System for screening high risk of lower limb arteriosclerosis

Abstract

The invention provides a system for screening high risk of lower limb arteriosclerosis, and relates to the technical field of risk assessment of lower limb arteriosclerosis. The system comprises a data acquisition module, an initial risk assessment module, a CHINA-PAR assessment module, an ABI assessment module and a Medicom-ArtRiskLegFlow model assessment module. Wherein the data acquisition module is used for collecting patient information, current medical history and inspection data, and generating structured data by using a natural language processing technology. And the initial risk assessment module determines diagnosed and undiagnosed patients according to the structured data. The CHINA-PAR module performs risk assessment on undiagnosed patients, and the risk assessment is divided into low risk, medium risk and high risk. The ABI module performs ABI measurements on middle and high risk patients to determine a further assessment. And finally, the Medicom-ArtRiskLegFlow module calculates the final risk level of the undiagnosed patient according to the weight score and the type score. The system solves the problems of high cost, low precision and inspection risk in the prior art.

Description

Technical Field

[0001] This invention relates to the field of lower extremity arteriosclerosis risk assessment technology, and in particular to a system for screening for high risk of lower extremity arteriosclerosis. Background Technology

[0002] Lower extremity arteriosclerosis (PAD) is a common and difficult-to-treat disease, primarily caused by the formation of atherosclerotic plaques in the lower extremity arteries, leading to stenosis and occlusion, and consequently chronic limb ischemia. With changes in dietary structure and the increasing aging of the population, the incidence of PAD in my country is on the rise. Related studies show that the incidence of peripheral artery disease in China in 2019 increased by 40% compared to 2000. PAD has a high disability rate, and most patients are diagnosed after missing the optimal treatment window, resulting in poor treatment outcomes. Therefore, early screening of asymptomatic high-risk individuals for PAD and early prevention and treatment are of significant practical importance for improving patients' quality of life and reducing the burden on families and society.

[0003] Lower extremity arteriography is the "gold standard" for diagnosing arteriosclerotic diseases of the lower extremities. However, this examination is invasive, complex, and carries a high risk, making it difficult to popularize as a routine examination. Doppler ultrasound is a commonly used method for examining lower extremity arterial diseases, but it requires well-trained sonographers to perform the procedure. Different doctors may interpret the data differently, and Doppler ultrasound is not ideal for examining distal small arteries; furthermore, there are no unified testing standards. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a system for screening for high risk of lower extremity arteriosclerosis. This invention solves the problems of high cost, low detection accuracy and bleeding risk of the examination method in the prior art.

[0005] To achieve the above objectives, the present invention provides the following solution: A system for screening for high risk of lower extremity arteriosclerosis includes: The data acquisition module, initial risk module, CHINA-PAR assessment module, ABI assessment module, and Medicom-ArtRiskLegFlow model assessment module are connected sequentially. The data acquisition module is used to collect personal information, current medical history, and test data of the target patient set and to parse the unstructured data using natural language processing technology to obtain structured data. The initial risk module is used to determine the first confirmed patient set and the first undiagnosed patient set based on the structured data. The CHINA-PAR assessment module is used to perform risk assessment on the first undiagnosed patient set based on the CHINA-PAR model and the structured data to obtain low-risk, intermediate-risk, and high-risk patients. The ABI assessment module is used to perform ABI assessment on the intermediate-risk and high-risk patients based on the structured data to obtain ABI results and to determine the second confirmed patient set and the second undiagnosed patient set among the intermediate-risk and high-risk patients based on the ABI results. The Medicom-ArtRiskLegFlow model assessment module is used to calculate the weight score and type score of the second undiagnosed patient set by assigning a preset weight score to each data point in the structured data and to determine the final risk level of each patient in the second undiagnosed patient set based on the weight score and type score.

[0006] Preferably, the personal information includes: Smoking history, gender, and age.

[0007] Preferably, the present medical history includes: Lower extremity arteriosclerosis, hypertension, diabetes, cerebral infarction, coronary heart disease.

[0008] Preferably, the test data includes: Serum creatinine, triglycerides, low-density lipoprotein, high-density lipoprotein, homocysteine, glycated hemoglobin, and total cholesterol.

[0009] Preferably, the ABI evaluation module includes: The module includes a data reading submodule, an ABI assessment submodule, an ABI result screening submodule, and a patient confirmation submodule. The data reading submodule is used to read the structured data. The ABI assessment submodule is used to measure the ratio of ankle blood pressure to brachial artery blood pressure in intermediate-risk and high-risk patients to obtain ABI results. The ABI result filtering submodule is used to determine patients with abnormal ABI, normal ABI, and missing ABI among intermediate-risk and high-risk patients according to a preset range value. The patient confirmation submodule is used to confirm patients with abnormal ABI as a second confirmed patient set and patients with normal ABI and missing ABI as a second undiagnosed patient set.

