Method and system for evaluating thrombus risk before internal arteriovenous fistula operation

Abstract

The invention provides a method and a system for evaluating thrombus risk before an internal arteriovenous fistula operation, and relates to the field of medical data processing, and the method comprises the following steps: based on the medical history information, treatment information and postoperative thrombus monitoring information of a plurality of internal arteriovenous fistula operation patients, evaluating the thrombus risk before the internal arteriovenous fistula operation; determining a plurality of thrombus risk related diseases, thrombus risk related operations and thrombus risk related drugs, and determining a plurality of internal arteriovenous fistula operation patient types; based on the medical history information and the treatment information of the patient to be subjected to the internal arteriovenous fistula operation, the type of the patient to be subjected to the internal arteriovenous fistula operation is determined, and preoperative detection information of the patient to be subjected to the internal arteriovenous fistula operation is obtained; and performing thrombus risk assessment before the internal arteriovenous fistula operation by combining the key preoperative detection factor corresponding to the type of the internal arteriovenous fistula operation patient and the influence weight corresponding to the key preoperative detection factor, thereby improving the efficiency and accuracy of thrombus risk assessment before the internal arteriovenous fistula operation.

Description

Technical Field

[0001] This invention relates to the field of medical data processing, and in particular to a method and system for assessing the risk of thrombosis before arteriovenous fistula surgery. Background Technology

[0002] Autologous arteriovenous fistulas (AVFs) are the primary dialysis access for patients undergoing maintenance hemodialysis (MHD). They ensure adequate blood flow and serve as a lifeline for these patients; therefore, their proper functioning is crucial for successful hemodialysis. While AVFs offer advantages such as a relatively longer lifespan and lower maintenance costs, their lifespan is not permanent. Fistula thrombosis and vascular stenosis are the primary and leading causes of AVF dysfunction.

[0003] In the absence of standardized assessment tools, doctors primarily rely on personal experience and intuition to assess the risk of thrombosis after arteriovenous fistula surgery. This assessment method is often influenced by the doctor's subjective judgment, which may lead to inaccurate or inconsistent risk assessments.

[0004] Therefore, there is a need to provide a method and system for assessing the risk of thrombosis before arteriovenous fistula surgery, in order to improve the efficiency and accuracy of thrombosis risk assessment before arteriovenous fistula surgery. Summary of the Invention

[0005] This invention provides a method and system for assessing thrombotic risk before arteriovenous fistula (AVF) surgery. The method includes: acquiring medical information from multiple AVF surgery patients, wherein the medical information includes at least medical history, treatment information, preoperative testing information, and postoperative thrombotic monitoring information; identifying multiple thrombotic risk-related diseases, multiple thrombotic risk-related surgeries, and multiple thrombotic risk-related medications based on the medical history, treatment information, and postoperative thrombotic monitoring information of the multiple AVF surgery patients; identifying multiple AVF surgery patient types based on the medical information of the multiple AVF surgery patients and the multiple thrombotic risk-related diseases, multiple thrombotic risk-related surgeries, and multiple thrombotic risk-related medications; and for each AVF surgery patient type, based on the arteriovenous fistula... The preoperative testing information and postoperative thrombosis monitoring information of multiple arteriovenous fistula (AVF) surgery patients were used to determine the key preoperative testing factors and their corresponding influence weights for each AVF surgery patient type. Based on the medical history and treatment information of patients to be operated on, the AVF surgery patient type was determined. Based on the AVF surgery patient type to which the patient to be operated on was identified, the preoperative testing information of the patients to be operated on was obtained. Based on the preoperative testing information of the patients to be operated on and the key preoperative testing factors and their corresponding influence weights for each AVF surgery patient type, a preoperative thrombosis risk assessment was performed for AVF surgery.

[0006] Furthermore, based on the medical history, treatment information, and postoperative thrombosis monitoring information of multiple arteriovenous fistula (AVF) surgeries, various thrombosis risk-related diseases, surgeries, and medications are identified, including: identifying multiple diseases, surgeries, and medications to be screened; for each disease to be screened, based on the medical history and postoperative thrombosis monitoring information of the multiple AVF surgeries, determining the thrombosis risk impact coefficient of the disease to be screened, and determining whether the disease to be screened is a thrombosis risk-related disease based on the thrombosis risk correlation coefficient of the disease to be screened; for each surgery to be screened, based on the treatment information and postoperative thrombosis monitoring information of the multiple AVF surgeries, determining the thrombosis risk impact coefficient of the surgery to be screened, and determining whether the surgery to be screened is a thrombosis risk-related surgery based on the thrombosis risk correlation coefficient of the surgery to be screened; for each medication to be screened, based on the treatment information and postoperative thrombosis monitoring information of the multiple AVF surgeries, determining the thrombosis risk impact coefficient of the medication to be screened, and determining whether the medication to be screened is a thrombosis risk-related medication based on the thrombosis risk correlation coefficient of the medication to be screened.

[0007] Furthermore, based on the medical information of multiple arteriovenous fistula (AVF) surgeries and various thrombosis-related diseases, surgeries, and medications, multiple patient types were identified, including: identifying the thrombosis-related surgeries and medications associated with each thrombosis-related disease; determining the combined thrombosis risk coefficients among various thrombosis-related diseases, surgeries, and medications based on the medical information of multiple AVF surgeries; and identifying multiple patient types based on these combined thrombosis risk coefficients among various thrombosis-related diseases, surgeries, and medications.

