Monitoring management system for anorectal postoperative care

Abstract

The invention discloses a monitoring management system for anorectal postoperative care, and relates to the technical field of medical monitoring management, and the monitoring management system comprises a cloud management center which is used for globally coordinating and storing various nursing data of anorectal postoperative patients; and the data access module is in communication connection with the cloud management center and is used for synchronously acquiring the physical sign monitoring information, the wound image data and the postoperative recovery progress of the patient. According to the monitoring management system for postoperative care of the anorectal department, efficient data acquisition, transmission and analysis can be realized, and the monitoring management system is effectively linked with a remote medical platform and an in-hospital information system, so that the collaborative level of data management and remote diagnosis and treatment is practically improved; safer and more convenient rehabilitation experience is provided for the patient.

Description

Technical Field

[0001] The present invention relates to the technical field of medical monitoring management, and in particular to a monitoring management system for anorectal postoperative care. Background Art

[0002] Patients after anorectal surgery usually need to go through a long recovery period, and because the wound is sensitive and easily affected by the external environment, the demand for nursing and monitoring is more stringent; however, the existing postoperative nursing monitoring system still has many shortcomings in data management and remote diagnosis and treatment collaboration: first, data collection outside the hospital mostly relies on patients' own records or intermittent uploads, and information updates are not timely and prone to errors; second, monitoring data is difficult to seamlessly connect with the hospital information system (HIS) and electronic medical record system (EMR), and the interface standards between the systems are not unified, which easily forms "data islands"; third, the remote diagnosis and treatment platform and the in-hospital system information exchange are not uniform. Due to poor communication, doctors find it difficult to timely integrate multi-dimensional data such as wound healing and fluctuations of vital signs of patients, and are unable to make effective interventions in the first place. These problems not only limit the efficiency and accuracy of anorectal postoperative care, but also lead to a lack of timely and convenient medical guidance for patients during their rehabilitation outside the hospital, and potential complications or abnormal conditions may not receive timely intervention. For this reason, there is an urgent need for an anorectal postoperative care monitoring and management system that can realize efficient data collection, transmission and analysis, and effectively link with telemedicine platforms and in-hospital information systems, so as to effectively improve the level of coordination between data management and remote diagnosis and treatment, and provide patients with a safer and more convenient rehabilitation experience. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] In view of the shortcomings of the existing technology, the present invention provides a monitoring and management system for anorectal postoperative care, which solves the existing problem that not only limits the efficiency and accuracy of anorectal postoperative care, but also leads to a lack of timely and convenient medical guidance for patients during out-of-hospital rehabilitation, and potential complications or abnormal conditions may not receive timely intervention.

[0005] (II) Technical solution

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a monitoring and management system for anorectal postoperative care, comprising:

[0007] The cloud management center is used to globally coordinate and store various nursing data of anorectal postoperative patients, specifically storing patients' personal information, medical records, postoperative examination results, and real-time monitoring data. The cloud management center is deployed with high-availability servers, which have data storage and computing capabilities, and provide database services, load balancing, and disaster recovery backup functions to ensure normal operation in the event of network fluctuations or server failures.

[0008] The data access module is connected to the cloud management center to synchronously obtain the patient's vital signs monitoring information, wound imaging data, and postoperative recovery progress; during the patient's postoperative hospitalization or post-discharge recovery process, it collects and uploads real-time vital signs and wound information through wireless or wired networks, and provides a standardized data processing interface to package the data into JSON format, which is convenient for the subsequent analysis service module to quickly call;

[0009] The analysis service module is deployed in the cloud management center and is used to grade and evaluate the patient's wound healing risk, infection probability, and pain level based on the patient's basic information, preoperative diagnostic data, and postoperative real-time data. Specifically, the analysis service module has multiple built-in algorithm models, including preliminary analysis based on threshold rules and in-depth evaluation based on machine learning or statistical methods. It comprehensively processes each patient's vital sign data, wound image features, and self-reported pain scores to output risk grading results.

[0010] The intelligent guidance module interacts with the data of the analysis service module to automatically generate personalized nursing guidance plans based on the evaluation results and trigger early warning reminders when abnormal trends occur. The intelligent guidance module maps the classification results to the pre-defined nursing plans. If a high risk level is identified, the early warning mechanism is automatically triggered. The nursing recommendations are personalized according to individual differences of patients (such as whether they have diabetes or blood system diseases).

[0011] The remote feedback interface is used to transmit nursing guidance plans and early warning reminder information to clinical doctors or professional nursing staff, and receive doctors' online diagnosis and treatment opinions or adjustment instructions; medical staff can view the patient's real-time status after logging into the system, and can initiate online video consultations, send text messages, and other operations. The remote feedback interface also has the ability to transmit data in both directions, and can instantly push doctors' prescriptions and nursing plans to patients or nursing staff for execution;

[0012] The patient interaction platform is connected to the cloud management center and is used by patients to upload self-reported pain scores, daily activity records, and pictures or video information, while displaying nursing instructions and review reminders generated by doctors and the system; the patient interaction platform supports multiple terminals, including mobile APP, tablet computers, and PC web pages. Patients can view personalized nursing tasks, review reminders, and rehabilitation progress at any time; and cooperate with mobile terminals or camera devices to realize the convenient uploading of wound images or videos.

