Whole course follow-up visit management system for gynecological and obstetrical patients

Abstract

The invention discloses a whole course follow-up visit management system for gynaecology and obstetrics patients, and belongs to the technical field of follow-up visit management. Comprising a patient full-period health archive management module, an intelligent personalized follow-up visit planning module, a dynamic risk assessment module, a health monitoring and early warning module, a personalized health guidance module, a remote intelligent monitoring emergency intervention module and a doctor-patient interaction remote cooperation module. The non-tampering and traceability of data are realized, unstructured data are converted into structured fields through natural language processing, the data utilization rate is improved, the block chain is encrypted through a national cryptographic algorithm, IPFS distributed storage and intelligent contract authority management are realized, a patient can authorize a medical institution to access data through a digital identity label, cross-institution seamless circulation is realized, and the medical institution can access to the medical institution. The repeated examination rate is obviously reduced, and the medical efficiency and the data security are improved.

Description

Technical Field

[0001] This invention belongs to the field of follow-up management technology, specifically referring to a full-process follow-up management system for obstetric and gynecological patients. Background Technology

[0002] Traditional obstetric and gynecological follow-up relies on manual methods, which is inefficient, prone to omissions, and results in scattered information. It lacks continuous monitoring and intelligent early warning, leading to poor patient compliance and making it difficult to achieve precise management of the entire pregnancy and gynecological disease cycle.

[0003] However, the existing patient follow-up management system still has certain shortcomings. It relies on scattered electronic medical records, paper records, and data from independent devices, lacking a unified data integration mechanism. Based on static medical history and fixed thresholds, it does not dynamically adjust risk levels by combining real-time vital sign data. Data collection relies on manual entry or single-point monitoring devices, which is delayed, and there is no mechanism to complete missing data, resulting in significant bias in risk prediction models. The detection of sudden abnormal indicators relies on doctors' experience judgment and lacks real-time data stream processing capabilities. Emergency reassessment processes need to be manually triggered, follow-up plans are mostly fixed cycles, and follow-up reminder methods are limited. Therefore, a full-process follow-up management system for obstetric and gynecological patients is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a comprehensive follow-up management system for obstetric and gynecological patients to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a full-process follow-up management system for obstetric and gynecological patients, including a patient full-cycle health record management module, an intelligent personalized follow-up plan module, a dynamic risk assessment module, a health monitoring and early warning module, a personalized health guidance module, a remote intelligent monitoring and emergency intervention module, and a doctor-patient interaction and remote collaboration module;

[0006] The patient full-cycle health record management module integrates data from the entire pregnancy and childbirth process with blockchain to dynamically store patient health information;

[0007] The dynamic risk assessment module combines the data from the patient's full-cycle health record management module with the real-time indicators from the health monitoring and early warning module to dynamically predict the risk level.

[0008] The intelligent personalized follow-up plan module generates a follow-up schedule and pushes multimodal reminders based on the risk tags and prediction results of the patient's full-cycle health record management module using AI algorithms.

[0009] The health monitoring and early warning module collects vital sign data through IoT devices, detects abnormal indicators and triggers graded early warnings, and synchronizes the data to the dynamic risk assessment module to update the risk level.

[0010] The personalized health guidance module pushes targeted health education content through VR based on abnormal results and consultation conclusions from the doctor-patient interaction remote collaboration module.

[0011] The remote intelligent monitoring and emergency intervention module receives abnormal signals, automatically dispatches emergency resources, and links with the medical staff-patient interaction remote collaboration module to implement remote collaboration.

[0012] The remote collaboration module for doctor-patient interaction conducts remote consultations and decisions based on early warning information and abnormal data.

[0013] Preferably, the patient's full-cycle health record management module connects to the hospital's electronic medical record system and image archive, and obtains data on pre-pregnancy examinations, pregnancy records, delivery process, postoperative rehabilitation and newborn follow-up in real time. It integrates smart bracelets, fetal heart monitors and blood glucose meters to collect vital sign data, extracts key fields from unstructured data through natural language processing, and stores them in a structured manner.

