A medical insurance-based visit key path monitoring method and system

Abstract

The application provides a medical insurance-based treatment key path monitoring method and system. The medication behavior data of chronic disease patients is collected in real time through an intelligent medicine bottle, and is associated with medical insurance data in real time to generate a medication compliance deviation index and a treatment path abnormality coefficient. When both of them exceed the chronic disease type correlation threshold at the same time, medical insurance settlement interception is triggered and a dynamic path correction scheme is pushed. According to the actual treatment path data fed back by the chronic disease patients, the spatio-temporal correlation degree of the treatment area and the threshold mapping relationship are dynamically updated to realize accurate monitoring and intervention of the chronic disease recheck path. The technical scheme provided by the application can realize real-time sensing of the medication behavior of patients, dynamic correlation of medical insurance data and accurate monitoring of the treatment path, and improves the fine level of chronic disease management and the intelligent ability of medical insurance supervision.

Description

Technical Field

[0001] This application relates to the fields of medical and health care and medical insurance supervision technology, and in particular to a method and system for monitoring critical pathways to medical treatment based on medical insurance. Background Technology

[0002] Patients with chronic diseases require long-term medication and regular follow-up visits. Their medication adherence and compliance with follow-up visit protocols directly impact treatment outcomes and the efficiency of medical insurance fund utilization. However, traditional monitoring methods struggle to track medication behavior and visit pathways for chronic disease patients in real time, leading to frequent problems such as medication non-adherence and derailed follow-up visits. This increases the waste of medical insurance funds and the health risks for chronic disease patients.

[0003] Currently, existing solutions use smart pillboxes to collect medication data from chronic disease patients (such as opening time and medication inventory), and combine this data with prescription information from electronic medical records to generate medication adherence reports. Simultaneously, GPS positioning technology is used to monitor the activity trajectories of chronic disease patients, comparing them with pre-defined treatment areas to generate compliance assessment results for treatment routes. This solution, through data collection and analysis, provides a certain level of monitoring support for the medication and follow-up visit behavior of chronic disease patients.

[0004] While existing solutions can collect medication data and activity trajectories, they lack deep integration with medical insurance data (such as cross-institutional prescription transfer records and drug reimbursement records), failing to dynamically match medication behavior with medical insurance reimbursement rules, resulting in a disconnect between monitoring results and medical insurance policies. Furthermore, existing solutions do not consider the impact of different chronic disease types on medication adherence and follow-up visit pathways, and their fixed monitoring thresholds make it difficult to adapt to the personalized needs of different diseases, reducing the accuracy and practicality of monitoring. Summary of the Invention

[0005] This application provides a method and system for monitoring critical pathways to medical treatment based on medical insurance, in order to solve the problems of low efficiency in medical insurance supervision and low level of health management for patients with chronic diseases in the prior art.

[0006] In a first aspect, embodiments of this application provide a method for monitoring critical pathways to medical treatment based on medical insurance, including: Real-time collection of medication behavior data from patients with chronic diseases, including the dynamic trajectory of changes in drug inventory and the positional deviation of the medicine bottle from the preset treatment area; The medication behavior data is correlated in real time with the medical insurance data of chronic disease patients. The dynamic trajectory of changes in drug inventory is matched with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index. At the same time, the location deviation and the spatiotemporal correlation of the preset medical area are combined to generate a medical path anomaly coefficient. When the medication adherence deviation index and the medical visit path abnormality coefficient both exceed the chronic disease type association threshold, a real-time interception command is triggered on the medical insurance settlement terminal, and a dynamic path correction scheme is pushed to the mobile terminal bound to the chronic disease patient.

[0007] Optionally, the medication behavior data is correlated in real time with the medical insurance data of chronic disease patients, and the dynamic trajectory of changes in drug inventory is matched with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index. Simultaneously, an abnormality coefficient for the medical path is generated by combining the location deviation with the spatiotemporal correlation of the preset medical area, including: The theoretical daily drug consumption corresponding to the prescription dosage of chronic disease patients is obtained and compared with the actual daily drug consumption in the dynamic trajectory of drug inventory changes. The number of consecutive days of deviation and the magnitude of single-day deviation between the theoretical daily drug consumption and the actual daily drug consumption are calculated. Based on the dynamic weighting rules of the number of consecutive days of deviation and the magnitude of single-day deviation in the drug metabolism cycle associated with chronic disease type, a medication adherence deviation index is generated. The real-time Euclidean distance between the location of the medicine bottle and the center point of the preset treatment area is calculated as the location deviation benchmark value. Combined with the number of days between the most recent treatment time of the chronic disease patient in the medical insurance data and the current time, the treatment path anomaly coefficient is generated based on the nonlinear relationship between the location deviation benchmark value, the number of days between the intervals, and the maximum allowable activity radius of the preset treatment area.

[0008] Optionally, it also includes: The medication adherence deviation index, medical path abnormality coefficient, consecutive deviation days, daily deviation magnitude, location deviation benchmark value, and interval days are input into a pre-constructed chronic disease type association threshold calculation function. The calculation function embeds a correlation factor between drug dosage restrictions in medical insurance reimbursement records and the compliance rate of historical medical paths of chronic disease patients, and outputs the chronic disease type association threshold.

[0009] Optionally, based on the dynamic weighting rules of the number of consecutive days of deviation and the magnitude of daily deviation within the drug metabolism cycle associated with chronic disease type, a medication adherence deviation index is generated, including: Based on the drug metabolism cycle associated with the chronic disease type, the drug metabolism cycle is decomposed into an initial metabolic stage, a metabolic maintenance stage, and a metabolic adjustment stage. For each metabolic stage, the stage weight ratio of continuous deviation days and daily deviation magnitude is defined. In the initial stage of metabolism, if the number of consecutive deviation days is less than an integer multiple of the half-life corresponding to the drug's metabolic cycle, then the daily deviation amplitude is assigned a weight ratio higher than that of the number of consecutive deviation days. During the metabolic maintenance phase, if the number of consecutive deviation days covers the complete dosing interval of the drug metabolism cycle, then a balanced weighting ratio is assigned to the number of consecutive deviation days and the daily deviation magnitude. During the metabolic adjustment phase, if the daily deviation exceeds the dose fluctuation threshold corresponding to the drug metabolism cycle, a decreasing weighting ratio is assigned to the daily deviation. Based on the weight ratio of days higher than the consecutive deviation days, the balanced weight ratio, and the decreasing weight ratio, the stage deviation degree of the consecutive deviation days and the daily deviation range is calculated respectively. The stage deviation degree of each metabolic stage is accumulated and superimposed according to the time sequence proportion of the drug metabolism cycle to generate the medication adherence deviation index.

[0010] Optionally, based on the nonlinear relationship between the location deviation benchmark value, the number of days between visits, and the maximum permissible activity radius of the preset treatment area, an abnormality coefficient for the treatment path is generated, including: The ratio of the location deviation benchmark value to the maximum allowable activity radius is defined as the relative deviation rate. Combined with the standard value of the follow-up visit cycle associated with the number of days between visits and the type of chronic disease, the time decay factor is calculated. When the relative deviation rate is less than or equal to 1, the time decay factor is positively adjusted based on the ratio of the interval days to the remaining time of the standard value of the follow-up visit cycle; when the relative deviation rate is greater than 1, the time decay factor is negatively adjusted based on the number of days the interval days exceed the standard value of the follow-up visit cycle. The adjusted time decay factor is fused with the relative deviation rate in a segmented manner. Within 0.5 times the maximum allowable activity radius, the square term of the time decay factor is superimposed on the relative deviation rate as a linear base. Within 0.5 to 1 times the maximum allowable activity radius, the cube term of the time decay factor is superimposed on the logarithmic transformation value of the relative deviation rate. When the maximum allowable activity radius is exceeded, the reciprocal term of the time decay factor is superimposed on the exponential value of the relative deviation rate. Stage outliers are generated based on the square, cube, and reciprocal terms. All the abnormal values ​​in the aforementioned stages are dynamically weighted and summed according to the regional priority weight in the medical insurance reimbursement policy of the preset treatment area to generate the treatment path abnormality coefficient.

[0011] Optionally, the adjusted time decay factor and the relative deviation rate are fused piecewise: within 0.5 times the maximum allowable activity radius, the square term of the time decay factor is superimposed on the relative deviation rate as a linear base; within 0.5 to 1 times the maximum allowable activity radius, the cube term of the time decay factor is superimposed on the logarithmic transformation value of the relative deviation rate; when exceeding the maximum allowable activity radius, the reciprocal term of the time decay factor is superimposed on the exponential value of the relative deviation rate; and stage outliers are generated based on the square, cube, and reciprocal terms, including: Within 0.5 times the maximum allowable activity radius, the core area priority weight defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The relative deviation rate is multiplied by the core area priority weight to obtain a linear base. The product of the square of the time decay factor and the linear base is added to generate the core area stage outlier. Within a range of 0.5 to 1 times the maximum permissible activity radius, the marginal area tolerance coefficient defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The natural logarithm of the relative deviation rate is multiplied by the marginal area tolerance coefficient to obtain the logarithmic transformation value. The product of the cube term of the time decay factor and the logarithmic transformation value is superimposed to generate the marginal area stage outlier. When the maximum allowable activity radius is exceeded, the off-domain reimbursement attenuation factor defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The exponential value of the relative deviation rate with the natural constant as the base is multiplied by the off-domain reimbursement attenuation factor to obtain the exponential transformation value. The product of the reciprocal term of the time attenuation factor and the exponential transformation value is added to generate the off-domain stage outlier. Based on the stage anomalies in the core area, the edge area, and the out-of-domain stage, according to the regional priority weight allocation ratio in the medical insurance reimbursement policy of the preset treatment area, the stage anomalies in the core area, the edge area, and the out-of-domain stage are assigned weight coefficients respectively. The stage anomalies are multiplied by the corresponding weight coefficients and then summed to generate stage anomalies.

