Intelligent reminding device based on use of inpatient

Abstract

The invention relates to the technical field of data management, in particular to an intelligent reminding device based on inpatient use, which comprises a patient risk analysis module, a patient screening module, a nursing early warning module, an injury analysis module, an action guidance module and a statistical analysis module. According to the method and the system, the stress injury risk level of each patient is calculated by integrating the medical record data with the lying time, age and basic illness state of the patient, so that the accurate assessment of the risk state of the patient is remarkably improved, the demand of manual intervention is reduced, the information transmission speed is also accelerated, and the risk assessment efficiency is improved. According to the method and the system, nursing personnel can obtain key early warning information in real time, take corresponding prevention measures and collect and identify injury photos to further make injury degree assessment more accurate and timely, provided nursing action guidance is formulated based on actual injury conditions, the nursing working process is optimized, and the overall efficiency and safety of medical services are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and particularly to an intelligent reminder device for inpatients. Background Art

[0002] Data management is a technical field that involves collecting, validating, storing, protecting, and processing data to ensure the accuracy, accessibility, and reliability of the data. It encompasses various technologies from database management, data integration, data security to big data and cloud services. The main goal of data management is to improve the quality and utilization efficiency of data while ensuring the security and privacy of information. In the context of the rapid development of information technology today, effective data management is crucial for enterprises and institutions, which can help gain an advantage in data-driven decision-making processes, improve operational efficiency, and business outcomes.

[0003] The definition of experimental diagnosis of heart diseases is that when the heart undergoes lesions, corresponding biochemical levels in the body will change. Therefore, laboratory diagnosis of cardiac biochemical markers and assessment of cardiovascular risk factors have good value for the diagnosis, risk classification, and prognosis estimation of heart diseases. Common heart diseases include coronary heart disease, heart failure, angina pectoris, etc.

[0004] The cardiovascular assessment system for patients with secondary hyperparathyroidism in chronic kidney disease with the patent number CN210896636U is provided with a cardiovascular detection module, a central processor, a wireless gateway, a big data module, a family member terminal, a display module, a storage module, a doctor terminal, and an alarm module. The cardiovascular detection module is electrically connected to the central processor, the central processor is electrically connected to the wireless gateway, and the wireless gateway is signal-connected to the big data module and the family member terminal. This cardiovascular assessment system for patients with secondary hyperparathyroidism in chronic kidney disease can not only compare and analyze the information collected by big data and the detected information, thereby automatically judging the patient's physical condition and automatically reminding medical staff in critical situations to improve the rescue efficiency, but also facilitate family members to understand the patient's situation. However, this design only has a single description and cannot well evaluate the patient's actual cardiovascular condition and provide corresponding treatment.

[0005] Therefore, how to evaluate the patient's cardiovascular condition and provide corresponding treatment is a technical problem that needs to be solved by those skilled in the art at present. Summary of the Invention

[0006] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose an intelligent reminder device for inpatients.

[0007] To achieve the above object, the present invention adopts the following technical solutions. An intelligent reminder device for inpatients includes:

[0008] The patient risk analysis module extracts the medical records of patients, combines the patient's bedridden time, age, and basic medical conditions, calculates the risk level of pressure injury for each patient, and generates a risk rating result.

[0009] The patient screening module screens the patient information with a high priority risk level according to the risk rating result, collates the patient's name, ward number, and contact nurse information, and obtains a list of dangerous patients.

[0010] The nursing warning module uses the list of dangerous patients to set a scheduled task to send emails and text messages to nursing staff, including the risk rating information of the patient's pressure injury and reminders of preventive measures, and obtains a warning push record.

[0011] The injury analysis module uses the warning push record, combines the patient's living habits and photos of the injury site, takes preventive measures for patients without pressure injury, and evaluates the injury degree and injury level of pressure injury patients to obtain an injury assessment result.

[0012] The action guidance module provides nursing action guidance through the injury level in the injury assessment result, analyzes the information of wound assessment and the optimal time for dressing change, and obtains a nursing guidance list.

[0013] The statistical analysis module conducts statistical analysis on the management of patients' pressure injuries through the nursing guidance list, records real-time risk and nursing quality control data, and generates a management statistical log.

[0014] As a further solution of the present invention, the risk rating result includes high and low scores of the risk and grade classification, the list of dangerous patients includes the patient's name, ward number, and contact nurse information, the warning push record includes the timestamp, sending status, and content summary of the emails and text messages sent to nursing staff, the injury assessment result includes the injury degree identification, injury level assessment, and severity classification of the patient's injury site photos, the nursing guidance list includes wound assessment information, recommended wound treatment measures, and the optimal time for dressing change, and the management statistical log includes the statistical analysis results of the management of patients' pressure injuries, the recorded real-time risk assessment data, and nursing quality control data.

[0015] As a further solution of the present invention, the patient risk analysis module includes:

[0016] The medical record extraction sub-module extracts the patient's personal basic information, medical records, and responses to pressure injury, screens eligible records, performs data processing, and standardizes and unifies them to generate patient medical record records.