[0014] Preferably, the patient full-cycle health record management module generates a blockchain-based digital identity for each patient, binding their full-cycle health data. Delivery records, vaccinations, and critical medical events related to the diagnosis of major diseases are encrypted using national cryptographic algorithms and then uploaded to the blockchain via a consortium blockchain. Each data retrieval and update operation generates a timestamp record on the blockchain. Unstructured data such as images and reports are stored via IPFS, with only the hash value uploaded to the blockchain. Each data update generates a new block, retaining historical versions. Access permissions for different roles are defined through smart contracts, and sensitive fields are processed using homomorphic encryption. When authorizing access, identity is verified using zero-knowledge proofs. Patients can authorize other medical institutions to access their health data through their digital identity. Smart contracts automatically execute data retrieval requests and integrate with regional medical information platforms to achieve cross-institutional data sharing through blockchain.

[0015] Preferably, the dynamic risk assessment module extracts medical history, previous surgical records, and static data on gestational age progression from the patient's full-cycle health record, receives real-time vital signs and abnormal events from the health monitoring and early warning module, performs structured processing on the data using the FHIR standard, unifies the timestamp format, and completes missing vital sign data using interpolation.

[0016] Preferably, the dynamic risk assessment module constructs a risk prediction model to calculate the risk probability, using the following formula:

[0017]

[0018] In the formula, P represents the risk probability, S represents the static risk factor, which is based on the static characteristics of the patient's full-cycle health record, D represents the dynamic risk factor, which is based on the real-time vital signs of the health monitoring and early warning module, T represents the time decay factor, which reflects the dynamic adjustment of risk over time, and α, β, γ represent weighting coefficients. The formula receives real-time data streams from the health monitoring and early warning module via Kafka. For sudden abnormal indicators, it triggers an emergency reassessment process, dynamically adjusts the risk threshold based on the patient's historical data, optimizes the risk level classification strategy in conjunction with the medical resource load, generates risk labels for patients, attaches key risk factors, and writes the labels into the patient's full-cycle health record.

[0019] Preferably, the intelligent personalized follow-up plan module obtains the patient's risk level and key risk factors from the dynamic risk assessment module, obtains the probability and time sensitivity of the patient's complications, extracts previous follow-up records, preferences, and geographical location from the patient's full-cycle health record, and generates a follow-up schedule through AI algorithms, as shown in the formula:

[0020]

[0021] In the formula, F represents the follow-up intensity index, Ph represents the patient preference coefficient, G represents the geographic adjustment factor, L represents the resource load coefficient, Ts represents the time sensitivity coefficient, E represents the time urgency weight, W represents the normalized commonality, A represents the abnormal response penalty factor, D represents the patient's compliance history, and represents the patient's past follow-up task completion coefficient. If the prediction result indicates the highest risk within a preset time period, an emergency follow-up will be prioritized in the schedule. The time will be automatically adjusted based on the patient's schedule and the hospital's bed and doctor scheduling resource load. If the patient prefers SMS + APP push, a link will be sent to the APP via SMS and a reminder will pop up in the APP. If the patient prefers telephone reminders, the AI ​​voice system will make a call at a specified time. If the patient does not respond to the reminder within the specified time, a second reminder will be triggered and the attending physician will be notified simultaneously. If the patient actively reports an abnormality, the follow-up strategy will be dynamically upgraded. When the dynamic risk assessment module updates the risk level, the follow-up plan will be automatically adjusted.

[0022] Preferably, the health monitoring and early warning module collects patients' vital sign data in real time through IoT devices, uploads the data to a local gateway via WiFi, pushes the data to a cloud-based health platform via LoRa network, performs preliminary anomaly screening at the local gateway via edge computing, detects the rate of change of indicators through a sliding window, judges composite anomalies by combining multiple indicators, evaluates the combined probability of multiple factors through a Bayesian network, and classifies the early warning level into Level 1, Level 2, and Level 3. Patients receive notifications via APP, SMS, and voice calls, while medical staff receive notifications via hospital HIS system pop-ups, WeChat work group messages, and remote consultation platform reminders. The abnormal detection results and original vital sign data are uploaded to the dynamic risk assessment module for updates in real time.

[0023] Preferably, the personalized health guidance module obtains abnormal results of the patient's vital signs, obtains the doctor's diagnostic conclusion on the abnormal results from the doctor-patient interaction remote collaboration module, determines the guidance focus based on the consultation conclusion, designs VR content format in combination with patient preferences, dynamically combines scene modules according to the consultation conclusion, simulates real scenes through VR, embeds sensors in VR to detect user operations in real time and provides correction suggestions, and immediately triggers the push of VR educational content when the health monitoring module detects abnormal indicators.