[0013] Preferably, the data access module includes:

[0014] Mobile vital sign collection components, worn by patients or used by nursing staff in and outside the hospital, are used to measure patients' blood pressure, heart rate, body temperature and local pain levels in real time;

[0015] Image acquisition component, used to collect local images or short videos of the wound, and automatically mark the shooting time and location to assist in subsequent analysis;

[0016] The data synchronization gateway encrypts the data obtained by the mobile vital signs acquisition component and the image acquisition component and uploads it to the cloud management center, and regularly checks the network environment to ensure the real-time nature of the data.

[0017] Preferably, the analysis service module analyzes the postoperative monitoring data through the following steps:

[0018] Obtain and compare the differences in key indicators before and after surgery to determine the patient's current postoperative recovery baseline;

[0019] Use multidimensional rule sets or algorithm models to aggregate and evaluate wound appearance changes, pain scores, and risk factors for common postoperative complications;

[0020] Based on the assessment results, the patient's status is divided into several risk or concern levels and passed to the intelligent guidance module.

[0021] Preferably, the analysis service module uses the following steps and formulas to perform multi-parameter fusion calculations on patient data:

[0022] A: Obtain the preoperative baseline index B0 and the postoperative current index vector V=(v1,v2,…v n ,), where v1 represents the wound redness and swelling area, v2 represents the pain numerical quantification score, and v n represents parameters of white blood cell count, blood sugar level, and bleeding volume;

[0023] B: According to the formula:

[0024] ;

[0025] where b i is the corresponding baseline value before surgery, α i is the weight coefficient, all α i satisfy:

[0026] ;

[0027] C: If the calculated S is greater than the first threshold T1, the patient is judged to have a moderate abnormal risk; if S is greater than the second threshold T2, the patient is judged to have a high abnormal risk, and the corresponding result is transmitted to the intelligent guidance module.

[0028] Preferably, the intelligent guidance module includes:

[0029] The nursing strategy generation unit formulates a phased nursing checklist, diet and medication recommendations for patients based on the risk level output by the analysis service module;

[0030] Automatic early warning unit, which identifies abnormal trends or sudden spikes in indicators, including abnormal body temperature and obvious bleeding signs, and sends alarms to the remote feedback interface in a timely manner to remind medical staff to intervene;

[0031] The decision support submodule is used to record and track the treatment measures and effects after each abnormal alarm, so as to continuously improve the early warning strategy and nursing recommendation algorithm.

[0032] Preferably, the remote feedback interface includes:

[0033] Online consultation channel, which opens a remote visual consultation interface for professional medical staff, allowing doctors to access patients' real-time data, wound images and give adjustment instructions;

[0034] The data bridging submodule establishes a docking interface with the hospital information system (HIS) or electronic medical record (EMR) system to achieve two-way synchronization of patient hospitalization records, surgical details, and nursing data inside and outside the hospital;

[0035] The medication and prescription management submodule receives online prescriptions or care plans and pushes them to the patient interaction platform to ensure that patients and their families receive the latest medical advice in a timely manner.

[0036] Preferably, the patient interaction platform includes:

[0037] The patient self-assessment interface uses a digital pain rating scale and wound self-examination prompt questionnaire tool to guide patients to fill in and upload symptom information;

[0038] Nursing task schedule, which visualizes the nursing tasks and doctor's review arrangements generated by the system and provides reminder functions;

[0039] The multi-terminal access function supports patients to log in and use it on smartphones, tablets or web pages to receive push information and video guidance.

[0040] Preferably, the cloud management center is provided with a security audit submodule for:

[0041] Encrypt and control access rights of patient privacy data in transmission;

[0042] Record and review all parties' operations on adding, deleting and modifying patient data, and generate security logs;

[0043] Ensure data integrity and prevent potential leaks during system updates or maintenance.

[0044] Preferably, the system further comprises a model self-learning engine, which performs model training based on the collected postoperative recovery data of multiple batches of patients, for:

[0045] Continuously optimize the wound healing prediction algorithm to improve the accuracy of abnormality detection;

[0046] Combine historical medication and nursing effect data to output nursing optimization suggestions that are more in line with individual differences;

[0047] By regularly reviewing actual intervention results, the assessment thresholds and decision-making logic can be dynamically adjusted.