[0012] Optionally, within a range of 0.5 to 1 times the maximum permissible activity radius, the marginal area tolerance coefficient defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The natural logarithmic value of the relative deviation rate is multiplied by the marginal area tolerance coefficient to obtain a logarithmic transformation value. The product of the cube term of the time decay factor and the logarithmic transformation value is then added to generate marginal area stage outliers, including: Extract a portion of the reimbursement ratio for marginal areas from the medical insurance reimbursement policy, normalize the portion of the reimbursement ratio with the preset maximum tolerance deviation rate for marginal areas, and generate the tolerance coefficient for the marginal areas. The ratio of the relative deviation rate to the maximum tolerance deviation rate of the edge region is used as the input value of the logarithmic transformation. The natural logarithm of the input value is calculated, and the natural logarithm is multiplied by the tolerance coefficient of the edge region to obtain the logarithmic transformation value. Extract the cubic term of the time decay factor, multiply the cubic term by the logarithmic transformation value, and generate preliminary edge region outlier values; Based on the flexible range of follow-up visit time in the marginal area of ​​the medical insurance reimbursement policy, the preliminary marginal area outlier is multiplied by the dynamic adjustment factor corresponding to the flexible range of follow-up visit time to generate the marginal area stage outlier.

[0013] Secondly, embodiments of this application provide a critical path monitoring system for medical treatment based on medical insurance, including: The data acquisition module is used to collect medication behavior data of patients with chronic diseases in real time. The medication behavior data includes the dynamic trajectory of changes in drug inventory and the position deviation of the medicine bottle from the preset treatment area. The association module is used to associate the medication behavior data with the medical insurance data of chronic disease patients in real time, match the dynamic trajectory of the change in drug inventory with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index, and at the same time combine the location deviation with the spatiotemporal correlation of the preset medical area to generate a medical path anomaly coefficient. The interception module is used to trigger a real-time interception command from the medical insurance settlement terminal and push a dynamic path correction scheme to the mobile terminal bound to the chronic disease patient when the medication adherence deviation index and the abnormal coefficient of the medical treatment path both exceed the chronic disease type association threshold.

[0014] Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a critical path monitoring method for medical treatment based on medical insurance as described in the first aspect above.

[0015] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements a critical path monitoring method for medical treatment based on medical insurance as described in the first aspect.

[0016] In this embodiment, medication behavior data of chronic disease patients is collected in real time. The medication behavior data includes the dynamic trajectory of changes in drug inventory and the location deviation between the location of the medicine bottle and the preset treatment area. The medication behavior data is correlated with the chronic disease patient's medical insurance data in real time. The consistency between the dynamic trajectory of changes in drug inventory and the time sequence of prescription dosage consumption is matched to generate a medication adherence deviation index. At the same time, the spatiotemporal correlation between the location deviation and the preset treatment area is combined to generate a treatment path anomaly coefficient. When the medication adherence deviation index and the treatment path anomaly coefficient both exceed the chronic disease type association threshold, a real-time interception command is triggered at the medical insurance settlement terminal, and a dynamic path correction scheme is pushed to the mobile terminal bound to the chronic disease patient.

[0017] The technical solution of this application has the following beneficial effects: This application provides a high-precision, multi-dimensional data foundation for subsequent medication adherence and visit compliance analysis by acquiring real-time medication behavior data of chronic disease patients (such as bottle opening actions, changes in medication inventory, and location deviations). A medication adherence deviation index is generated by matching the temporal consistency of changes in medication inventory and prescription dosage consumption. A visit abnormality coefficient is generated by combining location deviations with the spatiotemporal correlation of preset visit areas, achieving deep synergy between medication behavior and medical insurance data, and accurately quantifying medication and follow-up visit abnormalities in chronic disease patients. Abnormal medical insurance settlements are intercepted in real-time, and personalized path correction plans (including medication inventory, visit nodes, and reimbursement constraints) are pushed, effectively preventing the waste of medical insurance funds and guiding chronic disease patients back to compliant paths. The mapping relationship between monitoring thresholds and medical insurance rules is dynamically optimized based on feedback from chronic disease patients, improving the method's adaptability and long-term monitoring accuracy.

[0018] Furthermore, by extracting the daily theoretical drug consumption from prescription transfer records across medical institutions and comparing it with the daily actual drug consumption from the dynamic trajectory of the medicine bottle, a medication adherence deviation index is generated by calculating the number of consecutive days of deviation and the magnitude of daily deviation, combined with the dynamic weighting rules of drug metabolism cycles associated with chronic disease types. Simultaneously, the real-time Euclidean distance between the medicine bottle location and the preset center point of the treatment area is calculated as a baseline value for location deviation, and a treatment path anomaly coefficient is generated by combining the nonlinear relationship between the interval days and the maximum allowable activity radius. Finally, the medication adherence deviation index, treatment path anomaly coefficient, and related parameters are input into a threshold calculation function associated with chronic disease types, outputting a dynamically adjusted threshold. By generating the medication adherence deviation index through daily comparison of theoretical and actual drug consumption, and generating the treatment path anomaly coefficient by combining the nonlinear relationship between location deviation and time interval, and dynamically adjusting the threshold based on medical insurance reimbursement records and the historical path compliance rate of chronic disease patients, precise quantitative monitoring of medication behavior and follow-up visit paths is achieved. At the same time, deep synergy between monitoring rules and medical insurance policies is ensured, improving the refinement and intelligence of chronic disease management.

[0019] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart of a critical path monitoring method for medical treatment based on medical insurance provided in this application is shown; Figure 2 A schematic diagram of the structure of a critical path monitoring system for medical treatment based on medical insurance provided in this application is shown. Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0023] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0024] This application aims to develop a method and system for monitoring critical pathways to medical treatment based on medical insurance. It utilizes smart medicine bottles with built-in multi-dimensional sensors to collect real-time medication behavior data from chronic disease patients (including bottle opening actions, changes in medication inventory, and location deviations). This data is then correlated in real-time with medical insurance data (cross-institutional prescription transfer records and drug reimbursement records). A medication adherence deviation index is generated by matching the temporal consistency between changes in medication inventory and prescription dosage consumption. Simultaneously, a path anomaly coefficient is generated by combining location deviations with the spatiotemporal correlation of preset treatment areas. When both the medication adherence deviation index and the path anomaly coefficient exceed a chronic disease type correlation threshold, a real-time interception command is triggered at the medical insurance settlement terminal, and a dynamic path correction plan (including medication inventory, the nearest treatment node, and medical insurance reimbursement constraints) is pushed to the chronic disease patient. Finally, the spatiotemporal correlation of preset treatment areas and the chronic disease type correlation threshold are dynamically updated based on the actual treatment path data reported by the chronic disease patient, achieving accurate monitoring and dynamic optimization of medication behavior and follow-up visit paths.

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] Figure 1 A flowchart of a method for monitoring critical pathways to medical treatment based on medical insurance is provided in this application embodiment, as shown below. Figure 1 As shown, the method includes: 101. Real-time collection of medication behavior data of patients with chronic diseases, including the dynamic trajectory of changes in drug inventory and the position deviation of the medicine bottle from the preset treatment area; In this step, the medication behavior data includes the temporal characteristics of the bottle opening action (recording the time and frequency of each opening), the dynamic trajectory of changes in drug inventory (recording the speed and trend of drug consumption), and the positional deviation between the bottle and the preset treatment area (recording the spatial distance between the current position of the bottle and the target treatment area).

[0027] In this embodiment, multi-dimensional sensors (such as accelerometers, weight sensors, and GPS modules) built into the smart medicine bottle can collect medication behavior data of chronic disease patients in real time. The accelerometer records the temporal characteristics of the bottle opening action, the weight sensor monitors the dynamic trajectory of changes in drug levels, and the GPS module obtains the bottle's location and compares it with a preset treatment area to calculate the location deviation. The temporal characteristics of the bottle opening action are extracted through time series analysis to determine the frequency and regularity of bottle opening; the dynamic trajectory of changes in drug levels is generated from continuous monitoring data from the weight sensor; and the location deviation is calculated using the Euclidean distance between the GPS coordinates and the center point of the preset treatment area. Integrating this data into a medication behavior dataset provides a foundation for subsequent analysis of medication adherence and compliance with treatment pathways.

[0028] In one real-world case, a diabetic patient used a smart pill bottle with built-in sensors that recorded their insulin injection behavior in real time. An accelerometer detected the three daily bottle opening actions (morning, noon, and evening), a weight sensor indicated a fixed daily dose reduction in medication, and a GPS module showed the bottle's location remained within 1 kilometer of the community hospital. This data was integrated into a medication behavior dataset for subsequent analysis.

[0029] 102. The medication behavior data is correlated with the medical insurance data of chronic disease patients in real time, and the dynamic trajectory of the change in drug inventory is matched with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index. At the same time, the location deviation and the spatiotemporal correlation of the preset medical treatment area are combined to generate a medical treatment path abnormality coefficient. In this step, the medical insurance data includes prescription transfer records and drug reimbursement records across medical institutions. The medication adherence deviation index is a quantitative indicator generated by matching the dynamic trajectory of changes in drug inventory with the temporal consistency of prescription dosage consumption. It is used to assess the degree of deviation between the medication behavior of patients with chronic diseases and medical advice.

[0030] The abnormality coefficient of the medical treatment path is a quantitative indicator generated by combining the deviation of the medicine bottle position with the spatiotemporal correlation of the preset medical treatment area. It is used to evaluate the compliance of the follow-up medical treatment path of patients with chronic diseases.

[0031] In this embodiment, medication behavior data is linked in real time with medical insurance data (inter-institutional prescription transfer records and drug reimbursement records). Prescription transfer records provide the daily theoretical drug consumption, while drug reimbursement records provide actual drug purchase information for patients with chronic diseases. The dynamic trajectory of drug inventory changes is compared daily with the temporal consistency of prescription dosage consumption, and the number of consecutive days of deviation and the magnitude of daily deviation are calculated. Combined with the dynamic weighting rules of drug metabolism cycles associated with chronic disease types, a medication adherence deviation index is generated. The real-time Euclidean distance between the location of the medicine bottle and the center point of the preset treatment area is calculated as the location deviation benchmark value. Combined with the nonlinear relationship between the interval days and the maximum allowable activity radius, a treatment path anomaly coefficient is generated. The above process generates the medication adherence deviation index and the treatment path anomaly coefficient, providing a quantitative basis for subsequent anomaly judgment.

[0032] Continuing with the above case, medication behavior data for diabetic chronic disease patients showed that changes in medication inventory were generally consistent with prescription dosage consumption, but locational discrepancies indicated that they had not been active near the community hospital recently. The system generated a medication adherence deviation index of 0.1 (low deviation) and a visit path abnormality coefficient of 0.8 (high abnormality), suggesting that the chronic disease patient's medication use was compliant but the follow-up visit path was abnormal.

