Chronic disease family oxygen therapy intelligent management method and system based on IKAP theory

Abstract

The invention discloses a chronic disease family oxygen therapy intelligent management method and system based on an IKAP theory, and relates to the field of chronic disease family oxygen therapy. Patient baseline data is collected through a structured questionnaire and comprises oxygen therapy knowledge reserve data, caregiver adaptability data and family environment adaptability data; and obtaining an initial oxygen therapy knowledge mastery score, a caregiver adaptability score and a home environment risk label based on structured questionnaire analysis. According to the invention, through the fusion of the IKAP theory and the intelligent technology, the upgrading of chronic disease family oxygen therapy from extensive management to precision, individuation and closed loop is realized, the health income of patients is guaranteed, the medical resource allocation is optimized, and the method has significant clinical value and social significance. The core logic can be migrated to family management of other chronic diseases, rapid adaptation can be achieved by adjusting a knowledge graph and monitoring indexes, and the method has wide popularization value.

Description

Technical Field

[0001] The present invention relates to the technical field of home oxygen therapy for chronic diseases, and in particular to an intelligent management method and system for home oxygen therapy for chronic diseases based on IKAP theory. Background Art

[0002] Chronic obstructive pulmonary disease (COPD) is a common chronic disease characterized by partially reversible airflow limitation that often progressively worsens, primarily manifesting as dyspnea, chronic cough, and sputum production. The China Pulmonary Health Study (CPHstudy) showed that the prevalence of COPD in adults aged 20 years and older was 8.6%, and in those aged 40 years and older, the prevalence was as high as 13.7%, and this prevalence is expected to continue to rise in the future [3-4]. As the third most common chronic disease worldwide after diabetes and hypertension, COPD places tremendous pressure on individuals, families, and society, impacting quality of life and even leading to death.

[0003] COPD often causes hypoxemia, which compromises patients' physiological and psychological functions and reduces their quality of life. Therefore, home oxygen therapy, as a routine treatment for COPD patients, has been included in COPD prevention and treatment guidelines in various countries. Long-term home oxygen therapy (LTDOT) refers to long-term oxygen therapy implemented after patients leave the hospital and return to their homes or the community, requiring continuous oxygen inhalation for more than 15 hours per day. Home oxygen therapy can increase blood oxygen partial pressure, help reduce the degree of respiratory failure, improve patients' symptoms of dyspnea, avoid the occurrence or further deterioration of complications such as cor pulmonale, reduce readmission rates, and improve quality of life.

[0004] Currently, most COPD patients who are discharged from the hospital perform home oxygen therapy independently, lacking regular supervision and effective guidance from medical staff. The current implementation of long-term home oxygen therapy is worrying. Patients and primary caregivers often lack understanding of long-term home oxygen therapy, and oxygen inhalation is often haphazard. There are issues such as insufficient duration of oxygen therapy, non-standard oxygen inhalation procedures, lack of knowledge about oxygen therapy, and inadequate disinfection and humidification of oxygen therapy devices. Inadequate implementation of home oxygen therapy can prevent it from achieving its therapeutic and palliative effects, and can even lead to acute exacerbations and readmissions to the hospital due to inadequate disease control, resulting in significant personal and societal burdens.

[0005] With the rapid development of today's society, the health needs of the general public, especially those with chronic diseases like COPD, are growing. Due to limited space, personnel, and equipment, traditional models of post-hospital health guidance, outpatient follow-up visits, and telephone follow-up are no longer sufficient for COPD patients. They require in-person visits and health guidance from medical staff at home.

[0006] COPD is a chronic disease requiring long-term self-management. With the introduction of long-term home oxygen therapy, patients are shifting from recipients of medical services to active disease managers. Meeting the increasingly diverse and individualized needs of COPD patients receiving home oxygen therapy in this new landscape, while also improving their oxygen therapy effectiveness and self-management capabilities, has become a crucial challenge facing the professional nursing field. Summary of the Invention

[0007] In order to solve the above-mentioned technical problems of home oxygen therapy management for chronic obstructive pulmonary disease, the present invention provides a method and system for intelligent management of home oxygen therapy for chronic diseases based on the IKAP theory. The following technical solutions are adopted:

[0008] The intelligent management method for home oxygen therapy for chronic diseases based on IKAP theory includes the following steps:

[0009] Step 1: Collect patient baseline data through a structured questionnaire. The patient baseline data includes oxygen therapy knowledge reserve data, caregiver compliance data, and home environment adaptability data. Based on the structured questionnaire analysis, the initial oxygen therapy knowledge mastery score, caregiver compliance score, and home environment risk label are obtained.

[0010] Step 2: Collect the patient's real-time home oxygen therapy data through the IoT data collection gateway. The patient's real-time home oxygen therapy data includes the operation data of the oxygen therapy equipment, the physiological parameters of the wearable device, and the patient's terminal APP interaction data;

[0011] Step 3: Perform a fusion analysis on the patient baseline data from step 1 and the patient's real-time home oxygen therapy data from step 2 to output a personalized needs list for the patient;

[0012] Step 4: Build an oxygen therapy knowledge graph, combine the knowledge mastery score in step 1 and the personalized needs list in step 3, and generate a patient-personalized knowledge package based on the oxygen therapy knowledge graph. This package is then pushed through the patient terminal app.

[0013] Step 5: After the patient has studied the personalized knowledge package, the system will pop up a test question and update the oxygen therapy knowledge mastery score based on the test results;

[0014] Step 6: The system combines the oxygen therapy knowledge mastery score from step 5 and the patient's real-time home oxygen therapy data from step 2 to generate an incentive content package, push it to the patient's terminal APP, and collect incentive feedback;

[0015] Step 7: Set behavioral target thresholds based on the doctor's order data and the motivational feedback from step 6;

[0016] Step 8: Based on the deviation between the patient's real-time home oxygen therapy data in step 2 and the behavioral target threshold, a graded intervention action is triggered and an intervention record is output.