[0048] Preferably, the model self-learning engine performs retrospective analysis on the historical data of patients who have completed the nursing process and performs parameter iteration, and continuously improves the accuracy of risk determination in the following ways:

[0049] D: Compare the follow-up survey data within one month after the patient is discharged from the hospital with the corresponding hierarchical analysis process, and modify the α in the analysis service module according to the actual efficacy. i and specific values ​​of thresholds T1 and T2;

[0050] E: Statistics on intervention measures and final outcomes for high-risk patients, establish a mapping relationship between intervention timeliness and wound healing speed, and update the strategy library in the nursing strategy generation unit;

[0051] F: Output the iterated calculation model and write the model version number into the algorithm management list of the cloud management center to ensure the traceability of historical data.

[0052] (III) Beneficial effects

[0053] The present invention provides a monitoring and management system for anorectal postoperative care. It has the following beneficial effects:

[0054] The monitoring and management system for anorectal postoperative care can realize efficient data collection, transmission and analysis, and effectively link with the telemedicine platform and the hospital information system, thereby effectively improving the coordination level of data management and remote diagnosis and treatment, and providing patients with a safer and more convenient rehabilitation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the framework of a monitoring and management system for anorectal postoperative care according to the present invention;

[0056] Figure 2 A schematic diagram of a process for improving the accuracy of risk determination according to the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] See also Figure 1 and Figure 2 , the present invention provides a technical solution: a monitoring and management system for postoperative patient care in anorectal department, including a cloud management center, which is respectively connected to a data access module, an analysis service module, an intelligent guidance module, a remote feedback interface and a patient interaction platform through a data exchange channel;

[0059] The cloud management center is used to centrally store and encrypt the wound imaging data and physical sign measurement data of anorectal patients after surgery, and distribute nursing instructions based on the system's built-in policy executor;

[0060] The data access module is used to collect wound change information, vital signs information and self-assessment data of patients in and out of hospital environments;

[0061] The analysis service module uses a multi-parameter fusion algorithm based on the monitoring data obtained by the data access module to output the patient's wound healing status, infection tendency, and abnormal pain assessment results;

[0062] The intelligent guidance module generates nursing plans for different risk levels based on the evaluation results output by the analysis service module, and sends warning instructions to the remote feedback interface when the threshold is exceeded;

[0063] The remote feedback interface transmits nursing plans and early warning instructions to clinicians or professional nurses, and receives online diagnosis and treatment suggestions from doctors;

[0064] The patient interaction platform is used for self-reporting by patients, including records of self-perceived pain levels, daily activity records, and a fixed-point upload interface for wound images. It can also receive personalized nursing instructions and follow-up reminders forwarded by the remote feedback interface.

[0065] The mobile vital sign collection component uses a wristwatch-type wearable device with built-in sensors to detect the patient's blood pressure, heart rate and skin temperature. It also has an additional smart button for the patient to self-assess pain. When the patient feels obvious pain, he or she can enter a scale value of 0 to 10 points with one click. It also has a built-in Bluetooth or Wi-Fi module to communicate wirelessly with a mobile phone or data synchronization gateway. It collects the patient's heart rate, blood pressure, body temperature and perianal pain score through a fixed sensor device, and uploads the data to the cloud management center every 30 minutes.

[0066] The image acquisition component uses a camera with zoom and autofocus functions to ensure clear wound details; automatically embed timestamps and geographic location data, and use built-in algorithms to perform preliminary compression and encryption on images to reduce the upload burden; the camera performs wound image capture operations, collects no less than 3 partial wound images at a fixed time every day, and sends image data containing timestamps and pixel density information to the cloud management center;

[0067] The data synchronization gateway exists as an independent gateway device or mobile APP background service. It is responsible for encrypting the device-side data with AES or RSA, and then uploading it to the cloud through 4G / 5G or wired network; it also periodically checks the network connectivity, automatically caches the data when the signal is insufficient or the network is disconnected, and then uploads it after the network is restored. Through the mobile vital signs acquisition component and image acquisition component in the data access module, the patient's blood pressure, heart rate, body temperature, local pain level and wound imaging data can be collected in real time, and uploaded to the cloud management center with the help of the data synchronization gateway encryption, which effectively solves the problem of untimely information updates and errors caused by the reliance on self-recording or intermittent uploading of data collection outside the hospital, and realizes efficient data collection and stable transmission.

[0068] The analysis service module analyzes the postoperative monitoring data through the following steps: obtaining and comparing the differences between preoperative and postoperative key indicators to determine the patient's current postoperative recovery baseline; it should be further explained that in the specific implementation process, the patient's preoperative indicators, including blood routine, operation duration, and wound range recorded by the doctor, are retrieved in advance from the ward management system or hospital information system (HIS); the heart rate, blood pressure, body temperature, and wound healing degree in the images collected after the operation are compared to generate a recovery baseline;

[0069] A multi-dimensional rule set or algorithm model is used to aggregate and evaluate the changes in wound appearance, pain scores, and risk factors for common postoperative complications. It should be further explained that in the specific implementation process, a classifier or decision tree model is included to input the patient's pain score, wound image features (redness and swelling area, exudate volume), and comorbidities. If the model determines that the patient has suspected signs of infection, the risk level is increased, and intermediate analysis records are retained for manual review.