[0008] Furthermore, based on the medical information of multiple arteriovenous fistula (AVF) surgeries, the joint influence coefficients of thrombosis risk among various thrombosis-related diseases, the joint influence coefficients of thrombosis risk among various thrombosis-related surgeries, and the joint influence coefficients of thrombosis risk among various risk-related drugs are determined. This includes: determining the joint influence coefficient of disease risk between any two thrombosis-related diseases based on the medical history and postoperative thrombosis monitoring information of multiple AVF surgeries; determining multiple thrombosis-related disease units based on the joint influence coefficients of disease risk between any two thrombosis-related diseases, wherein each thrombosis-related disease unit includes two thrombosis-related diseases; and for each thrombosis-related disease unit, determining the joint influence coefficients of surgical risk for any two thrombosis-related surgeries and the joint influence coefficients of drug risk for any two thrombosis-related drugs corresponding to each thrombosis-related disease unit, based on the thrombosis-related surgeries and thrombosis-related drugs associated with each thrombosis-related disease, as well as the treatment information and postoperative thrombosis monitoring information of multiple AVF surgeries.

[0009] Furthermore, based on the joint influence coefficients of thrombosis risk among multiple thrombosis-related diseases, multiple thrombosis-related surgeries, and multiple thrombosis-related medications, various patient types are identified. This includes: for each thrombosis-related disease unit, determining the unit joint influence coefficient based on the joint influence coefficients of disease risk between the two thrombosis-related diseases included in the unit, the joint influence coefficients of surgical risk for any two thrombosis-related surgeries, and the joint influence coefficients of drug risk for any two thrombosis-related medications corresponding to the unit; grouping multiple thrombosis-related diseases according to the unit joint influence coefficients for each thrombosis-related disease unit to determine multiple thrombosis-related disease groups; and identifying multiple patient types based on the multiple thrombosis-related disease groups.

[0010] Furthermore, based on the preoperative testing information of multiple arteriovenous fistula (AVF) surgery patients included in the aforementioned AVF surgery patient types, key preoperative testing factors and their corresponding influence weights are determined for each AVF surgery patient type. This includes: determining multiple preoperative testing factors to be screened for each AVF surgery patient type based on the thrombosis risk-related disease group corresponding to that AVF surgery patient type; determining the thrombosis risk influence coefficient of each preoperative testing factor to be screened for each AVF surgery patient type based on the preoperative testing information and postoperative thrombosis monitoring information of multiple AVF surgery patients included in the AVF surgery patient types; and determining the key preoperative testing factors and their corresponding influence weights for each AVF surgery patient type based on the thrombosis risk influence coefficient of each preoperative testing factor to be screened for each AVF surgery patient type.

[0011] Furthermore, based on the thrombosis risk impact coefficient of each preoperative testing factor to be screened corresponding to the patient type of arteriovenous fistula surgery, the key preoperative testing factors and their corresponding impact weights are determined for each patient type of arteriovenous fistula surgery. This includes: determining the key preoperative testing factors for each patient type of arteriovenous fistula surgery based on the thrombosis risk impact coefficient and thrombosis risk impact threshold of each preoperative testing factor to be screened; and determining the impact weight of each key preoperative testing factor based on the thrombosis risk impact coefficient of each key preoperative testing factor for each patient type of arteriovenous fistula surgery.

[0012] Furthermore, based on the preoperative testing information of patients undergoing arteriovenous fistula (AVF) surgery and the key preoperative testing factors and their corresponding influence weights for each AVF patient type, a preoperative thrombotic risk assessment is conducted. This includes: for each AVF patient type, establishing a risk assessment model corresponding to the AVF patient type based on the key preoperative testing factors and their corresponding influence weights; and using the risk assessment model corresponding to the AVF patient type to which the patient belongs, a preoperative thrombotic risk assessment is conducted based on the preoperative testing information of the patients undergoing AVF surgery.

[0013] Furthermore, using a risk assessment model corresponding to the arteriovenous fistula (AVF) patient type to which the patient belongs, and based on the preoperative testing information of the patient, a preoperative thrombotic risk assessment for AVF surgery is performed. This includes: determining the matching value of the AVF patient type to which the patient belongs based on the patient's medical history and treatment information; determining the thrombotic risk value for AVF surgery using the risk assessment model corresponding to the AVF patient type to which the patient belongs, based on the preoperative testing information of the patient; and determining the comprehensive thrombotic risk value for AVF surgery using a comprehensive assessment model based on the matching value of each AVF patient type to which the patient belongs and the thrombotic risk value output by the corresponding risk assessment model.

[0014] This invention provides a system for assessing thrombotic risk before arteriovenous fistula (AVF) surgery. It utilizes the aforementioned method for assessing thrombotic risk before AVF surgery, comprising: an information acquisition module for acquiring medical information of multiple AVF surgery patients, wherein the medical information includes at least medical history, treatment information, preoperative testing information, and postoperative thrombotic monitoring information; a correlation analysis module for identifying multiple thrombotic risk-related diseases, multiple thrombotic risk-related surgeries, and multiple thrombotic risk-related medications based on the medical history, treatment information, and postoperative thrombotic monitoring information of multiple AVF surgery patients; a type analysis module for identifying multiple AVF surgery patient types based on the medical information of multiple AVF surgery patients and the multiple thrombotic risk-related diseases, surgeries, and medications; and a factor analysis module for analyzing each type of AVF surgery. The system comprises several modules: a patient type determination module, a type identification module, and a risk assessment module. The first module is used to determine the preoperative detection factors and their corresponding influence weights for each arteriovenous fistula (AVF) surgery patient type, based on the patient's medical history and treatment information. The second module is used to determine the AVF surgery patient type to which the patient belongs. The third module is used to acquire the preoperative detection information of the patient based on the AVF surgery patient type. The fourth module is used to conduct a preoperative thrombosis risk assessment for the AVF surgery based on the preoperative detection information and the key preoperative detection factors and their corresponding influence weights.