[0017] The risk calculation sub-module uses the patient's medical record to analyze the patient's age and bedridden time, calculates the pressure injury risk of the patient through a risk assessment model, evaluates the risk level of each patient, and obtains risk assessment data;

[0018] The data classification sub-module uses the risk assessment data to classify and process the data, sorts them according to the risk level, tracks and processes patients with a high risk level in real time, and generates a risk rating result.

[0019] As a further solution of the present invention, the patient screening module includes:

[0020] The risk ranking sub-module uses the risk rating result, adopts the comprehensive risk index model evaluation method to screen patients with a high risk level, sorts the data according to the condition of the patient's pressure injury and the admission time, excludes patients with stable conditions, and generates priority patient data;

[0021] The information integration sub-module extracts the patient's name and ward number based on the priority patient data, and at the same time queries the contact information of the associated nurse, verifies the extracted data, and generates verified patient information;

[0022] The list output sub-module uses the verified patient information, combines it with a predefined risk template, and automatically fills in the patient's information, including emergency contact information and nursing measures, to generate a list of dangerous patients.

[0023] As a further solution of the present invention, the formula of the comprehensive risk index model evaluation method is as follows:

[0024]

[0025] Among them, R z is the risk score of the patient, P represents the degree of the patient's pressure injury, T represents the patient's admission time, S represents the patient's real-time stability index, D represents the number of days since the patient was admitted, and w1, w2, and w3 are weight coefficients.

[0026] As a further solution of the present invention, the nursing warning module includes:

[0027] The task scheduling sub-module formulates a daily timed-triggered task plan based on the list of dangerous patients, checks and sends updated patient information to the nursing team at a fixed time every day, and generates a timed task configuration;

[0028] The warning sending sub-module uses the timed task configuration to send emails and text messages to the target nursing staff list through the hospital's internal communication server, including the pressure injury risk rating and prevention measures of the patient, and generates warning message content;

[0029] The information update sub-module uses the content of the warning message to record the sending time, recipient, and message content each time, synchronously updates the pressure injury information of the patient, verifies the consistency of the information, and generates a warning push record.

[0030] As a further solution of the present invention, the injury analysis module includes:

[0031] The habit analysis sub-module uses the warning push record to collect the daily activity data of the patient, screens the key behavior patterns through data collation, and obtains the analysis result of the living pattern;

[0032] The injury photo analysis sub-module uses the analysis result of the living pattern to scan and analyze the photos of the injury site, mark the injury characteristics, and classify them according to the severity of the characteristics to form a judgment result of the injury degree;

[0033] The prevention measure sub-module analyzes the patient characteristics according to the judgment result of the injury degree, identifies the successful prevention measures of the case, formulates a prevention plan, and predicts the effect of the plan to obtain an injury assessment result.

[0034] As a further solution of the present invention, the action guidance module includes:

[0035] The characteristic analysis sub-module uses the injury level information in the injury assessment result to analyze the characteristics of different-level injuries and the required nursing degree, provides treatment suggestions for different-level injuries, checks the matching degree of nursing measures and injury severity, and generates a draft of action suggestions;

[0036] The action suggestion sub-module calculates the optimal time interval for dressing change according to the draft of action suggestions, referring to the duration of pressure injury and dressing change requirements, optimizes the nursing actions, and generates an action guidance plan;

[0037] The plan execution sub-module formulates the key action steps and dressing change plan by using the decision tree algorithm according to the requirements of the action guidance plan, and generates a nursing guidance list.

[0038] As a further solution of the present invention, the formula of the decision tree algorithm is as follows:

[0039]

[0040] Among them, C s represents the classification result, P s represents the real-time health status index of the patient, Q s represents the allergic reaction index of the patient to the target dressing, R s represents the risk assessment associated with the patient, S sRepresents the urgency score of preventive measures, a s and b s are weight parameters.

[0041] As a further aspect of the present invention, the statistical analysis module includes:

[0042] The associated information analysis sub-module collects data associated with the patient's pressure injury information based on the nursing guidance list, including the implementation of nursing measures, dressing change frequency, and injury recovery progress, and generates a pressure injury management data set;

[0043] The risk factor identification sub-module analyzes the pressure injury management data set, evaluates the effectiveness of nursing measures and the timeliness of dressing changes, identifies potential risk factors, and generates risk and quality control analysis results;

[0044] The log recording sub-module uses the risk and quality control analysis results to record associated statistical data, conducts real-time monitoring and review through formatted storage, verifies the information transfer speed and archiving efficiency, and generates a management statistical log.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] In the present invention, by integrating medical record data with the patient's bedridden time, age, and basic medical conditions, the pressure injury risk level of each patient is calculated, and a risk rating result is generated, significantly improving the accurate assessment of the patient's risk status. The risk rating result can effectively screen out high-risk patients, and the screening mechanism ensures the priority allocation of medical resources to the most needed individuals. The timed tasks automatically send emails and text messages, which not only reduce the need for manual intervention but also speed up the information transfer speed, enabling nursing staff to obtain key warning information immediately and take corresponding preventive measures. Collecting and identifying injury photos further makes the assessment of the injury degree more accurate and timely, and the provided nursing action guidance is formulated based on the actual injury situation, making each nursing more personalized and targeted. The implementation of statistical analysis more accurately reflects the real-time status of pressure injury management, provides data support for continuous improvement of nursing quality, not only optimizes the nursing work process but also enhances the overall effectiveness and safety of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is the system flowchart of the present invention;