[0010] Preferably, the formula for calculating the weight score is: S W=X1+X2+X3+Y1+Y2+Y3+Y4+Z1+Z2+Z3+Z4+Z5+Z6+Z7; Among them, S W The total weighted score for risk factors is as follows: X1 is the weighted score for smoking history, X2 is the weighted score for gender, X3 is the weighted score for age, Y1 is the weighted score for hypertension, Y2 is the weighted score for diabetes, Y3 is the weighted score for cerebral infarction, Y4 is the weighted score for coronary heart disease, Z1 is the weighted score for serum creatinine, Z2 is the weighted score for triglycerides, Z3 is the weighted score for low-density lipoprotein, Z4 is the weighted score for high-density lipoprotein, Z5 is the weighted score for homocysteine, Z6 is the weighted score for glycated hemoglobin, and Z7 is the weighted score for total cholesterol.

[0011] Preferably, the formula for calculating the type score is: S T =β1T1+β2T2+β3T3; Among them, S T T1 is the total weighted score for the high-risk factor category, β1 is the fixed weighted score for the personal information category (2 points), and β1 is the number of factors with actual scores in the personal information category; T2 is the fixed weighted score for the current medical history category (3 points), and β2 is the number of factors with actual scores in the current medical history category; T3 is the fixed weighted score for the test data category (4 points), and β3 is the number of factors with actual scores in the test data category.

[0012] The present invention discloses the following technical effects: This invention provides a system for screening for high risk of lower extremity arteriosclerosis, comprising: The data acquisition module, initial risk module, CHINA-PAR assessment module, ABI assessment module, and Medicom-ArtRiskLegFlow model assessment module are connected sequentially. The data acquisition module collects personal information, current medical history, and laboratory data of the target patient set and uses natural language processing technology to parse the unstructured data to obtain structured data. The initial risk module determines a first confirmed patient set and a first undiagnosed patient set based on the structured data. The CHINA-PAR assessment module performs risk assessment on the first undiagnosed patient set based on the CHINA-PAR model and the structured data to obtain low-risk, intermediate-risk, and high-risk patients. The ABI assessment module performs ABI assessment on the intermediate-risk and high-risk patients based on the structured data to obtain ABI results and determines a second confirmed patient set and a second undiagnosed patient set from the intermediate-risk and high-risk patients based on the ABI results. The Medicom-ArtRiskLegFlow model assessment module assigns preset weight scores to each data point in the structured data to calculate the weight score and type score of the second undiagnosed patient set and determines the final risk level of each patient in the second undiagnosed patient set based on the weight score and type score. This invention effectively reduces equipment dependence and operational risks through non-invasive data acquisition and automated analysis. Traditional lower extremity arterial angiography requires catheter insertion through arterial puncture, posing risks such as infection and bleeding. In contrast, the screening method of this invention only requires collecting the patient's medical information, without any invasive procedures, thus greatly reducing patient suffering and risks. This invention automatically acquires the patient's medical information through an interface, including basic information, medication information, medical history, and test data, and can automatically complete CHINA-PAR model assessment, ABI test result analysis, and Medicom-ArtRiskLegFlow model assessment. This not only effectively reduces the workload of doctors but also significantly improves screening efficiency. The Medicom-ArtRiskLegFlow model in this invention collects 14 risk factors, assessing the patient's risk of lower extremity arteriosclerosis from multiple dimensions. Combined with ABI and CHINA-PAR assessments, the system can more comprehensively reflect the patient's condition. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of a system structure for screening high-risk lower extremity arteriosclerosis, provided by an embodiment of the present invention. Figure 2This is a flowchart of a system for screening for high risk of lower extremity arteriosclerosis, provided as an embodiment of the present invention.

