Thrombus prevention method and system based on digital clinical precise execution

Abstract

The invention provides a thrombus prevention method and system based on digital clinical precise execution, and relates to the technical field of digital medical treatment, and the method comprises the steps: obtaining clinical data of a patient, and determining a thromboembolism risk level, a bleeding risk level and a mechanical prevention taboo state of the patient based on a preset rule; based on the combined state of the three evaluation results, generating a thrombus prevention measure scheme for the patient through a preset prevention strategy decision rule; monitoring clinical data changes in the hospitalization process of the patient in real time, when it is detected that the risk level or the taboo state is changed, comparing the consistency of a current execution scheme with a scheme regenerated based on the changed state, and if the current execution scheme is not consistent with the scheme regenerated based on the changed state, generating preventive measure adjustment reminding information; and pushing the scheme or the adjustment prompt to a clinical medical care terminal. According to the invention, intelligent decision and dynamic adjustment of thrombus prevention can be realized, the accuracy and timeliness of prevention measures are improved, and medical risks are reduced.

Description

Technical Field

[0001] This invention relates to the field of digital medical technology, and in particular to a method and system for thrombosis prevention based on precise digital clinical execution. Background Technology

[0002] Thromboembolism is a serious clinical problem faced by hospitalized patients, including types such as deep vein thrombosis and pulmonary embolism, which can lead to death or serious complications. Clinical studies have shown that standardized risk assessment and preventive measures can effectively reduce the incidence of thromboembolism. Traditional thrombosis prevention management mainly relies on the clinical experience and subjective judgment of healthcare professionals, manually assessing patient risk and developing prevention plans.

[0003] With the development of medical informatization, some medical institutions have begun to explore establishing thrombosis prevention management systems, utilizing electronic medical record data to assist clinical decision-making. Some systems can perform risk scoring based on patients' basic information and clinical indicators, providing doctors with reference suggestions. However, existing thrombosis prevention management technologies still have significant limitations.

[0004] In current technologies, risk assessment for thrombosis prevention is often a static, single-assessment model. Healthcare professionals typically conduct a risk assessment and develop a prevention plan upon patient admission. However, a patient's condition changes dynamically during hospitalization. Various factors, such as surgery, medication, and mobility, can affect changes in thrombosis and bleeding risks. Current technologies lack the ability to continuously monitor and dynamically assess a patient's clinical status, making it impossible to detect changes in risk levels in a timely manner. This results in a mismatch between preventive measures and the patient's actual risk status. Summary of the Invention

[0005] The embodiments of the present invention provide a method and system for thrombosis prevention based on digital clinical precision execution, which can solve the problems in the prior art.

[0006] A first aspect of the present invention provides a thrombosis prevention method based on digital clinical precision execution, comprising: Acquire the patient's clinical data, analyze the clinical data based on preset thromboembolism risk assessment rules, and determine the patient's thromboembolism risk level; The patient's bleeding risk level is determined based on clinical data and pre-defined bleeding risk assessment rules; whether the patient has contraindications to mechanical prophylaxis is determined based on clinical data and pre-defined contraindications assessment rules. Based on the combined status of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, a thrombosis prevention plan for the patient is generated through preset prevention strategy decision rules; Real-time monitoring of changes in patients' clinical data during hospitalization; when changes are detected in a patient's thromboembolism risk level, bleeding risk level, or contraindications to mechanical prophylaxis, the currently implemented thromboprophylaxis plan is compared with a newly generated thromboprophylaxis plan based on the changed status; if inconsistencies are found, a reminder to adjust the prophylaxis measures is generated. The aforementioned preventive measures plan or reminders for adjusting the preventive measures will be pushed to the clinical medical staff terminal.

[0007] The clinical data includes medical records, examination and test results, medical orders, surgical records, and transfer records; The preventive measures program includes basic prevention, drug prevention, mechanical prevention, or a combination thereof.

[0008] The patient's bleeding risk level is determined based on clinical data and pre-defined bleeding risk assessment rules; contraindications to mechanical prophylaxis are determined based on clinical data and pre-defined contraindications assessment rules, including: Feature data related to bleeding risk is extracted from the clinical data, and the feature data is matched with risk factors in the preset bleeding risk assessment rules. The risk factors include active bleeding status, coagulation dysfunction status, anticoagulant use status, and high-risk surgical status to obtain matching results. Based on the presence status of each risk factor in the matching results, the patient's bleeding risk level is determined through a preset risk level determination logic, which defines the mapping relationship between the presence status of risk factors and the bleeding risk level. The process of determining whether a patient has contraindications to mechanical prophylaxis based on their clinical data and pre-defined contraindication assessment rules includes: From the clinical data, vital signs data related to contraindications for mechanical prophylaxis were extracted, including lower limb condition information, vascular condition information, and skin condition information. The vital signs data are compared with the contraindication criteria in the preset mechanical prevention contraindication assessment rules. The contraindication criteria include lower limb trauma, lower limb edema, arteriosclerosis, and skin damage. The presence or absence of contraindications for mechanical prophylaxis in the patient is determined by comparing the vital signs data with the contraindication criteria.

[0009] Based on the combined status of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, a thrombosis prevention plan for the patient is generated through preset prevention strategy decision rules, including: The combined status of the thromboembolic risk level, bleeding risk level, and contraindication to mechanical prophylaxis is encoded as a combined status identifier, which uniquely represents a specific combination of the thromboembolic risk level, bleeding risk level, and contraindication to mechanical prophylaxis. Retrieve the prevention strategy item corresponding to the combined state identifier from the preset prevention strategy decision rules. The prevention strategy item defines the type and intensity of prevention measures for a specific combined state. The prevention method of the thrombosis prevention plan is determined according to the type of prevention measure in the prevention strategy item. The prevention method includes basic prevention, pharmacological prevention, mechanical prevention, and combined prevention. The execution parameters for the thrombosis prevention plan are generated based on the intensity of the preventive measures in the prevention strategy items and the patient's clinical data. The execution parameters include the timing of initiation of the preventive measures, the duration, and the monitoring frequency. Based on the prevention method and the execution parameters, a thrombosis prevention plan for the patient is constructed.