[0048] Figure 2 is the schematic diagram of the system framework of the present invention;

[0049] Figure 3 is the flowchart of the patient risk analysis module of the present invention;

[0050] Figure 4Flowchart of the patient screening module of the present invention;

[0051] Figure 5 Flowchart of the nursing warning module of the present invention;

[0052] Figure 6 Flowchart of the injury analysis module of the present invention;

[0053] Figure 7 Flowchart of the action guidance module of the present invention;

[0054] Figure 8 Flowchart of the statistical analysis module of the present invention. Detailed implementation manners

[0055] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0056] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0057] Please refer to Figures 1 to 2 , an intelligent reminder device for inpatients includes:

[0058] The patient risk analysis module extracts the medical records of the patient, combines the patient's bedridden time, age and basic medical conditions, performs data operations using a risk assessment model, calculates the risk level of pressure injury for each patient, and generates a risk rating result;

[0059] The patient screening module screens the patient information with a high priority risk level according to the risk level data in the risk rating result, and collates the patient's name, ward number and contact nurse information to obtain a list of dangerous patients;

[0060] The nursing warning module uses the list of dangerous patients to set a scheduled task to send emails and text messages to the nursing staff, including the risk rating information of the patient's pressure injury and reminders of preventive measures, to perform scheduled push of information and obtain a warning push record;

[0061] The injury analysis module uses the aforementioned warning push records, combines the patient's living habits and photos of the injury site, takes preventive measures for patients without pressure injuries, evaluates the injury degree and injury level of pressure injury patients, and obtains the injury assessment results;

[0062] The action guidance module provides nursing action guidance through the injury level in the injury assessment results, analyzes the information of wound assessment and the time of dressing change, and obtains the nursing guidance list;

[0063] The statistical analysis module conducts statistical analysis on the management of patients' pressure injuries through the nursing guidance list, records real-time risk and nursing quality control data, and generates a management statistical log.

[0064] The risk rating result includes the high and low scores of the risk and the grade classification. The list of dangerous patients includes the patient's name, ward number, and contact nurse information. The warning push record includes the timestamp, sending status, and content summary of the emails and text messages sent to the nursing staff. The injury assessment result includes the injury degree identification, injury level assessment, and severity classification of the patient's injury site photos. The nursing guidance list includes wound assessment information, recommended wound treatment measures, and the optimal time for dressing change. The management statistical log includes the statistical analysis results of the management of patients' pressure injuries, the recorded real-time risk assessment data, and nursing quality control data.

[0065] Please refer to Figure 2 、 3 , the patient risk analysis module includes:

[0066] The medical record extraction sub-module extracts the patient's personal basic information, case history, and reactions to pressure injuries, screens eligible records, and performs data processing to standardize the formatting, and generates the patient medical record record according to the following execution process;

[0067] Extract the patient's personal basic information, case history, and reactions to pressure injuries, screen out records that meet specific conditions, such as key information such as age, gender, and course of disease, and through the data processing function of the database, format the information, unify the data format, ensure that all data meet the requirements of further analysis and storage, check and verify the integrity and accuracy of the data, and generate a patient medical record record with a unified format and easy to retrieve.

[0068] The risk calculation sub-module uses the patient medical record record, analyzes the patient's age and bedridden time, and calculates the pressure injury risk of the patient through a risk assessment model, and evaluates the risk level of each patient. The execution process of obtaining the risk assessment data is as follows;

[0069] Use the patient medical record record, and calculate the pressure injury risk through a risk assessment model. According to the formula Calculate the risk level of each patient. In the formula, R a represents the risk level of pressure injury, a represents the age coefficient, and t represents the bedridden time.

[0070] Detailed explanation of the formula and the derivation process of the formula calculation: Set the age coefficient a, and its value can be obtained according to the correlation between age and pressure injury in previous studies. t is the bedridden time of the patient, in days. For example, for a patient with an age coefficient of 0.03 and a bedridden time of 10 days, substitute the values into the formula to calculate R≈0.406. The result shows that the risk level of pressure injury of the patient is 0.406, indicating that the patient is at a medium risk level, providing an important basis for evaluating the nursing level and preventive measures required by the patient.