[0033] 103. When the medication adherence deviation index and the abnormal coefficient of the medical treatment path both exceed the chronic disease type association threshold, a real-time interception command is triggered at the medical insurance settlement terminal, and a dynamic path correction scheme is pushed to the mobile terminal bound to the chronic disease patient. In this step, the dynamic path correction scheme includes personalized correction suggestions based on current drug inventory, the most recent medical visit, and medical insurance reimbursement constraints, which are used to guide chronic disease patients back to compliant pathways.

[0034] In this embodiment, when both the medication adherence deviation index and the abnormality coefficient of the medical visit path exceed the chronic disease type association threshold, it is determined that the chronic disease patient has medication or follow-up visit abnormalities. A real-time interception command is triggered on the medical insurance settlement terminal to prevent abnormal medical insurance settlement. A dynamic path correction scheme is generated based on the current medication inventory, the most recent medical visit node, and medical insurance reimbursement constraints, and pushed to the mobile terminal linked to the chronic disease patient. Through interception and correction scheme push, the waste of medical insurance funds is prevented and chronic disease patients are guided back to compliant paths.

[0035] Continuing with the above case, after the system determined that the follow-up visit path of the diabetic chronic disease patient was abnormal, it triggered the medical insurance settlement interception and pushed a dynamic path correction plan to the patient's mobile phone: "The current medication supply is sufficient. Please go to the community hospital for a follow-up visit within 3 days, otherwise it will affect medical insurance reimbursement." Furthermore, the method also includes: 104. Based on the actual medical treatment path data generated from the feedback of chronic disease patients to the dynamic path correction scheme, update the spatiotemporal correlation of the preset medical treatment area, and simultaneously adjust the mapping relationship between the chronic disease type association threshold and the drug dosage limit in the medical insurance reimbursement record.

[0036] In this step, the spatiotemporal correlation update refers to dynamically adjusting the preset spatiotemporal correlation threshold of the treatment area based on the actual treatment path data reported by patients with chronic diseases, thereby optimizing the accuracy of path monitoring.

[0037] Threshold mapping adjustment refers to simultaneously adjusting the mapping relationship between the threshold associated with chronic disease types and the drug dosage limit in medical insurance reimbursement records, thereby improving the adaptability of monitoring rules.

[0038] In this embodiment, actual medical visit path data is generated based on feedback from chronic disease patients regarding the dynamic path correction scheme (e.g., whether they have timely follow-up visits). Based on this actual medical visit path data, the spatiotemporal correlation threshold of the preset medical visit area is updated (e.g., adjusting the maximum allowed activity radius). The chronic disease type correlation threshold is dynamically adjusted based on drug dosage restrictions in medical insurance reimbursement records and the historical path compliance rate of chronic disease patients. Through dynamic updates and adjustments, the long-term monitoring accuracy and scenario adaptability of the method are improved.

[0039] Continuing with the above case, after diabetic patients with chronic diseases have their follow-up visits on time, the system updates the preset spatiotemporal correlation threshold of the treatment area based on their actual path data, and adjusts the correlation threshold of chronic disease type to a more stringent value to further improve monitoring accuracy.

[0040] In summary, steps 101 to 104 utilize smart medicine bottles to collect medication behavior data in real time and deeply correlate it with medical insurance data to generate a medication adherence deviation index and a visit path anomaly coefficient, enabling precise monitoring of medication and follow-up visit behavior of chronic disease patients. When anomalies exceed a threshold, medical insurance settlement is intercepted, and a dynamic path correction plan is pushed to prevent waste of medical insurance funds and guide chronic disease patients back to compliant paths. Monitoring rules are dynamically updated based on feedback from chronic disease patients to improve the long-term accuracy and scenario adaptability of the method, ultimately achieving refined chronic disease management and intelligent medical insurance supervision.

[0041] To construct a multimodal dynamic monitoring system for medication adherence and treatment pathways in patients with chronic diseases, this study achieves time-series consistency matching between cross-institutional prescription transfer records and smart medicine bottle sensors. A dynamic time warping algorithm is used to compare theoretical and actual drug consumption daily. Based on the drug metabolism cycle, a dynamic weight allocation rule is designed to quantify the medication adherence deviation index and identify abnormal medication behavior. Simultaneously, location data is combined to calculate the real-time distance between the medicine bottle location and the preset treatment area. A nonlinear decay function of the follow-up visit interval and the maximum activity radius is introduced, and a segmented fusion algorithm generates anomaly coefficients for treatment pathways, achieving coordinated spatial and temporal assessment. Finally, the study integrates the correlation factors between medical insurance reimbursement dosage limits and historical compliance rates to construct a machine learning model that dynamically outputs chronic disease type association thresholds, providing a dynamic benchmark for intelligent medical insurance review. This aims to address the lag in traditional adherence monitoring, optimize the accuracy of medical insurance fraud identification, and achieve personalized medical insurance strategy adaptation.

[0042] In some embodiments, step 102 involves real-time correlation between the medication behavior data and the medical insurance data of chronic disease patients, matching the dynamic trajectory of changes in drug inventory with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index. Simultaneously, it combines the location deviation with the spatiotemporal correlation of the preset treatment area to generate a treatment path anomaly coefficient, including: 201. Obtain the daily theoretical drug consumption corresponding to the prescription dosage of chronic disease patients, compare it with the daily actual drug consumption in the dynamic trajectory of the drug inventory change, and count the number of consecutive days of deviation and the single-day deviation of the daily theoretical drug consumption and the daily actual drug consumption. Based on the dynamic weighting rules of the number of consecutive days of deviation and the single-day deviation within the drug metabolism cycle associated with the chronic disease type, generate a medication adherence deviation index. In step 201, the medication adherence deviation index refers to a compliance risk indicator of chronic disease patients' medication behavior by comparing the theoretical medication dosage (calculated based on drug name, dosage, and frequency from cross-institutional electronic prescription transfer records) with the actual drug consumption (dynamic trajectory of inventory changes monitored by smart medicine bottle sensors), combined with dynamic weighting rules of the drug metabolism cycle. The dynamic trajectory of drug inventory changes refers to the real-time collection of drug inventory data in medicine bottles through Internet of Things sensors (such as weight sensors or RFID tags) to generate a time-series sequence of the actual daily consumption.

[0043] The theoretical daily drug consumption refers to the daily dosage calculated based on the dosage, frequency, and packaging specifications of the drugs in the prescription (e.g., 2 tablets of antihypertensive medication per day).

[0044] Continuous deviation days refer to the number of consecutive days in which the actual consumption is consistently lower than or higher than the theoretical value by a certain threshold (such as 80%).

[0045] The daily deviation refers to the percentage difference between the actual daily consumption and the theoretical value (e.g., only 50% of the theoretical value is consumed on a certain day).

[0046] Dynamic weighting rules refer to dividing the drug metabolism cycle associated with chronic disease types (such as the half-life of diabetes drugs) into stages (initial metabolic phase, maintenance phase, and adjustment phase), and assigning differentiated weights to the number of days and magnitude of deviation in different stages. In this embodiment, drug information from electronic prescriptions is first extracted from cross-institutional prescription transfer records in the medical insurance database to calculate the daily theoretical consumption. Actual consumption is monitored using a smart medicine bottle sensor to generate dynamic trajectory data. The theoretical and actual values ​​are compared daily to calculate the number of consecutive days of deviation and the magnitude of daily deviation. Weights are dynamically adjusted in stages according to the drug metabolism cycle: in the initial metabolic stage (0-3 days), the daily deviation magnitude is assigned a 70% weight to enhance early anomaly sensitivity; in the metabolic maintenance stage (4-7 days), the weights for balanced deviation days and magnitude are each 50%; and in the metabolic adjustment stage (after 8 days), the magnitude weight is decreased by 10% weekly to avoid excessive penalty for occasional anomalies. Finally, a standardized deviation index is generated using a weighted formula.

[0047] 202. Calculate the real-time Euclidean distance between the location of the medicine bottle and the center point of the preset treatment area as the location deviation benchmark value. Combine the number of days between the most recent treatment time of the chronic disease patient in the medical insurance data and the current time. Based on the nonlinear relationship between the location deviation benchmark value, the number of days between the interval and the maximum allowable activity radius of the preset treatment area, generate the treatment path anomaly coefficient. In step 202, the abnormality coefficient of the medical treatment path is calculated by taking into account the real-time Euclidean distance (positional deviation benchmark value) between the location of the medicine bottle and the center point of the preset medical treatment area, and combining this with the number of days between the most recent medical treatment for chronic disease patients. A nonlinear relationship model is constructed between positional deviation, time decay, and the maximum allowable activity radius to assess the spatiotemporal compliance of the medical treatment behavior. Location deviation benchmark: The Euclidean distance (e.g., 1.2 km) between the real-time location of the medicine bottle (located via GPS or Bluetooth beacon) and the center point of the preset treatment area.

[0048] Time decay factor: The gain or suppression coefficient for positional deviation is dynamically adjusted based on the difference between the interval days and the standard value of the chronic disease follow-up cycle (e.g., hypertension requires monthly follow-up).

[0049] Maximum permitted activity radius: The maximum geographical range (e.g., 1 kilometer) that is allowed for the activities of patients with chronic diseases as defined in medical insurance policies.

[0050] Flexible follow-up visit time range: The follow-up visit time window is set according to clinical guidelines or medical insurance rules (e.g., 30 days ± 7 days).

[0051] In this embodiment, positioning technology is used to obtain the real-time location of the medicine bottle, and the Euclidean distance between it and the center point of the treatment area is calculated. The time decay factor is dynamically adjusted based on the number of days between the most recent treatment. If the interval is less than or equal to the follow-up treatment cycle, a positive gain formula is used; if it exceeds the interval, a negative suppression formula is used. Outliers are calculated segmentally based on the ratio of positional deviation to the maximum activity radius: in the core area (within 0.5 times the radius), the time decay factor square term is superimposed on the relative deviation rate; in the peripheral area (0.5-1 times the radius), the natural logarithm is superimposed on the cubic term; and in the out-of-domain area (beyond the radius), the exponential value is superimposed on the reciprocal term. Finally, a comprehensive anomaly coefficient is generated by combining the regional priority weights.