[0010] Retrieving the prevention strategy item corresponding to the combined state identifier from the preset prevention strategy decision rules includes: The thromboembolism risk level status value is extracted from the combined status identifiers. The status value is compared with the first-level judgment condition of the preset prevention strategy decision rule. Based on the comparison result, the corresponding first-level decision branch is selected. Extract the bleeding risk level status value from the first-level decision branch, compare the status value with the second-level judgment conditions under that branch, and select the corresponding second-level decision branch based on the comparison result; Extract the state value of mechanical prevention taboo in the second-level decision branch, compare the state value with the third-level decision condition under the branch, and reach the corresponding termination node according to the comparison result; The prevention strategy item is read from the termination node. The prevention strategy item records the type and strength of the prevention measures for the combined state.

[0011] The currently implemented thrombosis prevention plan is compared with the regenerated thrombosis prevention plan based on the changed state. If they are inconsistent, a prevention measure adjustment reminder message is generated, including: Extract the current prevention method and current execution parameters from the currently implemented thrombosis prevention measure plan, and extract the regenerated prevention method and regenerated execution parameters from the thrombosis prevention measure plan regenerated based on the changed state; The current prevention method is compared with the regenerated prevention method to determine whether the prevention methods are consistent, and the comparison result of the prevention methods is obtained. The current execution parameters are compared with the regenerated execution parameters to determine whether the execution parameters are consistent, and the execution parameter comparison result is obtained. A consistency determination result is determined based on the comparison results of the prevention methods and the comparison results of the execution parameters. When the comparison results of the prevention methods or the comparison results of the execution parameters show inconsistency, the consistency determination result is determined to be inconsistent. When the consistency determination result is inconsistent, a preventive measure adjustment reminder message is generated based on the prevention method comparison result and the execution parameter comparison result. The preventive measure adjustment reminder message indicates the preventive measure content that needs to be adjusted.

[0012] A second aspect of the present invention provides a thrombosis prevention system based on digital clinical precision execution, comprising: The first unit is used to acquire the patient's clinical data, analyze the clinical data based on preset thromboembolism risk assessment rules, and determine the patient's thromboembolism risk level. The second unit is used to determine the patient's bleeding risk level based on the patient's clinical data and pre-set bleeding risk assessment rules; and to determine whether the patient has contraindications to mechanical prophylaxis based on the patient's clinical data and pre-set contraindications assessment rules. The third unit is used to generate a thrombosis prevention plan for the patient based on the combination of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, using preset prevention strategy decision rules. The fourth unit is used to monitor changes in patients' clinical data in real time during hospitalization. When changes are detected in a patient's thromboembolism risk level, bleeding risk level, or contraindications to mechanical prophylaxis, the currently implemented thromboprophylaxis plan is compared with a newly generated thromboprophylaxis plan based on the changed status. If there is a discrepancy, a reminder message for adjusting the prophylaxis measures is generated. The fifth unit is used to push the prevention measure plan or the prevention measure adjustment reminder information to the clinical medical care terminal.

[0013] A third aspect of the present invention, An electronic device is provided, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0014] Fourth aspect of the embodiments of the present invention, A computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0015] The beneficial effects of this application are as follows: This invention establishes rules for thromboembolism risk assessment, bleeding risk assessment, and mechanical prophylaxis contraindication assessment. It enables multi-dimensional comprehensive analysis of patients' clinical data, automatically determines patients' risk levels and contraindications, and generates personalized thrombosis prevention plans based on multi-factor combinations. This achieves precision and standardization in thrombosis prevention, avoids the subjectivity and inconsistency of manual assessment, and improves the scientificity and accuracy of thrombosis prevention plans.

[0016] This invention can dynamically capture changes in a patient's risk status by monitoring clinical data changes during hospitalization in real time, and automatically compare the current implementation plan with the plan generated based on the new status. When inconsistencies are found, adjustment reminders are generated in a timely manner, realizing dynamic and closed-loop management of thrombosis prevention. This ensures that preventive measures can be adjusted in a timely manner according to changes in the patient's condition, effectively reducing the risk of inappropriate preventive measures due to changes in the patient's condition.

[0017] This invention enables precise digital clinical execution of thrombosis prevention by pushing preventive measures and adjustment reminders to clinical medical staff terminals. This allows medical staff to receive timely and standardized prevention advice and adjustment reminders, reducing their workload, improving the efficiency and compliance of thrombosis prevention work, and helping to reduce the incidence of thromboembolism in hospitalized patients, thereby improving patient prognosis and quality of medical care. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the thrombosis prevention method based on digital clinical precision execution, as described in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0021] Figure 1 This is a flowchart illustrating the thrombosis prevention method based on digital clinical precision execution according to an embodiment of the present invention, as shown below. Figure 1As shown, the method includes: Acquire the patient's clinical data, analyze the clinical data based on preset thromboembolism risk assessment rules, and determine the patient's thromboembolism risk level; The patient's bleeding risk level is determined based on clinical data and pre-defined bleeding risk assessment rules; whether the patient has contraindications to mechanical prophylaxis is determined based on clinical data and pre-defined contraindications assessment rules. Based on the combined status of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, a thrombosis prevention plan for the patient is generated through preset prevention strategy decision rules; Real-time monitoring of changes in patients' clinical data during hospitalization; when changes are detected in a patient's thromboembolism risk level, bleeding risk level, or contraindications to mechanical prophylaxis, the currently implemented thromboprophylaxis plan is compared with a newly generated thromboprophylaxis plan based on the changed status; if inconsistencies are found, a reminder to adjust the prophylaxis measures is generated. The aforementioned preventive measures plan or reminders for adjusting the preventive measures will be pushed to the clinical medical staff terminal.

[0022] In one optional implementation, the clinical data includes medical record information, examination and test results, medical orders, surgical records, and transfer records; The preventive measures program includes basic prevention, drug prevention, mechanical prevention, or a combination thereof.