[0015] Compared with existing technologies, the method and system for assessing the risk of thrombosis before arteriovenous fistula surgery provided by this invention have at least the following beneficial effects: 1. By analyzing the medical information of multiple patients, different patient types and their specific thrombosis risk factors can be identified. For patients awaiting surgery, more personalized thrombosis risk assessments can be conducted based on their patient type, thereby improving the accuracy of the assessment. By identifying key preoperative testing factors and their influence weights, targeted testing can be performed on patients before surgery. If a patient is found to have a high thrombosis risk, intervention measures can be taken in advance, such as adjusting medication, changing the surgical plan, or strengthening postoperative monitoring, thereby effectively preventing the occurrence of thrombosis and ensuring the safety and success rate of the surgery. By preventing the occurrence of thrombosis, surgical complications can be reduced, thereby improving the success rate of surgery and the speed of patient recovery.

[0016] 2. By determining the thrombosis risk impact coefficients of the selected diseases, surgeries, and medications, this method employs a scientific and quantitative approach to assess their association with thrombosis risk. It is more accurate than traditional experience-based judgments or subjective assumptions, and can more realistically reflect the impact of diseases, surgeries, and medications on thrombosis risk. It systematically considers the influence of multiple diseases, surgeries, and medications on thrombosis risk, rather than focusing on a single factor. This comprehensive assessment approach helps to gain a more holistic understanding of the patient's thrombosis risk status, providing stronger support for subsequent surgical decisions and risk management.

[0017] 3. By determining the combined influence coefficients among multiple thrombosis risk-related diseases, surgeries, and medications, this method enables more refined patient classification. This classification approach considers the interactions between multiple factors, making patient categorization more accurate and targeted. It considers not only the impact of a single disease, surgery, or medication on thrombosis risk but also their combined effects. This comprehensive assessment method more accurately reflects a patient's thrombosis risk status, providing more comprehensive information for surgical decisions and risk management. By identifying patient types and their corresponding key preoperative testing factors and influence weights, more accurate thrombosis risk assessment can be performed.

[0018] 4. By establishing specific risk assessment models for each type of arteriovenous fistula surgery patient, personalized assessment of thrombotic risk can be achieved. This personalized assessment considers the specific characteristics of the patient type and key preoperative detection factors, making the assessment results more accurate and targeted. Using the risk assessment model corresponding to the patient type, combined with the patient's preoperative detection information, the patient's thrombotic risk can be assessed more accurately. This assessment method reduces subjective judgment and reliance on experience, improving the objectivity and accuracy of the assessment. Through a comprehensive assessment model, the matching values ​​of multiple patient types and the thrombotic risk values ​​output by the corresponding risk assessment models can be considered together. This comprehensive assessment method can more comprehensively reflect the patient's thrombotic risk status, providing more comprehensive information for surgical decisions. Attached Figure Description

[0019] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein: Figure 1 This is a flowchart illustrating a method for assessing the risk of thrombosis before arteriovenous fistula surgery, according to some embodiments of this specification. Figure 2 This is a schematic diagram of a module for assessing the risk of thrombosis before arteriovenous fistula surgery, according to some embodiments of this specification. Detailed Implementation

[0020] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0021] Figure 1 This is a flowchart illustrating a method for assessing the risk of thrombosis before arteriovenous fistula surgery, as shown in some embodiments of this specification. Figure 1 As shown, a method for assessing the risk of thrombosis before arteriovenous fistula surgery may include the following procedure.

[0022] S101. Obtain medical information of multiple arteriovenous fistula surgery patients.

[0023] The medical information includes at least medical history, treatment information, preoperative testing information, and postoperative thrombosis monitoring information. Medical history may include past medical history and family medical history. Treatment information may include information on medications received for past medical history (e.g., drug name, dosage, and cumulative duration of medication) and surgical treatment information (e.g., surgical method, surgical date). Preoperative testing information may include biochemical tests, complete blood count, and vascular ultrasound results. Postoperative thrombosis monitoring information may include whether postoperative thrombosis occurred after arteriovenous fistula surgery.

[0024] As an example only, the medical information of patient A undergoing arteriovenous fistula surgery may include: 1. Medical history information: Past medical history: Male, 52 years old, with chronic renal failure for 5 years, which has progressed to uremia and requires long-term hemodialysis. He also has a history of hypertension for 12 years, which is currently well controlled.

[0025] Family medical history: Father died of diabetic nephropathy.

[0026] 2. Treatment Information: Medication information: Long-term use of antihypertensive drugs (such as amlodipine, 1 tablet daily, for a total of 12 years) and blood sugar control drugs (discontinued due to recent normalization of blood sugar).

[0027] Surgical treatment information: None.

[0028] 3. Preoperative testing information: Biochemical tests showed elevated serum creatinine and blood urea nitrogen, consistent with uremia; blood glucose was normal.

[0029] Blood routine test: Hemoglobin level is low, indicating mild anemia.

[0030] Vascular ultrasound findings: The radial artery and cephalic vein in the forearm have normal internal diameters, with no thrombosis or stenosis, making them suitable candidates for arteriovenous fistula surgery.

[0031] 4. Postoperative thrombosis monitoring information: No postoperative arteriovenous fistula thrombosis occurred, the fistula thrill and murmur were normal, and blood flow was unobstructed.

[0032] S102. Based on the medical history, treatment information, and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgeries, identify multiple thrombosis risk-related diseases, multiple thrombosis risk-related surgeries, and multiple thrombosis risk-related drugs.