[0024] Preferably, the remote intelligent monitoring and emergency intervention module receives the pushed abnormal indicators and warning levels, simultaneously obtains the dynamic risk level label, calculates the shortest path based on the patient's GPS location and the location of emergency resources, prioritizes the allocation of idle resources, notifies the patient and family members through SMS, APP push, and telephone, guides the family members to perform basic first aid through video conferencing, and synchronizes the patient's abnormal data to the doctor-patient interaction remote collaboration module.

[0025] Preferably, the remote collaboration module for doctor-patient interaction acquires and pushes abnormal indicators and warning levels in real time, integrates patient basic information and real-time monitoring data, and automatically initiates remote consultation when the warning level is level one or two. It matches the corresponding specialist doctor according to the type of abnormality, dynamically allocates tasks according to the doctor's current work status, and notifies the doctor through APP, SMS and telephone. The doctor makes a preliminary diagnosis based on the data, and the specialist doctors supplement their opinions to form a unified treatment plan. The consultation report is automatically generated and embedded into the patient's electronic health record.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. This invention integrates data from the entire pregnancy and childbirth process and combines it with blockchain technology to achieve data immutability and traceability. It transforms unstructured data into structured fields through natural language processing, improving data utilization. The blockchain uses national cryptographic algorithms for encryption, IPFS distributed storage, and smart contract permission management. Patients can authorize medical institutions to access their data through digital identity identifiers, enabling seamless cross-institutional transfer, significantly reducing duplicate examination rates, and improving medical efficiency and data security.

[0028] 2. This invention unifies structured data through the FHIR standard, combines interpolation to fill in missing data to ensure data integrity, introduces time decay factors and dynamic risk factors to build a dynamic risk prediction model, updates risk probabilities in real time, and uses Kafka real-time data stream processing and an emergency reassessment mechanism for sudden abnormal indicators to make risk level classification more accurate and avoid missed diagnoses.

[0029] 3. This invention uses AI algorithms to integrate multiple parameters such as risk level, patient preference, geographical factors, and resource load to generate a dynamic follow-up schedule, prioritizes emergency follow-ups, and dynamically adjusts the schedule based on hospital bed availability and doctor scheduling to avoid resource conflicts. The design of a secondary reminder mechanism and an abnormal response penalty factor significantly improves patient compliance. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the structure of the obstetrics and gynecology patient follow-up management system of the present invention;

[0031] Figure 2 The present invention describes the operation flow of the obstetrics and gynecology patient full-process follow-up management system. Figure 1 ;

[0032] Figure 3 The present invention describes the operation flow of the obstetrics and gynecology patient full-process follow-up management system. Figure 2 . Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Example

[0035] Please see Figures 1-3As shown, the present invention provides a technical solution including a patient full-cycle health record management module, an intelligent personalized follow-up plan module, a dynamic risk assessment module, a health monitoring and early warning module, a personalized health guidance module, a remote intelligent monitoring and emergency intervention module, and a doctor-patient interaction and remote collaboration module;

[0036] The patient full-cycle health record management module integrates data from the entire pregnancy and childbirth process with blockchain to dynamically store patient health information;

[0037] The dynamic risk assessment module combines the data from the patient's full-cycle health record management module with the real-time indicators from the health monitoring and early warning module to dynamically predict the risk level.

[0038] The intelligent personalized follow-up plan module generates a follow-up schedule and pushes multimodal reminders based on the risk tags and prediction results of the patient's full-cycle health record management module using AI algorithms.

[0039] The health monitoring and early warning module collects vital sign data through IoT devices, detects abnormal indicators and triggers graded early warnings, and synchronizes the data to the dynamic risk assessment module to update the risk level.

[0040] The personalized health guidance module pushes targeted health education content through VR based on abnormal results and consultation conclusions from the doctor-patient interaction remote collaboration module.

[0041] The remote intelligent monitoring and emergency intervention module receives abnormal signals, automatically dispatches emergency resources, and links with the medical staff-patient interaction remote collaboration module to implement remote collaboration.