[0023] In one optional implementation, the patient's bleeding risk level is determined based on the patient's clinical data and pre-defined bleeding risk assessment rules; determining whether the patient has contraindications to mechanical prophylaxis is based on the patient's clinical data and pre-defined contraindications assessment rules, including: Feature data related to bleeding risk is extracted from the clinical data, and the feature data is matched with risk factors in the preset bleeding risk assessment rules. The risk factors include active bleeding status, coagulation dysfunction status, anticoagulant use status, and high-risk surgical status to obtain matching results. Based on the presence status of each risk factor in the matching results, the patient's bleeding risk level is determined through a preset risk level determination logic, which defines the mapping relationship between the presence status of risk factors and the bleeding risk level. The process of determining whether a patient has contraindications to mechanical prophylaxis based on their clinical data and pre-defined contraindication assessment rules includes: From the clinical data, vital signs data related to contraindications for mechanical prophylaxis were extracted, including lower limb condition information, vascular condition information, and skin condition information. The vital signs data are compared with the contraindication criteria in the preset mechanical prevention contraindication assessment rules. The contraindication criteria include lower limb trauma, lower limb edema, arteriosclerosis, and skin damage. The presence or absence of contraindications for mechanical prophylaxis in the patient is determined by comparing the vital signs data with the contraindication criteria.

[0024] Once the medical system receives the patient's electronic medical record data, the data preprocessing module parses the structured data and unstructured text within the record. For bleeding risk assessment, coagulation function indicators, including prothrombin time, activated partial thromboplastin time (APPT), and international normalized ratio (INR), are retrieved from the electronic medical record's laboratory report unit. For example, if a patient's prothrombin time is 18 seconds and their INR is 2.3, these values ​​are compared to preset normal reference ranges. If the INR exceeds 1.5, the patient is marked as having an abnormal coagulation function.

[0025] Simultaneously, disease diagnosis codes are extracted from the diagnostic information field of the medical record and identified using the International Classification of Diseases (ICD-1). When diagnostic codes such as gastrointestinal bleeding, intracranial hemorrhage, and urinary tract bleeding are detected, the patient is determined to have active bleeding. For a specific patient whose medical record records a diagnosis of upper gastrointestinal bleeding with a bleeding volume of 400 ml, the patient's active bleeding status is marked as positive.

[0026] In assessing medication use, the patient's medication record database is accessed to retrieve the names and dosages of currently used medications. A built-in drug knowledge base contains the standard names, alternative names, and brand names of various anticoagulants. A positive anticoagulant use status is identified when the patient's medication record shows they are taking 5 mg of warfarin daily or receiving 4000 units of low molecular weight heparin subcutaneously twice daily. The dosing time is also checked; if the last dose was within 24 hours of the current time, the medication is considered to still be within its effective period.

[0027] For surgery-related risk assessment, surgery type information is extracted from surgery request forms and surgical records. A pre-stored list of high-risk surgeries is maintained, including categories such as craniocerebral surgery, cardiovascular surgery, and hepatobiliary and pancreatic surgery. When a patient is about to undergo or has just completed a surgery on the list, the high-risk surgery status is marked as present. For example, if a patient plans to undergo abdominal aortic aneurysm resection, this surgery type is marked as extremely high-risk on the high-risk surgery list, and this status is immediately recorded in the risk assessment data structure.

[0028] After obtaining the presence status of all four risk factors, the risk level determination module is invoked for comprehensive assessment. The determination logic adopts a hierarchical determination rule, prioritizing the assessment of the most severe risk factor. When active bleeding is detected, the bleeding risk level is directly determined to be high-risk, regardless of the status of other risk factors. When active bleeding is absent but coagulation dysfunction is present and the international normalized ratio (INR) is greater than 2.0, the risk level is determined to be intermediate-high-risk. If a patient is simultaneously using anticoagulants and undergoing high-risk surgery, but has no active bleeding or severe coagulation dysfunction, the risk level is determined to be intermediate-risk. When only a single risk factor is present and its severity is mild, it is determined to be low-risk. For a specific patient with an INR of 2.3, currently using warfarin, with no recent active bleeding and no planned high-risk surgery, the overall bleeding risk level is determined to be intermediate-high-risk.

[0029] The assessment process for contraindications to mechanical prophylaxis begins with the collection of physical examination data. Nursing staff assess lower limb signs upon patient admission, obtaining information about the lower limb status through observation and palpation. The assessment includes skin integrity, degree of limb swelling, skin temperature, and color. Nursing staff enter the assessment results into the physical examination module of the electronic medical record, storing this information in structured fields. For example, if a patient's lower limb assessment record shows an 8 cm long skin laceration on the lateral side of the left calf, with the wound reaching the dermis and surrounding skin redness and swelling extending approximately 12 cm in diameter, this information would be stored as a positive lower limb injury status field, with the injury location being the lateral side of the left calf and the severity of the injury being moderate.

[0030] Lower extremity edema was assessed using a pitting edema grading system. Caregivers pressed the skin in front of the patient's tibia with their thumb for approximately 5 seconds and observed the indentation. A pit depth of 4 mm and a recovery time exceeding 30 seconds were recorded as grade II edema. One patient exhibited a 6 mm deep pit after pressing on both lower legs, with a recovery time of 45 seconds, and the skin was taut and shiny; this indicated significant lower extremity edema. The severity of the edema was also assessed by considering the patient's weight changes; a weight gain exceeding 2 kg within 24 hours accompanied by lower extremity edema was graded higher.

[0031] Vascular status assessment requires obtaining the patient's past medical history and imaging results. Lower extremity vascular ultrasound reports are extracted from the medical records, and the descriptive text regarding the degree of arterial stenosis and wall calcification is analyzed. When the report describes extensive arterial wall calcification and luminal stenosis exceeding 50%, the arteriosclerosis is marked as severe. For example, a patient's lower extremity arterial ultrasound report showed irregular thickening of the femoral and popliteal artery walls, with multiple calcified plaques, and the narrowest point of the lumen diameter was only 40% of the normal value, confirming severe arteriosclerosis.