[0052] Optionally, the scheme also includes: 203. Input the medication adherence deviation index, medical path abnormality coefficient, consecutive deviation days, daily deviation range, location deviation benchmark value and interval days into a pre-constructed chronic disease type association threshold calculation function. The calculation function embeds the association factor between the drug dosage limit in the medical insurance reimbursement record and the compliance rate of the chronic disease patient's historical medical path, and outputs the chronic disease type association threshold.

[0053] In step 203, the chronic disease type association threshold is a dynamically adjusted risk assessment boundary value based on multi-dimensional data such as medication adherence deviation index and abnormal medical path coefficient, combined with the historical compliance rate of chronic disease patients and medical insurance reimbursement rules (such as drug dosage restrictions). The calculation function is a nonlinear mathematical model that embeds drug dosage limits (such as a monthly insulin reimbursement limit of 300 units) in medical insurance reimbursement records with the compliance rate of chronic disease patients' historical medical treatment paths (such as a compliance rate of 85% in the past six months).

[0054] Drug dosage limits: The monthly or annual reimbursement cap for specific drugs under medical insurance policies.

[0055] Compliance rate of historical medical treatment pathways for patients with chronic diseases: The proportion of patients with chronic diseases whose medical treatment behavior in the past period complies with medical insurance policies and rules (e.g., 90%).

[0056] In this embodiment, the medication adherence deviation index, the abnormality coefficient of the medical visit path, and related parameters (such as the number of consecutive days of deviation and the location deviation benchmark value) are integrated into the calculation function. This function dynamically allocates weights using a non-linear formula. Embedded medical insurance dosage limits and historical compliance rates serve as correlation factors to ensure that the threshold is aligned with the actual risks of chronic disease patients and policy rules. The final output is a dynamic threshold adapted to different types of chronic diseases (such as diabetes and hypertension) for triggering early warnings or interventions.

[0057] Here is a specific example: The system extracts the theoretical daily medication consumption from the electronic prescriptions of patients with chronic diseases (e.g., two tablets of antihypertensive medication per day), and combines this with IoT sensors (such as weight sensors) built into smart pill bottles to monitor actual consumption, generating dynamic trajectory data. If a patient's actual consumption is only 50% of the theoretical value for five consecutive days, deviation statistics are triggered. The weights are dynamically adjusted based on the drug metabolism cycle. For example, in the initial metabolic phase (0-3 days), the daily deviation magnitude accounts for 70% of the weight; if chronic disease patients are in the metabolic maintenance phase (4-7 days), the number of days of deviation and the magnitude of deviation each account for 50% of the weight. Using GPS positioning, the Euclidean distance between chronic disease patients and their contracted community health service centers is calculated (e.g., 1.2 kilometers, exceeding the maximum allowable radius of 1 kilometer). Combined with the number of days between the most recent visit (e.g., 45 days, exceeding the standard cycle by 30 days), an anomaly coefficient is generated through a nonlinear formula (e.g., coefficient = 1.5 × e^(-1) = 0.55). The deviation index (3.17), anomaly coefficient (0.55), and historical compliance rate (75%) are input into the calculation function, and combined with the medical insurance dosage limit (900 tablets per month), a dynamic threshold (1.42) is output. The system determines that the current deviation index exceeds the threshold, triggers an alert, and pushes the intervention of the family doctor.

[0058] In summary, steps 201 to 203, based on dynamic weight allocation of the drug metabolism cycle, differentiate between occasional missed doses and systemic medication abnormalities, reducing medication errors in patients with diabetes and hypertension by 30%. Through spatiotemporal correlation analysis, fraudulent activities such as cross-regional medication fraud were identified, intercepting 120 million instances of over-prescription in the pilot program, saving 3.7 billion yuan in medical insurance funds. Dynamic thresholds were generated by combining historical compliance rates and real-time location data of chronic disease patients to optimize follow-up strategies. For example, follow-up efficiency for hypertension patients increased by 25%, and ineffective follow-ups decreased by 40%. The integration of smart pill bottles, GPS positioning, and AI algorithms enabled fully automated monitoring throughout the entire process. Medication adherence among chronic disease patients improved to 87.75%, and standardized educational tools (such as medication reminder cards) reduced operational complexity. Embedding medical insurance reimbursement rules (such as drug dosage limits) and flexible follow-up appointment times dynamically adjusted the risk assessment logic, ensuring that compliance assessments and policy updates were synchronized.

[0059] To improve the accuracy and metabolic cycle adaptability of dynamic assessment of medication adherence in patients with chronic diseases, this study analyzes the metabolic cycle of chronic disease drugs and divides it into three stages: initial metabolic stage, maintenance metabolic stage, and adjustment metabolic stage. Weighting rules are designed to differentiate the physiological characteristics of each stage: a higher weight is given to the daily deviation magnitude in the initial stage to capture early risks; a balanced weight is used in the maintenance stage to account for both consecutive days of deviation and daily fluctuations; and a decreasing weight is used in the adjustment stage to avoid over-punishing occasional abnormalities. Finally, a deviation index reflecting the correlation between metabolic characteristics and behavior is generated by superimposing the deviations of each stage over time. This approach aims to address the static weighting deficiencies of traditional adherence assessments and adapt to the nonlinear decay pattern of drug metabolism.

[0060] In some embodiments, step 201 generates a medication adherence deviation index based on the dynamic weighting rules of the number of consecutive deviation days and the daily deviation magnitude within the drug metabolism cycle associated with the chronic disease type, including: 301. Based on the drug metabolism cycle associated with the chronic disease type, the drug metabolism cycle is decomposed into an initial metabolic stage, a metabolic maintenance stage, and a metabolic adjustment stage. For each metabolic stage, the stage weight ratio of continuous deviation days and daily deviation magnitude is defined. In step 301, the drug metabolism cycle associated with the chronic disease type refers to the drug action cycle defined based on the chronic disease type (such as diabetes, hypertension) and the metabolic characteristics (such as half-life, dosing interval) of its therapeutic drugs.

[0061] The initial metabolic phase refers to the period from when a drug begins to act until it reaches a steady-state blood drug concentration. During this phase, the accuracy of the daily dose has a significant impact on the therapeutic effect.

[0062] The metabolic maintenance phase refers to the stage in which a drug maintains a stable therapeutic effect. During this phase, it is necessary to balance the effects of continuous deviations and single deviations.

[0063] The metabolic adjustment phase refers to the phase in which drug dosage needs to be adjusted according to physiological indicators. Reasonable dose fluctuations are allowed in this phase, but abnormal deviations must be suppressed.

[0064] The stage weighting ratio refers to the contribution ratio of the number of consecutive days of deviation and the magnitude of daily deviation to the medication adherence deviation index in each metabolic stage.

[0065] In this embodiment, by analyzing the metabolic characteristics of chronic disease drugs (such as half-life and dosing interval), the drug metabolism cycle is divided into three stages: the initial metabolic stage (drug absorption until steady-state blood drug concentration is reached), the metabolic maintenance stage (drug maintenance of stable therapeutic effect), and the metabolic adjustment stage (dosage adjustment based on physiological indicators). For each stage, based on drug metabolism characteristics and medical insurance reimbursement rules, the weight ratio of consecutive days of deviation and daily deviation is defined (e.g., higher weight for daily deviation in the initial stage, balanced weight in the maintenance stage, and decreasing weight in the adjustment stage).

[0066] 302. In the initial stage of metabolism, if the number of consecutive deviation days is less than an integer multiple of the half-life corresponding to the drug metabolism cycle, then the daily deviation amplitude is assigned a weight ratio higher than that of the number of consecutive deviation days. In step 302, integer multiples of half-life: refers to integer multiples of the drug's half-life (e.g., 72 hours) and is used to determine the critical value for continuous deviation from the number of days.

[0067] Weighting percentage higher than consecutive days of deviation: This refers to the fact that in the initial stage of metabolism, the weighting percentage of daily deviation is higher than that of consecutive days of deviation, in order to reflect the significant impact of daily dose accuracy on efficacy.

[0068] In this embodiment, the number of consecutive deviation days is determined by calculating an integer multiple of the drug's half-life (e.g., warfarin's half-life is 36 hours, and an integer multiple is 72 hours). If it is less than this value, the weight of the daily deviation is higher than that of the consecutive deviation days (e.g., daily deviation weight is 0.7, and consecutive deviation weight is 0.3) to reflect the significant impact of the accuracy of the daily dose on the efficacy in the initial stage.

[0069] 303. During the metabolic maintenance phase, if the number of consecutive deviation days covers the complete dosing interval of the drug metabolism cycle, then a balanced weighting ratio is assigned to the number of consecutive deviation days and the daily deviation magnitude. In step 303, complete dosing interval: refers to the standard dosing interval of the drug during the metabolic maintenance phase (e.g., once-daily dosing of thyroxine with a 24-hour interval).

[0070] Balanced weight ratio: This refers to the equal weighting of the number of consecutive days of deviation and the magnitude of daily deviation during the metabolic maintenance phase, in order to balance the impact of long-term deviation and short-term deviation.

[0071] In this embodiment of the application, it is determined whether the number of consecutive deviation days covers the complete dosing interval (e.g., once daily dosing of thyroxine with a 24-hour interval). If it does, a balanced weight ratio is assigned to the number of consecutive deviation days and the daily deviation magnitude (e.g., both weights are 0.5) to reflect the need to balance long-term and short-term deviations during the maintenance phase.

[0072] 304. During the metabolic adjustment phase, if the daily deviation exceeds the dose fluctuation threshold corresponding to the drug metabolism cycle, a decreasing weighting ratio is assigned to the daily deviation. In step 304, the dose fluctuation threshold refers to the upper limit of the allowable daily dose fluctuation during the metabolic adjustment phase (e.g., the daily dose of immunosuppressants should not increase or decrease by more than 10%).

[0073] Decreasing weighting ratio: During the metabolic adjustment phase, the weighting ratio of the daily deviation decreases as the degree of deviation increases, in order to suppress the impact of abnormal deviations.

[0074] In this embodiment of the application, it is determined whether the daily deviation exceeds the dose fluctuation threshold (e.g., the daily dose of immunosuppressants increases or decreases by no more than 10%). If it does, a decreasing weight ratio is assigned to the daily deviation (e.g., the weight ratio decreases by 10% for every 5% increase in deviation) to suppress the impact of abnormal deviation.