[0042] The remote collaboration module for doctor-patient interaction conducts remote consultations and decisions based on early warning information and abnormal data.

[0043] Preferably, the patient's full-cycle health record management module connects to the hospital's electronic medical record system and image archive, and obtains data on pre-pregnancy examinations, pregnancy records, delivery process, postoperative rehabilitation and newborn follow-up in real time. It integrates smart bracelets, fetal heart monitors and blood glucose meters to collect vital sign data, extracts key fields from unstructured data through natural language processing, and stores them in a structured manner.

[0044] Preferably, the patient full-cycle health record management module generates a blockchain-based digital identity for each patient, binding their full-cycle health data. Delivery records, vaccinations, and critical medical events related to the diagnosis of major diseases are encrypted using national cryptographic algorithms and then uploaded to the blockchain via a consortium blockchain. Each data retrieval and update operation generates a timestamp record on the blockchain. Unstructured data such as images and reports are stored via IPFS, with only the hash value uploaded to the blockchain. Each data update generates a new block, retaining historical versions. Access permissions for different roles are defined through smart contracts, and sensitive fields are processed using homomorphic encryption. When authorizing access, identity is verified using zero-knowledge proofs. Patients can authorize other medical institutions to access their health data through their digital identity. Smart contracts automatically execute data retrieval requests and integrate with regional medical information platforms to achieve cross-institutional data sharing through blockchain.

[0045] Preferably, the dynamic risk assessment module extracts medical history, previous surgical records, and static data on gestational age progression from the patient's full-cycle health record, receives real-time vital signs and abnormal events from the health monitoring and early warning module, performs structured processing on the data using the FHIR standard, unifies the timestamp format, and completes missing vital sign data using interpolation.

[0046] Preferably, the dynamic risk assessment module constructs a risk prediction model to calculate the risk probability, using the following formula:

[0047]

[0048] In the formula, P represents the risk probability, S represents the static risk factor, which is based on the static characteristics of the patient's full-cycle health record, D represents the dynamic risk factor, which is based on the real-time vital signs of the health monitoring and early warning module, T represents the time decay factor, which reflects the dynamic adjustment of risk over time, and α, β, γ represent weighting coefficients. The formula receives real-time data streams from the health monitoring and early warning module via Kafka. For sudden abnormal indicators, it triggers an emergency reassessment process, dynamically adjusts the risk threshold based on the patient's historical data, optimizes the risk level classification strategy in conjunction with the medical resource load, generates risk labels for patients, attaches key risk factors, and writes the labels into the patient's full-cycle health record.

[0049] Preferably, the intelligent personalized follow-up plan module obtains the patient's risk level and key risk factors from the dynamic risk assessment module, obtains the probability and time sensitivity of the patient's complications, extracts previous follow-up records, preferences, and geographical location from the patient's full-cycle health record, and generates a follow-up schedule through AI algorithms, as shown in the formula:

[0050]

[0051] In the formula, F represents the follow-up intensity index, Ph represents the patient preference coefficient, G represents the geographic adjustment factor, L represents the resource load coefficient, Ts represents the time sensitivity coefficient, E represents the time urgency weight, W represents the normalized commonality, A represents the abnormal response penalty factor, D represents the patient's compliance history, and represents the patient's past follow-up task completion coefficient. If the prediction result indicates the highest risk within a preset time period, an emergency follow-up will be prioritized in the schedule. The time will be automatically adjusted based on the patient's schedule and the hospital's bed and doctor scheduling resource load. If the patient prefers SMS + APP push, a link will be sent to the APP via SMS and a reminder will pop up in the APP. If the patient prefers telephone reminders, the AI ​​voice system will make a call at a specified time. If the patient does not respond to the reminder within the specified time, a second reminder will be triggered and the attending physician will be notified simultaneously. If the patient actively reports an abnormality, the follow-up strategy will be dynamically upgraded. When the dynamic risk assessment module updates the risk level, the follow-up plan will be automatically adjusted.