[0032] Skin condition assessment focuses on skin integrity and infection status. Skin examination data is extracted from nursing records, including the presence of ulcers, sores, and foci of infection. A skin breakdown is defined as a ulcer larger than 3 cm in diameter on the patient's lower leg, with visible granulation tissue at the ulcer base, surrounded by redness and exudate. For example, a diabetic patient had a 5 cm diameter ulcer on their foot, reaching the muscle layer, accompanied by purulent discharge; this was recorded as a severe skin breakdown.

[0033] All collected vital sign data were compared item by item with the contraindication criteria. The comparison rules set threshold standards for each vital sign data point; when any vital sign data point reached or exceeded the contraindication threshold, the corresponding contraindication was determined to exist. In lower limb trauma, a contraindication was triggered when the injury involved full-thickness skin rupture or a wound length exceeding 5 cm. In lower limb edema, a contraindication was triggered when the edema reached grade II or higher and its extent exceeded one-third of the calf circumference. In arteriosclerosis, a contraindication was triggered when vascular stenosis exceeded 50% or extensive calcification was present. In skin rupture, a contraindication was triggered when there was an open wound larger than 2 cm in diameter or active skin infection.

[0034] The comprehensive assessment process for a particular patient was as follows: The patient had an 8 cm laceration on their left calf, meeting contraindications; grade II edema covering the entire circumference of both lower extremities, meeting contraindications; 60% stenosis of the lower extremity arteries, meeting contraindications; and a 5 cm foot ulcer, meeting contraindications. Therefore, the patient was deemed contraindicated for mechanical prophylaxis and was not suitable for the use of intermittent pneumatic compression devices or graduated compression stockings. A contraindication assessment report was generated, detailing the specific data triggering each contraindication, providing a basis for clinicians to develop preventative treatment plans.

[0035] In one optional implementation, based on the combined status of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, a thromboembolism prevention plan for the patient is generated through preset prevention strategy decision rules, including: The combined status of the thromboembolic risk level, bleeding risk level, and contraindication to mechanical prophylaxis is encoded as a combined status identifier, which uniquely represents a specific combination of the thromboembolic risk level, bleeding risk level, and contraindication to mechanical prophylaxis. Retrieve the prevention strategy item corresponding to the combined state identifier from the preset prevention strategy decision rules. The prevention strategy item defines the type and intensity of prevention measures for a specific combined state. The prevention method of the thrombosis prevention plan is determined according to the type of prevention measure in the prevention strategy item. The prevention method includes basic prevention, pharmacological prevention, mechanical prevention, and combined prevention. The execution parameters for the thrombosis prevention plan are generated based on the intensity of the preventive measures in the prevention strategy items and the patient's clinical data. The execution parameters include the timing of initiation of the preventive measures, the duration, and the monitoring frequency. Based on the prevention method and the execution parameters, a thrombosis prevention plan for the patient is constructed.

[0036] During the generation of thrombosis prevention measures, the system receives the results of the thromboembolic risk level, bleeding risk level, and contraindication for mechanical prophylaxis calculated by the risk assessment module. The thromboembolic risk level is divided into three categories: low, intermediate, and high risk; the bleeding risk level is also divided into three categories: low, intermediate, and high risk; and the contraindication for mechanical prophylaxis is a binary state of yes or no. The assessment results of these three dimensions are combined and coded using a three-digit coding method to generate a combined state identifier. The specific coding rules are as follows: low-risk thromboembolic risk level is coded as 1, intermediate as 2, and high as 3; low-risk bleeding risk level is coded as 1, intermediate as 2, and high as 3; and contraindication for mechanical prophylaxis is coded as no (0) or yes (1). When a patient's thromboembolic risk level is high, bleeding risk level is low, and contraindication for mechanical prophylaxis is no, the generated combined state identifier is 310. This coding method maps all possible risk combinations to a unique three-digit identifier, representing a total of eighteen different combinations.

[0037] A pre-configured prevention strategy decision rule base is established, using the combined state identifier as the index key to store the prevention strategy items corresponding to each combined state. Each prevention strategy item includes a prevention measure type field and a prevention measure intensity field. The prevention measure type field defines four types: basic prevention, pharmacological prevention, mechanical prevention, and combined prevention. Basic prevention guides patients to engage in early mobilization and ensure adequate hydration; pharmacological prevention uses anticoagulants or antiplatelet drugs; mechanical prevention uses intermittent pneumatic compression devices or graduated compression stockings; and combined prevention is a combination of pharmacological and mechanical prevention. The prevention measure intensity field uses three levels: standard intensity, enhanced intensity, and reduced intensity. When a patient with a combined state identifier of 310 is retrieved, the matched prevention strategy item is the combined prevention type and the enhanced intensity level.

[0038] After determining the type of prophylactic measure, execution parameters are generated based on the intensity of the measure and the patient's clinical data. These parameters include three core elements: initiation timing, duration, and monitoring frequency. Initiation timing is determined based on the patient's surgery or admission time. For standard-intensity prophylaxis, the initiation time is set at 12 hours post-surgery; for enhanced-intensity prophylaxis, it is set at 6 hours post-surgery; and for reduced-intensity prophylaxis, it is set at 24 hours post-surgery. The duration is calculated by comprehensively considering the patient's length of hospital stay, recovery of mobility, and type of underlying disease. For major orthopedic surgery with enhanced-intensity prophylaxis, the duration is set at 35 days; for general surgery with standard-intensity prophylaxis, the duration is set at 7 to 14 days. Monitoring frequency is dynamically adjusted based on the prophylactic intensity and bleeding risk level. For patients using enhanced-intensity prophylaxis with a medium-risk bleeding condition, coagulation function monitoring is set every three days, and complete blood count monitoring is set every two days.