[0015] Explanation of reference numerals in the attached figures: 1-Data Acquisition Module, 2-Initial Risk Module, 3-CHINA-PAR Assessment Module, 4-ABI Assessment Module, 5-Medicom-ArtRiskLegFlow Model Assessment Module. Detailed Implementation

[0016] 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.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, the present invention provides a system for screening high risk of lower extremity arteriosclerosis, comprising: The data acquisition module 1, the initial risk module 2, the CHINA-PAR assessment module 3, the ABI assessment module 4, and the Medicom-ArtRiskLegFlow model assessment module 5 are connected in sequence. The data acquisition module 1 is used to collect personal information, current medical history, and test data of the target patient set and to perform unstructured data parsing using natural language processing technology to obtain structured data. The initial risk module 2 is used to determine the first confirmed patient set and the first undiagnosed patient set based on the structured data. The CHINA-PAR assessment module 3 is used to perform risk assessment on the first undiagnosed patient set based on the CHINA-PAR model and the structured data to obtain low-risk, intermediate-risk, and high-risk patients. The ABI assessment module 4 is used to perform ABI assessment on the intermediate-risk and high-risk patients based on the structured data to obtain ABI results and to determine the second confirmed patient set and the second undiagnosed patient set among the intermediate-risk and high-risk patients based on the ABI results. The Medicom-ArtRiskLegFlow model assessment module 5 is used to calculate the weight score and type score of the second undiagnosed patient set by assigning a preset weight score to each data point in the structured data and to determine the final risk level of each patient in the second undiagnosed patient set based on the weight score and type score.

[0019] Specifically, such as Figure 2 As shown, the system obtains the patient's personal information, current medical history, and laboratory data. The personal information includes: smoking history, gender, and age; the current medical history includes lower extremity arteriosclerosis, hypertension, diabetes, cerebral infarction, and coronary heart disease; and the laboratory data includes serum creatinine, triglycerides, low-density lipoprotein, high-density lipoprotein, homocysteine, glycated hemoglobin, and total cholesterol.

[0020] Real-time patient diagnosis and treatment information is obtained from the Hospital Information System (HIS) and Laboratory Information Management System (LIS) through interfaces. The sources of patient diagnosis and treatment information are shown in Table 1. Table 1. Data and its corresponding source table

[0021] Furthermore, natural language processing (NLP) techniques (such as Bi-LSTM+CRF models or pre-trained language models BERT) are used to extract key information (such as smoking history and disease stage) and other structured data from free text in electronic medical records (such as chief complaint, present illness, past medical history, personal history, and progress notes in admission records).

[0022] The analysis results are shown in Table 2: Table 2. Example of Electronic Medical Record Parsing

[0023] This system is designed to screen high-risk individuals for lower extremity arteriosclerosis among undiagnosed patients. Therefore, if a patient has already been diagnosed, meaning that the patient's disease information matches "lower extremity arteriosclerosis," the system will automatically stop the assessment process and indicate that the patient has been "diagnosed."

[0024] Furthermore, CHINA-PAR assessment module 3 also includes a self-testing function, specifically: Create a new article in the "Material Management" section of the hospital's official WeChat account backend and enter the article editing page. First, enter the article title "Lower Extremity Arteriosclerosis Risk Assessment Tool" and a brief introduction to the CHINA-PAR tool in the body text. Second, use the "Link" icon in the editor, enter the official website URL of the CHINA-PAR model (https: / / www.cvdrisk.com.cn / ) in the pop-up dialog box, and enter "Click Rating" in the "Display Content" box as the hyperlink content. Complete the article editing and publish. Finally, select the "Auto Reply" function in the WeChat account backend and set the "Reply to Followers" to automatically push this article to users who follow the hospital's official WeChat account.

[0025] After following the hospital's official WeChat account, patients will automatically receive a dialog box prompt. Clicking this dialog box will redirect patients to an official mini-program page containing the CHINA-PAR assessment, where they will be invited to complete a self-assessment. The CHINA-PAR model's assessment results categorize 10-year risk into low, intermediate, and high risk groups. After completing the self-assessment, the CHINA-PAR website will automatically display the patient's 10-year risk stratification results. Once the results are displayed, patients must exit the CHINA-PAR website and return to the hospital's WeChat account to select their desired outcome.

[0026] Low risk: If the patient is assessed as low risk, no further monitoring is required and the procedure can be terminated directly.

[0027] Intermediate and high risk: For individuals at intermediate and high risk, further evaluation of the ankle-brachial index (ABI) is required.

[0028] Furthermore, ABI assessment module 4 includes: The module includes a data reading submodule, an ABI assessment submodule, an ABI result screening submodule, and a patient confirmation submodule. The data reading submodule is used to read the structured data. The ABI assessment submodule is used to measure the systolic blood pressure ratio of the ankle artery to the brachial artery in intermediate-risk and high-risk patients to obtain ABI results. The ABI result screening submodule is used to determine patients with abnormal ABI, normal ABI, and missing ABI among intermediate-risk and high-risk patients according to a preset range value. The patient confirmation submodule is used to confirm the patients with abnormal ABI as a second confirmed patient set and the patients with normal ABI and missing ABI as a second undiagnosed patient set.