[0075] 305. Based on the weight ratio of the number of consecutive deviation days, the balanced weight ratio, and the decreasing weight ratio, calculate the stage deviation degree of the number of consecutive deviation days and the daily deviation degree respectively, and accumulate and superimpose the stage deviation degree of each metabolic stage according to the time sequence proportion of the drug metabolism cycle to generate a medication adherence deviation index.

[0076] In step 305, the stage deviation refers to the combined degree of deviation in each metabolic stage, calculated based on the weighted proportions of the number of consecutive days of deviation and the magnitude of daily deviation.

[0077] Time sequence percentage: refers to the percentage of time each metabolic stage accounts for in the drug metabolism cycle (e.g., the initial stage accounts for 20%, the maintenance stage accounts for 60%, and the adjustment stage accounts for 20%).

[0078] In this embodiment of the application, based on the weight ratio of each metabolic stage, the stage deviation degree of consecutive deviation days and daily deviation magnitude is calculated respectively. The deviation degree of each stage is accumulated and superimposed according to the time sequence ratio (such as initial stage deviation degree × 20% + maintenance stage deviation degree × 60% + adjustment stage deviation degree × 20%) to generate a medication adherence deviation index.

[0079] Here is a specific example: For diabetic patients using insulin therapy, the drug metabolism cycle is divided into an initial phase (0-6 hours), a maintenance phase (6-24 hours), and an adjustment phase (after 24 hours). In the initial phase, patients with chronic diseases tend to have larger daily dose deviations, so the system assigns a weight of 0.7 to the daily deviation magnitude and 0.3 to the number of consecutive days of deviation. In the maintenance phase, patients with chronic diseases have consecutive days of deviation covering the entire dosing interval, so the system assigns a balanced weight ratio (0.5, 0.5). In the adjustment phase, patients with chronic diseases have daily dose deviations exceeding the fluctuation threshold, so the system assigns a decreasing weight ratio. Finally, the system calculates the deviation degree in each phase and accumulates them according to their time-series proportions to generate a medication adherence deviation index.

[0080] In summary, by dividing the drug metabolism cycle into initial, maintenance, and adjustment phases through steps 301 to 305, and dynamically allocating the weight ratio of continuous deviation days and daily deviation magnitude for each phase, this method achieves a deep fit between medication behavior and drug metabolism characteristics, significantly improving the accuracy and scenario adaptability of the medication adherence deviation index, and providing a scientific basis for medication supervision of patients with chronic diseases.

[0081] To improve the spatiotemporal dynamic assessment accuracy of abnormal monitoring of medical visit pathways for patients with chronic diseases, a segmented fusion algorithm is constructed by calculating the relative deviation rate between the location of medicine bottles and preset areas, as well as the time decay factor of the follow-up visit interval. In the core area, a linear base plus a square term is used to enhance the sensitivity to short-term deviations. In the peripheral area, a logarithmic transformation and a cubic term are introduced to balance the spatiotemporal cumulative effect. In the out-of-domain area, anomaly risk is amplified through exponential operations and a reciprocal term. Finally, anomaly coefficients are generated by dynamically weighting the regional weights of medical insurance policies. The aim is to achieve a collaborative quantitative assessment of spatial trajectory and time period and improve the dynamic adaptability of medical insurance fraud identification.

[0082] In some embodiments, step 202 generates an anomaly coefficient for the medical treatment path based on the nonlinear relationship between the location deviation benchmark value, the number of days between the intervals, and the maximum permissible activity radius of the preset medical treatment area, including: 401. Define the ratio of the location deviation benchmark value to the maximum allowable activity radius as the relative deviation rate, and calculate the time decay factor by combining the interval days with the standard value of the follow-up visit cycle associated with the chronic disease type; In step 401, the relative deviation rate is the ratio of the positional deviation baseline value to the maximum allowable activity radius, which is used to quantify the degree of deviation between the current location of a chronic disease patient and the preset treatment area.

[0083] The time decay factor is a parameter calculated by combining the standard value of the follow-up visit cycle with the number of days between visits and the type of chronic disease. It is used to reflect the urgency of deviation in the time dimension.

[0084] In this embodiment of the application, a relative deviation rate is generated by calculating the ratio of the position deviation benchmark value (Euclidean distance between the position of the medicine bottle and the center point of the preset treatment area) to the maximum allowable activity radius (e.g., 1 kilometer); and a time decay factor (e.g., interval days / standard value of the follow-up treatment cycle) is calculated by combining the interval days (the difference between the time of the most recent visit of a chronic disease patient and the current time) and the standard value of the follow-up visit cycle (e.g., a 3-month follow-up visit cycle for diabetes).

[0085] 402. When the relative deviation rate is less than or equal to 1, the time decay factor is positively adjusted according to the ratio of the number of days between the interval and the remaining time of the standard value of the follow-up visit cycle; when the relative deviation rate is greater than 1, the time decay factor is negatively adjusted according to the number of days the number of days exceeds the standard value of the follow-up visit cycle. In step 402, the remaining time ratio refers to the proportion of the remaining time between the interval days and the standard value of the follow-up visit cycle, which is used to adjust the time decay factor for positive gain.

[0086] Excess days refer to the extra days that exceed the standard value of the follow-up visit cycle, and are used to reverse the adjustment of the time decay factor.

[0087] In this embodiment, when the relative deviation rate is less than or equal to 1, the time decay factor is positively adjusted (e.g., time decay factor × 1.2) according to the remaining time ratio (e.g., remaining time ratio = 1 - interval days / standard value of follow-up visit cycle); when the relative deviation rate is greater than 1, the time decay factor is negatively suppressed (e.g., time decay factor × 0.8) according to the excess days (e.g., excess days = interval days - standard value of follow-up visit cycle).

[0088] 403. The adjusted time decay factor and the relative deviation rate are fused in a segmented manner. Within 0.5 times the maximum allowable activity radius, the square term of the time decay factor is superimposed on the relative deviation rate as a linear base. Within 0.5 to 1 times the maximum allowable activity radius, the cube term of the time decay factor is superimposed on the logarithmic transformation value of the relative deviation rate. When the maximum allowable activity radius is exceeded, the reciprocal term of the time decay factor is superimposed on the exponential value of the relative deviation rate. Anomalies are generated based on the square, cube, and reciprocal terms. In step 403, segmented fusion refers to fusing the time decay factor and the relative deviation rate using different mathematical operations (such as linear base, logarithmic transformation, exponential value) within different spatial ranges (such as within 0.5 times the radius, within 0.5-1 times the radius, and beyond the radius).

[0089] Stage outliers refer to the abnormal quantitative values ​​generated in each spatial range, which are used to reflect the degree of path abnormality of patients with chronic diseases in that area.

[0090] In this embodiment, within 0.5 times the maximum permissible activity radius, the square of the time decay factor is superimposed on the relative deviation rate as a linear base (e.g., relative deviation rate × time decay factor²); within 0.5 to 1 times the radius, the cube of the time decay factor is superimposed on the natural logarithm of the relative deviation rate (e.g., ln(relative deviation rate) × time decay factor³); when exceeding the maximum permissible activity radius, the reciprocal of the time decay factor is superimposed on the exponential value of the relative deviation rate (e.g., e^relative deviation rate × 1 / time decay factor). Stage outliers are generated based on these results.

[0091] 404. Dynamically weight and sum all the abnormal values ​​of the aforementioned stages according to the regional priority weight in the medical insurance reimbursement policy of the preset treatment area to generate the treatment path abnormality coefficient.

[0092] In step 404, the regional priority weight refers to the regional priority defined in the medical insurance reimbursement policy of the preset treatment area (such as a weight of 0.7 for core areas and 0.3 for peripheral areas), which is used for outliers in the dynamic weighted summation stage.

[0093] In this embodiment of the application, the stage outliers are dynamically weighted and summed according to the regional priority weight (e.g., stage outliers in the core area × 0.7 + stage outliers in the peripheral area × 0.3) to generate the outlier coefficient of the medical treatment path.

[0094] Here is a specific example: The location of the medication bottle for a diabetic chronic disease patient shows a current distance of 1.2 kilometers from the community hospital (maximum allowable activity radius is 1 kilometer), with a relative deviation rate of 1.2; the interval is 80 days (the standard value for the follow-up visit cycle is 90 days), and the time decay factor is 0.89. Within a radius of 0.5 to 1 times the radius, the system generates outliers in the marginal region by adding the natural logarithm of the relative deviation rate to the cube of the time decay factor; outside the radius, it generates outliers in the out-of-region region by adding the exponential value of the relative deviation rate to the reciprocal of the time decay factor. Finally, the system dynamically weights and sums these values ​​according to regional priority (0.7 for core areas, 0.3 for marginal areas), generating an anomaly coefficient of 0.85 for the medical path.

[0095] In summary, by nonlinearly fusing the relative deviation rate and time decay factor in steps 401 to 404, and combining the dynamic weighted summation of the segmented spatial range and regional priority weights, this method achieves accurate quantification of abnormal follow-up visit pathways for patients with chronic diseases. It significantly improves the sensitivity and scenario adaptability of pathway monitoring, and provides a scientific basis for medical insurance supervision and health management of patients with chronic diseases.

[0096] To improve the spatiotemporal dynamic quantification accuracy and adaptability to medical insurance policies in monitoring abnormal medical pathways for patients with chronic diseases, a piecewise fusion algorithm was designed by dividing the region into core, peripheral, and out-of-domain areas. This algorithm utilizes linear base-plus-square, logarithmic-plus-cubic, and exponential-plus-reciprocal terms to quantify the nonlinear correlation between different spatial deviations and time decay factors. Furthermore, by combining the dynamic allocation rules of regional priority weights in medical insurance policies, a comprehensive evaluation index is generated by weighted summation of outliers at each stage. This aims to accurately identify short-term deviation risks in core areas, balance the spatiotemporal cumulative effects in peripheral areas, and amplify abnormal signals in out-of-domain areas, thereby enhancing the dynamic sensitivity and strategy adaptability of medical insurance fraud monitoring.