[0039] A specific thromboprophylaxis plan was generated for the patient with a combined status marker of 310. This patient had a high risk of thromboembolism, a low risk of bleeding, and no contraindications to mechanical prophylaxis. The prophylactic approach was determined to be combined prophylaxis, employing both pharmacological and mechanical methods. For pharmacological prophylaxis, low molecular weight heparin calcium was administered subcutaneously at a dose of 5,000 IU once daily, initiated six hours post-surgery. For mechanical prophylaxis, an intermittent pneumatic compression device was used on both lower extremities for at least 18 hours daily, initiated immediately post-surgery. The duration of pharmacological prophylaxis was 35 days, continuing after discharge, while mechanical prophylaxis continued until the patient was able to ambulate independently for more than six hours daily. Monitoring frequency was set as follows: coagulation function was tested every three days, including activated partial thromboplastin time and prothrombin time; complete blood count was performed every two days, focusing on platelet count and hemoglobin levels; and lower extremity vascular ultrasound was performed every seven days to assess for deep vein thrombosis.

[0040] For patients with a combined status of 231, the risk of thromboembolism is intermediate, the risk of bleeding is high, and mechanical prophylaxis is contraindicated. The retrieved prophylaxis strategies were basic prophylaxis type and standard intensity level. Basic prophylaxis was determined, instructing patients to begin ankle pump exercises six hours post-surgery, ten times per hour, and maintain a daily fluid intake of at least 2000 ml. Due to the contraindication to mechanical prophylaxis and the high risk of bleeding, pharmacological and mechanical prophylaxis measures were not recommended. The initiation time was set to begin immediately after the patient is conscious and vital signs are stable, and the duration was set to the entire hospital stay. The monitoring frequency was set to daily coagulation function testing, close observation of lower limb swelling, and daily measurement of lower limb circumference; the prophylaxis strategy was adjusted promptly if abnormalities were observed. An early warning mechanism was also set up to automatically trigger a reassessment and adjustment process for the prophylaxis strategy when the patient's bleeding risk level decreases or thromboembolism risk level increases, ensuring that the prophylactic measures are always matched to the patient's current risk status.

[0041] In one optional implementation, retrieving the prevention strategy item corresponding to the combined state identifier from the preset prevention strategy decision rules includes: The thromboembolism risk level status value is extracted from the combined status identifiers. The status value is compared with the first-level judgment condition of the preset prevention strategy decision rule. Based on the comparison result, the corresponding first-level decision branch is selected. Extract the bleeding risk level status value from the first-level decision branch, compare the status value with the second-level judgment conditions under that branch, and select the corresponding second-level decision branch based on the comparison result; Extract the state value of mechanical prevention taboo in the second-level decision branch, compare the state value with the third-level decision condition under the branch, and reach the corresponding termination node according to the comparison result; The prevention strategy item is read from the termination node. The prevention strategy item records the type and strength of the prevention measures for the combined state.

[0042] When retrieving prevention strategy items corresponding to combined status identifiers from the preset prevention strategy decision rules, a multi-level decision tree structure is required for step-by-step matching and filtering. This decision tree structure contains three levels of decision nodes, with each level branching based on different clinical indicator status values, ultimately reaching a termination node containing a specific prevention strategy item.

[0043] Upon receiving the combined status identifier, which contains the status values ​​of multiple clinical risk indicators for the patient, the data structure of the combined status identifier is in string form, where the status values ​​are arranged in a fixed order and connected by delimiters. Assuming the received combined status identifier is "High Risk - Moderate Bleeding - No Contraindications," this identifier contains three key status values, corresponding to the thromboembolism risk level, bleeding risk level, and contraindication to mechanical prophylaxis, respectively.

[0044] The first state value, representing the risk level of thromboembolism, is extracted from the combined state identifiers. This state value is extracted using a string parsing module, which locates the characters before the first delimiter and identifies the "high risk" state value. After extraction, this state value is compared with the first-level judgment criteria in the preset prevention strategy decision rules. The first-level judgment criteria include all possible values ​​for the thromboembolism risk level, such as "extremely high risk," "high risk," "moderate risk," and "low risk." The decision rules set corresponding first-level decision branches for each risk level, with each branch representing the path for further judgment under that risk level.

[0045] The comparison process is implemented through string matching, performing a complete match between the extracted status value "high risk" and each value in the first-level judgment criteria. When a "high risk" option is matched, the comparison result is confirmed as a successful match, and the user proceeds to the first-level decision branch marked "high risk" based on this result. This branch contains the subsequent judgment logic and decision nodes for high-risk patients. In another case, if the extracted status value is "moderate risk," the user will proceed to the "moderate risk branch," where the judgment criteria and decision path differ from those of the high-risk branch.

[0046] After entering the first-level decision branch, the second state value, namely the bleeding risk level, is extracted from the combined state identifier. The parsing module locates the character content between the first and second separators and identifies the state value "moderate bleeding". This state value is then compared with the second-level judgment criteria set under the current first-level decision branch. The second-level judgment criteria are specifically categorized based on bleeding risk, including value options such as "high bleeding risk", "moderate bleeding risk", and "low bleeding risk". It should be noted that the specific content and weight of the second-level judgment criteria may differ under different first-level decision branches.

[0047] Under the "High-Risk Branch," the second-level criteria for classifying bleeding risk are more detailed because patients at high thrombotic risk need to balance thrombosis prevention and bleeding risk. Matching "Moderate Bleeding" with the second-level criteria under this branch, if a match is successful, indicates selection into the second-level decision branch labeled "High-Risk - Moderate Bleeding." This branch represents a clinical situation where the patient has both high thrombotic risk and moderate bleeding risk, requiring a prevention strategy that considers both risks. If another patient's bleeding risk status is "High Bleeding Risk," they would enter the "High-Risk - High Bleeding Branch," where the prevention strategy will more cautiously consider bleeding risk.

[0048] Upon entering the second-level decision branch, the third state value, namely the state value of the mechanical taboo, is extracted from the combined state identifier. The parsing module locates the characters after the second separator and identifies the state value as "no taboo". The mechanical taboo state typically includes three values: "absolute taboo exists", "relative taboo exists", and "no taboo". The extracted "no taboo" state value is compared with the third-level decision condition under the current second-level decision branch. The third-level decision condition is the last layer of decision logic in the decision tree, used to determine whether mechanical preventive measures can be used and the restrictions on their use.