[0029] Specifically, ABI refers to the ratio of systolic blood pressure in the ankle artery to that in the brachial artery, and it is an important indicator for assessing the health of the lower extremity arteries. ABI results are divided into three categories: normal (0.9-1.3), abnormal (<0.9 or >1.3), and missing.

[0030] Abnormal ABI (<0.9 or >1.3): If the ABI value is below 0.9 or above 1.3, it indicates that the patient is at risk of lower extremity arteriosclerosis. Further evaluation in the predictive model is not required, and the doctor can be directly advised to diagnose "lower extremity arteriosclerosis".

[0031] ABI normal (0.9-1.3): If the ABI value is within this range, the lower extremity arteries are generally considered to be in good health. However, patients with vascular calcification may also present with a normal ABI. To avoid missed diagnoses, patients with an ABI result between 0.9 and 1.3 should undergo a more detailed risk assessment using the Medicom-ArtRiskLegFlow prediction model.

[0032] ABI data missing: If no ABI data is read, the system will automatically proceed to the Medicom-ArtRiskLegFlow model for risk assessment.

[0033] Furthermore, the formula for calculating the total weight of high-risk factors in the Medicom-ArtRiskLegFlow model is as follows: S W =X1+X2+X3+Y1+Y2+Y3+Y4+Z1+Z2+Z3+Z4+Z5+Z6+Z7.

[0034] X1: Read the patient's X1 standardized result, determine the number of cigarettes smoked per day for patients with a smoking history, and record their corresponding weighted score (e.g., patient A's X1 standardized result is 10+ years of smoking, 30 cigarettes per day, scoring 6.5 points). Patients for whom no cigarette count was extracted are uniformly recorded as "<10 cigarettes per day," scoring 4.5 points. X2: Gender information is determined directly through the interface, and scores are calculated according to rules. Patients for whom no "gender" information is extracted are uniformly scored as 0 points.

[0035] X3: Directly obtain the patient's age information through the interface, determine their age group, and record the corresponding weight score. If no age information is available, record 1 point.

[0036] Y1: Read the patient's Y1 standardized results. If a matching record for "hypertension" exists, determine the patient's hypertension level according to the hypertension grading criteria and record the corresponding weighted score. If no matching record exists, record 0 points.

[0037] Y2: Read the patient's Y2 standardized results. If a matching record for "diabetes" exists, score 3 points. If no matching record exists, score 0 points.

[0038] Y3: Read the patient's Y3 standardized results. If a matching record for "cerebral infarction" is found, score 0.5 points. If no matching record is found, score 0 points.

[0039] Y4: Read the patient's Y4 standardized results. If a matching record for "coronary artery disease" is found, 1 point is awarded. If no matching record is found, 0 points are awarded.

[0040] Z1: The patient's Z1 result is read directly through the interface, and a corresponding weighted score is recorded based on the eGFR value. If no Z1 record exists, the electronic medical record is searched for "renal insufficiency." If a matching record for "renal insufficiency" is found, 1 point is awarded; otherwise, 0 points are awarded. (For example, patient B did not have a serum creatinine test, so there is no Z1 test result, but the disease information in the electronic medical record includes "renal insufficiency," so 1 point is awarded.) Z2: Read the patient's Z2 result directly through the interface. If a matching record is found, score 2 points. If no matching record is found, score 0 points.

[0041] Z3: Read the patient's Z3 result directly through the interface. If a matching record exists, score 2.5 points. If no matching record exists, score 0 points.

[0042] Z4: The patient's Z4 results are read directly through the interface. If a matching record exists, the corresponding factor weight score is recorded according to the rules. If no matching record exists, a score of 0 is recorded.

[0043] Z5: The patient's Z5 result is read directly through the interface. If a matching record exists, the corresponding factor weight score is recorded according to the rules. If no matching record exists, a score of 0 is recorded.

[0044] Z6: Read the patient's Z6 result directly through the interface. If a matching record exists, score 2.5 points. If no matching record exists, score 0 points.

[0045] Z7: Read the patient's Z7 result directly through the interface. If it exists, record the corresponding weight score according to the rules. If the Z7 result does not exist, record 0 points.

[0046] Furthermore, the formula for calculating the type score is as follows: S T =β1T1+β2T2+β3T3; Where T1 is the fixed weight score for the personal information category, β1 is the number of factors with actual scores in the personal information category, T2 is the fixed weight score for the current medical history category, β2 is the number of factors with actual scores in the current medical history category, T3 is the fixed weight score for the test data category, and β3 is the number of factors with actual scores in the test data category.

[0047] Specifically, T1: 2 points; T2: 3 points; T3: 4 points; β1: Filter for X1≠0, X2≠0, X3≠0. If there are n matching records, then β1=n.