[0033] Specifically, it includes: Identify multiple diseases to be screened (e.g., diabetes, hypertension, atherosclerosis, etc.), multiple surgeries to be screened (e.g., vascular reconstruction, vascular bypass, etc.), and multiple drugs to be screened (e.g., erythropoietin, anticoagulants, etc.). As an example only, multiple diseases to be screened, multiple surgeries to be screened, and multiple drugs to be screened can be identified manually or through big data analysis. For each disease to be screened, the thrombosis risk impact coefficient of the disease to be screened is determined based on the medical history information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients. The thrombosis risk correlation coefficient of the disease to be screened is used to determine whether the disease to be screened is a thrombosis risk-related disease. For example, the disease to be screened with a thrombosis risk impact coefficient greater than the first thrombosis risk impact coefficient threshold can be classified as a thrombosis risk-related disease. For each type of surgery to be screened, the thrombosis risk impact coefficient of the surgery to be screened is determined based on the treatment information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients. The thrombosis risk correlation coefficient of the surgery to be screened is used to determine whether the surgery to be screened is a thrombosis risk-related surgery. For example, the surgery to be screened with a thrombosis risk impact coefficient greater than the second thrombosis risk impact coefficient threshold can be classified as a thrombosis risk-related surgery. For each drug to be screened, the thrombosis risk impact coefficient of the drug is determined based on the treatment information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients. The thrombosis risk correlation coefficient of the drug is used to determine whether the drug is a thrombosis risk-related drug. For example, drugs with a thrombosis risk impact coefficient greater than the third thrombosis risk impact coefficient threshold can be classified as thrombosis risk-related drugs.

[0034] For example, the thrombosis risk impact coefficient for the disease to be screened can be calculated using the following formula: in, Let be the thrombosis risk impact coefficient for the i-th disease to be screened. Let be the label value for the i-th disease to be screened in the nth arteriovenous fistula surgery patient. If the i-th disease to be screened exists in the medical history information of the nth arteriovenous fistula surgery patient, then... =1, otherwise 0 Let be the marker value for the postoperative thrombosis in the nth arteriovenous fistula surgery patient. If thrombosis is present in the postoperative thrombosis monitoring information of the nth arteriovenous fistula surgery patient, then... =1, otherwise 0, representing the total number of patients who underwent arteriovenous fistula surgery.

[0035] The calculation methods for the thrombosis risk impact coefficient of the surgery to be screened and the thrombosis risk impact coefficient of the drug to be screened are the same as those for the disease to be screened, and will not be repeated here.

[0036] S103. Based on the medical information of multiple arteriovenous fistula surgery patients and multiple thrombosis risk-related diseases, multiple thrombosis risk-related surgeries, and multiple thrombosis risk-related drugs, identify multiple types of arteriovenous fistula surgery patients.

[0037] Specifically, it includes: Identify the thrombosis-related surgeries and medications associated with each thrombosis-related disease. Specifically, thrombosis-related surgeries and medications can be identified as procedures used to treat thrombosis-related diseases, and can be determined manually or through big data analysis. Based on the medical information of multiple arteriovenous fistula patients, we determined the joint influence coefficient of thrombosis risk among various thrombosis risk-related diseases, the joint influence coefficient of thrombosis risk among various thrombosis risk-related surgeries, and the joint influence coefficient of thrombosis risk among various risk-related drugs. Based on the combined influence coefficients of thrombosis risk among various thrombosis-related diseases, various thrombosis-related surgeries, and various thrombosis-related medications, multiple patient types were identified.

[0038] In some embodiments, based on medical information from multiple arteriovenous fistula surgeries, the joint influence coefficient of thrombosis risk among multiple thrombosis risk-related diseases, the joint influence coefficient of thrombosis risk among multiple thrombosis risk-related surgeries, and the joint influence coefficient of thrombosis risk among multiple risk-related medications are determined, including: Based on the medical history and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients, the joint risk coefficient between any two thrombosis-related diseases was determined. Based on the joint risk coefficient between any two thrombosis-related diseases, multiple thrombosis-related disease units are identified, wherein each thrombosis-related disease unit includes two thrombosis-related diseases. For each thrombosis-related disease unit, based on the thrombosis-related surgeries and medications associated with each thrombosis-related disease, as well as the treatment information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgeries, the joint surgical risk coefficient and the joint drug risk coefficient of any two thrombosis-related surgeries corresponding to the thrombosis-related disease unit are determined. For example, if the thrombosis-related disease unit includes thrombosis-related disease A and thrombosis-related disease B, and the thrombosis-related surgeries associated with thrombosis-related disease A include thrombosis-related surgeries C1 and C2, and the thrombosis-related surgeries associated with thrombosis-related disease B include thrombosis-related surgery C3, then the joint surgical risk coefficients of thrombosis-related surgeries C1 and C3, as well as the joint surgical risk coefficients of thrombosis-related surgeries C2 and C3, need to be calculated.

[0039] For example, based on postoperative thrombosis monitoring information of arteriovenous fistula (AVF) surgery patients, patients who develop thrombosis can be identified. Based on the medical history information of AVF surgery patients who develop thrombosis, the joint risk coefficient between any two thrombosis-related diseases can be determined.

[0040] The joint risk coefficient between two thrombosis-related diseases can be calculated using the following formula: in, Let be the joint risk coefficient between the i-th thrombosis-related disease and the j-th thrombosis-related disease. Let be the marker value for the i-th disease to be screened in the m-th arteriovenous fistula surgery patient with thrombosis. If the i-th disease to be screened exists in the medical history information of the m-th arteriovenous fistula surgery patient with thrombosis, then... =1, otherwise 0 Let be the marker value for the j-th disease to be screened in the m-th arteriovenous fistula surgery patient with thrombosis. If the j-th disease to be screened exists in the medical history information of the m-th arteriovenous fistula surgery patient with thrombosis, then... =1, otherwise 0, representing the total number of arteriovenous fistula surgeries that resulted in thrombosis.

[0041] The calculation methods for the joint surgical risk coefficient of any two thrombosis-related surgeries and the joint drug risk coefficient of any two thrombosis-related drugs are the same as those for the joint disease risk coefficient between two thrombosis-related diseases, and will not be repeated here.