[0070] Based on the evaluation results, the patient status is divided into several risk or concern levels and transmitted to the intelligent guidance module. It should be further explained that in the specific implementation process, the risk levels can be divided into three warning levels: low risk, medium risk, and high risk. Different risk levels trigger different levels of nursing operations, including low risk only sending daily reminders by the system, medium risk suggesting to increase the frequency of reexamination, and high risk immediately informing the clinician to deal with it.

[0071] Risk levels can also be divided into three warning levels: green, yellow, and red. Different risk levels trigger different levels of nursing operations, including green where the system only sends daily reminders, yellow where it recommends increasing the frequency of reexaminations, and red where the clinician is immediately notified to handle the situation.

[0072] The analysis service module uses the following steps and formulas to perform multi-parameter fusion calculations on patient data:

[0073] A: Obtain the preoperative baseline index B0 and the postoperative current index vector V=(v1,v2,…v n ,), where v1 represents the wound redness and swelling area, v2 represents the pain numerical quantification score, and v n represents parameters of white blood cell count, blood sugar level, and bleeding volume;

[0074] B: According to the formula:

[0075] ;

[0076] where b i is the corresponding baseline value before surgery, α i is the weight coefficient, all α i satisfy:

[0077] ;

[0078] C: If the calculated S is greater than the first threshold value T1, the patient is judged to have a moderate abnormal risk; if S is greater than the second threshold value T2, the patient is judged to have a high abnormal risk, and the corresponding result is transmitted to the intelligent guidance module;

[0079] Among them, v1 represents the wound redness and swelling area, which is obtained by analyzing the pixels in the image taken by the image acquisition component, and the unit is square millimeter;

[0080] The pain score represented by v2 is the average or highest value of the somatosensory score buttons pressed by the patient;

[0081] Weight coefficient α i The weight of the pain score can be set to 0.3, the wound redness and swelling area can be set to 0.4, and the weight of the other indicators is 0.3 in total. It can also be automatically corrected with the accumulation of cases with the help of the model self-learning engine.

[0082] The values ​​of T1 and T2 are set by statistical results or clinical expert experience as T1=1.0 and T2=2.0; when S exceeds T2, it means that the patient has a significant abnormality and should be examined immediately or the doctor should be notified;

[0083] The intelligent guidance module includes:

[0084] The nursing strategy generation unit selects intervention plans from the rule base based on the risk results passed in by the analysis service module and outputs a personalized nursing schedule based on individual patient factors, and formulates a phased nursing list, diet and medication recommendations for the patient; it should be further explained that in the specific implementation process, the nursing strategy generation unit pre-designs multiple "nursing templates" and assigns corresponding task lists to nursing staff or patients based on the patient's risk level: such as "changing dressings twice a day" and "resting in bed for 2 hours", etc. Combined with dietary guidance (increasing dietary fiber and reducing spicy and irritating foods) and routine medication reminders (such as oral analgesics and antibiotic time), these guidance plans are automatically pushed to the patient interaction platform;

[0085] The automatic early warning unit identifies abnormal trends or sudden spikes in indicators, which include abnormal body temperature and obvious signs of bleeding, and promptly sends alarms to the remote feedback interface to remind medical staff to intervene. Specifically, after obtaining real-time vital sign data and wound image analysis results, a logical judgment is performed on the degree of deviation of indicators: when the body temperature of three consecutive monitoring points deviates from the baseline temperature by more than 0.6 degrees Celsius, an infection warning signal is triggered; when the pain score increases by more than 3 points within 1 hour, a pain relief intervention prompt is triggered. In the specific implementation process, the automatic early warning unit monitors the fluctuation amplitude or speed of body temperature, pain score, and wound appearance data. When it detects that "body temperature is higher than 38.5℃ for three consecutive measurements" or "wound bleeding is too dark and the area exceeds the warning line", a red alarm is immediately pushed to the doctor. It also uses multiple channels such as SMS, APP push, and system background pop-up windows to avoid delays.

[0086] The decision support submodule is used to record and track the treatment measures and effects after each abnormal alarm, so as to continuously improve the early warning strategy and nursing recommendation algorithm. Specifically, when an emergency risk is detected, an image link and a risk briefing are sent to the doctor through the remote feedback interface. In the specific implementation process, the decision support submodule records the diagnosis and treatment measures taken by the doctor after each abnormal trigger and the patient's subsequent recovery; based on the actual effect, the system's early warning judgment threshold is iterated, the time interval for the next early warning trigger is shortened or extended, and the algorithm accuracy is gradually improved.