[0048] β2: Filter for Y1≠0, Y2≠0, Y3≠0, Y4≠0. If there are n matching records, then β2=n.

[0049] β3: Filter Z1≠0, Z2≠0, Z3≠0, Z4≠0, Z5≠0, Z6≠0, Z7≠0. If there are n matching records, then β3=n.

[0050] The correlation between Medicom-ArtRiskLegFlow score and severity is analyzed, and the results are output based on the correlation:

[0051] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0052] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A system for screening high-risk lower extremity arteriosclerosis, characterized in that, include: The data acquisition module, initial risk module, CHINA-PAR assessment module, ABI assessment module, and Medicom-ArtRiskLegFlow model assessment module are connected sequentially. The data acquisition module is used to collect personal information, current medical history, and test data of the target patient set and to parse the unstructured data using natural language processing technology to obtain structured data. The initial risk module is used to determine the first confirmed patient set and the first undiagnosed patient set based on the structured data. The CHINA-PAR assessment module is used to perform risk assessment on the first undiagnosed patient set based on the CHINA-PAR model and the structured data to obtain low-risk, intermediate-risk, and high-risk patients. The ABI assessment module is used to perform ABI assessment on the intermediate-risk and high-risk patients based on the structured data to obtain ABI results and to determine the second confirmed patient set and the second undiagnosed patient set among the intermediate-risk and high-risk patients based on the ABI results. The Medicom-ArtRiskLegFlow model assessment module is used to calculate the weight score and type score of the second undiagnosed patient set by assigning a preset weight score to each data point in the structured data and to determine the final risk level of each patient in the second undiagnosed patient set based on the weight score and type score.

2. The system for screening high risk of lower extremity arteriosclerosis according to claim 1, characterized in that, The personal information includes: Smoking history, gender, and age.

3. The system for screening high risk of lower extremity arteriosclerosis according to claim 1, characterized in that, The present medical history includes: Lower extremity arteriosclerosis, hypertension, diabetes, cerebral infarction, coronary heart disease.

4. The system for screening high risk of lower extremity arteriosclerosis according to claim 1, characterized in that, The test data includes: Serum creatinine, triglycerides, low-density lipoprotein, high-density lipoprotein, homocysteine, glycated hemoglobin, and total cholesterol.

5. The system for screening high risk of lower extremity arteriosclerosis according to claim 1, characterized in that, The ABI assessment module includes: The module includes a data reading submodule, an ABI assessment submodule, an ABI result screening submodule, and a patient confirmation submodule. The data reading submodule is used to read the structured data. The ABI assessment submodule is used to measure the ratio of ankle artery systolic blood pressure to brachial artery systolic blood pressure in intermediate-risk and high-risk patients to obtain ABI results. The ABI result filtering submodule is used to determine patients with abnormal ABI, normal ABI, and missing ABI among intermediate-risk and high-risk patients according to a preset range value. The patient confirmation submodule is used to confirm patients with abnormal ABI as a second confirmed patient set and patients with normal ABI and missing ABI as a second undiagnosed patient set.

6. The system for screening high risk of lower extremity arteriosclerosis according to claim 1, characterized in that, The formula for calculating the weighted score is: S W =X1+X2+X3+Y1+Y2+Y3+Y4+Z1+Z2+Z3+Z4+Z5+Z6+Z7; Among them, S W The total weighted score for risk factors is as follows: X1 is the weighted score for smoking history, X2 is the weighted score for gender, X3 is the weighted score for age, Y1 is the weighted score for hypertension, Y2 is the weighted score for diabetes, Y3 is the weighted score for cerebral infarction, Y4 is the weighted score for coronary heart disease, Z1 is the weighted score for serum creatinine, Z2 is the weighted score for triglycerides, Z3 is the weighted score for low-density lipoprotein, Z4 is the weighted score for high-density lipoprotein, Z5 is the weighted score for homocysteine, Z6 is the weighted score for glycated hemoglobin, and Z7 is the weighted score for total cholesterol.

7. The system for screening high risk of lower extremity arteriosclerosis according to claim 1, characterized in that, The formula for calculating the type score is: S T =β1T1+β2T2+β3T3; Among them, S T T1 is the total weighted score for the high-risk factor category, β1 is the fixed weighted score for the personal information category, and β1 is the number of factors with actual scores in the personal information category; T2 is the fixed weighted score for the present medical history category, and β2 is the number of factors with actual scores in the present medical history category; T3 is the fixed weighted score for the test data category, and β3 is the number of factors with actual scores in the test data category.

Images