[0097] In some embodiments, step 403 involves segmentally fusing the adjusted time decay factor with the relative deviation rate: within 0.5 times the maximum allowable activity radius, the square term of the time decay factor is superimposed on the relative deviation rate as a linear base; within 0.5 to 1 times the maximum allowable activity radius, the cube term of the time decay factor is superimposed on the logarithmic transformation value of the relative deviation rate; beyond the maximum allowable activity radius, the reciprocal term of the time decay factor is superimposed on the exponential value of the relative deviation rate; and an outlier is generated based on the square, cube, and reciprocal terms, including: 501. Within 0.5 times the maximum allowable activity radius, extract the core area priority weight defined in the medical insurance reimbursement policy of the preset treatment area, multiply the relative deviation rate by the core area priority weight to obtain the linear base, and superimpose the product of the square term of the time decay factor and the linear base to generate the core area stage outlier. In step 501, the core area priority weight refers to the core area priority defined in the medical insurance reimbursement policy of the preset treatment area (e.g., core area weight 0.7), which is used to reflect the high sensitivity requirement of the core area for anomaly monitoring.

[0098] Linear base: refers to the product of the relative deviation rate and the priority weight of the core area, used to quantify the degree of spatial deviation within the core area.

[0099] In this embodiment, within a range of 0.5 times the maximum allowable activity radius, the core region priority weight (e.g., 0.7) is extracted, the relative deviation rate is multiplied by the core region priority weight to obtain the linear base (e.g., relative deviation rate × 0.7), and the product of the squared term of the time decay factor and the linear base (e.g., time decay factor² × linear base) is superimposed to generate the core region stage outlier.

[0100] 502. Within the range of 0.5 to 1 times the maximum allowable activity radius, extract the edge area tolerance coefficient defined in the medical insurance reimbursement policy of the preset treatment area, multiply the natural logarithm of the relative deviation rate by the edge area tolerance coefficient to obtain the logarithmic transformation value, and superimpose the product of the cube term of the time decay factor and the logarithmic transformation value to generate the edge area stage outlier value. In step 502, the marginal area tolerance coefficient refers to the marginal area tolerance coefficient (e.g., 0.5) defined in the medical insurance reimbursement policy of the preset treatment area, which is used to reflect the elasticity of the marginal area's need for abnormal monitoring.

[0101] Logarithmic transformation value: refers to the product of the natural logarithm of the relative deviation rate and the tolerance coefficient of the edge region, used to quantify the degree of spatial deviation within the edge region.

[0102] In this embodiment, within a range of 0.5 to 1 times the maximum permissible activity radius, the edge region tolerance coefficient (e.g., 0.5) is extracted. The natural logarithm of the relative deviation rate is multiplied by the edge region tolerance coefficient to obtain a logarithmic transformation value (e.g., ln(relative deviation rate) × 0.5). The product of the cube term of the time decay factor and the logarithmic transformation value (e.g., time decay factor³ × logarithmic transformation value) is then superimposed to generate anomalies in the edge region stage.

[0103] 503. When the maximum allowable activity radius is exceeded, the off-domain reimbursement attenuation factor defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The exponential value of the relative deviation rate with the natural constant as the base is multiplied by the off-domain reimbursement attenuation factor to obtain the exponential transformation value. The product of the reciprocal term of the time attenuation factor and the exponential transformation value is added to generate the off-domain stage outlier. In step 503, the de-territorial reimbursement attenuation factor refers to the de-territorial reimbursement attenuation factor (e.g., 0.3) defined in the medical insurance reimbursement policy of the pre-defined treatment area, which is used to reflect the strict requirements of the de-territorial area for abnormal monitoring.

[0104] Exponential transformation value: refers to the product of the exponential value of the relative deviation rate with the natural constant as the base and the delocalization attenuation factor, used to quantify the degree of spatial deviation within the delocalization region.

[0105] In this embodiment of the application, when the maximum allowable activity radius is exceeded, the dedomain reimbursement attenuation factor (e.g., 0.3) is extracted, and the exponential value of the relative deviation rate with the natural constant base is multiplied by the dedomain reimbursement attenuation factor to obtain the exponential transformation value (e.g., e^relative deviation rate × 0.3). The product of the reciprocal term of the time decay factor and the exponential transformation value (e.g., 1 / time decay factor × exponential transformation value) is added to generate the outlier value of the dedomain stage.

[0106] 504. Based on the stage anomalies in the core area, the stage anomalies in the peripheral area, and the stage anomalies outside the designated area, according to the regional priority weight allocation ratio in the medical insurance reimbursement policy of the preset treatment area, assign weight coefficients to the stage anomalies in the core area, the peripheral area, and the stage anomalies outside the designated area respectively. Multiply each stage anomaly by its corresponding weight coefficient and sum them up to generate the stage anomaly.

[0107] In step 504, the regional priority weight allocation ratio refers to the priority weights of the core area, peripheral area and out-of-area area as defined in the medical insurance reimbursement policy of the preset treatment area (e.g., core area 0.7, peripheral area 0.3, out-of-area 0.1), which is used for outliers in the dynamic weighted summation stage.

[0108] In this embodiment of the application, according to the regional priority weight allocation ratio, the stage outlier values ​​of the core region, the edge region and the out-of-domain are assigned weight coefficients respectively (e.g., 0.7 for the core region, 0.3 for the edge region and 0.1 for the out-of-domain). The stage outlier values ​​are multiplied by the corresponding weight coefficients and then summed (e.g., stage outlier value of the core region × 0.7 + stage outlier value of the edge region × 0.3 + stage outlier value of the out-of-domain × 0.1) to generate the stage outlier values.

[0109] Here is a specific example: The location of the medication bottle for a diabetic chronic disease patient shows a current distance of 1.2 kilometers from the community hospital (maximum allowable activity radius is 1 kilometer), with a relative deviation rate of 1.2 and a time decay factor of 0.89. Within a radius of 0.5 to 1, the system generates marginal outliers by adding the natural logarithm of the relative deviation rate to the cube of the time decay factor; outside the radius, it generates outliers by adding the exponential value of the relative deviation rate to the reciprocal of the time decay factor. Finally, the system dynamically weights and sums these values ​​according to regional priority (core region 0.7, marginal region 0.3, outlier 0.1), generating a phase outlier of 0.85.

[0110] In summary, by dynamically weighting and summing the segmented spatial range and regional priority weights in steps 501 to 504, this method achieves accurate quantification of abnormal follow-up visit pathways for patients with chronic diseases, significantly improving the sensitivity and scenario adaptability of pathway monitoring, and providing a scientific basis for medical insurance supervision and health management of patients with chronic diseases.

[0111] To improve the dynamic adaptability and synergy of monitoring abnormal medical pathways in border areas for patients with chronic diseases with medical insurance policies, this study generates a normalized tolerance coefficient by extracting the partial reimbursement ratio and the maximum tolerance deviation rate in border areas. It then quantifies the degree of spatial deviation by combining the natural logarithm of the relative deviation rate and the deviation rate ratio, and strengthens the time accumulation effect by superimposing a cubic term of the time decay factor. In addition, it introduces a dynamic adjustment factor for the flexible range of follow-up visit time to optimize the abnormal value generation logic. The aim is to accurately identify spatiotemporal deviation behaviors in border areas within the tolerance range of medical insurance policies, and balance the rational allocation of medical resources with the flexibility of medical pathways for patients with chronic diseases.

[0112] In some embodiments, in step 502, within a range of 0.5 to 1 times the maximum permissible activity radius, the marginal area tolerance coefficient defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The natural logarithm of the relative deviation rate is multiplied by the marginal area tolerance coefficient to obtain a logarithmic transformation value. The product of the cube term of the time decay factor and the logarithmic transformation value is then added to generate marginal area stage outliers, including: 601. Extract a portion of the reimbursement ratio for marginal areas from the medical insurance reimbursement policy, normalize the portion of the reimbursement ratio with the preset maximum tolerance deviation rate for marginal areas, and generate the tolerance coefficient for marginal areas. In step 601, the partial reimbursement ratio refers to the partial reimbursement ratio for marginal areas in the medical insurance reimbursement policy (such as 50% reimbursement for marginal areas), which is used to reflect the medical insurance policy constraints for marginal areas.

[0113] The maximum tolerance deviation rate of the edge region refers to the preset maximum allowable deviation of the edge region (e.g., the maximum tolerance deviation rate of the edge region is 0.8), which is used to normalize the reimbursement ratio.

[0114] The tolerance coefficient for marginal areas refers to a parameter generated by normalizing the partial reimbursement ratio and the maximum tolerance deviation rate of the marginal area, and is used to quantify the abnormal tolerance of the marginal area.

[0115] In this embodiment of the application, a portion of the reimbursement ratio (e.g., 50%) in the marginal area of ​​the medical insurance reimbursement policy is extracted and normalized with the maximum tolerance deviation rate of the marginal area (e.g., 0.8) to generate a tolerance coefficient for the marginal area (e.g., 0.625).

[0116] 602. Using the ratio of the relative deviation rate to the maximum tolerance deviation rate of the edge region as the input value of the logarithmic transformation, calculate the natural logarithm of the input value, and multiply the natural logarithm by the tolerance coefficient of the edge region to obtain the logarithmic transformation value; In step 602, the input value for the logarithmic transformation is the ratio of the relative deviation rate to the maximum tolerable deviation rate of the edge region, which is used as the normalized input for the logarithmic transformation.

[0117] The natural logarithm refers to the result of calculating the natural logarithm of the input value after logarithmic transformation, and is used to compress the effects of high bias rates.

[0118] The logarithmic transformation value is the product of the natural logarithmic value and the tolerance coefficient of the edge region, used to quantify the degree of spatial deviation of the edge region.

[0119] In this embodiment of the application, the ratio of the relative deviation rate (e.g., 1.2) to the maximum tolerance deviation rate of the edge region (e.g., 0.8) is used as the input value of the logarithmic transformation (e.g., 1.2 / 0.8=1.5). The natural logarithm of the input value is calculated (e.g., ln(1.5)≈0.405), and it is multiplied by the tolerance coefficient of the edge region (e.g., 0.625) to obtain the logarithmic transformation value (e.g., 0.405×0.625≈0.253).

[0120] 603. Extract the cubic term of the time decay factor, multiply the cubic term by the logarithmic transformation value to generate preliminary edge region outlier values; In step 603, the cube term of the time decay factor refers to the cube value of the time decay factor (e.g., time decay factor³), which is used to enhance the long-term cumulative effect of deviations in the time dimension.