[0017] Optionally, step 9 is also included, generating a chronic disease home oxygen therapy report and an optimized intervention plan for the next cycle every set period. The chronic disease home oxygen therapy report includes the degree of improvement in knowledge mastery and the rate of behavioral compliance, which are pushed to the nurse terminal APP. The nurse adjusts the push frequency of the patient's personalized knowledge package in step 4 based on the report.

[0018] Optionally, in step 1, the initial oxygen therapy knowledge mastery score is calculated as follows:

[0019] ;

[0020] in is the initial oxygen therapy knowledge mastery score, is the score of question i, is the weight of question i, and n is the total number of questions in the structured questionnaire.

[0021] Optionally, in step 5, the calculation method for the oxygen therapy knowledge mastery update score is:

[0022] ;

[0023] in is the updated oxygen therapy knowledge mastery score, The last oxygen therapy knowledge mastery score, This is the test score. is the historical weighted score.

[0024] Optionally, the method for generating the personalized requirements list in step 3 is:

[0025] The patient baseline data in step 1 and the patient's real-time home oxygen therapy data in step 2 are standardized, and personalized demand association features are extracted from the standardized data. Based on the preset rule library and the preset personalized demand dictionary, the personalized demand association features are matched with the patient's personalized demand list in step 3, and the initial demand set is output. The intensity value is calculated for each label in the initial demand set, and the mechanical energy demand priority is sorted based on the threshold range of the intensity value. The patient's personalized demand list is output in the form of demand priority, demand label, specific demand description and association features.

[0026] Optionally, personalized demand-related features include the initial score of knowledge mastery, weak links in knowledge, caregiver cooperation score, home environment risk level, ability to use digital devices, underlying diseases, average daily duration of oxygen therapy in the previous cycle, number of equipment operation errors in the previous cycle, average blood oxygen saturation in the previous cycle, cumulative duration of blood oxygen saturation less than the set saturation threshold in the previous cycle, high-frequency question keywords in the patient terminal APP, and completion rate of patients' personalized knowledge package viewing.

[0027] Optionally, the method for calculating the intensity value of the tag is:

[0028] ;in is the tag strength value, is the severity score of the fth personalized demand-related feature, is the weight of the f-th personalized demand-related feature, and k is the total number of personalized demand-related features.

[0029] Optionally, the method for generating the incentive content package in step 6 is:

[0030] Input the updated oxygen therapy knowledge mastery score from step 5 and the real-time oxygen therapy data from the previous cycle in step 2. Extract the knowledge improvement extent, behavior compliance rate, physiological indicator changes, operation proficiency, and knowledge application relevance from the input data as core features. Preset a standardized incentive content library. Based on the secondary mapping rules from core features to content types to sub-templates, match incentive content templates from the standardized incentive content library. Deeply customize the matched incentive content templates to fit the individual characteristics of the patients. Combine the customized content into the final incentive content package based on the core content and auxiliary content.

[0031] Optionally, in step 8, the deviation between each real-time data item and the corresponding target threshold is calculated to output a set of deviations of various indicators. The intervention level is divided into mild, moderate, and severe levels according to the deviation and indicator importance. The corresponding intervention action is triggered according to the determined intervention level. The intervention actions are divided into mild intervention, moderate intervention, and severe intervention.

[0032] Mild interventions include pop-up reminders on the patient terminal app, push notifications of related knowledge packages, and smart device linkage;

[0033] Moderate intervention includes strong vibration reminders on the patient terminal app, SMS notifications, synchronization deviation details and operation instructions on the family terminal app, and nurse prepared responses;

[0034] Severe intervention includes full-screen alarms on the patient APP, sound and light alarms on smart devices, strong reminders on the nurse-side APP, display of real-time patient data, and automatic calls to the responsible nurse.

[0035] An intelligent management system for home oxygen therapy for chronic diseases based on the IKAP theory is used to implement an intelligent management method for home oxygen therapy for chronic diseases. The system includes a home oxygen therapy intelligent management server, an Internet of Things data acquisition gateway, a patient terminal, a family terminal, a nurse terminal, oxygen therapy equipment, and a wearable device. The home oxygen therapy intelligent management server is wirelessly connected to the Internet of Things data acquisition gateway, the patient terminal, the family terminal, and the nurse terminal, respectively. The patient's baseline data and interaction data are collected through the patient terminal, and the operating data of the oxygen therapy equipment and the physiological parameters of the wearable device are collected through the Internet of Things data acquisition gateway. The home oxygen therapy intelligent management server controls the execution actions of the patient terminal, the family terminal, the nurse terminal, the oxygen therapy equipment, and the wearable device through the Internet.

[0036] In summary, the present invention includes at least one of the following beneficial technical effects:

[0037] The present invention can provide an intelligent management method and system for home oxygen therapy for chronic diseases based on the IKAP theory. By integrating the IKAP theory with intelligent technology, this technical solution realizes the upgrade of home oxygen therapy for chronic diseases from extensive management to precise, personalized, and closed-loop management, which not only ensures the health benefits of patients but also optimizes the allocation of medical resources. It has significant clinical value and social significance.

[0038] The core logic can be transferred to the home management of other chronic diseases and can be quickly adapted by adjusting the knowledge graph and monitoring indicators, which has broad promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of the intelligent management method for chronic disease home oxygen therapy based on the IKAP theory of the present invention;

[0040] Figure 2 This is a schematic diagram of the component connection principle of the intelligent management system for chronic disease home oxygen therapy based on the IKAP theory of the present invention;

[0041] Explanation of the accompanying symbols: 1. Home oxygen therapy intelligent management server; 2. Internet of Things data acquisition gateway; 3. Patient terminal; 4. Family terminal; 5. Nurse terminal; 6. Oxygen therapy equipment; 7. Wearable device. DETAILED DESCRIPTION

[0042] The present invention will be further described in detail below with reference to the accompanying drawings.

[0043] The embodiments of the present invention disclose a method and system for intelligent management of chronic disease home oxygen therapy based on the IKAP theory.