[0049] Under the "High-Risk - Moderate Bleeding Branch," the Level 3 decision criterion assesses the applicability of mechanical prophylaxis. Matching "No Contraindications" with the Level 3 decision criterion for this branch shows that the patient has no contraindications to mechanical prophylaxis. Based on this comparison, the decision tree path leads to the corresponding termination node, marked as the "High-Risk - Moderate Bleeding - No Contraindications Termination Node." If the patient has an absolute contraindication to mechanical prophylaxis, with a status value of "Absolute Contraindication Exists," the path will reach the "High-Risk - Moderate Bleeding - Absolute Contraindications Termination Node," where the prophylaxis strategy will exclude all mechanical prophylaxis measures.

[0050] Upon reaching the termination node, the pre-stored prevention strategy items are read from that node. These prevention strategy items are stored in the termination node as structured data, containing two core fields: Prevention Measure Type and Prevention Measure Intensity. The Prevention Measure Type field records the classification of prevention measures applicable to this combined state, such as "pharmacological prevention plus mechanical prevention," "pharmacological prevention only," or "mechanical prevention only." The Prevention Measure Intensity field records the implementation intensity level of the prevention measure, such as "standard intensity," "moderate intensity," or "enhanced intensity."

[0051] In the "High Risk - Moderate Bleeding - No Contraindications Termination Node" section, the prevention strategy item shows the prevention measure type as "pharmacological prophylaxis plus mechanical prophylaxis," and the prevention measure intensity as "standard intensity." This strategy item means that for patients with both high thrombotic risk and moderate bleeding risk who have no contraindications to mechanical prophylaxis, a combination of pharmacological and mechanical prophylaxis should be used, with the pharmacological dosage and mechanical prophylaxis equipment used at standard intensity. The prevention strategy item may also include more detailed implementation parameters, such as a specific drug recommendation of "low molecular weight heparin," a dosage of "once daily subcutaneous injection," and a mechanical prophylaxis recommendation of "intermittent pneumatic compression device," with a usage duration of "at least eighteen hours daily."

[0052] In contrast, the prevention strategy retrieved at the "High Risk - Moderate Bleeding - Absolute Contraindication Termination Node" section showed the prevention measure type as "Pharmaceutical Prophylaxis Only" and the prevention measure intensity as "Moderate Intensity." Due to the absolute contraindication of mechanical prophylaxis, this strategy completely excludes mechanical prophylaxis, relying solely on medication for thrombosis prevention, with a relatively reduced medication intensity to balance the bleeding risk. The entire search process, through a three-level progressive screening, ensured that the most suitable prevention strategy was matched for each specific combination of patient conditions, achieving precision and individualization in clinical decision-making.

[0053] In one optional implementation, the currently implemented thrombosis prevention plan is compared with the thrombosis prevention plan regenerated based on the changed state. If they are inconsistent, a prevention measure adjustment reminder message is generated, including: Extract the current prevention method and current execution parameters from the currently implemented thrombosis prevention measure plan, and extract the regenerated prevention method and regenerated execution parameters from the thrombosis prevention measure plan regenerated based on the changed state; The current prevention method is compared with the regenerated prevention method to determine whether the prevention methods are consistent, and the comparison result of the prevention methods is obtained. The current execution parameters are compared with the regenerated execution parameters to determine whether the execution parameters are consistent, and the execution parameter comparison result is obtained. A consistency determination result is determined based on the comparison results of the prevention methods and the comparison results of the execution parameters. When the comparison results of the prevention methods or the comparison results of the execution parameters show inconsistency, the consistency determination result is determined to be inconsistent. When the consistency determination result is inconsistent, a preventive measure adjustment reminder message is generated based on the prevention method comparison result and the execution parameter comparison result. The preventive measure adjustment reminder message indicates the preventive measure content that needs to be adjusted.

[0054] In the dynamic adjustment of thrombosis prevention measures, when a patient's condition changes, a consistency comparison needs to be performed between the currently implemented thrombosis prevention plan and a plan regenerated based on the changed condition. The plan record is read from the database of currently implemented thrombosis prevention plans. This record contains a prevention method identifier field and an execution parameter set field. The prevention method identifier field stores numerical codes that identify different types of prevention measures; for example, value 1 indicates mechanical prevention, value 2 indicates pharmacological prevention, and value 3 indicates comprehensive prevention. The execution parameter set field stores specific execution details in structured data format. For mechanical prevention, this includes equipment type parameters, pressure value parameters, usage duration parameters, and usage frequency parameters; for pharmacological prevention, this includes drug name parameters, dosage parameters, route of administration parameters, and administration frequency parameters.

[0055] The current plan's data structure is parsed, and the prevention method identifier is extracted and stored in the current prevention method variable. Assuming the extracted current prevention method identifier is 2, it indicates that drug prevention is currently used. The execution parameter set is then extracted, and the values ​​of each parameter are parsed. Taking drug prevention as an example, the drug name parameter is extracted from the execution parameter set to obtain low molecular weight heparin, the dosage parameter to obtain 4,000 units, the route of administration parameter to obtain subcutaneous injection, and the frequency of administration parameter to obtain once daily. These parameter values ​​are stored in their corresponding current execution parameter variables, forming the current execution parameter group.

[0056] Data is extracted from the thrombosis prevention protocol regenerated based on the modified state in the same manner. The data record of the newly generated protocol is accessed, and its prevention method identifier field is read to obtain the regenerated prevention method. Assuming the regenerated prevention method identifier is 2, it also indicates drug prevention. The execution parameter set field of the new protocol is read, and the drug name parameter is extracted to obtain low molecular weight heparin, the dosage parameter to obtain 6,000 units, the route of administration parameter to obtain subcutaneous injection, and the frequency of administration parameter to obtain twice daily. These parameter values ​​are stored in the regenerated execution parameter variable, forming the regenerated execution parameter set.