[0042] In some embodiments, multiple patient types are identified based on the joint influence coefficient of thrombosis risk among multiple thrombosis-related diseases, the joint influence coefficient of thrombosis risk among multiple thrombosis-related surgeries, and the joint influence coefficient of thrombosis risk among multiple risk-related medications. These include: For each thrombosis-related disease unit, the unit risk joint influence coefficient is determined based on the joint influence coefficient of disease risk between the two thrombosis-related diseases included in the thrombosis-related disease unit, the joint influence coefficient of surgical risk of any two thrombosis-related surgeries and the joint influence coefficient of drug risk of any two thrombosis-related drugs corresponding to the thrombosis-related disease unit. Based on the joint influence coefficient of unit risk corresponding to each thrombosis risk-related disease unit, multiple thrombosis risk-related diseases can be grouped to determine multiple thrombosis risk-related disease groups. For example, a clustering algorithm can be used to group multiple thrombosis risk-related diseases based on the joint influence coefficient of unit risk corresponding to each thrombosis risk-related disease unit to determine multiple thrombosis risk-related disease groups. Multiple patient types were identified based on various thrombosis risk-related disease groups.

[0043] The joint risk coefficient of the unit risk for the disease unit related to thrombosis risk can be calculated using the following formula: in, Let be the joint risk coefficient of the unit corresponding to the k-th thrombosis risk-related disease unit. Let be the joint risk coefficient of the two thrombosis-related diseases included in the k-th thrombosis-related disease unit. The combined surgical risk coefficient for the k-th thrombosis-related disease unit, which includes the e-th type of thrombosis-related surgery associated with one thrombosis-related disease and the f-th type of thrombosis-related surgery associated with another thrombosis-related disease. Let E be the combined drug risk coefficient of the g-th thrombosis-related drug associated with one thrombosis-related disease and the h-th thrombosis-related drug associated with another thrombosis-related disease in the k-th thrombosis-related disease unit; let E be the total number of thrombosis-related surgeries associated with one thrombosis-related disease in the k-th thrombosis-related disease unit; let F be the total number of thrombosis-related surgeries associated with another thrombosis-related disease in the k-th thrombosis-related disease unit; let G be the total number of thrombosis-related drugs associated with one thrombosis-related disease in the k-th thrombosis-related disease unit; let H be the total number of thrombosis-related drugs associated with another thrombosis-related disease in the k-th thrombosis-related disease unit; and let w be the total number of thrombosis-related drugs associated with one thrombosis-related disease in the k-th thrombosis-related disease unit. 11 w 12 and w 13 For weights, w 11 w 12 and w 13 Greater than 0, w 11 +w 12 +w 13 =1.

[0044] Patient types can include patient types corresponding to each thrombosis risk-related disease group and patient types that do not correspond to any of the thrombosis risk-related disease groups. For example, multiple thrombosis risk-related disease groups include thrombosis risk-related disease group A (atherosclerosis, diabetic vascular disease, systemic lupus erythematosus), thrombosis risk-related disease group B (anticoagulant deficiency, hyperlipidemia), and thrombosis risk-related disease group C (congestive heart failure, valvular heart disease). Then, multiple patient types can include patient type A corresponding to thrombosis risk-related disease group A, patient type B corresponding to thrombosis risk-related disease group B, patient type C corresponding to thrombosis risk-related disease group C, and patient type D. When a patient's medical history includes atherosclerosis, diabetic vascular disease, systemic lupus erythematosus, etc. If a patient has at least one of the following conditions: systemic lupus erythematosus, then the patient's patient type includes patient type A. If a patient's medical history includes at least one of anticoagulant deficiency or hyperlipidemia, then the patient's patient type includes patient type B. If a patient's medical history includes at least one of congestive heart failure or valvular disease, then the patient's patient type includes patient type C. If a patient's medical history does not include atherosclerosis, diabetic vascular disease, systemic lupus erythematosus, anticoagulant deficiency, hyperlipidemia, congestive heart failure, or valvular disease, then the patient's patient type is patient type D.

[0045] Understandably, by comprehensively considering the combined influence coefficients of disease, surgery, and medication, this method can more precisely stratify the risk of thrombosis-related diseases. This stratification approach considers not only the risk contribution of a single factor but also the interaction between multiple factors, making risk stratification more accurate and detailed.

[0046] S104. For each type of arteriovenous fistula surgery patient, based on the preoperative testing information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients included in the arteriovenous fistula surgery patient type, determine the key preoperative testing factors corresponding to the arteriovenous fistula surgery patient type and the influence weights of the key preoperative testing factors.

[0047] Specifically, it includes: Based on the thrombosis risk-related disease groups corresponding to the types of patients undergoing arteriovenous fistula (AVF) surgery, multiple preoperative testing factors to be screened for AVF surgery patients are identified (e.g., prothrombin time, activated partial thromboplastin time, thrombin time, fibrinogen, whole blood viscosity, hematocrit, von Willebrand factor, soluble thrombomodulin, cardiac ejection fraction, etc.). These multiple preoperative testing factors to be screened for AVF surgery patients can be identified manually or through big data analysis. Based on the preoperative testing information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients, including the patient types of arteriovenous fistula surgery, the thrombosis risk impact coefficient of each preoperative testing factor to be screened for the corresponding patient type of arteriovenous fistula surgery was determined. Based on the thrombosis risk impact coefficient of each preoperative detection factor to be screened for arteriovenous fistula surgery patient type, the key preoperative detection factors and their corresponding impact weights for the key preoperative detection factors are determined.

[0048] As an example, the various preoperative detection factors to be screened for different types of arteriovenous fistula surgery patients can be shown in Table 1.