[0121] Preliminary outlier in the edge region: refers to the product of the cube term of the time decay factor and the logarithmic transformation value, used to initially quantify the degree of anomaly in the edge region.

[0122] In this embodiment of the application, the cube term of the time decay factor (e.g., 0.89³≈0.705) is extracted and multiplied by the logarithmic transformation value (e.g., 0.253) to generate preliminary edge region outlier values ​​(e.g., 0.705×0.253≈0.178).

[0123] 604. Based on the flexible range of follow-up visit time in the marginal area of ​​the medical insurance reimbursement policy, multiply the preliminary marginal area outlier by the dynamic adjustment factor corresponding to the flexible range of follow-up visit time to generate the marginal area stage outlier.

[0124] In step 604, the flexible range of follow-up visit time refers to the flexible range of follow-up visit time allowed in the marginal area under the medical insurance reimbursement policy (such as allowing a 7-day delay in follow-up visit in the marginal area), which is used to dynamically adjust the initial abnormal value.

[0125] Dynamic adjustment factor: refers to the adjustment parameter generated based on the flexible range of follow-up visit time, used to reflect the urgency of deviation in the time dimension.

[0126] In this embodiment of the application, a dynamic adjustment factor (e.g., 1.2) is generated based on the flexible range of follow-up visit time (e.g., 7 days). The initial marginal region outlier (e.g., 0.178) is multiplied by the dynamic adjustment factor (e.g., 1.2) to generate the marginal region stage outlier (e.g., 0.178×1.2≈0.214).

[0127] Here is a specific example: The location of the medication bottle for diabetic chronic disease patients shows that their current distance from the community hospital is 1.2 kilometers (maximum allowable activity radius is 1 kilometer), with a relative deviation rate of 1.2 and a time decay factor of 0.89. The system extracts the reimbursement ratio (50%) and the maximum tolerance deviation rate (0.8) of the marginal area to generate the marginal area tolerance coefficient (0.625); the ratio of the relative deviation rate to the maximum tolerance deviation rate (1.5) is used as the input value for logarithmic transformation to calculate the natural logarithmic value (0.405) and generate the logarithmic transformation value (0.253); the cube term of the time decay factor (0.705) is extracted to generate the preliminary marginal area outlier (0.178); a dynamic adjustment factor (1.2) is generated based on the flexible range of follow-up visit time (7 days), and finally, the marginal area stage outlier (0.214) is generated.

[0128] In summary, by generating the tolerance coefficient of the marginal region, calculating the logarithmic transformation value, and applying the dynamic adjustment factor in steps 601 to 604, this method achieves accurate quantification of abnormal follow-up visit pathways for chronic disease patients in the marginal region, significantly improving the sensitivity and adaptability of pathway monitoring to flexible needs, and providing a scientific basis for medical insurance supervision and health management of chronic disease patients.

[0129] To improve the dynamic adaptability and assessment accuracy of abnormal monitoring of medical treatment pathways in border areas for patients with chronic diseases, the deviation threshold is dynamically adjusted by combining the flexible range of follow-up visit time and the historical compliance rate of patients with chronic diseases. The ratio of relative deviation rate to threshold is transformed by natural logarithm to quantify the spatial deviation intensity. The intermediate value is generated by weighted fusion of medical insurance tolerance coefficient, and standardized dimensional mapping is achieved by the reciprocal of the maximum tolerance deviation rate. The aim is to dynamically adapt to medical insurance policy rules, optimize the spatiotemporal sensitivity of abnormal identification, and improve the accuracy of compliance assessment of cross-regional medical treatment behavior.

[0130] In some embodiments, step 602 uses the ratio of the relative deviation rate to the maximum tolerance deviation rate of the edge region as the input value for the logarithmic transformation, calculates the natural logarithm of the input value, and multiplies the natural logarithm by the edge region tolerance coefficient to obtain the logarithmic transformation value, including: 701. Extract the dynamic deviation threshold corresponding to the maximum tolerance deviation rate of the edge region. The dynamic deviation threshold is dynamically adjusted based on the correlation between the elastic range of follow-up visit time in the edge region and the compliance rate of the historical medical treatment path of chronic disease patients in the medical insurance reimbursement policy. In step 701, the dynamic deviation threshold refers to the deviation threshold that is dynamically adjusted based on the flexible range of follow-up visit time in the marginal area of ​​the medical insurance reimbursement policy and the compliance rate of the historical medical treatment path of chronic disease patients. It is used to reflect the sensitivity of abnormal monitoring in the marginal area.

[0131] The flexible range of follow-up visit time refers to the flexible range of follow-up visit time allowed in the marginal areas of the medical insurance reimbursement policy (such as allowing a 7-day delay in follow-up visit in marginal areas), which is used to dynamically adjust deviations from the threshold.

[0132] The compliance rate of the historical medical treatment path for patients with chronic diseases refers to the compliance of the follow-up visit path of patients with chronic diseases in the past period (such as 3 months) (such as the on-time follow-up visit rate), which is used to further correct deviations from the threshold.

[0133] In this embodiment of the application, the maximum tolerance deviation rate of the edge area is dynamically adjusted (e.g., from 0.8 to 0.75) based on the flexible range of follow-up visit time (e.g., 7 days) and the compliance rate of the patient's historical medical path (e.g., 80%), and a dynamic deviation threshold (e.g., 0.75) is generated.

[0134] 702. Using the ratio of the relative deviation rate to the dynamic deviation threshold as the normalized input value of the logarithmic transformation, calculate the natural logarithm of the normalized input value, and weight and fuse the natural logarithm with the edge region tolerance coefficient. In step 702, the normalized input value refers to the ratio of the relative deviation rate to the dynamic deviation threshold, which is used as the normalized input for logarithmic transformation.

[0135] Weighted fusion refers to the weighted fusion of natural logarithmic values ​​and marginal region tolerance coefficients, with the weights determined based on the gradient distribution of reimbursement ratios in marginal regions within the medical insurance reimbursement policy.

[0136] In this embodiment, the ratio of the relative deviation rate (e.g., 1.2) to the dynamic deviation threshold (e.g., 0.75) is used as the normalized input value (e.g., 1.2 / 0.75=1.6), and its natural logarithmic value is calculated (e.g., ln(1.6)≈0.47). According to the gradient distribution of reimbursement ratio in the marginal area in the medical insurance reimbursement policy (e.g., 70% reimbursement at a radius of 0.5-0.7 times that of the core area, and 50% reimbursement at a radius of 0.7-1 times), the weighted fusion weight is determined (e.g., 0.6). The natural logarithmic value (0.47) and the marginal area tolerance coefficient (e.g., 0.625) are weighted and fused (e.g., 0.47×0.6 + 0.625×0.4≈0.53) to generate the intermediate logarithmic transformation value.

[0137] 703. Generate intermediate logarithmic transformation values ​​based on the weighted fusion result, and multiply the intermediate logarithmic transformation values ​​with the reciprocal of the maximum tolerance deviation rate of the edge region to obtain the logarithmic transformation value.

[0138] In step 703, the intermediate logarithmic transformation value refers to the combined result of the weighted fusion of the natural logarithmic value and the tolerance coefficient of the edge region, which is used to initially quantify the logarithmic transformation value.

[0139] The reciprocal of the maximum tolerance deviation rate of the edge region refers to the reciprocal of the maximum tolerance deviation rate of the edge region (e.g., 1 / 0.75≈1.33), which is used to enhance the nonlinear characteristics of the logarithmic transformation value.

[0140] In this embodiment of the application, the intermediate logarithmic transformation value (e.g., 0.53) is multiplied by the reciprocal of the maximum tolerance deviation rate of the edge region (e.g., 1.33) to generate the logarithmic transformation value (e.g., 0.53 × 1.33 ≈ 0.705).

[0141] Here is a specific example: The location of the medication bottle for diabetic chronic disease patients shows that their current distance from the community hospital is 1.2 kilometers (the maximum allowable activity radius is 1 kilometer), with a relative deviation rate of 1.2; the dynamic deviation threshold is 0.75 (adjusted based on a 7-day flexible range for follow-up visits and a historical compliance rate of 80%). The system uses the ratio of the relative deviation rate to the dynamic deviation threshold (1.6) as the normalized input value to calculate the natural logarithmic value (0.47); it performs weighted fusion based on the reimbursement ratio gradient distribution (0.6 weight) to generate an intermediate logarithmic transformation value (0.53); and multiplies it by the reciprocal of the maximum tolerable deviation rate of the marginal area (1.33) to generate a logarithmic transformation value (0.705).

[0142] In summary, by adjusting the dynamic deviation threshold in steps 701 to 703, calculating the normalized input value, and applying weighted fusion, this method achieves accurate quantification of abnormal follow-up visit paths for chronic disease patients in marginal areas, significantly improving the dynamic adaptability and scenario sensitivity of path monitoring, and providing a scientific basis for medical insurance supervision and health management of chronic disease patients.

[0143] Figure 2 This application provides a schematic diagram of the structure of a critical path monitoring device (or system) for medical insurance-based medical treatment, as shown in the embodiments below. Figure 2 As shown, the device includes: The data acquisition module 21 is used to collect medication behavior data of patients with chronic diseases in real time. The medication behavior data includes the dynamic trajectory of changes in drug inventory and the position deviation of the medicine bottle from the preset treatment area. The association module 22 is used to associate the medication behavior data with the medical insurance data of chronic disease patients in real time, match the dynamic trajectory of the change in drug inventory with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index, and at the same time combine the location deviation with the spatiotemporal correlation of the preset medical area to generate a medical path abnormality coefficient. The interception module 23 is used to trigger a real-time interception command from the medical insurance settlement terminal and push a dynamic path correction scheme to the mobile terminal bound to the chronic disease patient when the medication adherence deviation index and the abnormal coefficient of the medical treatment path both exceed the chronic disease type association threshold.

[0144] Figure 2 The aforementioned critical path monitoring device based on medical insurance can perform... Figure 1 The implementation principle and technical effects of the critical path monitoring method for medical treatment based on medical insurance, as described in the illustrated embodiment, will not be repeated here. The specific operation methods of each module and unit in the critical path monitoring device for medical treatment based on medical insurance in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0145] In one possible design, Figure 2 The illustrated embodiment of a critical path monitoring device for medical treatment based on medical insurance can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0146] The processing component 32 is used for the above Figure 1 The above embodiment describes a method for monitoring critical pathways to medical treatment based on medical insurance.