[0044] Reference Figure 1 and Figure 2 Example 1: An intelligent management method for chronic disease home oxygen therapy based on IKAP theory, comprising the following steps:

[0045] Step 1: Collect patient baseline data through a structured questionnaire. The patient baseline data includes oxygen therapy knowledge reserve data, caregiver compliance data, and home environment adaptability data. Based on the structured questionnaire analysis, the initial oxygen therapy knowledge mastery score, caregiver compliance score, and home environment risk label are obtained.

[0046] Step 2: Collect the patient's real-time home oxygen therapy data through the IoT data collection gateway 2. The patient's real-time home oxygen therapy data includes the operation data of the oxygen therapy equipment 6, the physiological parameters of the wearable device 7, and the patient terminal 3 APP interaction data;

[0047] Step 3: Perform a fusion analysis on the patient baseline data from step 1 and the patient's real-time home oxygen therapy data from step 2 to output a personalized needs list for the patient;

[0048] Step 4: Build an oxygen therapy knowledge graph, combine the knowledge mastery score in step 1 and the personalized needs list in step 3, and generate a patient-personalized knowledge package based on the oxygen therapy knowledge graph. This package is then pushed through the patient terminal 3 APP.

[0049] Step 5: After the patient has studied the personalized knowledge package, the system will pop up a test question and update the oxygen therapy knowledge mastery score based on the test results;

[0050] Step 6: The system combines the oxygen therapy knowledge mastery score from step 5 and the patient's real-time home oxygen therapy data from step 2 to generate an incentive content package and push it to the patient's terminal 3 APP, and collects incentive feedback;

[0051] Step 7: Set behavioral target thresholds based on the doctor's order data and the motivational feedback from step 6;

[0052] Step 8: Based on the deviation between the patient's real-time home oxygen therapy data in step 2 and the behavioral target threshold, a graded intervention action is triggered and an intervention record is output.

[0053] By adopting the above technical solutions, IKAP theory emphasizes that healthy behavior change requires a progressive process of information acquisition, knowledge mastery, belief reinforcement, and behavior formation. This solution precisely matches each step with the theoretical stages, forming a logical closed loop:

[0054] Information phase (I): Through step 1, baseline data collection, and step 2, real-time data collection, comprehensive basic information about the patient is obtained: oxygen therapy knowledge, family environment, and caregiver capabilities; dynamic status: oxygen therapy behavior, physiological indicators, and interaction needs, providing accurate input for subsequent interventions.

[0055] Knowledge Stage (K): Through step 3: needs analysis, step 4: knowledge package generation, and step 5: knowledge mastery update, the needs identified in the information stage are transformed into personalized knowledge delivery. Based on the oxygen therapy knowledge map and the patient's knowledge gaps, adaptive content is delivered, and knowledge mastery is dynamically assessed through test questions to ensure that patients understand both the facts and the reasons behind them.

[0056] Belief stage (A): Through step 6, content generation is encouraged, knowledge mastery is linked to real-time oxygen therapy data, and visual feedback on knowledge improvement, behavior improvement, and health benefits is generated. Combined with similar patient cases, patients' recognition of the value of oxygen therapy is strengthened, solving the gap between knowing and not doing, and shifting from passive acceptance to active adherence.

[0057] Behavioral stage (P): Through step 7 goal setting and step 8 graded intervention, beliefs are transformed into executable behavioral goals. By monitoring the deviation between real-time data and goals, progressive intervention is triggered to ensure that behaviors continue to meet standards and ultimately achieve health benefits.

[0058] The solution uses technologies such as the Internet of Things, big data analysis, and knowledge graphs to break through the lag and homogeneity limitations of traditional home oxygen therapy management:

[0059] By integrating structured questionnaires with real-time IoT data, and through feature extraction and demand matching, we can accurately identify patients' shortcomings and avoid blind intervention.

[0060] Based on differences in patients' knowledge, digital capabilities, family environment, etc., the knowledge package format, incentive content and intervention intensity are dynamically adjusted to achieve customized management for each individual.

[0061] Improve the standardization and effectiveness of home oxygen therapy and improve patient health; optimize the allocation of medical resources, improve management efficiency, reduce the burden on medical staff, free nurses from repeated reminders and basic guidance, and focus on the precise management of high-risk patients. Based on a personalized needs list, medical staff can allocate resources in a targeted manner to avoid waste of resources.

[0062] Different from traditional one-size-fits-all health guidance, the program ensures that each patient receives knowledge, incentives and interventions that are tailored to their actual situation through knowledge mapping and needs matching.

[0063] The solution is built based on the IKAP theory, and its core logic can be transferred to the family management of other chronic diseases. It can be quickly adapted by adjusting the knowledge graph and monitoring indicators, and has broad promotion value.

[0064] This technical solution, through the integration of IKAP theory and intelligent technology, has achieved the upgrade of home oxygen therapy for chronic diseases from extensive management to precise, personalized, and closed-loop management. It not only ensures the health benefits of patients, but also optimizes the allocation of medical resources. It has significant clinical value and social significance.

[0065] Example 2 also includes step 9, generating a chronic disease home oxygen therapy report and an optimized intervention plan for the next cycle every set period. The chronic disease home oxygen therapy report includes the degree of improvement in knowledge mastery and the rate of behavioral compliance, which are pushed to the nurse terminal APP. The nurse adjusts the push frequency of the patient's personalized knowledge package in step 4 based on the report.

[0066] By adopting the above technical solution, step 9 regularly generates quantitative indicators such as the degree of improvement in knowledge mastery and the rate of behavioral compliance, which directly reflect the patient's response to the previous intervention. Nurses can use this data to accurately judge the patient's learning pace and behavioral improvement trends:

[0067] If a patient's knowledge mastery improves by less than 10%, it indicates that the current knowledge package push frequency is insufficient. The nurse can adjust it to twice a day to strengthen knowledge input.

[0068] If the patient's knowledge level increases by more than 30% and the behavioral compliance rate remains stable at more than 80%, the push frequency can be reduced to avoid resistance caused by information overload.

[0069] This data-driven frequency adjustment makes the knowledge package push more in line with the actual needs of patients and significantly improves the accuracy of personalized management.