[0049] Table 1: Multiple preoperative detection factors to be screened for different types of arteriovenous fistula surgeries

[0050] The thrombosis risk influence coefficient of the preoperative testing factor to be screened can be calculated using the following formula: in, The thrombosis risk impact coefficient of the r-th preoperative detection factor to be screened, corresponding to the patient type for arteriovenous fistula surgery. Let r be the label value of the preoperative screening factor corresponding to the t-th arteriovenous fistula surgery patient, which is included in the patient types for arteriovenous fistula surgery. If there is an abnormal value of the r-th preoperative screening factor in the preoperative testing information of the t-th arteriovenous fistula surgery patient, then... =1, otherwise 0 Let be the marker value for the t-th arteriovenous fistula (AVF) surgery patient among the AVF surgery patient types. If a thrombus is present in the postoperative thrombosis monitoring information of the t-th AVF surgery patient among the AVF surgery patient types, then... =1, otherwise 0, representing the total number of arteriovenous fistula (AVF) patients included in the AVF patient types.

[0051] In some embodiments, based on the thrombosis risk impact coefficient of each preoperative screening factor corresponding to the patient type of arteriovenous fistula surgery, the key preoperative screening factors and their corresponding impact weights are determined, including: Based on the thrombosis risk impact coefficient and thrombosis risk impact threshold of each preoperative detection factor to be screened for arteriovenous fistula surgery patient type, the key preoperative detection factors corresponding to the arteriovenous fistula surgery patient type are determined. For example, the preoperative detection factors to be screened with a thrombosis risk impact coefficient greater than the thrombosis risk impact threshold can be used as key preoperative detection factors. Based on the thrombosis risk impact coefficient of key preoperative detection factors corresponding to each type of arteriovenous fistula surgery patient, the impact weight of each key preoperative detection factor is determined.

[0052] The influence weights of key preoperative testing factors can be calculated using the following formula: Where is the influence weight corresponding to the i-th key preoperative detection factor. Let be the thrombosis risk impact coefficient of the i-th key preoperative detection factor. denoted as the thrombosis risk impact coefficient of the d-th key preoperative detection factor, and denoted as the total number of key preoperative detection factors.

[0053] S105. Based on the medical history and treatment information of patients awaiting arteriovenous fistula surgery, determine the type of arteriovenous fistula surgery patient to which the patients awaiting arteriovenous fistula surgery belong.

[0054] Specifically, if the medical history of a patient undergoing arteriovenous fistula surgery contains a disease corresponding to a certain type of arteriovenous fistula surgery patient, then the patient is classified into that type of arteriovenous fistula surgery patient.

[0055] S106. Based on the type of arteriovenous fistula surgery patient to which the patient to be operated on is, obtain the preoperative testing information of the patient to be operated on.

[0056] Specifically, the key preoperative testing factors corresponding to the type of arteriovenous fistula surgery patient to which the patient to be operated on are identified, and the preoperative testing items are determined, thereby determining the preoperative testing information of the patient to be operated on. The preoperative testing information of the patient to be operated on may include the value of each key preoperative testing factor corresponding to the type of arteriovenous fistula surgery patient to which the patient to be operated on.

[0057] S107. Based on the preoperative testing information of patients undergoing arteriovenous fistula surgery and the key preoperative testing factors corresponding to the patient type of arteriovenous fistula surgery, as well as the influence weights of the key preoperative testing factors, conduct a preoperative thrombosis risk assessment for arteriovenous fistula surgery.

[0058] Specifically, it includes: For each type of arteriovenous fistula surgery patient, a risk assessment model is established based on the key preoperative detection factors and their corresponding influence weights. The risk assessment model can be a convolutional neural network model. Using a risk assessment model corresponding to the type of arteriovenous fistula (AVF) patient to which the patient to undergo AVF surgery belongs, and based on the preoperative test information of the patient to be AVF surgery, the risk of thrombosis before AVF surgery is assessed.

[0059] In some embodiments, a risk assessment model corresponding to the arteriovenous fistula (AVF) patient type to which the patient to undergo AVF surgery belongs is used to perform a preoperative thrombosis risk assessment of the AVF patient, based on the preoperative testing information of the patient, including: Based on the medical history and treatment information of patients undergoing arteriovenous fistula surgery, a matching value is determined to identify the patient type of arteriovenous fistula surgery to which the patient belongs. For example, a matching value determination model can be used to determine the matching value of the patient type of arteriovenous fistula surgery to which the patient belongs, based on the medical history and treatment information of the patient. The matching value determination model can be a convolutional neural network model. The risk of thrombosis during arteriovenous fistula surgery is determined by using a risk assessment model corresponding to the type of arteriovenous fistula surgery patient to which the patient to be undergoing arteriovenous fistula surgery belongs, based on the preoperative test information of the patients to be undergoing arteriovenous fistula surgery. The comprehensive risk value of arteriovenous fistula surgery is determined by using a comprehensive evaluation model based on the matching value of each arteriovenous fistula surgery patient type to which the patient to be operated on belongs and the thrombosis risk value of arteriovenous fistula surgery output by the corresponding risk assessment model. The comprehensive evaluation model can be a convolutional neural network model.

[0060] Figure 2This is a schematic diagram of a module for assessing the risk of thrombosis before arteriovenous fistula surgery, as shown in some embodiments of this specification. Figure 2 As shown, a system for assessing the risk of thrombosis before arteriovenous fistula surgery may include an information acquisition module, a correlation analysis module, a type analysis module, a factor analysis module, a type determination module, and a risk assessment module.