[0071] The data classification sub-module uses the risk assessment data, classifies and processes the data, sorts them according to the risk level, and tracks and processes the patients with high risk levels in real time. The execution process of generating the risk rating result is as follows;

[0072] Using the risk assessment data, classifying and processing the data requires the use of data classification algorithms and real-time data processing technologies. It is necessary to classify and sort all patients' risk assessment data according to the risk level. The process involves complex data sorting algorithms such as quicksort or heapsort. Through programming, the automatic classification and sorting of data are realized to ensure that patients with high risk levels can be immediately identified, and the data can be tracked in real time. By setting thresholds, high-risk patients who need urgent treatment or continuous monitoring can be identified, and the priority treatment of patients with high risk in the medical system can be realized to ensure timely intervention and recovery. The generated risk rating result will be used to guide the allocation of hospital resources and the management strategy of patients.

[0073] Please refer to Figure 2 、 4 , the patient screening module includes:

[0074] The risk ranking sub-module uses the risk rating result and adopts the comprehensive risk index model evaluation method to screen patients with high risk levels, sorts the data according to the condition of the patient's pressure injury and the admission time, excludes patients with stable conditions, and the execution process of generating the priority patient data is as follows;

[0075] The formula of the comprehensive risk index model evaluation method is as follows:

[0076]

[0077] Among them, R zis the risk score of the patient, P represents the degree of pressure injury of the patient, T represents the admission time of the patient, S represents the real-time stability index of the patient, D represents the number of days since the patient was admitted, and w1, w2, and w3 are weight coefficients.

[0078] Formula:

[0079]

[0080] Detailed explanation of the formula and the derivation process of formula calculation:

[0081] The formula is used to evaluate the risk level of the patient, where P represents the degree of pressure injury. It is scored by medical staff according to the clinical assessment form, and the score range is from 1 to 10. A high score indicates a severe degree of injury.

[0082] Suppose a patient's pressure injury score is 8, and T represents the admission time, calculated in hours and obtained from the hospital records. For example, 24 hours have passed since a patient was admitted.

[0083] S is the stability index, which is also scored by medical staff, ranging from 1 to 5. A low score indicates unstable condition. Set the stability index of this patient to 3.

[0084] D is the number of days since the patient was admitted, directly obtained from the hospital's electronic health record, and set to 1 day.

[0085] The values of the weight coefficients w1, w2, and w3 need to be determined based on clinical experience analysis. According to statistical data, the degree of pressure injury has the greatest impact on the disease risk. Therefore, set w1 = 0.5. The square root of the admission time is used to reduce the direct impact of the time length, so set w2 = 0.3. The absolute value change of the stability index also has a significant impact on the risk level, so set w3 = 0.2.

[0086] Substitute specific values for calculation:

[0087]

[0088] The calculation result is 2.835. The score represents the risk level after comprehensively considering the patient's degree of pressure injury, admission time, and stability. A higher score indicates that the patient's condition requires closer monitoring and timely intervention. The results show that this patient is at a high risk level, and the team should give priority to nursing to ensure the reasonable allocation of resources, improve the recovery efficiency and patient safety.

[0089] Based on the priority patient data, the information integration sub-module extracts the patient's name and ward number, and at the same time queries the contact information of the associated nurse. The execution process of verifying the extracted data to generate verified patient information is as follows;

[0090] Based on the priority patient data, set the priority rules for data extraction. According to the urgency and recovery needs of patients, prioritize the extraction of patient data that requires immediate care or monitoring. The process involves the use of database query languages, such as SQL statements for effective data extraction. For example, SELECT name, ward number FROM patient database WHERE emergency level > 7. At the same time, the nurse information to be contacted is also obtained through a similar query statement. Further data verification processing ensures the accuracy and timeliness of the extracted information. By comparing the existing information in the database, verify the consistency and accuracy of the newly extracted data, ensure the accurate transmission and effective utilization of information, which is of great significance for rapid response in emergency situations, and generate verified patient information.

[0091] The list output sub-module uses the verified patient information and combines it with a predefined risk template to automatically fill in the patient's information, including emergency contact information and care measures. The execution process of generating the list of high-risk patients is as follows;

[0092] Use the verified patient information to automatically fill in the patient's information. According to the formula N = f(P b , E, M), generate the list of high-risk patients. In the formula, N represents the list of high-risk patients, P b represents the basic patient information, E represents the emergency contact information, and M represents the care measures. Explanation of the formula and the derivation process of the formula calculation: The basic patient information P b includes the name and ward number. The emergency contact information E is obtained through database query, and the care measures M are predefined according to the patient's specific condition and risk level. For example, for patient John Doe, ward number 123, emergency contact phone number 8888 - 555 - 0000, the required care measure is 24-hour electrocardiogram monitoring. Substitute the information into the formula N = f("JohnDoe123", "8888 - 555 - 0000", "24-hour electrocardiogram monitoring"), and the result generates an entry for John Doe in the list of high-risk patients. The results show that according to the verified patient information and the predefined risk template, generating a list for high-risk patients is crucial for the rational allocation of medical resources and emergency response management.