[0070] Home oxygen therapy for chronic diseases is a long-term process, and patients' knowledge needs and behavioral status will change over time. The periodic assessment in step 9 can capture these changes in a timely manner:

[0071] If a patient experiences nasal dryness due to seasonal changes, leading to a decrease in oxygen compliance, the sudden drop in reported behavioral compliance will trigger the nurse to adjust the knowledge package content, adding winter oxygen therapy moisturizing techniques and simultaneously increasing the frequency of push notifications to strengthen memory;

[0072] If the patient has mastered the operation of the equipment through early intervention, the report can prompt the nurse to shift the focus from operating skills to long-term effect monitoring.

[0073] This dynamic adjustment avoids the drawbacks of static intervention plans and ensures that management effects are continuously optimized over time.

[0074] The report generated in step 9 can simultaneously open key information to patients, allowing them to intuitively see their own progress and strengthen their motivation to persist. At the same time, when nurses adjust the push strategy based on the report, they can explain the reasons for the adjustment to patients, such as "You have become more proficient in the operation recently, so we will reduce the frequency of knowledge push to avoid disturbing you", thereby enhancing patients' understanding and cooperation with the management plan.

[0075] This transparent collaborative model enables patients to shift from passively accepting intervention to actively participating in management, improves the efficiency of doctor-patient communication, and significantly increases patient satisfaction.

[0076] In Example 3, in step 1, the calculation method for the initial oxygen therapy knowledge mastery score is:

[0077] ;

[0078] in is the initial oxygen therapy knowledge mastery score, is the score of question i, is the weight of question i, and n is the total number of questions in the structured questionnaire.

[0079] In Example 4, in step 5, the calculation method for the oxygen therapy knowledge mastery update score is:

[0080] ;

[0081] in is the updated oxygen therapy knowledge mastery score, The last oxygen therapy knowledge mastery score, This is the test score. is the historical weighted score.

[0082] By adopting this technical solution, traditional knowledge scoring often uses a simple calculation method of 1 point per question and dividing the total score by the number of questions. This method fails to distinguish the importance of different knowledge points, such as the difference in importance between equipment disinfection and oxygen therapy history. The weighted calculation formula for the initial oxygen therapy knowledge mastery score assigns high weight to core oxygen therapy knowledge directly impacting treatment outcomes, such as operational safety, blood oxygen saturation monitoring, and equipment disinfection, while assigning low weight to basic theory. This ensures that the score directly reflects the patient's mastery of key knowledge and avoids misjudgment. For example, if a patient incorrectly answers a high-weighted operational question, such as how to adjust oxygen flow, and incorrectly answers a low-weighted theoretical question, such as the global prevalence of COPD, the traditional average score might be deducted by 1 point. However, this formula uses a 10-point deduction for the former and a 5-point deduction for the latter, more accurately reflecting the patient's risk profile. This precise scoring provides a clear basis for generating personalized knowledge packages in step 4, avoiding wasted resources on non-core knowledge.

[0083] The updated scoring formula solves the drawback of traditional methods of using only single test scores to cover historical scores by introducing historical weights. If a patient's test score drops sharply due to accidental factors (such as fatigue or inattention), the traditional method will directly update the score to 60 points. However, this formula uses historical weights (60% historical score + 40% current score) to calculate (80×0.6+60×0.4=72) points, which more objectively reflects the patient's long-term knowledge level and avoids misadjustment of intervention strategies due to single errors.

[0084] The scoring calculation methods of Examples 3 and 4, through innovations in weighted differentiation, historical dynamic balance, and standardized quantification, upgrade the assessment of oxygen therapy knowledge mastery from vague and subjective to precise and objective. This not only provides a basis for precise intervention for medical staff, but also enhances the patient's sense of participation, ultimately promoting the conversion efficiency from knowledge transfer to behavioral improvement, and is the core support for the personalized and refined realization of the entire intelligent management solution.

[0085] In Example 5, the method for generating the personalized demand list in step 3 is:

[0086] The patient baseline data in step 1 and the patient's real-time home oxygen therapy data in step 2 are standardized, and personalized demand association features are extracted from the standardized data. Based on the preset rule library and the preset personalized demand dictionary, the personalized demand association features are matched with the patient's personalized demand list in step 3, and the initial demand set is output. The intensity value is calculated for each label in the initial demand set, and the mechanical energy demand priority is sorted based on the threshold range of the intensity value. The patient's personalized demand list is output in the form of demand priority, demand label, specific demand description and association features.

[0087] Example 6, personalized demand-related features include the initial score of knowledge mastery, weak links in knowledge, caregiver cooperation score, home environment risk level, ability to use digital devices, underlying diseases, average daily duration of oxygen therapy in the previous cycle, number of equipment operation errors in the previous cycle, average blood oxygen saturation in the previous cycle, cumulative duration of blood oxygen saturation less than the set saturation threshold in the previous cycle, high-frequency question keywords in the patient terminal 3 APP, and completion rate of viewing of the patient's personalized knowledge package.

[0088] In Example 7, the method for calculating the intensity value of a tag is:

[0089] ;in is the tag strength value, is the severity score of the fth personalized demand-related feature, is the weight of the f-th personalized demand-related feature, and k is the total number of personalized demand-related features.

[0090] By adopting the above technical solution, the 12 personalized demand-related features clearly defined in Example 6 include both the baseline data of step 1, such as the initial score of knowledge mastery, the risk level of the home environment, and the degree of cooperation of the caregiver, and the real-time dynamic data of step 2, such as the duration of oxygen therapy in the previous cycle, the number of equipment operation errors, and the high-frequency question keywords of the APP, thereby achieving a three-dimensional characterization of basic capabilities, real-time behaviors, and interactive needs.

[0091] Regularized mapping reduces subjective bias and improves consistency. Traditional demand identification relies on the subjective judgment of medical staff. Due to differences in individual experience, the same patient may be judged to have different needs, resulting in poor consistency.