[0061] The information acquisition module is used to acquire medical information of multiple arteriovenous fistula surgery patients. The medical information includes at least medical history, treatment information, preoperative test information, and postoperative thrombosis monitoring information. The correlation analysis module is used to identify various thrombosis-related diseases, thrombosis-related surgeries, and thrombosis-related drugs based on the medical history, treatment information, and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients. The type analysis module is used to determine the type of multiple arteriovenous fistula surgery patients based on their medical information, multiple thrombosis-related diseases, multiple thrombosis-related surgeries, and multiple thrombosis-related medications. The factor analysis module is used to determine the key preoperative detection factors and their corresponding influence weights for each type of arteriovenous fistula surgery patient, based on the preoperative detection information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients included in the arteriovenous fistula surgery patient type. The type determination module is used to determine the type of arteriovenous fistula surgery patient to which the patient belongs based on the patient's medical history and treatment information. The information acquisition module is also used to acquire the preoperative test information of patients undergoing arteriovenous fistula surgery based on the type of arteriovenous fistula surgery patients to which the patients to be operated on belong; The risk assessment module is used to assess the risk of thrombosis before arteriovenous fistula surgery based on the preoperative test information of patients undergoing arteriovenous fistula surgery, the key preoperative test factors corresponding to the patient type of arteriovenous fistula surgery, and the influence weights of the key preoperative test factors.

[0062] A system for assessing the risk of thrombosis before arteriovenous fistula surgery can be used to perform a method for assessing the risk of thrombosis before arteriovenous fistula surgery, which will not be elaborated here.

[0063] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for assessing the risk of thrombosis before arteriovenous fistula surgery, characterized in that, include: Obtain medical information from multiple patients undergoing arteriovenous fistula surgery, wherein the medical information includes at least medical history, treatment information, preoperative testing information, and postoperative thrombosis monitoring information; Based on the medical history, treatment information, and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgeries, various thrombosis risk-related diseases, various thrombosis risk-related surgeries, and various thrombosis risk-related drugs were identified. Based on the medical information of multiple arteriovenous fistula surgery patients, as well as multiple thrombosis-related diseases, multiple thrombosis-related surgeries, and multiple thrombosis-related drugs, we identified multiple types of arteriovenous fistula surgery patients. For each type of arteriovenous fistula surgery patient, based on the preoperative testing information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients included in the arteriovenous fistula surgery patient type, the key preoperative testing factors corresponding to the arteriovenous fistula surgery patient type and the influence weights of the key preoperative testing factors are determined. Based on the medical history and treatment information of patients awaiting arteriovenous fistula surgery, the type of arteriovenous fistula surgery patient to which the patients awaiting arteriovenous fistula surgery belong is determined. Based on the type of arteriovenous fistula (AVF) patient to which the patient to undergo AVF surgery belongs, obtain the preoperative testing information of the patient to undergo AVF surgery. Based on the preoperative testing information of patients undergoing arteriovenous fistula surgery, the key preoperative testing factors corresponding to the patient type of arteriovenous fistula surgery, and the influence weights of the key preoperative testing factors, a thrombosis risk assessment is performed before arteriovenous fistula surgery.

2. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 1, characterized in that, Based on the medical history, treatment information, and postoperative thrombosis monitoring information of multiple patients who underwent arteriovenous fistula surgery, several thrombosis-related diseases, surgeries, and medications were identified, including: Identify multiple diseases, surgeries, and drugs to be screened; For each disease to be screened, the thrombosis risk impact coefficient of the disease to be screened is determined based on the medical history information and postoperative thrombosis monitoring information of the multiple arteriovenous fistula surgery patients, and the thrombosis risk correlation coefficient of the disease to be screened is used to determine whether the disease to be screened is a thrombosis risk-related disease. For each type of surgery to be screened, based on the treatment information and postoperative thrombosis monitoring information of the multiple arteriovenous fistula surgery patients, the thrombosis risk impact coefficient of the surgery to be screened is determined, and based on the thrombosis risk correlation coefficient of the surgery to be screened, it is determined whether the surgery to be screened is a thrombosis risk-related surgery. For each drug to be screened, the thrombosis risk impact coefficient of the drug to be screened is determined based on the treatment information and postoperative thrombosis monitoring information of the multiple arteriovenous fistula surgery patients. The thrombosis risk correlation coefficient of the drug to be screened is used to determine whether the drug to be screened is a thrombosis risk-related drug.

3. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 1, characterized in that, Based on medical information from multiple arteriovenous fistula surgeries, various thrombosis-related diseases, surgeries, and medications, several patient types were identified, including: Identify the thrombosis-related surgeries and medications associated with each thrombosis-related disease; Based on the medical information of multiple arteriovenous fistula patients, we determined the joint influence coefficient of thrombosis risk among various thrombosis risk-related diseases, the joint influence coefficient of thrombosis risk among various thrombosis risk-related surgeries, and the joint influence coefficient of thrombosis risk among various risk-related drugs. Based on the combined influence coefficients of thrombosis risk among various thrombosis-related diseases, various thrombosis-related surgeries, and various thrombosis-related medications, multiple patient types were identified.

4. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 3, characterized in that, Based on medical information from multiple arteriovenous fistula surgeries, the joint influence coefficients of thrombosis risk among various thrombosis-related diseases, various thrombosis-related surgeries, and various risk-related medications were determined, including: Based on the medical history and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients, the joint risk coefficient between any two thrombosis-related diseases was determined. Based on the joint risk coefficient between any two thrombosis-related diseases, multiple thrombosis-related disease units are identified, wherein each thrombosis-related disease unit includes two thrombosis-related diseases. For each thrombosis-related disease unit, based on the thrombosis-related surgeries and medications associated with each thrombosis-related disease, as well as the treatment information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgeries, the combined surgical risk coefficient of any two thrombosis-related surgeries and the combined drug risk coefficient of any two thrombosis-related medications corresponding to the thrombosis-related disease unit are determined.

5. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 4, characterized in that, Based on the combined thrombosis risk coefficients among various thrombosis-related diseases, various thrombosis-related surgeries, and various risk-related medications, multiple patient types were identified, including: For each thrombosis-related disease unit, the unit risk joint influence coefficient is determined based on the joint influence coefficient of disease risk between the two thrombosis-related diseases included in the thrombosis-related disease unit, the joint influence coefficient of surgical risk of any two thrombosis-related surgeries and the joint influence coefficient of drug risk of any two thrombosis-related drugs corresponding to the thrombosis-related disease unit. Based on the joint risk factor coefficient of each thrombosis-related disease unit, multiple thrombosis-related diseases are grouped to determine multiple thrombosis-related disease groups; Multiple patient types were identified based on various thrombosis risk-related disease groups.

6. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 4, characterized in that, Based on the preoperative testing information of multiple arteriovenous fistula (AVF) surgery patients included in the aforementioned AVF surgery patient types, key preoperative testing factors corresponding to each AVF surgery patient type and their corresponding influence weights are determined, including: Based on the thrombosis risk-related disease groups corresponding to the patient types of arteriovenous fistula surgery, multiple preoperative detection factors to be screened were identified for the patient types of arteriovenous fistula surgery. Based on the preoperative testing information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients, including the patient types of arteriovenous fistula surgery, the thrombosis risk impact coefficient of each preoperative testing factor to be screened for the corresponding patient type of arteriovenous fistula surgery was determined. Based on the thrombosis risk impact coefficient of each preoperative detection factor to be screened for arteriovenous fistula surgery patient type, the key preoperative detection factors and their corresponding impact weights for the key preoperative detection factors are determined.

7. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 6, characterized in that, Based on the thrombosis risk impact coefficient of each preoperative screening factor corresponding to the patient type of arteriovenous fistula surgery, the key preoperative screening factors and their corresponding impact weights for each patient type of arteriovenous fistula surgery are determined, including: Based on the thrombosis risk impact coefficient and thrombosis risk impact threshold of each preoperative detection factor to be screened for arteriovenous fistula surgery patient type, the key preoperative detection factors corresponding to the arteriovenous fistula surgery patient type are determined. Based on the thrombosis risk impact coefficient of key preoperative detection factors corresponding to each type of arteriovenous fistula surgery patient, the impact weight of each key preoperative detection factor is determined.

8. The method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 1, characterized in that, Based on the preoperative testing information of patients undergoing arteriovenous fistula (AVF) surgery, the key preoperative testing factors corresponding to the patient type of AVF surgery, and the influence weights of the key preoperative testing factors, a preoperative thrombotic risk assessment is performed for AVF surgery, including: For each type of arteriovenous fistula surgery patient, a risk assessment model is established based on the key preoperative detection factors and their corresponding influence weights. Using a risk assessment model corresponding to the type of arteriovenous fistula (AVF) patient to which the patient to undergo AVF surgery belongs, and based on the preoperative test information of the patient to be AVF surgery, the risk of thrombosis before AVF surgery is assessed.

9. A method for assessing the risk of thrombosis before arteriovenous fistula surgery according to claim 8, characterized in that, Using a risk assessment model corresponding to the patient type of arteriovenous fistula (AVF) surgery, and based on the preoperative testing information of patients undergoing AVF surgery, a preoperative thrombotic risk assessment for AVF surgery is performed, including: Based on the medical history and treatment information of patients awaiting arteriovenous fistula surgery, the matching value of the arteriovenous fistula surgery patient type to which the patients awaiting arteriovenous fistula surgery belong is determined; The risk of thrombosis during arteriovenous fistula surgery is determined by using a risk assessment model corresponding to the type of arteriovenous fistula surgery patient to which the patient to be undergoing arteriovenous fistula surgery belongs, based on the preoperative test information of the patients to be undergoing arteriovenous fistula surgery. The comprehensive risk value of arteriovenous fistula surgery (AVF) is determined by using a comprehensive evaluation model based on the matching value of each AVF patient type to which the patient to undergo AVF surgery belongs and the corresponding risk assessment model output AVF surgery thrombosis risk value.

10. A system for assessing the risk of thrombosis before arteriovenous fistula surgery, characterized in that, A method for assessing the risk of thrombosis before arteriovenous fistula surgery according to any one of claims 1-9, comprising: The information acquisition module is used to acquire medical information of multiple arteriovenous fistula surgery patients, wherein the medical information includes at least medical history information, treatment information, preoperative test information and postoperative thrombosis monitoring information; The correlation analysis module is used to identify various thrombosis-related diseases, thrombosis-related surgeries, and thrombosis-related drugs based on the medical history, treatment information, and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients. The type analysis module is used to determine the type of multiple arteriovenous fistula surgery patients based on their medical information, multiple thrombosis-related diseases, multiple thrombosis-related surgeries, and multiple thrombosis-related medications. The factor analysis module is used to determine the key preoperative detection factors and their corresponding influence weights for each type of arteriovenous fistula surgery patient, based on the preoperative detection information and postoperative thrombosis monitoring information of multiple arteriovenous fistula surgery patients included in the arteriovenous fistula surgery patient type. The type determination module is used to determine the type of arteriovenous fistula surgery patient to which the patient belongs based on the patient's medical history and treatment information. The information acquisition module is also used to acquire preoperative test information of patients undergoing arteriovenous fistula surgery based on the type of arteriovenous fistula surgery patients to which the patients undergoing arteriovenous fistula surgery belong. The risk assessment module is used to assess the risk of thrombosis before arteriovenous fistula surgery based on the preoperative test information of patients undergoing arteriovenous fistula surgery, the key preoperative test factors corresponding to the patient type of arteriovenous fistula surgery, and the influence weights of the key preoperative test factors.

Images