[0087] The online consultation channel in the remote feedback interface opens a remote visual consultation interface for professional medical staff. Doctors can access patients' real-time data and wound images and give adjustment instructions, so that clinical doctors can log in to the cloud management center and conduct visual inquiries on patients' current and historical monitoring data. In the specific implementation process, the online consultation channel allows doctors with legal account permissions to log in to the cloud management center and view patients' current monitoring parameters and wound images through the Web or mobile terminal, supporting real-time voice or video consultations. The secure encrypted tunnel TLS / SSL encryption ensures that the diagnosis and treatment data will not be tampered with or stolen. The data bridging submodule in the remote feedback interface establishes a docking interface with the hospital information system or electronic medical record system, realizing the two-way synchronization of patients' hospitalization records, surgical details, and nursing data inside and outside the hospital, breaking the information barriers between the systems and avoiding the "data island" phenomenon. Doctors can access patients' real-time data and wound images through the online consultation channel and give adjustment instructions. The medication and prescription management submodule can push medical advice to the patient interaction platform in a timely manner, greatly improving the coordination level of data management and remote diagnosis and treatment, enabling doctors to integrate multi-dimensional data and make effective interventions in a timely manner.

[0088] The data bridging submodule establishes a docking interface with the hospital information system (HIS) or electronic medical record (EMR) system, writes the latest wound healing analysis results in the system into the hospital information system, and imports surgical records and drug orders from the electronic medical record into the cloud management center, realizing two-way synchronization of patient hospitalization records, surgical details, and nursing data inside and outside the hospital; in the specific implementation process, the hospital's existing HIS or EMR system is used to communicate with this system, including patient medical record query, surgical record access, and system-generated nursing data write-back; HL7 or FHIR medical data exchange standards are used to ensure unified data format and strong compatibility;

[0089] The medication and prescription management submodule pushes the corresponding prescription information to the patient interaction platform after the doctor issues a drug dosage adjustment instruction, and automatically synchronizes it to the nursing strategy generation unit for merged updates to ensure that patients and their families obtain the latest medical advice in a timely manner; in the specific implementation process, after the doctor issues a new prescription online, the system automatically sends the prescription details to the patient interaction platform, including the drug name, dosage, usage, and contraindications. If an existing prescription needs to be changed, the system automatically updates and generates a prescription change log to ensure the integrity of subsequent traceability.

[0090] The patient self-assessment interface in the patient interaction platform obtains the patient's subjective assessment of pain intensity through segmented numerical input, and automatically compares the numerical difference recorded last time; in the specific implementation process, the patient self-assessment interface uses a digital scale NRS or VAS to allow patients to directly input pain values ​​using a sliding bar, and simultaneously displays the key points of wound self-examination and common complications prompts to facilitate patients to conduct self-examination, and patients can select the "redness, swelling, exudation, and odor" option in the questionnaire; thus, patients can conveniently upload self-reported pain scores, daily activity records, and wound-related information, and can promptly check nursing instructions and reexamination reminders, so that patients can obtain timely and convenient medical guidance during outpatient rehabilitation, thereby enhancing patients' participation in and management of the rehabilitation process.

[0091] The nursing task schedule visualizes the nursing tasks generated by the system and the doctor's review arrangements, and provides a reminder function. Specifically, it lists the execution time and completion status of nursing tasks such as fixed-point dressing changes and maintaining smooth bowel movements in a calendar format, and marks a red warning when the execution time is exceeded; in the specific implementation process, the nursing task schedule lists the "dressing change time", "review time", and "sitz bath frequency" nodes in the form of a calendar or timeline. After completion, you can click the "Finish" button to let the system record it; it provides countdown reminders and message push functions to reduce the situation where patients forget or delay execution;

[0092] The multi-terminal access function supports patients to log in and use the system on smartphones, tablets or web pages to receive push information and video guidance, and guide patients to upload text records of daily sitz baths, diet and exercise intensity, and combine the vital sign data of the mobile monitoring subcomponent to provide input parameters for the analysis service module; it is convenient for patients to view nursing instructions and upload pain changes through mobile apps or web pages outside the hospital; if the patient is not good at operating smart devices, family caregivers can assist in the operation.

[0093] The cloud management center is equipped with a security audit submodule, which scans and records all data writing and reading operations at fixed time intervals G, and uses a digital signature verification mechanism to verify the legitimacy of each operation; when an abnormal access request occurs, the corresponding account is immediately blocked, and the alarm information is archived; in the specific implementation process, the AES-256 or RSA-2048 encryption algorithm is used to encrypt patient information and vital sign data in sensitive data; a user role permission table is established, including "administrator", "doctor", "patient" and "nurse", and the data accessible to different roles varies; and any read and write operations on the database automatically generate records with timestamps and role information, and the audit administrator can retrieve logs according to conditions; among them, read and write operations include browsing, editing, and deleting; and when upgrading the server software or replacing the server, data backup and verification are performed first; ensure that the backup files are also encrypted and stored, and perform consistency verification after the update is completed, and only after passing the verification will external services be restored.