[0092] Clearly match based on the preset rule base and personalized demand dictionary, and convert abstract demands into quantifiable rules. For example, if the initial knowledge mastery score is less than 60 points and the number of equipment operation errors is greater than 3 times / week, the operation skill improvement demand label will be matched to ensure that the demand judgment results of different medical staff and the same patient at different times are consistent.

[0093] The 12 personalized demand-related features are mapped one-to-one to demand labels. For example, if the blood oxygen saturation is less than 88% for a cumulative duration of more than 2 hours, the hypoxemia intervention label will be matched.

[0094] The app matches frequently asked disinfection questions with equipment maintenance tags to avoid overly generalized requests. This regularized matching shifts demand identification from being experience-driven to data-driven, improving matching accuracy.

[0095] Quantified intensity values ​​distinguish urgency and ensure that high-risk needs are given priority. Traditional demand lists are often presented in a list format, which cannot distinguish between importance and urgency, resulting in wasted medical resources on low-priority needs.

[0096] The intensity value calculation formula solves the problem that traditional demand lists are often presented in a listed form and cannot distinguish between priorities. The weighted quantification of urgency assigns high weights to safety-related features, such as the cumulative duration of blood oxygen saturation less than 88%, medium weights to effect-related features, such as insufficient oxygen therapy duration, and low weights to experience-related features, such as nasal catheter discomfort. Combined with the severity score of the features, the calculated intensity value directly reflects the urgency of the demand.

[0097] Prioritization by threshold interval: Using intensity thresholds—e.g., 8 or higher for Level 1 needs, 5-7 for Level 2 needs, and less than 5 for Level 3 needs—determines which needs should be addressed first and which should be addressed later, ensuring that high-risk needs receive priority intervention resources. This quantitative prioritization allows medical staff to quickly focus on core issues and improve resource utilization efficiency.

[0098] Associated features can be traced to improve targeted intervention. Traditional demand lists are often vague. If a patient needs help, medical staff need to communicate again to clarify the specific problem.

[0099] The needs list includes a detailed description of the need and associated features, making the reasons for the need transparent. For example, the need label for operational skill improvement describes five flow adjustment errors in the past seven days, with a knowledge mastery score of 58 points (the operational module is only 40 points). The associated features clearly indicate that the weak link is device operation, and the number of operational errors is five per week. This description allows medical staff to directly identify the root cause of the problem and tailor an intervention plan.

[0100] In Example 8, the method for generating the incentive content package in step 6 is:

[0101] Input the updated oxygen therapy knowledge mastery score from step 5 and the real-time oxygen therapy data from the previous cycle in step 2. Extract the knowledge improvement extent, behavior compliance rate, physiological indicator changes, operation proficiency, and knowledge application relevance from the input data as core features. Preset a standardized incentive content library. Based on the secondary mapping rules from core features to content types to sub-templates, match incentive content templates from the standardized incentive content library. Deeply customize the matched incentive content templates to fit the individual characteristics of the patients. Combine the customized content into the final incentive content package based on the core content and auxiliary content.

[0102] In Example 9, in step 8, the deviation between each real-time data item and the corresponding target threshold is first calculated to output a set of deviations for each indicator. The intervention level is divided into three levels: mild, moderate, and severe based on the deviation and indicator importance. The corresponding intervention action is triggered based on the determined intervention level. The intervention actions are divided into mild intervention, moderate intervention, and severe intervention.

[0103] Mild intervention includes pop-up reminders on the patient terminal 3APP, push of related knowledge packages, and linkage with smart devices;

[0104] Moderate intervention includes a strong vibration reminder on the patient terminal 3 APP, SMS notification, synchronization deviation details and operation instructions on the family terminal 4 APP, and a nurse's prepared response;

[0105] Severe intervention includes full-screen alarms on the patient APP, sound and light alarms on smart devices, strong reminders on the nurse-side APP, display of real-time patient data, and automatic calls to the responsible nurse.

[0106] By adopting the above technical solutions, traditional incentives often use a unified template, such as "Please adhere to oxygen therapy", ignoring individual differences among patients, resulting in weak incentive effects. Through the process of core feature extraction, secondary mapping, and personalized customization, we can achieve deep adaptation of incentive content to patient status:

[0107] The extracted knowledge improvement rate, such as from 60 points to 80 points, behavioral compliance rate, and changes in physiological indicators, such as blood oxygen saturation from 88% to 92%, directly reflect the patient's learning effect, behavioral performance and health benefits, ensuring that the incentive content has clear data support. For example, if your knowledge improves by 20 points and your blood oxygen improves by 4%, the effect of insisting on oxygen inhalation is significant.

[0108] Through the mapping rules of core features, content types, and sub-templates, such as the data feedback category for rapid knowledge improvement plus behavioral standard matching plus the emotional encouragement category, adaptive templates are screened from the standardized content library to avoid pushing the same content to all patients. For example, success stories are pushed to patients with slow knowledge progress, and step-by-step tasks are pushed to patients with poor behavioral standard matching.

[0109] By embedding patients' real data, adapting language style, and integrating into family scenes, patients can feel that the incentives are tailored for them, making them more receptive and converting them into actions.

[0110] In chronic disease management, patients often give up because they don't see immediate benefits, such as "I don't know what oxygen is for." By designing motivational content, we can intuitively present the three connections:

[0111] For example, the weekly report on oxygen therapy benefits uses a line graph to simultaneously display the trends of improved knowledge scores, increased oxygen therapy duration, and improved blood oxygen saturation, allowing patients to clearly perceive that learning knowledge can help regulate behavior and thus bring health benefits.

[0112] Push content such as "Why haven't I done what I learned?" to patients who are out of sync with learning, analyze obstacles such as forgetting operating steps, and provide solutions, such as setting daily operation reminders, to lower the threshold for conversion from knowing to doing.

[0113] The strengthening of this associative cognition enables patients to shift from passively following doctor's orders to actively pursuing health benefits, and their compliance with long-term oxygen therapy is improved.

[0114] Traditional incentives rely on manual writing by medical staff, such as nurses manually sending encouraging text messages, which is inefficient and difficult to standardize. Example 8: Standardized content library plus automated matching mechanism:

[0115] Medical staff do not need to repeatedly create content, but only need to maintain the content library template;

[0116] The system automatically completes feature extraction, mapping matching and customization, and can process incentive push notifications for hundreds of patients per day with stable quality.

