutions or doctors according to the distance. The closer the medical resources are, the more preferentially they are recommended; specifically, define the recommendation priority of each medical resource as the reciprocal of the distance. The formula is as follows: , By sorting the priorities

[0027] of all optional medical resources, the system preferentially recommends medical services with a relatively short physical distance to users, thereby reducing the user's travel time and improving the convenience of the user's medical treatment. The real-time physiological data monitoring module includes: A data acquisition unit, which is used to continuously collect the user's physiological data through a wearable device, including heart rate, blood pressure, respiratory rate, and blood oxygen saturation. The system synchronizes these physiological data in real time to ensure that the user's health status can be continuously monitored, and stores the collected data in the user data acquisition module for management and update as part of the multimodal data; An anomaly detection unit, which is used to perform real-time analysis on the collected physiological data according to preset health thresholds to detect whether there are any abnormal situations; when any physiological index exceeds the health threshold (such as an abnormal increase in heart rate or too low blood pressure), a potential health risk mark is added; A data update and analysis trigger unit. Once an anomaly is detected, the data update and analysis trigger unit will automatically update the multimodal data in the user data acquisition module to ensure that the latest status of the physiological data is recorded. The data update and analysis trigger unit also includes updating the original physiological data and triggering the re-analysis process of the latent demand analysis module according to the latest abnormal data; by inputting the new health data into the latent demand analysis process, the system can identify new latent health needs or re-evaluate existing health risks, and adjust the medical service recommendations according to the new analysis results.

[0028] A dynamic adjustment module, which is used to real-time adjust the content of the medical resource recommendation list based on the user's physiological indicators and the list of latent demands after triggering data update.

[0029] The dynamic adjustment module includes: An abnormal data input unit, which is used to receive the updated multimodal health data, extract the user's latest physiological data, and input it into the latent demand analysis module to re-evaluate the user's health status; A demand update analysis unit, which is used to re-analyze the user's health status and re-evaluate the risk level of the user's latent health needs. The priority update formula for the latent health needs is: ; where represents the updated priority of the latent health needs, is the newly detected abnormal risk value, and are the weight coefficients for priority adjustment; A recommended list dynamic adjustment unit, which is used to dynamically adjust the medical resource recommendation list according to the updated priority of the latent health needs provided by the demand update analysis unit.

[0030] Priority is given to recommending medical services related to high-priority needs. The recommendation priority calculation formula is: ; Among them, is the recommended priority of medical resources, is the geographical distance between medical resources and users, is the real-time availability time window of medical resources. This formula comprehensively considers the priority of latent health needs, geographical location, and availability, ensuring that the system can provide an optimal list of medical service recommendations.

[0031] For example, Figure 2 as shown, an intelligent search method for medical services, the method includes: S1. Collect and preprocess the user's multimodal health data, including health questionnaires, medical records, physiological data and behavioral data of wearable devices; S2. Through behavior pattern analysis and risk assessment models, mine and quantify the user's latent health needs, and generate a list of latent needs with priority ranking; S3. Combine the medical knowledge graph and big data analysis, match medical services, and dynamically generate a list of medical resource recommendations according to geographical location and resource availability; S4. Continuously monitor the user's physiological indicators, trigger data updates and reanalysis of latent needs; S5. Based on the user's physiological indicators and the list of latent needs after triggering data updates, real-time adjust the content of the medical resource recommendation list.

[0032] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0033] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation to the scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the appended claims.

[0034] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent search system for medical services, characterized in that, The system includes: A user data collection module, which is used to collect and preprocess the user's multimodal health data, including health questionnaires, medical records, physiological data and behavioral data from wearable devices; A latent demand analysis module, which is used to mine and quantify the user's latent health needs through behavior pattern analysis and risk assessment models, and generate a prioritized list of latent needs; A medical service recommendation module, which is used to match medical services by combining a medical knowledge graph and big data analysis, and dynamically generate a medical resource recommendation list according to geographical location and resource availability; A physiological data real-time monitoring module, which is used to continuously monitor the user's physiological indicators, trigger data updates and re-analysis of latent needs; A dynamic adjustment module, which is used to real-time adjust the content of the medical resource recommendation list based on the user's physiological indicators and the list of latent needs after triggering data updates.