[0094] It further includes a model self-learning engine, which trains the model based on the collected postoperative recovery data of multiple batches of patients, and is used to:

[0095] Continuously optimize the wound healing prediction algorithm to improve the accuracy of abnormality detection;

[0096] Combined with historical medication and nursing effect data, we can output nursing optimization suggestions that are more in line with individual differences; in the specific implementation process, we can count the differences between patients with the same condition or similar physical signs, and identify the impact of nursing or medication plans, such as whether the healing period is significantly shortened or the incidence of complications is reduced. We can output personalized suggestions, such as using antibacterial drugs earlier and appropriately strengthening pain management for patients with diabetes and anorectal surgery;

[0097] By regularly reviewing the actual intervention results, the evaluation threshold and decision logic are dynamically adjusted. In the specific implementation process, the model is retrained every 30 days to make subtle adjustments to the T1 and T2 thresholds to ensure that the system still has high accuracy when the seasons change or new surgical procedures emerge.

[0098] During the specific implementation process, the model self-learning engine runs on the high-performance computing nodes of the cloud management center, periodically extracting patients' historical monitoring data and outcome information from the database; using regression, random forest, and neural network algorithms, it trains wound healing speed, infection incidence, and pain index curves.

[0099] The model self-learning engine performs retrospective analysis on the historical data of patients who have completed the nursing process and performs parameter iteration, continuously improving the accuracy of risk assessment through the following methods:

[0100] D: Compare the follow-up survey data within one month after the patient is discharged from the hospital with the corresponding hierarchical analysis process, and modify the α in the analysis service module according to the actual efficacy. i and the specific values ​​of thresholds T1 and T2; in the specific implementation process, the pain scores, wound healing images and other information obtained by telephone or online follow-up within 1 month after discharge from the hospital are compared with the original analysis service module judgment results one by one; if analysis deviation is found, that is, the system has judged it as a moderate risk but the actual situation is more serious, the α of the corresponding indicator is appropriately increased. i Weight;

[0101] E: Statistics on intervention measures and final outcomes for high-risk patients, establish a mapping relationship between intervention timeliness and wound healing speed, and update the strategy library in the nursing strategy generation unit; in the specific implementation process, record the specific intervention content, medication types, dressing change frequency, and final wound healing time or complication probability taken by doctors for high-risk patients; construct a relationship diagram between "intervention timeliness (time from alarm to first intervention)" and "healing time" through regression analysis, and fine-tune the processing priority rules in the nursing strategy generation unit;

[0102] F: Output the iterated calculation model and write the model version number into the algorithm management list of the cloud management center to ensure the traceability of historical data. In the specific implementation process, after completing an update, assign a new version number to the model and save the old version for comparison and audit; the subsequent analysis service module automatically calls the latest version of the model to perform more accurate risk assessment for new patients;

[0103] It should be further explained that in the specific implementation process: Patient Li, male, 45 years old, underwent anorectal surgery due to hemorrhoids and anal fissures; the baseline indicators measured before the operation included: blood pressure 120 / 80mmHg, heart rate 76 times / minute, fasting blood glucose 5.6mmol / L, local redness and swelling area of ​​the preoperative wound was 0 (no obvious wound surface), and the preoperative pain score was 2; 48 hours after the operation, the patient began to access this system during hospitalization, and the mobile vital sign collection component automatically measured blood pressure, heart rate, body temperature, etc., and sent the data to the cloud management center every 30 minutes;

[0104] Nurses use dedicated terminals to take photos of wounds at 8:00 and 16:00 every day. The image acquisition component automatically embeds timestamps and upload tags. The current indicator vector V=(v1,v2,… )is obtained through the analysis service module, where v1 is the redness and swelling area of ​​10mm², v2 is the pain score of 4, and the baseline value b is used to calculate the pain score. i =0, b2=2, by formula:

[0105] ;

[0106] The postoperative risk score of the current patient is obtained; if S exceeds the set threshold T1, it is determined that the patient has a moderate abnormal risk; after determining that the patient has a moderate risk, the system automatically recommends increasing the number of sitting baths, observing the amount of wound exudate, and taking oral antibiotics based on Li's medical records and diabetes history; if the subsequent monitoring shows that the body temperature rises to 38.7℃ and does not drop for two consecutive measurements, a red alarm can be immediately sent to the clinician, indicating "suspected infection"; the system records whether the doctor adjusts the antibiotics or surgical treatment in time, and automatically includes it in the analysis when the algorithm is upgraded next time, combined with the patient's subsequent recovery status;