[0093] Please refer to Figure 2 、 5 , the care warning module includes:

[0094] The task scheduling sub-module, based on the list of high-risk patients, formulates a task plan triggered daily at a fixed time, checks and sends updated patient information to the nursing team at a fixed time every day. The execution process of generating the timed task configuration is as follows;

[0095] Based on the list of high-risk patients, scheduled tasks are set up programmatically. The task schedule is configured on the server using task scheduling software such as Cron. For example, a task is set to trigger at 3 am every day. SQL queries are used to retrieve updated patient information, such as "SELECT name, condition update FROM patient list WHERE update time >= yesterday's date", and the information is packaged into a task. The task scheduler is responsible for triggering the task at the set time to ensure that the nursing team can obtain the latest patient information. The process also includes monitoring the execution of the task and logging, in order to detect and resolve problems that occur during execution, including the execution time, execution frequency, and details of the target task, to ensure the timeliness of information transmission and the accuracy of nursing response, and generate a scheduled task configuration.

[0096] The early warning sending sub-module adopts a scheduled task configuration. Through the hospital's internal communication server, emails and text messages are sent to the list of target nursing staff, including the risk rating of the patient's pressure injury and preventive measures. The execution process of generating the early warning message content is as follows;

[0097] Adopting a scheduled task configuration, the configured scheduled task is used to trigger the message sending process. The content of the early warning message includes the risk rating of the patient's pressure injury and corresponding preventive measures. The message is sent through the email server and the SMS gateway, involving the formatting of emails and the simplification of SMS content, to ensure the adaptability and reading efficiency of information on different platforms. At the same time, the system needs to optimize the sending time according to the working hours and job arrangements of the nursing staff to improve the efficiency and timeliness of information transmission, ensuring that the nursing team can obtain important patient information and action guidelines in a timely manner. The generated early warning message content aims to improve the response speed and efficiency of patient care.

[0098] The information update sub-module uses the content of the early warning message to record the time of each sending, the recipients, and the message content, synchronously updates the pressure injury information of the patient, and verifies the consistency of the information. The execution process of generating the early warning push record is as follows;

[0099] Using the content of the early warning message, synchronously update the pressure injury information of the patient. According to the formula U = g(T s , R, C), generate an early warning push record. In the formula, U represents the early warning push record, T s represents the time of each sending, R represents the recipient, and C represents the message content.

[0100] Detailed explanation of the formula and the derivation process of the formula calculation: The time of each sending T sIt is recorded as a specific date and time, such as 2023-09-23-14:00. The receiving person R is matched from the employee database, such as the name of a nurse or doctor. The message content C includes the patient's risk rating and preventive measures. Substitute the information into the formula U = g("2023-09-23-14:00", "NurseA", "High risk rating, enhanced monitoring required"), and as a result, a specific warning push record is generated. The results show that by accurately recording the sending time, receiving person, and message content, the system can effectively track and manage the process of pushing warning messages, providing detailed records and a basis for subsequent patient monitoring.

[0101] Please refer to Figure 2 、 6 , the injury analysis module includes:

[0102] The habit analysis sub-module uses the warning push record to collect the patient's daily activity data, sorts and filters the key behavior patterns through data collation, and the execution process of obtaining the life pattern analysis result is as follows;

[0103] The warning push record is used to collect key behavior patterns. The system analyzes the patient's daily activity data, sorts and filters out the behaviors that affect the quality of life, identifies important behavior patterns through data collation technology, and screens out the behaviors that have the greatest impact on the patient's life by calculating parameters such as the frequency and duration of different activities. This process involves a large amount of data comparison and pattern recognition technology. The system can effectively extract valuable information from the massive data, and then through algorithm optimization, ensure that the collected data can accurately reflect the patient's actual life pattern, and the obtained life pattern analysis result can provide a scientific basis for subsequent injury risk assessment.

[0104] The injury photo analysis sub-module uses the life pattern analysis result to scan and analyze the photos of the injured part, mark the injury characteristics, and classify them according to the severity of the characteristics to form the execution process of the injury degree judgment result as follows;

[0105] Scanning and analysis of the photos of the injured part, calculate the dispersion degree of the injury characteristics according to the formula In the formula, S s represents the standard deviation of the injury characteristics, x s represents the measured value of each injury characteristic, x represents the average value of the injury characteristic measured values, and n s represents the number of injury characteristics.

[0106] Detailed explanation of the formula and the formula calculation derivation process: The formula is used to calculate the standard deviation of the injury characteristic data, indicating the variability size between different injury characteristics. If the injury characteristic values are 15, 20, 22, 20, 15, then the average value x = 18.4, substitute it into the formula to calculate and get S s≈3.13, indicating that the dispersion degree of damage characteristics on data points is 3.13. The results help evaluate the severity of the damage and conduct grading.