2. The system according to claim 1, wherein The latent demand analysis module includes: A behavior pattern analysis unit, which is used to mine the user's multimodal health data, analyze the user's activity patterns, sleep patterns, eating habits, exercise frequencies and health correlations, and identify the user's latent health needs; A health risk assessment unit, which is used to quantify the risks of latent health needs through a risk assessment model, and calculate the potential threat degree of each latent health need to the user's health; A prioritization unit, which is used to prioritize the latent health needs based on the risk quantification results.

3. The system according to claim 2, wherein The health risk assessment unit includes: Based on the identified latent health needs, combined with the user's multimodal health data, determine potential health risks, analyze the user's age, gender, past medical history, current health status and living habits, and identify risk factors related to the latent health needs; Quantify each latent health need according to a predefined risk assessment model, and define the risk level of each health need as a weighted combination of multiple risk factors, and its calculation formula is: ; Among them, represents the overall risk value of a certain latent health need, represents the weight of the th risk factor, represents the risk value of this risk factor, is the total number of risk factors related to the latent health need corresponding to the risk level ; Calculate the priority of each latent health need according to the predetermined rules in the risk assessment model: ; Among them, represents the priority of latent health needs, represents the overall risk tolerance of the user's health.

4. The system according to claim 1, characterized in that, The medical service recommendation module includes: A knowledge graph matching unit, which is used to match based on the latent health needs through a medical knowledge graph. Each medical resource in the knowledge graph is represented by a node, and the relationship between nodes is represented by an edge; A big data analysis unit, which is used to perform retrieval analysis using medical big data and recommend medical services and doctors that match the latent health needs; A geographical location optimization unit, which is used to screen medical resources based on the user's current geographical location by analyzing the physical distances of medical institutions and doctors, and generate a medical resource recommendation list with proximity as the priority recommendation rule; A time dynamic adjustment unit, which is used to real-time obtain the idle time of doctors and equipment by docking with the internal system of medical institutions, and dynamically adjust the medical resource recommendation list according to the appointment time window of medical service providers and the available status of equipment.

5. The system according to claim 1, wherein The physiological data real-time monitoring module includes: A data collection unit, which is used to continuously collect the user's physiological data through wearable devices, including heart rate, blood pressure, respiratory rate and blood oxygen saturation; An anomaly detection unit, which is used to perform real-time analysis on the collected physiological data according to preset health thresholds to detect whether there are abnormal conditions; when any physiological indicator exceeds the health threshold, a potential health risk mark is added; A data update and analysis trigger unit, which is used to automatically update the multimodal health data when there is a potential health risk mark.

6. The system according to claim 1, wherein The dynamic adjustment module includes: An abnormal data input unit, which is used to receive the updated multimodal health data, extract the user's latest physiological data, and input it into the implicit demand analysis module to re-evaluate the user's health status; A demand update analysis unit, which is used to re-analyze the user's health status and re-evaluate the risk level of the user's implicit health needs. The priority update formula for implicit health needs is: , where represents the updated priority of latent health needs, is the newly detected abnormal risk value, and is the weight coefficient for priority adjustment; A recommended list dynamic adjustment unit, which is used to dynamically adjust the medical resource recommendation list according to the updated implicit health need priority provided by the demand update analysis unit.

7. An intelligent search method for medical services, characterized in that, The method includes: S1. Collect and preprocess the user's multimodal health data, including health questionnaires, medical records, physiological data and behavior data of wearable devices; S2. Through behavior pattern analysis and risk assessment models, mine and quantify the user's implicit health needs, and generate a list of implicit needs with priority ranking; S3. Combine the medical knowledge graph and big data analysis to match medical services, and dynamically generate a medical resource recommendation list according to geographical location and resource availability; S4. Continuously monitor the user's physiological indicators, trigger data update and re-analysis of implicit needs; S5. Based on the user's physiological indicators and implicit need list after triggering data update, real-time adjust the content of the medical resource recommendation list.

Images