[0057] The prevention method comparison module is activated, comparing the current prevention method variable with the newly generated prevention method variable. The comparison logic determines whether the values ​​of the two variables are equal. If the values ​​are equal, the prevention methods are consistent; otherwise, the prevention methods are inconsistent. In this case, the current prevention method identifier is 2, and the newly generated prevention method identifier is also 2. Since their values ​​are equal, the prevention methods are determined to be consistent, and the comparison result is marked as consistent, specifically stored as a Boolean value of true.

[0058] The execution parameter comparison module is activated to compare the current execution parameter group with the regenerated execution parameter group item by item. The comparison module iterates through each parameter field in the execution parameter group and compares them sequentially. For the drug name parameter, the current value is low molecular weight heparin, and the regenerated value is also low molecular weight heparin; the two strings match exactly, so the comparison result for this parameter is consistent. For the dosage parameter, the current value is 4,000 units, and the regenerated value is 6,000 units; the two values ​​are different, so the comparison result for this parameter is inconsistent. For the route of administration parameter, the current value is subcutaneous injection, and the regenerated value is also subcutaneous injection; the two match, so the comparison result for this parameter is consistent. For the frequency of administration parameter, the current value is once daily, and the regenerated value is twice daily; the two values ​​are different, so the comparison result for this parameter is inconsistent.

[0059] The comparison results of each parameter are summarized. The parameter comparison logic determines the overall comparison result as inconsistent if any single parameter is inconsistent. In this case, both the dosage parameter and the dosing frequency parameter show inconsistencies, so the comparison result is marked as inconsistent and stored as a Boolean value (false). The parameter comparison module also records the identifiers of the specific inconsistent parameter items and their corresponding current and new values, generating a detailed list of inconsistent parameters. This list includes the current value of the dosage parameter (4,000 units) and the new value (6,000 units), as well as the current value of the dosing frequency parameter (once daily) and the new value (twice daily).

[0060] Upon entering the consistency determination phase, two status values ​​are read: the prevention method comparison result and the execution parameter comparison result. The determination logic uses a logical OR operation rule. When either the prevention method comparison result or the execution parameter comparison result is inconsistent, the consistency determination result is inconsistent. In this case, a true prevention method comparison result indicates consistency, but a false execution parameter comparison result indicates inconsistency. According to the logical OR operation rule, the final consistency determination result is determined to be inconsistent. This result is stored and triggers the subsequent reminder information generation process.

[0061] Upon detecting an inconsistency in the consistency assessment result, the preventive measure adjustment reminder information generation module is activated. This module reads the prevention method comparison result, the execution parameter comparison result, and the detailed list of inconsistent parameters. The information generation logic determines whether the prevention method type needs to be changed based on the prevention method comparison result. In this case, the prevention methods are consistent, so no prevention method change prompt is generated. The module continues to process the execution parameter comparison result. Due to the inconsistency in the execution parameters, it reads the difference information item by item from the inconsistent parameter detail list.

[0062] The information builder creates adjustment prompt entries for each inconsistent parameter. For dosage parameter differences, the generated prompt text includes the parameter name (dosage), current execution value (4,000 units), suggested adjustment value (6,000 units), and a change description (increase from 4,000 units to 6,000 units). For dosing frequency parameter differences, the generated prompt text includes the parameter name (dosing frequency), current execution value (once daily), suggested adjustment value (twice daily), and a change description (increase from once daily to twice daily). These prompt entries are ordered according to parameter importance, with drug dosage listed first as the key parameter.

[0063] All generated prompts are integrated into a complete preventative measure adjustment reminder data package. This data package includes a reminder type field indicating parameter adjustment, a patient identifier field linking to a specific patient record, original protocol identifier and new protocol identifier fields for traceability, an adjustment item list field storing all parameters requiring adjustment and their detailed information, and a generation timestamp field recording the reminder creation time. This data package is pushed to the reminder queue at the healthcare workstation, and the reminder message is displayed prominently on the user interface, prompting healthcare personnel that the currently implemented thrombosis prevention protocol requires parameter adjustments, specifically adjusting the drug dosage from 4,000 units to 6,000 units and the dosing frequency from once daily to twice daily, ensuring that the thrombosis prevention measures are compatible with the patient's current condition.

[0064] This invention relates to a thrombosis prevention system based on digital clinical precision execution, comprising: The first unit is used to acquire the patient's clinical data, analyze the clinical data based on preset thromboembolism risk assessment rules, and determine the patient's thromboembolism risk level. The second unit is used to determine the patient's bleeding risk level based on the patient's clinical data and pre-set bleeding risk assessment rules; and to determine whether the patient has contraindications to mechanical prophylaxis based on the patient's clinical data and pre-set contraindications assessment rules. The third unit is used to generate a thrombosis prevention plan for the patient based on the combination of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, using preset prevention strategy decision rules. The fourth unit is used to monitor changes in patients' clinical data in real time during hospitalization. When changes are detected in a patient's thromboembolism risk level, bleeding risk level, or contraindications to mechanical prophylaxis, the currently implemented thromboprophylaxis plan is compared with a newly generated thromboprophylaxis plan based on the changed status. If there is a discrepancy, a reminder message for adjusting the prophylaxis measures is generated. The fifth unit is used to push the prevention measure plan or the prevention measure adjustment reminder information to the clinical medical care terminal.

[0065] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0066] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0067] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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 or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A thrombosis prevention method based on digital clinical precision execution, characterized in that, include: Acquire the patient's clinical data, analyze the clinical data based on preset thromboembolism risk assessment rules, and determine the patient's thromboembolism risk level; The patient's bleeding risk level is determined based on clinical data and pre-defined bleeding risk assessment rules; whether the patient has contraindications to mechanical prophylaxis is determined based on clinical data and pre-defined contraindications assessment rules. Based on the combined status of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, a thrombosis prevention plan for the patient is generated through preset prevention strategy decision rules; Real-time monitoring of changes in patients' clinical data during hospitalization; when changes are detected in a patient's thromboembolism risk level, bleeding risk level, or contraindications to mechanical prophylaxis, the currently implemented thromboprophylaxis plan is compared with a newly generated thromboprophylaxis plan based on the changed status; if inconsistencies are found, a reminder to adjust the prophylaxis measures is generated. The aforementioned preventive measures plan or reminders for adjusting the preventive measures will be pushed to the clinical medical staff terminal.