[0107] According to the judgment result of the damage degree, the preventive measure sub-module analyzes the patient characteristics, identifies the successful preventive measures in the cases, formulates a prevention plan, and predicts the effect of the plan. The execution process for obtaining the injury assessment result is as follows;

[0108] According to the judgment result of the damage degree, analyze the patient characteristics, identify and formulate a highly targeted prevention plan. Using the data of historical successful cases and combining with the specific situation of the patient, predict the effect of the plan through a deep learning model. The prediction process includes the input and processing of numerous variables, such as the patient's past medical records, living habits, and responses to changes, etc. After the data is trained by the model, it can output the patient's injury risk and the effect evaluation of preventive measures, which directly affects future treatment and nursing plans, ensuring that the proposed prevention plan can effectively reduce the patient's injury risk and improve the quality of life, and obtaining the injury assessment result.

[0109] Please refer to Figure 2 、 7 , and the action guidance module includes:

[0110] The characteristic analysis sub-module uses the injury level information in the injury assessment result to analyze the characteristics of different-level injuries and the required nursing degree, provides treatment suggestions for different-level injuries, checks the matching degree between nursing measures and injury severity, and generates the initial draft of action suggestions. The execution process is as follows;

[0111] Use the injury level information in the injury assessment result. According to the formula D g =i g (L g , N g ), generate the initial draft of action suggestions. In the formula, D g represents the initial draft of action suggestions, L g represents the injury level, and N g represents the degree of nursing need.

[0112] Detailed explanation of the formula and the derivation process of formula calculation: The injury level L g is obtained from the injury assessment result. Different levels represent injuries of different severities. The degree of nursing need N g is formulated according to the injury level. For example, for high-level injuries, centralized monitoring and recovery are required. For example, if the patient's injury level is 3, high-level nursing measures are needed. Substitute the information into the formula D = i g(3, High-level care), the results generate specific nursing action suggestions. The results show that by systematically analyzing the matching of injury levels and nursing needs, targeted treatment suggestions can be provided for the medical team, optimizing the patient's nursing plan, increasing the recovery effect while reducing the patient's pain.

[0113] Based on the initial draft of the action suggestions, referring to the duration of pressure injury and the dressing change requirements, the action suggestion sub-module calculates the optimal time interval for dressing changes, optimizes nursing actions, and generates the implementation process of the action guidance plan as follows;

[0114] Based on the initial draft of the action suggestions, analyze the duration of pressure injury and relevant nursing records, determine the service life and replacement frequency of the dressing, use data analysis methods such as statistical analysis or mathematical modeling to calculate the optimal dressing change time interval to ensure the patient's wound recovery speed and prevent infection. The calculation process considers the patient's condition changes, wound healing speed, and the work efficiency of the nursing team, details the dressing change time interval, such as changing once every 48 hours, and relevant nursing measures, such as regularly cleaning and disinfecting the wound, to ensure that the patient receives coherent and scientific nursing services, optimize nursing actions, improve the recovery effect, and generate the action guidance plan.

[0115] The plan execution sub-module, based on the requirements of the action guidance plan, uses the decision tree algorithm to formulate key action steps and dressing change plans, and generates the implementation process of the nursing guidance list as follows;

[0116] The formula of the decision tree algorithm is as follows:

[0117]

[0118] Among them, C s represents the classification result, P s represents the patient's real-time health status index, Q s represents the patient's allergic reaction index to the target dressing, R s represents the risk assessment associated with the patient, S s represents the urgency score of preventive measures, a s and b s are weight parameters.

[0119] Formula:

[0120]

[0121] Detailed explanation of the formula and the formula calculation derivation process:

[0122] The formulas used in the decision tree algorithm include the patient's health status index P s , allergic reaction index Q s , risk assessment Rs , and the urgency score S of preventive measures s . The weight parameter a s and b s are adjusted according to the actual situation to balance the influence of each index on the decision-making.

[0123] Parameter setting and quantification process:

[0124] P s (Health status index) is calculated by continuously monitoring the patient's physiological parameters (such as heart rate, blood pressure, etc.). For example, it is set that within a certain period, the average heart rate is 80 bpm and the average blood pressure is 120 / 80 mmHg. According to the preset standard, P s is calculated as (80 / 100) × 1.2 = 0.96.

[0125] Q s (Allergy reaction index) is obtained based on the patient's response data to specific drugs or materials. It is set that there is a record of mild reaction in the past, and through the scoring system, Q s = 0.3.

[0126] R s (Risk assessment) is determined by a professional team based on the patient's medical records and recovery suggestions. For example, if the patient has mild heart disease and the risk assessment is 4 (scale 1 - 10), then R s = 4.

[0127] S s (Urgency score of preventive measures) is assigned according to the patient's current medical condition and the emergencies that occur. For example, the urgency score is 3 (scale 1 - 5), that is, S s = 3.

[0128] Weight a s and b s are obtained through data analysis. It is set that the weight of the health status is 0.7 and the weight of the allergy reaction is 0.3.