[0117] This mechanism transforms incentive management from labor-intensive to technology-driven, improves medical care efficiency, and facilitates promotion and replication in multiple centers and regions.

[0118] Traditional interventions often adopt a one-size-fits-all approach, such as triggering a nurse call for all abnormalities, resulting in a waste of resources. For example, a minor case of insufficient duration takes up nurse time or causes delays, or a case of severe hypoxemia is not treated promptly. Example 9 achieves a precise match between risk and intervention intensity through the logic of deviation calculation, grading, and graded actions:

[0119] Deviation quantification risk: Calculate the deviation of indicators such as oxygen flow error and duration of low blood oxygen saturation, such as a flow error of 40% and low blood oxygen saturation lasting 30 minutes, to objectively assess the severity of the abnormality and avoid subjective judgment.

[0120] Three-level intervention graded response:

[0121] Mild intervention, such as insufficient oxygen therapy duration but normal blood oxygen saturation, is only delivered via an app pop-up window and a knowledge package, without disturbing family members and nurses, thus reducing patient resistance.

[0122] Moderate intervention, such as large flow deviation or short-term hypoxemia, involves prompting family members to remind them, while the nurse prepares to respond and uses family support to solve the problem;

[0123] Severe intervention, such as blood oxygen saturation less than 85% for 30 minutes, triggers a full-screen alarm and calls the nurse at the same time to ensure rapid intervention in emergencies.

[0124] This tiered mechanism allows intervention resources, such as nurse time, to be focused on high-risk situations, improving resource utilization while reducing unnecessary disruption to patients.

[0125] Mild intervention relies on the patient’s self-correction;

[0126] Moderate intervention activates family members as the first responders, and the family terminal 4APP synchronizes deviation details and operation instructions, using family care to quickly intervene;

[0127] Severe intervention directly triggers a nurse response, and the nurse terminal 5APP provides strong reminders and automatic dialing to ensure that professionals intervene within 15 minutes.

[0128] This synergistic mechanism shortens the time it takes to correct abnormalities compared to traditional interventions, and especially reduces the health risks caused by patients staying alone at home and failing to receive timely treatment.

[0129] It is required to output intervention records, including deviation details, intervention actions, and correction results. These data can be used to evaluate intervention effects, optimize intervention rules, and provide a basis for periodic reporting.

[0130] Example 10, a chronic disease home oxygen therapy intelligent management system based on IKAP theory, is used to implement a chronic disease home oxygen therapy intelligent management method. The system includes a home oxygen therapy intelligent management server 1, an Internet of Things data acquisition gateway 2, a patient terminal 3, a family terminal 4, a nurse terminal 5, an oxygen therapy device 6 and a wearable device 7. The home oxygen therapy intelligent management server 1 is wirelessly connected to the Internet of Things data acquisition gateway 2, the patient terminal 3, the family terminal 4 and the nurse terminal 5 respectively, collects patient baseline data and interaction data through the patient terminal 3, and collects operating data of the oxygen therapy device 6 and physiological parameters of the wearable device 7 through the Internet of Things data acquisition gateway 2. The home oxygen therapy intelligent management server 1 controls the execution actions of the patient terminal 3, the family terminal 4, the nurse terminal 5, the oxygen therapy device 6 and the wearable device 7 through the Internet.

[0131] The following specific embodiments are used to illustrate the implementation principle of the present invention:

[0132] Patient: Li XX, male, 65 years old, diagnosed with COPD (GOLD grade 3), requires long-term home oxygen therapy (doctor's order: oxygen inhalation ≥15 hours per day, oxygen flow 2L / min, maintain blood oxygen saturation ≥90%).

[0133] Family situation: Lives alone, son (Wang X, 35 years old) visits twice a week and serves as the primary caregiver; home oxygen therapy equipment is a smart oxygen concentrator, and the patient can use basic smartphone functions.

[0134] System composition and connection relationship:

[0135] Home oxygen therapy intelligent management server 1: Deployed in the hospital cloud, it is responsible for data processing, knowledge graph storage, and intervention decision-making, and communicates wirelessly with other components via 4G / 5G.

[0136] IoT data collection gateway 2: installed in the patient's home, connected to oxygen therapy equipment 6 (smart oxygen concentrator) and wearable device 7 (smart bracelet) via Bluetooth, and uploads data to server 1 in real time.

[0137] Patient Terminal 3: A smartphone app (“Oxygen Therapy Manager”) used by patients to fill out questionnaires, study knowledge packages, and receive reminders.

[0138] Family terminal 4: The caregiver’s (son’s) mobile phone APP synchronizes the patient’s abnormal data and intervention reminders.

[0139] Nurse Terminal 5: Tablet APP of responsible nurse Zhang X, used to view reports and adjust knowledge package push strategies.

[0140] Oxygen therapy equipment 6: Oxygen concentrator with IoT module, which can record flow rate, usage time, disinfection records, and support remote status query.

[0141] Wearable device 7: Smart bracelet, which monitors blood oxygen saturation, heart rate, and body position changes (to determine whether the nasal cannula has fallen off).

[0142] Specific implementation process:

[0143] Step 1: Baseline data collection and analysis: The patient completes a structured questionnaire through Terminal 3, which includes: oxygen therapy knowledge: 10 questions (e.g., "How long do I need to inhale oxygen daily?" and "How do I disinfect the humidifier bottle?"); caregiver cooperation: 5 questions (e.g., "Can my son help adjust the flow rate?"); home environment adaptability: 5 questions, and uploads 3 photos (e.g., oxygen concentrator placement, power interface, and surrounding environment).

[0144] Analysis results: Initial oxygen therapy knowledge mastery score: 6 correct answers to operational questions (10 points each), 2 correct answers to theoretical questions (5 points each), ≈46.7; Caregiver cooperation score: 3 points (out of 5 points, "the son can remind the child to take oxygen but cannot operate the equipment"); Home environment risk label: IoT Gateway 2 uses image recognition to find that "the oxygen concentrator is 2 meters away from the kitchen stove (open flame source)", and combines the questionnaire answers to generate a "medium risk" label.