[0052] Preferably, the health monitoring and early warning module collects patients' vital sign data in real time through IoT devices, uploads the data to a local gateway via WiFi, pushes the data to a cloud-based health platform via LoRa network, performs preliminary anomaly screening at the local gateway via edge computing, detects the rate of change of indicators through a sliding window, judges composite anomalies by combining multiple indicators, evaluates the combined probability of multiple factors through a Bayesian network, and classifies the early warning level into Level 1, Level 2, and Level 3. Patients receive notifications via APP, SMS, and voice calls, while medical staff receive notifications via hospital HIS system pop-ups, WeChat work group messages, and remote consultation platform reminders. The abnormal detection results and original vital sign data are uploaded to the dynamic risk assessment module for updates in real time.

[0053] Preferably, the personalized health guidance module obtains abnormal results of the patient's vital signs, obtains the doctor's diagnostic conclusion on the abnormal results from the doctor-patient interaction remote collaboration module, determines the guidance focus based on the consultation conclusion, designs VR content format in combination with patient preferences, dynamically combines scene modules according to the consultation conclusion, simulates real scenes through VR, embeds sensors in VR to detect user operations in real time and provides correction suggestions, and immediately triggers the push of VR educational content when the health monitoring module detects abnormal indicators.

[0054] Preferably, the remote intelligent monitoring and emergency intervention module receives the pushed abnormal indicators and warning levels, simultaneously obtains the dynamic risk level label, calculates the shortest path based on the patient's GPS location and the location of emergency resources, prioritizes the allocation of idle resources, notifies the patient and family members through SMS, APP push, and telephone, guides the family members to perform basic first aid through video conferencing, and synchronizes the patient's abnormal data to the doctor-patient interaction remote collaboration module.

[0055] Preferably, the remote collaboration module for doctor-patient interaction acquires and pushes abnormal indicators and warning levels in real time, integrates patient basic information and real-time monitoring data, and automatically initiates remote consultation when the warning level is level one or two. It matches the corresponding specialist doctor according to the type of abnormality, dynamically allocates tasks according to the doctor's current work status, and notifies the doctor through APP, SMS and telephone. The doctor makes a preliminary diagnosis based on the data, and the specialist doctors supplement their opinions to form a unified treatment plan. The consultation report is automatically generated and embedded into the patient's electronic health record.

[0056] Working Principle: By integrating data from the entire pregnancy and childbirth process with blockchain technology, a dynamic storage system is constructed. This system connects to the hospital's electronic medical records and image archiving systems, acquiring patient data in real time. Unstructured vital signs data is collected via smart wearable devices, and key fields are extracted through natural language processing before being structured and stored. Blockchain technology generates a digital identity for each patient, binding their full-cycle health data. Key medical events are encrypted using a consortium blockchain, and unstructured data is stored using IPFS, retaining only the hash value on the chain. Each data retrieval or update generates a timestamp. Patients can authorize other institutions to access their data through their digital identity, and smart contracts execute automatically. Cross-institutional sharing integrates static data from patients' full-cycle health records with real-time vital sign data from health monitoring and early warning modules to dynamically predict risk levels. Data formats are unified using the FHIR standard, and missing data is supplemented using interpolation. Based on static and dynamic risk factors, combined with time decay factors, a risk prediction model is constructed to calculate risk probabilities in real time. Real-time data streams are received via Kafka. Emergency reassessment processes are triggered for sudden abnormal indicators. Risk thresholds are dynamically adjusted based on patients' historical data, and risk level classification strategies are optimized in conjunction with medical resource load. Risk labels and key risk factors are generated and written into patient records.

[0057] Based on the risk labels and prediction results of the dynamic risk assessment module, combined with patient preferences, geographical location, resource load, etc., a personalized follow-up schedule is generated through AI algorithms. Taking into account risk weight, time sensitivity, and patient compliance history, the follow-up intensity is dynamically adjusted. If the risk level is high, emergency follow-up is prioritized, and the time is automatically optimized according to hospital bed availability and doctor's schedule. Patients are reminded through multiple channels such as SMS, APP push, or AI voice call. If there is no response, a second reminder is triggered and the attending physician is notified. When the dynamic risk assessment module updates the risk level, the follow-up plan is adjusted in real time. Patient vital sign data is collected in real time through IoT devices, uploaded to the local gateway via WiFi, and then transmitted to the cloud platform edge computing via LoRa network. Preliminary abnormality screening is performed at the local gateway. The rate of change of indicators is detected using a sliding window, and multi-factor judgment is combined. The probability of multi-factor combination is evaluated through Bayesian network, and the warning is divided into three levels. Patients receive warnings through APP, SMS, or voice call. Medical staff synchronize abnormal data through hospital HIS system pop-ups and remote consultation platforms, and update the results to the dynamic risk assessment module in real time, forming a closed-loop management.