[0129] Calculation example:

[0130] Set a s = 0.7, b s = 0.3, P s = 0.96, Q s = 0.3, R s = 4, S s = 3

[0131] The calculation process is as follows:

[0132] Calculate the numerator: a s ×P s +b s ×Q s= 0.7 × 0.96 + 0.3 × 0.3 = 0.672 + 0.09 = 0.762

[0133] Calculate the denominator:

[0134] Final C s Value:

[0135] The results show that according to the current parameters, the comprehensive score of the patient is 0.1524, and the value reflects the patient's recovery needs and urgency at a specific time point. According to the level of the C value, the nursing team can decide to take corresponding nursing measures and prevention strategies. A high C value indicates the need for immediate and active recovery interventions, while a lower value indicates that the current recovery plan is appropriate or requires minor adjustments.

[0136] Please refer to Figure 2 、 8 , the statistical analysis module includes:

[0137] The associated information analysis sub-module collects data associated with the patient's pressure injury information based on the nursing guidance checklist, including the implementation of nursing measures, dressing change frequency, and injury recovery progress. The execution process of generating the pressure injury management dataset is as follows;

[0138] Based on the nursing guidance checklist, by integrating various nursing records and patient monitoring data, such as detailed records of the implementation of nursing measures, specific times and frequencies of dressing changes, and observation data on wound recovery, using data collection techniques and database management to systematically collect and store information, cleaning and formatting the collected data to ensure the accuracy and consistency of the data, through data correlation analysis, revealing the correlation between nursing measures and patient recovery, evaluating the impact of different nursing methods on the progress of injury recovery, providing basic data for subsequent risk assessment and nursing optimization, ensuring the comprehensiveness of information and the effectiveness of application, and generating the pressure injury management dataset.

[0139] The risk factor identification sub-module analyzes the pressure injury management dataset, evaluates the effectiveness of nursing measures and the timeliness of dressing changes, identifies potential risk factors, and the execution process of generating the risk and quality control analysis results is as follows;

[0140] Analyze the pressure injury management dataset, according to the formula F s = v(E s , T s ), generate the risk and quality control analysis results. In the formula, F s represents the risk and quality control analysis results, E s represents the effectiveness of nursing measures, and T s represents the timeliness of dressing changes.

[0141] Detailed Explanation of Formulas and Derivation Process of Formula Calculations: Effectiveness E of Nursing Measures s Evaluated by the patient's recovery speed and wound healing condition, Timeliness T of dressing change s Measured according to the recorded replacement frequency and time. For example, if the nursing measure is rated as highly effective and the dressing is changed on time without error, substitute the information into the formula F = v(highly effective, on time), and the result will generate a detailed analysis of risk and quality control. The results show that by quantitatively analyzing the implementation quality of nursing measures and dressing changes, potential risk factors can be effectively identified, providing accurate data support for improving the quality of patient management and reducing future risks.

[0142] The log recording sub-module uses the results of risk and quality control analysis to record associated statistical data, conducts real-time monitoring and review through formatted storage, verifies the information transfer speed and archiving efficiency, and the execution process of generating management statistical logs is as follows;

[0143] Using the results of risk and quality control analysis, ensure that all collected data and analysis results are formatted and stored in an automated manner, such as using stored procedures and triggers in a database system to ensure real-time data update and secure backup. Set up a data warehouse for long-term data storage and management, provide interfaces for data query and review to achieve fast data retrieval and analysis. At the same time, the system will monitor the speed of the data transfer and storage process, evaluate the system performance and efficiency, and provide important real-time information for system maintenance by recording the timestamp, data size, and processing time of each data event, supporting the continuous optimization and efficiency improvement of the system, and generating management statistical logs.

[0144] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent reminder device for inpatients, characterized in that: The system comprises: The patient risk analysis module extracts the patient's medical history, combines the patient's bed rest time, age and basic condition, calculates the risk level of pressure injury for each patient, and generates a risk rating result; The patient screening module screens the patient information with high risk priority according to the risk rating result, sorts the patient's name, ward number and contact nurse information, and obtains a list of dangerous patients; The nursing warning module uses the list of dangerous patients to set a scheduled task to send emails and text messages to nursing staff, including risk rating information of pressure injuries of patients and reminders of preventive measures, and obtains warning push records; The injury analysis module uses the warning push record, combined with the patient's living habits and photos of the injured part, to take preventive measures for patients who have not suffered pressure injuries, evaluate the degree of injury and injury grade of patients with pressure injuries, and obtain injury assessment results; The action guidance module provides nursing action guidance based on the injury level in the injury assessment result, analyzes the wound assessment information and the time for dressing change, and obtains a nursing guidance list; The statistical analysis module performs statistical analysis on the patient's pressure injury management through the nursing guidance list, records real-time risk and nursing quality control data, and generates a management statistical log.

2. The intelligent reminder device for inpatients according to claim 1, characterized in that: The risk rating results include high and low risk scores and level classifications, the list of dangerous patients includes the patient's name, ward number and contact nurse information, the early warning push records include the timestamp, sending status and content summary of the emails and text messages sent to nursing staff, the injury assessment results include injury degree identification, injury level assessment and severity classification of patient injury site photos, the nursing instruction list includes wound assessment information, recommended wound treatment measures and the optimal time for dressing changes, and the management statistics log includes statistical analysis results of patient pressure injury management, recorded real-time risk assessment data and nursing quality control data.