2. The method according to claim 1, characterized in that, The clinical data includes medical records, examination and test results, medical orders, surgical records, and transfer records; The preventive measures program includes basic prevention, drug prevention, mechanical prevention, or a combination thereof.

3. The method according to claim 1, characterized in that, The patient's bleeding risk level is determined based on clinical data and pre-defined bleeding risk assessment rules; contraindications to mechanical prophylaxis are determined based on clinical data and pre-defined contraindications assessment rules, including: Feature data related to bleeding risk is extracted from the clinical data, and the feature data is matched with risk factors in the preset bleeding risk assessment rules. The risk factors include active bleeding status, coagulation dysfunction status, anticoagulant use status, and high-risk surgical status to obtain matching results. Based on the presence status of each risk factor in the matching results, the patient's bleeding risk level is determined through a preset risk level determination logic, which defines the mapping relationship between the presence status of risk factors and the bleeding risk level. The process of determining whether a patient has contraindications to mechanical prophylaxis based on their clinical data and pre-defined contraindication assessment rules includes: From the clinical data, vital signs data related to contraindications for mechanical prophylaxis were extracted, including lower limb condition information, vascular condition information, and skin condition information. The vital signs data are compared with the contraindication criteria in the preset mechanical prevention contraindication assessment rules. The contraindication criteria include lower limb trauma, lower limb edema, arteriosclerosis, and skin damage. The presence or absence of contraindications for mechanical prophylaxis in the patient is determined by comparing the vital signs data with the contraindication criteria.

4. The method according to claim 1, characterized in that, Based on the combined status of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, a thrombosis prevention plan for the patient is generated through preset prevention strategy decision rules, including: The combined status of the thromboembolic risk level, bleeding risk level, and contraindication to mechanical prophylaxis is encoded as a combined status identifier, which uniquely represents a specific combination of the thromboembolic risk level, bleeding risk level, and contraindication to mechanical prophylaxis. Retrieve the prevention strategy item corresponding to the combined state identifier from the preset prevention strategy decision rules. The prevention strategy item defines the type and intensity of prevention measures for a specific combined state. The prevention method of the thrombosis prevention plan is determined according to the type of prevention measure in the prevention strategy item. The prevention method includes basic prevention, pharmacological prevention, mechanical prevention, and combined prevention. The execution parameters for the thrombosis prevention plan are generated based on the intensity of the preventive measures in the prevention strategy items and the patient's clinical data. The execution parameters include the timing of initiation of the preventive measures, the duration, and the monitoring frequency. Based on the prevention method and the execution parameters, a thrombosis prevention plan for the patient is constructed.

5. The method according to claim 4, characterized in that, Retrieving the prevention strategy item corresponding to the combined state identifier from the preset prevention strategy decision rules includes: The thromboembolism risk level status value is extracted from the combined status identifiers. The status value is compared with the first-level judgment condition of the preset prevention strategy decision rule. Based on the comparison result, the corresponding first-level decision branch is selected. Extract the bleeding risk level status value from the first-level decision branch, compare the status value with the second-level judgment conditions under that branch, and select the corresponding second-level decision branch based on the comparison result; Extract the state value of mechanical prevention taboo in the second-level decision branch, compare the state value with the third-level decision condition under the branch, and reach the corresponding termination node according to the comparison result; The prevention strategy item is read from the termination node. The prevention strategy item records the type and strength of the prevention measures for the combined state.

6. The method according to claim 1, characterized in that, The currently implemented thrombosis prevention plan is compared with the regenerated thrombosis prevention plan based on the changed state. If they are inconsistent, a prevention measure adjustment reminder message is generated, including: Extract the current prevention method and current execution parameters from the currently implemented thrombosis prevention measure plan, and extract the regenerated prevention method and regenerated execution parameters from the thrombosis prevention measure plan regenerated based on the changed state; The current prevention method is compared with the regenerated prevention method to determine whether the prevention methods are consistent, and the comparison result of the prevention methods is obtained. The current execution parameters are compared with the regenerated execution parameters to determine whether the execution parameters are consistent, and the execution parameter comparison result is obtained. A consistency determination result is determined based on the comparison results of the prevention methods and the comparison results of the execution parameters. When the comparison results of the prevention methods or the comparison results of the execution parameters show inconsistency, the consistency determination result is determined to be inconsistent. When the consistency determination result is inconsistent, a preventive measure adjustment reminder message is generated based on the prevention method comparison result and the execution parameter comparison result. The preventive measure adjustment reminder message indicates the preventive measure content that needs to be adjusted.

7. A thrombosis prevention system based on digital clinical precision execution, used to implement the method as described in any one of claims 1-6, characterized in that, include: The first unit is used to acquire the patient's clinical data, analyze the clinical data based on preset thromboembolism risk assessment rules, and determine the patient's thromboembolism risk level. The second unit is used to determine the patient's bleeding risk level based on the patient's clinical data and pre-set bleeding risk assessment rules; and to determine whether the patient has contraindications to mechanical prophylaxis based on the patient's clinical data and pre-set contraindications assessment rules. The third unit is used to generate a thrombosis prevention plan for the patient based on the combination of the thromboembolic risk level, bleeding risk level, and contraindications to mechanical prophylaxis, using preset prevention strategy decision rules. The fourth unit is used to monitor changes in patients' clinical data in real time during hospitalization. When changes are detected in a patient's thromboembolism risk level, bleeding risk level, or contraindications to mechanical prophylaxis, the currently implemented thromboprophylaxis plan is compared with a newly generated thromboprophylaxis plan based on the changed status. If there is a discrepancy, a reminder message for adjusting the prophylaxis measures is generated. The fifth unit is used to push the prevention measure plan or the prevention measure adjustment reminder information to the clinical medical care terminal.

8. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.

Images