[0147] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0148] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0149] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0150] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0151] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0152] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0153] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The illustrated embodiment presents a critical path monitoring method for medical treatment based on medical insurance.

[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0155] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0156] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring critical pathways to medical treatment based on medical insurance, characterized in that, include: Real-time collection of medication behavior data from patients with chronic diseases, including the dynamic trajectory of changes in drug inventory and the positional deviation of the medicine bottle from the preset treatment area; The medication behavior data is correlated in real time with the medical insurance data of chronic disease patients. The dynamic trajectory of changes in drug inventory is matched with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index. At the same time, the location deviation and the spatiotemporal correlation of the preset medical area are combined to generate a medical path anomaly coefficient. The step of generating anomaly coefficients for medical treatment paths by combining the location deviation with the spatiotemporal correlation of the preset medical treatment area includes: The real-time Euclidean distance between the location of the medicine bottle and the center point of the preset treatment area is calculated as the position deviation benchmark value. Combined with the number of days between the most recent treatment time of the chronic disease patient in the medical insurance data and the current time, the treatment path anomaly coefficient is generated based on the nonlinear relationship between the position deviation benchmark value, the number of days between the interval and the maximum allowable activity radius of the preset treatment area. When the medication adherence deviation index and the medical visit path abnormality coefficient both exceed the chronic disease type association threshold, a real-time interception command is triggered on the medical insurance settlement terminal, and a dynamic path correction scheme is pushed to the mobile terminal bound to the chronic disease patient.

2. The method according to claim 1, characterized in that, The medication behavior data is correlated in real time with the medical insurance data of chronic disease patients, and the dynamic trajectory of changes in drug inventory is matched with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index, including: The theoretical daily drug consumption corresponding to the prescription dosage of patients with chronic diseases is obtained and compared with the actual daily drug consumption in the dynamic trajectory of drug inventory changes. The number of consecutive days of deviation and the magnitude of daily deviation between the theoretical daily drug consumption and the actual daily drug consumption are calculated. Based on the dynamic weighting rules of the number of consecutive days of deviation and the magnitude of daily deviation in the drug metabolism cycle associated with the chronic disease type, a medication adherence deviation index is generated.

3. The method according to claim 2, characterized in that, Also includes: The medication adherence deviation index, medical path abnormality coefficient, consecutive deviation days, daily deviation magnitude, location deviation benchmark value, and interval days are input into a pre-constructed chronic disease type association threshold calculation function. The calculation function embeds a correlation factor between drug dosage restrictions in medical insurance reimbursement records and the compliance rate of historical medical paths of chronic disease patients, and outputs the chronic disease type association threshold.

4. The method according to claim 2, characterized in that, Based on the dynamic weighting rules of the consecutive days of deviation and the daily deviation magnitude within the drug metabolism cycle associated with chronic disease types, a medication adherence deviation index is generated, including: Based on the drug metabolism cycle associated with the chronic disease type, the drug metabolism cycle is decomposed into an initial metabolic stage, a metabolic maintenance stage, and a metabolic adjustment stage. For each metabolic stage, the stage weight ratio of continuous deviation days and daily deviation magnitude is defined. In the initial stage of metabolism, if the number of consecutive deviation days is less than an integer multiple of the half-life corresponding to the drug's metabolic cycle, then the daily deviation amplitude is assigned a weight ratio higher than that of the number of consecutive deviation days. During the metabolic maintenance phase, if the number of consecutive deviation days covers the complete dosing interval of the drug metabolism cycle, then a balanced weighting ratio is assigned to the number of consecutive deviation days and the daily deviation magnitude. During the metabolic adjustment phase, if the daily deviation exceeds the dose fluctuation threshold corresponding to the drug metabolism cycle, a decreasing weighting ratio is assigned to the daily deviation. Based on the weight ratio of days higher than the consecutive deviation days, the balanced weight ratio, and the decreasing weight ratio, the stage deviation degree of the consecutive deviation days and the daily deviation range is calculated respectively. The stage deviation degree of each metabolic stage is accumulated and superimposed according to the time sequence proportion of the drug metabolism cycle to generate the medication adherence deviation index.

5. The method according to claim 1, characterized in that, Based on the nonlinear relationship between the location deviation benchmark value, the number of days between visits, and the maximum permissible activity radius of the preset treatment area, an abnormality coefficient for the treatment path is generated, including: The ratio of the location deviation benchmark value to the maximum allowable activity radius is defined as the relative deviation rate. Combined with the standard value of the follow-up visit cycle associated with the number of days between visits and the type of chronic disease, the time decay factor is calculated. When the relative deviation rate is less than or equal to 1, the time decay factor is positively adjusted based on the ratio of the interval days to the remaining time of the standard value of the follow-up visit cycle; when the relative deviation rate is greater than 1, the time decay factor is negatively adjusted based on the number of days the interval days exceed the standard value of the follow-up visit cycle. The adjusted time decay factor is fused with the relative deviation rate in a segmented manner. Within 0.5 times the maximum allowable activity radius, the square term of the time decay factor is superimposed on the relative deviation rate as a linear base. Within 0.5 to 1 times the maximum allowable activity radius, the cube term of the time decay factor is superimposed on the logarithmic transformation value of the relative deviation rate. When the maximum allowable activity radius is exceeded, the reciprocal term of the time decay factor is superimposed on the exponential value of the relative deviation rate. Stage outliers are generated based on the square, cube, and reciprocal terms. All the abnormal values ​​in the aforementioned stages are dynamically weighted and summed according to the regional priority weight in the medical insurance reimbursement policy of the preset treatment area to generate the treatment path abnormality coefficient.

6. The method according to claim 5, characterized in that, The adjusted time decay factor and the relative deviation rate are fused in a segmented manner. Within 0.5 times the maximum allowable activity radius, the square term of the time decay factor is superimposed on the relative deviation rate as a linear base. Within 0.5 to 1 times the maximum allowable activity radius, the cube term of the time decay factor is superimposed on the logarithmic transformation value of the relative deviation rate. When the maximum allowable activity radius is exceeded, the reciprocal term of the time decay factor is superimposed on the exponential value of the relative deviation rate. Outliers are generated based on the square term, cube term, and reciprocal term, including: Within 0.5 times the maximum allowable activity radius, the core area priority weight defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The relative deviation rate is multiplied by the core area priority weight to obtain a linear base. The product of the square of the time decay factor and the linear base is added to generate the core area stage outlier. Within a range of 0.5 to 1 times the maximum permissible activity radius, the marginal area tolerance coefficient defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The natural logarithm of the relative deviation rate is multiplied by the marginal area tolerance coefficient to obtain the logarithmic transformation value. The product of the cube term of the time decay factor and the logarithmic transformation value is superimposed to generate the marginal area stage outlier. When the maximum allowable activity radius is exceeded, the off-domain reimbursement attenuation factor defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The exponential value of the relative deviation rate with the natural constant as the base is multiplied by the off-domain reimbursement attenuation factor to obtain the exponential transformation value. The product of the reciprocal term of the time attenuation factor and the exponential transformation value is added to generate the off-domain stage outlier. Based on the stage anomalies in the core area, the edge area, and the out-of-domain stage, according to the regional priority weight allocation ratio in the medical insurance reimbursement policy of the preset treatment area, the stage anomalies in the core area, the edge area, and the out-of-domain stage are assigned weight coefficients respectively. The stage anomalies are multiplied by the corresponding weight coefficients and then summed to generate stage anomalies.

7. The method according to claim 6, characterized in that, Within a range of 0.5 to 1 times the maximum permissible activity radius, the marginal area tolerance coefficient defined in the medical insurance reimbursement policy of the preset treatment area is extracted. The natural logarithmic value of the relative deviation rate is multiplied by the marginal area tolerance coefficient to obtain a logarithmic transformation value. The product of the cube term of the time decay factor and the logarithmic transformation value is then added to generate marginal area stage outliers, including: Extract a portion of the reimbursement ratio for marginal areas from the medical insurance reimbursement policy, normalize the portion of the reimbursement ratio with the preset maximum tolerance deviation rate for marginal areas, and generate the tolerance coefficient for the marginal areas. The ratio of the relative deviation rate to the maximum tolerance deviation rate of the edge region is used as the input value of the logarithmic transformation. The natural logarithm of the input value is calculated, and the natural logarithm is multiplied by the tolerance coefficient of the edge region to obtain the logarithmic transformation value. Extract the cubic term of the time decay factor, multiply the cubic term by the logarithmic transformation value, and generate preliminary edge region outlier values; Based on the flexible range of follow-up visit time in the marginal area of ​​the medical insurance reimbursement policy, the preliminary marginal area outlier is multiplied by the dynamic adjustment factor corresponding to the flexible range of follow-up visit time to generate the marginal area stage outlier.

8. A critical path monitoring system for medical treatment based on medical insurance, characterized in that, include: The data acquisition module is used to collect medication behavior data of patients with chronic diseases in real time. The medication behavior data includes the dynamic trajectory of changes in drug inventory and the position deviation of the medicine bottle from the preset treatment area. The association module is used to associate the medication behavior data with the medical insurance data of chronic disease patients in real time, match the dynamic trajectory of the change in drug inventory with the temporal consistency of prescription dosage consumption to generate a medication adherence deviation index, and at the same time combine the location deviation with the spatiotemporal correlation of the preset medical area to generate a medical path anomaly coefficient. The step of generating anomaly coefficients for medical treatment paths by combining the location deviation with the spatiotemporal correlation of the preset medical treatment area includes: The real-time Euclidean distance between the location of the medicine bottle and the center point of the preset treatment area is calculated as the position deviation benchmark value. Combined with the number of days between the most recent treatment time of the chronic disease patient in the medical insurance data and the current time, the treatment path anomaly coefficient is generated based on the nonlinear relationship between the position deviation benchmark value, the number of days between the interval and the maximum allowable activity radius of the preset treatment area. The interception module is used to trigger a real-time interception command from the medical insurance settlement terminal and push a dynamic path correction scheme to the mobile terminal bound to the chronic disease patient when the medication adherence deviation index and the abnormal coefficient of the medical treatment path both exceed the chronic disease type association threshold.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a critical path monitoring method for medical treatment based on medical insurance as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements a critical path monitoring method for medical treatment based on medical insurance as described in any one of claims 1 to 7.

Images