3. The intelligent reminder device for inpatients according to claim 1, characterized in that: The patient risk analysis module includes: The medical record extraction submodule extracts the patient's basic personal information, case history, and pressure injury response, screens eligible records, processes the data, unifies the formatting, and generates patient medical records; The risk calculation submodule uses the patient's medical records to analyze the patient's age and bed rest time, calculates the patient's pressure injury risk through a risk assessment model, assesses the risk level of each patient, and obtains risk assessment data; The data classification submodule uses the risk assessment data to classify the data, sort the data by risk level, track and process patients with high risk levels in real time, and generate risk rating results.

4. The intelligent reminder device for inpatients according to claim 1, characterized in that: The patient screening module includes: The risk ranking submodule uses the risk rating results and the comprehensive risk index model evaluation method to screen patients with high risk levels, sort the data according to the patient's pressure injury status and admission time, exclude patients with stable conditions, and generate priority patient data; The information integration submodule extracts the patient's name and ward number based on the priority patient data, queries the contact information of the associated nurse, verifies the extracted data, and generates verified patient information; The list output submodule uses the verified patient information in combination with a predefined risk template to automatically fill in the patient's information, including emergency contact information and nursing measures, to generate a list of dangerous patients.

5. The intelligent reminder device for inpatients according to claim 4, characterized in that: The formula of the comprehensive risk index model evaluation method is as follows: Among them, R z is the patient's risk score, P represents the patient's pressure injury degree, T represents the patient's admission time, S represents the patient's real-time stability index, D represents the number of days since the patient was admitted, and w1, w2 and w3 are weight coefficients.

6. The intelligent reminder device for inpatients according to claim 1, characterized in that: The nursing early warning module comprises: The task scheduling submodule formulates a daily scheduled triggering task plan based on the list of dangerous patients, checks and sends updated patient information to the nursing team on a daily basis, and generates a scheduled task configuration; The warning sending submodule adopts the timed task configuration to send emails and text messages to the target nursing staff list through the hospital's internal communication server, including the patient's pressure injury risk rating and preventive measures, and generates the warning message content; The information update submodule uses the warning message content to record the time of each transmission, the recipient and the message content, synchronously updates the patient's pressure injury information, verifies the consistency of the information, and generates a warning push record.

7. The intelligent reminder device for inpatients according to claim 1, characterized in that: The damage analysis module includes: The habit analysis submodule uses the warning push record to collect the patient's daily activity data, sorts and screens key behavior patterns through data, and obtains life pattern analysis results; The injury photo analysis submodule uses the life pattern analysis results to scan and analyze photos of the injury site, annotate the injury features, and classify the features according to their severity to form an injury degree judgment result; The preventive measures submodule determines the result of the injury degree, analyzes the patient's characteristics, identifies the successful preventive measures of the case, formulates a preventive plan, and predicts the effect of the plan to obtain the injury assessment result.

8. The intelligent reminder device for inpatients according to claim 1, characterized in that: The action guidance module includes: The characteristic analysis submodule uses the injury grade information in the injury assessment results to analyze the characteristics of the differentiated grade injuries and the required nursing degree, provides treatment suggestions for the differentiated grade injuries, checks the matching degree between the nursing measures and the severity of the injury, and generates a preliminary draft of the action suggestion; The action suggestion submodule calculates the optimal time interval for dressing change based on the draft action suggestion, the duration of the pressure injury and the dressing change requirement, optimizes the nursing action, and generates an action guidance plan; The plan execution submodule adopts a decision tree algorithm to formulate key action steps and dressing change plans based on the needs of the action guidance plan and generates a nursing guidance list.

9. The intelligent reminder device for inpatients according to claim 8, characterized in that: The formula of the decision tree algorithm is as follows: Among them, C s Represents the classification result, P s Represents the patient's real-time health status index, Q s represents the patient's allergic reaction index to the target dressing, R s represents the risk assessment associated with the patient, S s represents the urgency score of the preventive measures, a s and b s It is a weight parameter. If the index iPTH is greater than 88, it is judged as a relapse and an alarm will be issued to remind the patient and doctor.

10. The intelligent reminder device for inpatients according to claim 1, characterized in that: The statistical analysis module includes: The associated information analysis submodule collects data associated with the patient's pressure injury information based on the nursing guidance list, including the implementation of nursing measures, the frequency of dressing changes, and the progress of injury recovery, and generates a pressure injury management data set; The risk factor identification submodule analyzes the pressure injury management data set, evaluates the effectiveness of nursing measures and the timeliness of dressing changes, identifies potential risk factors, and generates risk and quality control analysis results; The logging submodule utilizes the risk and quality control analysis results to record the associated statistical data, conducts real-time monitoring and review through formatted storage, verifies the speed of information transmission and archiving efficiency, and generates management statistical logs.

Images