[0145] Step 2: Real-time oxygen therapy data collection. IoT gateway 2 continuously collects data and uploads it to server 1. The data for the first week is as follows: Oxygen therapy device 6 (oxygen concentrator): average daily flow rate of 2.3 L / min (slightly exceeding the doctor's order), average daily duration of 10 hours (5 hours short of the target), and disinfection records twice a week (less than once a day). Wearable device 7 (bracelet): average blood oxygen saturation of 88%, daily blood oxygen saturation below 88% for a cumulative 1.5 hours, and frequent changes in body position at night (suggesting possible nasal cannula dislocation).

[0146] Patient terminal 3 (APP): Clicked on the "equipment disinfection" knowledge 4 times, asked the key words "inaccurate flow adjustment" and "nasal congestion at night", and the knowledge package viewing completion rate was 50%.

[0147] Step 3: Generate a Personalized Needs List. Server 1 analyzes baseline and real-time data, extracting 12 relevant features (5 in the example): an initial knowledge score of 46.7, a weak area of ​​knowledge related to "equipment operation / disinfection," three equipment operation errors in the previous week (adjusting the flow rate to 3 L / min), frequent app questions about "flow adjustment," an average blood oxygen saturation of 88%, and risk in the home environment. Matching needs tags: Based on a pre-set rule base, "Knowledge score <60 + operation errors ≥3 / week" is matched to the "Operation Skills Improvement" tag.

[0148] The calculation intensity score is 9 points, a first-level requirement. Output requirement list: Operational skills improvement, 3 flow adjustment errors in the past week, knowledge score 46.7 points (operation module only 30 points); (Related characteristics: Number of operation errors = 3 times / week, weak knowledge link is equipment operation);

[0149] Step 4: Generate and push personalized knowledge packages. Server 1 calls the oxygen therapy knowledge graph (including nodes such as "flow adjustment" and "disinfection steps") and generates a knowledge package based on the requirements list: Core content: AR animation: 2L / min flow adjustment steps, short video: 3 steps for daily disinfection;

[0150] Push strategy: Push information through patient terminal 3 at 8 am every day. Since the patient has medium digital ability, video plus text and pictures are preferred.

[0151] Step 5: After the patient's knowledge package is updated, Terminal 3 pops up three test questions (such as "What should be used to disinfect the humidifier bottle every day?"), and the test score is 75 points.

[0152] The updated score is 58.02 points (an increase of 11.3 points from the initial score).

[0153] Step 6: Generate Incentive Content Package (Example 8) Server 1 extracts core features: knowledge improvement of 11.3 points, behavioral compliance rate (10 / 15 = 67%), and average blood oxygen saturation increased to 89% (a 1% increase from the previous level). Match the incentive template: "Data Feedback" + "Emotional Encouragement"; Personalization: Terminal 3 pushes: Oxygen Therapy Weekly Report: "Uncle Li, your operational knowledge has improved by 11.3 points. You've been using oxygen for 10 hours a day this week (67% of the target), and your blood oxygen saturation has increased from 88% to 89%! Keep it up! Another 5 hours of oxygen will meet the target."

[0154] The message is synchronized to the family member's terminal 4, and a prompt is given: "Dad's operating errors have decreased this week, so you can remind him to take supplemental oxygen therapy at night."

[0155] Step 7: Behavioral target threshold setting is combined with medical advice (15 hours / day, 2L / min) and incentive feedback (patients have positive feedback on the "ladder task"). Server 1 sets: mandatory threshold: blood oxygen saturation ≥90%, flow 1.5-2.5L / min; optimization threshold: ≥15 hours per day, disinfection ≥1 time / day.

[0156] Step 8: Graded intervention: On a certain day in the second week, real-time data showed: oxygen therapy duration was 8 hours (target 15 hours, deviation 46.7%), flow rate was 2.2 L / min (normal), and blood oxygen saturation was 89% (normal).

[0157] Server 1 determines "mild intervention", triggering: a pop-up window on patient terminal 3: "Uncle Li, you have been taking oxygen for 8 hours today, 7 hours away from the target. It is recommended that you take oxygen before going to bed (the device has preset a reminder at 22:00 for you)"; the smart bracelet (wearable device 7) gives a voice reminder: "It's time to take oxygen, hold on."

[0158] Step 9: Periodic reporting and program optimization: After two weeks, Server 1 generates a chronic disease home oxygen therapy report and pushes it to Nurse Terminal 5: Knowledge mastery improves by 11.3 points (46.7 to 58.02); the behavioral compliance rate reaches 67%, and operational errors are reduced to 1 per week.

[0159] Based on the report, nurse Zhang × adjusted the frequency of knowledge package push in step 4 from once a day to twice a day, and added content on comfortable nighttime oxygen inhalation techniques.

[0160] One month after implementation, patient data improved: knowledge mastery score rose to 72 points (an increase of 25.3 points); behavioral compliance rate was 80% (12 / 15 hours), and the average blood oxygen saturation was 91%; no unplanned hospitalizations occurred, and family members and patients were 90% satisfied with the management plan.

[0161] This example verifies the feasibility of the technical solution through specific scenarios: a four-stage closed-loop design based on the IKAP theory, combined with technologies such as the Internet of Things and knowledge graphs, realizes full-process management from data collection, personalized intervention, and effect optimization; the various components of the system work together to ensure accurate and efficient intervention, ultimately improving the standardization and health benefits of patients' home oxygen therapy, and demonstrating the clinical value and scalability of the solution.