[0107] After receiving the system alert, the doctor turned on the remote video mode to view Li's wound images and vital signs, and communicated with Li's family members by voice to confirm whether there was significant swelling or bleeding; the doctor updated the prescription information in the EMR system, and the system automatically pulled and synchronized it to the cloud management center, so that Li's care plan was consistent on both the hospital and patient sides; if the doctor decided to increase the dosage of antibiotics or switch to painkillers, the new medication instructions would be pushed to Li's mobile patient interaction platform at the same time; Li filled out the pain scale (value range 0~10) through the mobile phone APP every morning and evening, and uploaded wound photos for comparison; the system automatically generated nursing tasks such as dressing changes, cleaning, and sitz baths. After completion, Li or his family members would punch in through the "Completed" button, and the system would record the execution time; if Li was recuperating at home, his family members could use a tablet device to log in to the same account to check the instructions and precautions pushed by the system; the cloud management center would log each time Li's data was uploaded or read, including the time of uploading the picture, the identity of the visitor, and the content of the change;

[0108] If the platform is upgraded, the security audit submodule will encrypt all operations and save backups to ensure that no data is lost during the update process. When Li completes postoperative recovery and is discharged from the hospital one month later, the system will obtain his true recovery progress and recurrence status through online or telephone follow-up visits.

[0109] The model self-learning engine compares the initial analysis service module evaluation results with the actual situation. If the analysis error is found to be too large, α is adjusted accordingly. i Or thresholds T1, T2; the system counts the time difference from triggering the alarm to implementing the intervention, and the correlation between the time required for wound healing, modifies the trigger frequency of the "high-risk reminder" in the nursing strategy generation unit, and upgrades the model version, recording the new version number; through the model self-learning engine, based on a large amount of patient postoperative recovery data, it continuously optimizes the wound healing prediction algorithm, improves the accuracy of abnormal detection, combines historical medication and nursing effects to output more individual nursing suggestions, and dynamically adjusts the evaluation threshold and decision logic. Through retrospective analysis and parameter iteration of patient data that have completed the nursing process, the system performance is continuously improved to ensure that the system can adapt to different situations and provide patients with better services.

[0110] With the assistance of this system, Li's pain index dropped significantly around the 7th day after the operation, and the wound healed well. The doctor corrected a potential infection sign at an earlier stage through remote consultation, avoiding an extended hospitalization.

[0111] Through the description of the above technical solutions and representative embodiments, the application process of the present invention in the postoperative nursing scenario of the anorectal department can be clearly seen; the system includes a cloud management center, a data access module, an analysis service module, an intelligent guidance module, a remote feedback interface and a patient interaction platform, and with specific monitoring indicators, algorithm formulas, safety mechanisms and self-learning logic, it realizes all-weather dynamic monitoring, risk assessment, personalized intervention and subsequent model optimization of postoperative patients in the anorectal department; the system can provide more timely intervention and nursing guidance for postoperative patients, and also provide scientific and convenient technical means for medical staff in postoperative management and decision-making;

[0112] Through the continuous iteration of the model self-learning engine, this system can continuously improve the accuracy of risk assessment, the targetedness of nursing plans, and the rationality of resource scheduling as more case data is accumulated, thereby demonstrating significant clinical value in anorectal clinical care.

[0113] The analysis service module conducts a comprehensive assessment based on various patient data. The intelligent guidance module generates a personalized nursing guidance plan based on the evaluation results and triggers early warning reminders in a timely manner when abnormal trends occur. Adjusting nursing recommendations based on individual differences of different patients (such as whether they have other diseases) can effectively prevent the occurrence of potential complications or abnormal conditions. Taking a specific patient as an example, the system can recommend appropriate nursing measures based on the patient's medical history information and history of diabetes, such as increasing the number of sitz baths, observing the amount of wound exudate and taking oral antibiotics, etc. When abnormal temperature rises, the doctor is notified immediately to ensure the safety of the patient.

[0114] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device 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 device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0115] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A monitoring and management system for anorectal postoperative care, characterized in that: include: The cloud management center is used to globally coordinate and store various nursing data of anorectal postoperative patients; The data access module is connected to the cloud management center to synchronously obtain the patient's vital sign monitoring information, wound imaging data, and postoperative recovery progress; The analysis service module is deployed in the cloud management center and is used to grade and evaluate the patient's wound healing risk, infection probability, and pain level based on the patient's basic information, preoperative diagnostic data, and postoperative real-time data; The intelligent guidance module interacts with the data of the analysis service module to automatically generate personalized nursing guidance plans based on the evaluation results and trigger early warning reminders when abnormal trends occur; Remote feedback interface, used to transmit nursing guidance plans and early warning reminder information to clinicians or professional nursing staff, and receive doctors' online diagnosis and treatment opinions or adjustment instructions; The patient interaction platform is connected to the cloud management center and is used by patients to upload self-reported pain scores, daily activity records, and pictures or video information, while displaying nursing instructions and review reminders generated by doctors and the system.