[0058] Based on abnormal results from the health monitoring and early warning module and consultation conclusions from the remote collaboration module, targeted health education content is generated. Virtual reality scenarios are designed in conjunction with patient preferences, and sensors are embedded in the VR to detect user actions in real time and provide corrective suggestions. When abnormal indicators are detected, VR education content is immediately pushed to help patients intuitively understand health problems and coping measures. Immersive interaction improves patient compliance, and the learning effect is fed back to the dynamic risk assessment module. The system receives abnormal indicators and warning levels from the health monitoring and early warning module, and, combined with risk tags from the dynamic risk assessment module, initiates emergency resource dispatch. Based on the patient's GPS location and the location of emergency resources, the optimal route is calculated and available resources are prioritized. Patients and their families are notified via SMS, APP push, and telephone. Basic first aid is guided to family members through video conferencing. Upon receiving a level one or two warning, a remote consultation is automatically initiated. The system integrates basic patient information and real-time monitoring data, matches corresponding specialists according to the type of abnormality, and dynamically allocates tasks based on the doctor's work status. Doctors are notified via APP, SMS, and telephone to participate in the consultation. Doctors provide preliminary diagnoses based on the data, and after supplementary opinions from multiple specialists, a unified treatment plan is formed. The consultation report is automatically generated and embedded in the patient's electronic health record.

[0059] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their likenesses.

[0060] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A comprehensive follow-up management system for obstetric and gynecological patients, characterized by: It includes a patient full-cycle health record management module, an intelligent personalized follow-up plan module, a dynamic risk assessment module, a health monitoring and early warning module, a personalized health guidance module, a remote intelligent monitoring and emergency intervention module, and a doctor-patient interaction and remote collaboration module; The patient full-cycle health record management module integrates data from the entire pregnancy and childbirth process with blockchain to dynamically store patient health information; The dynamic risk assessment module combines the data from the patient's full-cycle health record management module with the real-time indicators from the health monitoring and early warning module to dynamically predict the risk level. The intelligent personalized follow-up plan module generates a follow-up schedule and pushes multimodal reminders based on the risk tags and prediction results of the patient's full-cycle health record management module using AI algorithms. The health monitoring and early warning module collects vital sign data through IoT devices, detects abnormal indicators and triggers graded early warnings, and synchronizes the data to the dynamic risk assessment module to update the risk level. The personalized health guidance module pushes targeted health education content through VR based on abnormal results and consultation conclusions from the doctor-patient interaction remote collaboration module. The remote intelligent monitoring and emergency intervention module receives abnormal signals, automatically dispatches emergency resources, and links with the medical staff-patient interaction remote collaboration module to implement remote collaboration. The remote collaboration module for doctor-patient interaction conducts remote consultations and decisions based on early warning information and abnormal data.

2. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The patient's full-cycle health record management module connects to the hospital's electronic medical record system and image archives, and obtains data on pre-pregnancy examinations, pregnancy records, delivery process, postoperative rehabilitation and newborn follow-up in real time. It integrates smart bracelets, fetal heart monitors and blood glucose meters to collect vital sign data, extracts key fields from unstructured data through natural language processing, and stores them in a structured manner.

3. The obstetrics and gynecology patient follow-up management system according to claim 2, characterized in that: The patient full-cycle health record management module generates a blockchain-based digital identity for each patient, binding their full-cycle health data. Delivery records, vaccinations, and critical medical events related to the diagnosis of major diseases are encrypted using national cryptographic algorithms and then uploaded to the blockchain via a consortium blockchain. Each data retrieval and update operation generates a timestamp record on the blockchain. Unstructured data such as images and reports are stored via IPFS, with only the hash value uploaded to the blockchain. Each data update generates a new block, preserving historical versions. Access permissions for different roles are defined through smart contracts, and sensitive fields are processed using homomorphic encryption. When authorizing access, identity is verified using zero-knowledge proofs. Patients can authorize other medical institutions to access their health data through their digital identity. Smart contracts automatically execute data retrieval requests and integrate with regional medical information platforms to achieve cross-institutional data sharing through blockchain.

4. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The dynamic risk assessment module extracts medical history, previous surgical records, and static data on gestational age progression from the patient's full-cycle health records. It receives real-time vital signs and abnormal events from the health monitoring and early warning module, performs structured processing on the data using the FHIR standard, standardizes the timestamp format, and completes missing vital sign data using interpolation.

5. The obstetrics and gynecology patient follow-up management system according to claim 4, characterized in that: The dynamic risk assessment module constructs a risk prediction model to calculate the risk probability, using the following formula: In the formula, P represents the risk probability, S represents the static risk factor, T represents the time decay factor, and α, β, and γ represent the weighting coefficients. The system receives real-time data streams from the health monitoring and early warning module via Kafka. For sudden abnormal indicators, it triggers an emergency reassessment process, dynamically adjusts the risk threshold based on the patient's historical data, optimizes the risk level classification strategy in conjunction with the medical resource load, generates risk labels for patients, attaches key risk factors, and writes the labels into the patient's full-cycle health record.

6. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The intelligent personalized follow-up plan module obtains the patient's risk level and key risk factors from the dynamic risk assessment module, acquires the probability and time sensitivity of the patient's complications, extracts past follow-up records, preferences, and geographical location from the patient's full-cycle health record, and generates a follow-up schedule through AI algorithms, using the following formula: In the formula, F represents the follow-up intensity index, Ph represents the patient preference coefficient, G represents the geographical adjustment factor, L represents the resource load coefficient, Ts represents the time sensitivity coefficient, E represents the time urgency weight, W represents the normalized commonality, A represents the abnormal response penalty factor, and D represents the patient's compliance history. If the prediction results indicate the highest risk within the preset time period, then emergency follow-ups will be prioritized in the schedule, and the time will be automatically adjusted based on the patient's schedule.

7. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The health monitoring and early warning module collects patients' vital signs data in real time through IoT devices, uploads the data to a local gateway via WiFi, pushes the data to a cloud-based health platform via LoRa network, performs preliminary anomaly screening at the local gateway using edge computing, detects the rate of change of indicators using a sliding window, judges composite anomalies by combining multiple indicators, and evaluates the combined probability of multiple factors using a Bayesian network. The early warning level is divided into Level 1, Level 2, and Level 3. Patients receive notifications via APP, SMS, and voice calls, while medical staff receive notifications via hospital HIS system pop-ups, WeChat work group messages, and remote consultation platform reminders. The abnormal detection results and original vital signs data are uploaded to the dynamic risk assessment module for updates in real time.

8. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The personalized health guidance module obtains abnormal results of the patient's vital signs, obtains the doctor's diagnostic conclusion on the abnormal results from the doctor-patient interaction remote collaboration module, determines the guidance focus based on the consultation conclusion, designs VR content format in combination with patient preferences, dynamically combines scene modules according to the consultation conclusion, simulates real scenes through VR, embeds sensors in VR to detect user operations in real time and provide correction suggestions, and immediately triggers the push of VR educational content when the health monitoring module detects abnormal indicators.

9. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The remote intelligent monitoring and emergency intervention module receives pushed abnormal indicators and warning levels, simultaneously obtains dynamic risk level labels, calculates the shortest path based on the patient's GPS location and the location of emergency resources, prioritizes the allocation of idle resources, and notifies the patient and family members through SMS, APP push, and telephone. It also guides family members to perform basic first aid through video conferencing and synchronizes the patient's abnormal data to the doctor-patient interaction remote collaboration module.

10. The obstetrics and gynecology patient follow-up management system according to claim 1, characterized in that: The remote collaboration module for doctor-patient interaction obtains and pushes abnormal indicators and warning levels in real time, integrates basic patient information and real-time monitoring data, and automatically initiates remote consultation when the warning level is level one or two. It matches the corresponding specialist doctor according to the type of abnormality, dynamically allocates tasks according to the doctor's current work status, and notifies the doctor through APP, SMS and telephone. The doctor makes a preliminary diagnosis based on the data, and the specialist doctors supplement their opinions to form a unified treatment plan. The consultation report is automatically generated and embedded into the patient's electronic health record.

Images