[0162] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent management method for home oxygen therapy for chronic diseases based on the IKAP theory, characterized by: The following steps are involved: Step 1: Collect patient baseline data through a structured questionnaire. The patient baseline data includes oxygen therapy knowledge reserve data, caregiver compliance data, and home environment adaptability data. Based on the structured questionnaire analysis, the initial oxygen therapy knowledge mastery score, caregiver compliance score, and home environment risk label are obtained. Step 2, collecting the patient's home real-time oxygen therapy data through the Internet of Things data collection gateway (2), the patient's home real-time oxygen therapy data includes the operation data of the oxygen therapy equipment (6), the physiological parameters of the wearable device (7) and the patient terminal (3) APP interaction data; Step 3: Perform a fusion analysis of the patient baseline data from step 1 and the patient's real-time home oxygen therapy data from step 2 to output a personalized needs list for the patient; Step 4: Construct an oxygen therapy knowledge graph, combine the knowledge mastery score in step 1 and the personalized demand list in step 3, generate a patient personalized knowledge package based on the oxygen therapy knowledge graph, and push it through the patient terminal (3) APP; Step 5: After the patient has studied the personalized knowledge package, the system will pop up a test question and update the oxygen therapy knowledge mastery score based on the test results; Step 6: The system combines the oxygen therapy knowledge mastery score from step 5 and the patient's real-time home oxygen therapy data from step 2) to generate an incentive content package and pushes it to the patient's terminal (3) APP, and collects incentive feedback; Step 7: Set behavioral target thresholds based on the doctor's order data and the motivational feedback from step 6; Step 8: Based on the deviation between the patient's real-time home oxygen therapy data in step 2 and the behavioral target threshold, a graded intervention action is triggered and an intervention record is output.

2. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 1, characterized in that: It also includes step 9, generating a chronic disease home oxygen therapy report and an optimized intervention plan for the next cycle every set period. The chronic disease home oxygen therapy report includes the degree of improvement in knowledge mastery and the rate of behavioral compliance, which are pushed to the nurse terminal APP. The nurse adjusts the push frequency of the patient's personalized knowledge package in step 4 based on the report.

3. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 2, characterized in that: In step 1, the initial oxygen therapy knowledge score is calculated as follows: ;in is the initial oxygen therapy knowledge mastery score, is the score of question i, is the weight of question i, and n is the total number of questions in the structured questionnaire.

4. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 3 is characterized in that: In step 5, the updated oxygen therapy knowledge score is calculated as follows: ;in is the updated oxygen therapy knowledge mastery score, The last oxygen therapy knowledge mastery score, This is the test score. is the historical weighted score.

5. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 1, characterized in that: The method for generating the personalized needs list in step 3 is: standardize the patient baseline data in step 1 and the patient's real-time home oxygen therapy data in step 2, extract personalized needs-related features from the standardized data, match the personalized needs-related features with the patient's personalized needs list in step 3) based on a preset rule base and a preset personalized needs dictionary, output an initial needs set, calculate the intensity value for each label in the initial needs set, prioritize the mechanical energy needs based on the threshold range of the intensity value, and output the patient's personalized needs list in the form of need priority, need label, specific need description, and associated features.

6. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 5, characterized in that: The characteristics associated with personalized needs include the initial score of knowledge mastery, weak links in knowledge, caregiver cooperation score, home environment risk level, ability to use digital devices, underlying diseases, average daily duration of oxygen therapy in the previous cycle, number of device operation errors in the previous cycle, average blood oxygen saturation in the previous cycle, cumulative duration of blood oxygen saturation less than the set saturation threshold in the previous cycle, high-frequency question keywords in the patient terminal (3) APP, and completion rate of patients' personalized knowledge package viewing.

7. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 5, characterized in that: The method for calculating the label strength value is: ;in is the tag strength value, is the severity score of the fth personalized demand-related feature, is the weight of the f-th personalized demand-related feature, and k is the total number of personalized demand-related features.

8. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 5, characterized in that: The method for generating the incentive content package in step 6 is: Input the updated oxygen therapy knowledge mastery score from step 5 and the real-time oxygen therapy data from the previous cycle in step 2. Extract the knowledge improvement extent, behavior compliance rate, physiological indicator changes, operation proficiency, and knowledge application relevance from the input data as core features. Preset a standardized incentive content library. Based on the secondary mapping rules from core features to content types to sub-templates, match incentive content templates from the standardized incentive content library. Deeply customize the matched incentive content templates to fit the individual characteristics of the patients. Combine the customized content into the final incentive content package based on the core content and auxiliary content.

9. The intelligent management method for chronic disease home oxygen therapy based on IKAP theory according to claim 8, characterized in that: In step 8, the deviation between each real-time data and the corresponding target threshold is calculated to output the deviation set of each indicator. The intervention level is divided into mild, moderate, and severe levels according to the deviation and indicator importance. The corresponding intervention action is triggered according to the determined intervention level. The intervention actions are divided into mild intervention, moderate intervention, and severe intervention. Mild intervention includes pop-up reminders on the patient terminal (3) APP, push of related knowledge packages, and smart device linkage; Moderate intervention includes (3) APP strong vibration reminder and SMS notification on the patient terminal, (4) APP synchronization deviation details and operation instructions and nurse prepared response on the family terminal; Severe intervention includes full-screen alarms on the patient APP, sound and light alarms on smart devices, strong reminders on the nurse-side APP, display of real-time patient data, and automatic calls to the responsible nurse.

10. The intelligent management system for chronic disease home oxygen therapy based on IKAP theory is characterized by: A method for intelligent management of home oxygen therapy for chronic diseases according to any one of claims 1 to 9 is provided, wherein the system comprises a home oxygen therapy intelligent management server (1), an Internet of Things data acquisition gateway (2), a patient terminal (3), a family terminal (4), a nurse terminal (5), an oxygen therapy device (6), and a wearable device (7), wherein the home oxygen therapy intelligent management server (1) is wirelessly connected to the Internet of Things data acquisition gateway (2), the patient terminal (3), the family terminal (4), and the nurse terminal (5), respectively, collects patient baseline data and interaction data through the patient terminal (3), collects operating data of the oxygen therapy device (6) and physiological parameters of the wearable device (7) through the Internet of Things data acquisition gateway (2), and controls the execution actions of the patient terminal (3), the family terminal (4), the nurse terminal (5), the oxygen therapy device (6), and the wearable device (7) through the Internet.

Images