2. A monitoring and management system for anorectal postoperative care according to claim 1, characterized in that: The data access module comprises: Mobile vital sign collection components, worn by patients or used by nursing staff in and outside the hospital, are used to measure patients' blood pressure, heart rate, body temperature and local pain levels in real time; Image acquisition component, used to collect local images or short videos of the wound, and automatically mark the shooting time and location to assist in subsequent analysis; The data synchronization gateway encrypts the data obtained by the mobile vital signs acquisition component and the image acquisition component and uploads it to the cloud management center, and regularly checks the network environment to ensure the real-time nature of the data.

3. A monitoring and management system for anorectal postoperative care according to claim 2, characterized in that: The analysis service module analyzes the postoperative monitoring data through the following steps: Obtain and compare the differences in key indicators before and after surgery to determine the patient's current postoperative recovery baseline; Use multidimensional rule sets or algorithm models to aggregate and evaluate wound appearance changes, pain scores, and risk factors for common postoperative complications; Based on the assessment results, the patient's status is divided into several risk or concern levels and passed to the intelligent guidance module.

4. A monitoring and management system for anorectal postoperative care according to claim 3, characterized in that: The analysis service module uses the following steps and formulas to perform multi-parameter fusion calculations on patient data: A: Obtain the preoperative baseline index B0 and the postoperative current index vector V=(v1,v2,…v n ,), where v1 represents the wound redness and swelling area, v2 represents the pain numerical quantification score, and v n represents parameters of white blood cell count, blood sugar level, and bleeding volume; B: According to the formula: ； where b i is the corresponding baseline value before surgery, α i is the weight coefficient, all α i satisfy: ； C: If the calculated S is greater than the first threshold T1, the patient is judged to have a moderate abnormal risk; if S is greater than the second threshold T2, the patient is judged to have a high abnormal risk, and the corresponding result is transmitted to the intelligent guidance module.

5. A monitoring and management system for anorectal postoperative care according to claim 4, characterized in that: The intelligent guidance module comprises: The nursing strategy generation unit formulates a phased nursing checklist, diet and medication recommendations for patients based on the risk level output by the analysis service module; Automatic early warning unit, which identifies abnormal trends or sudden spikes in indicators, including abnormal body temperature and obvious bleeding signs, and sends alarms to the remote feedback interface in a timely manner to remind medical staff to intervene; The decision support submodule is used to record and track the treatment measures and effects after each abnormal alarm, so as to continuously improve the early warning strategy and nursing recommendation algorithm.

6. A monitoring and management system for anorectal postoperative care according to claim 5, characterized in that: The remote feedback interface comprises: Online consultation channel, which opens a remote visual consultation interface for professional medical staff, allowing doctors to access patients' real-time data, wound images and give adjustment instructions; The data bridging submodule establishes a docking interface with the hospital information system (HIS) or electronic medical record (EMR) system to achieve two-way synchronization of patient hospitalization records, surgical details, and nursing data inside and outside the hospital; The medication and prescription management submodule receives online prescriptions or care plans and pushes them to the patient interaction platform to ensure that patients and their families receive the latest medical advice in a timely manner.

7. A monitoring and management system for anorectal postoperative care according to claim 6, characterized in that: The patient interaction platform includes: The patient self-assessment interface uses a digital pain rating scale and wound self-examination prompt questionnaire tool to guide patients to fill in and upload symptom information; Nursing task schedule, which visualizes the nursing tasks and doctor's review arrangements generated by the system and provides reminder functions; The multi-terminal access function supports patients to log in and use it on smartphones, tablets or web pages to receive push information and video guidance.

8. A monitoring and management system for anorectal postoperative care according to claim 7, characterized in that: The cloud management center is provided with a security audit submodule for: Encrypt and control access rights of patient privacy data in transmission; Record and review all parties' operations on adding, deleting and modifying patient data, and generate security logs; Ensure data integrity and prevent potential leaks during system updates or maintenance.

9. A monitoring and management system for anorectal postoperative care according to claim 8, characterized in that: It further includes a model self-learning engine, which trains the model based on the collected postoperative recovery data of multiple batches of patients, and is used to: Continuously optimize the wound healing prediction algorithm to improve the accuracy of abnormality detection; Combine historical medication and nursing effect data to output nursing optimization suggestions that are more in line with individual differences; By regularly reviewing actual intervention results, the assessment thresholds and decision-making logic can be dynamically adjusted.

10. A monitoring and management system for anorectal postoperative care according to claim 9, characterized in that: The model self-learning engine performs retrospective analysis on the historical data of patients who have completed the nursing process and performs parameter iteration, continuously improving the accuracy of risk determination in the following ways: D: Compare the follow-up survey data within one month after the patient is discharged from the hospital with the corresponding hierarchical analysis process, and modify the α in the analysis service module according to the actual efficacy. i and specific values ​​of thresholds T1 and T2; E: Statistics on intervention measures and final outcomes for high-risk patients, establish a mapping relationship between intervention timeliness and wound healing speed, and update the strategy library in the nursing strategy generation unit; F: Output the iterated calculation model and write the model version number into the algorithm management list of the cloud management center to ensure the traceability of historical data.

Images