TWM552639U - Risk management system - Google Patents
Risk management system Download PDFInfo
- Publication number
- TWM552639U TWM552639U TW106209139U TW106209139U TWM552639U TW M552639 U TWM552639 U TW M552639U TW 106209139 U TW106209139 U TW 106209139U TW 106209139 U TW106209139 U TW 106209139U TW M552639 U TWM552639 U TW M552639U
- Authority
- TW
- Taiwan
- Prior art keywords
- risk
- coefficient
- user
- value
- blood pressure
- Prior art date
Links
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
Description
本創作是關於一種風險管理系統，特別是關於一種心血管疾病風險管理系統。 This creation is about a risk management system, especially about a cardiovascular disease risk management system.
心血管疾病位居國人前十大死因的第二位，如何對於心血管疾病進行精準的風險評估及管理，是現代醫學的重要課題。以量化且系統化的方式建立心血管風險的評估模型成為顯學，目前實用的評估模型，包括：Framingham危險評估、PROCAM模型評估、REYNODS風險評估、歐洲SCORE危險評估、WHO/ISH風險預測及中國缺血性心血管病危險評估等。 Cardiovascular disease ranks second among the top ten causes of death among Chinese people. How to conduct accurate risk assessment and management of cardiovascular diseases is an important topic in modern medicine. An assessment model for establishing cardiovascular risk in a quantitative and systematic manner has become a prominent, current practical assessment model, including: Framingham risk assessment, PROCAM model assessment, REYNODS risk assessment, European SCORE risk assessment, WHO/ISH risk prediction, and China deficiency Blood cardiovascular disease risk assessment, etc.
然而，無論是何種評估模型，皆因時間、地區、種族、族群(例如男女、老、少)等差異而產生誤差，對於風險評估及管理產生影響。這種誤差仰賴完整的臨床實驗及充分的統計數據加以修正。 However, no matter what kind of evaluation model, errors occur due to differences in time, region, race, ethnic group (such as male and female, old and young), which have an impact on risk assessment and management. This error is corrected by a complete clinical trial and sufficient statistical data.
有鑑於此，根據本創作的一種觀點，是提出一種風險管理系統，包括：用戶資料庫、統計資料庫、運算器、轉換器及第一分類器。用戶資料庫儲存至少一個用戶的多個生理特徵；統計資料庫儲存多個統計值，對應於多個 生理特徵；運算器是連結至用戶資料庫及統計資料庫，並結合多個生理特徵及多個統計值以運算出風險因子；轉換器是連結至運算器，並將風險因子轉換成風險百分比；第一分類器是連結至轉換器，並根據風險百分比將用戶分類至多個族群的一個族群，並輸出分類標籤。 In view of this, according to one aspect of the present creation, a risk management system is proposed, including: a user database, a statistical database, an operator, a converter, and a first classifier. The user database stores a plurality of physiological characteristics of at least one user; the statistical database stores a plurality of statistical values corresponding to the plurality of Physiological characteristics; the operator is linked to the user database and the statistical database, and combines multiple physiological features and a plurality of statistical values to calculate a risk factor; the converter is linked to the operator and converts the risk factor into a risk percentage; The first classifier is linked to the converter and classifies the user into a group of multiple ethnic groups according to the percentage of risk, and outputs the classification tag.
進一步地，本創作的風險管理系統可更包括回授器，其連結至統計資料庫，並根據多個用戶的多個臨床資料更新多個統計值。 Further, the risk management system of the present invention may further include a feedback device that is linked to the statistical database and updates a plurality of statistical values according to the plurality of clinical data of the plurality of users.
進一步地，本創作的風險管理系統可更包括第二分類器，其連結至第一分類器，並根據分類標籤及調整指數將用戶重新分類至多個族群的一個族群。 Further, the risk management system of the present invention may further include a second classifier coupled to the first classifier and reclassifying the user to a group of the plurality of ethnic groups according to the classification tag and the adjustment index.
進一步地，在本創作的風險管理系統中，第一分類器是組態成：若風險百分比大於20%，則將用戶分類至高度風險族群；若風險百分比介於10%至20%，則將用戶分類至中度風險族群；且若風險百分比小於10%，則將用戶分類至低度風險族群。調整指數是動脈硬化指數。第二分類器是組態成：若動脈硬化指數小於0.9，則將用戶自低度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至高度風險族群；且若動脈硬化指數介於1.1至1.4，則將用戶自高度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至低度風險族群。 Further, in the risk management system of the present creation, the first classifier is configured to classify the user into a high risk group if the risk percentage is greater than 20%; if the risk percentage is between 10% and 20%, Users are classified to moderate risk groups; and if the risk percentage is less than 10%, the users are classified into low risk groups. The adjusted index is the arteriosclerosis index. The second classifier is configured to: if the arteriosclerosis index is less than 0.9, reclassify the user from a low risk group to a moderate risk group; or reclassify from a moderate risk group to a high risk group; and if the arteriosclerosis index Between 1.1 and 1.4, users are reclassified from a high-risk group to a moderate risk group; or from a moderate risk group to a low-risk group.
進一步地，在本創作的風險管理系統中，多個生理特徵包括：性別、年齡數值、總膽固醇數值、高密度膽固醇數值、收縮壓數值、抽菸布林值、糖尿病布林值及高敏感性發炎指標；且多個統計值包括：年齡係數、總膽固醇係數、高密度膽固醇係數、第一收縮壓因子、第二收縮壓因子、抽菸係數、糖尿病係數、高敏感性發炎指標係數及偏移值。 Further, in the risk management system of the present creation, a plurality of physiological characteristics include: gender, age value, total cholesterol value, high-density cholesterol value, systolic blood pressure value, smoking boll value, diabetic boolean value, and high sensitivity. Inflammatory index; and multiple statistical values include: age coefficient, total cholesterol coefficient, high-density cholesterol coefficient, first systolic blood pressure factor, second systolic blood pressure factor, smoking coefficient, diabetes coefficient, high-sensitivity inflammatory index coefficient and migration value.
進一步地，在本創作的風險管理系統中，運算器是組態成依序執行下列步驟：將年齡數值取自然對數乘以年齡係數；加上總膽固醇值取自然對數乘以總膽固醇係數；減去高密度膽固醇數值取自然對數乘以高密度膽固醇係數；加上收縮壓數值取自然對數乘以第一收縮壓因子或第二收縮壓因子，其中，若用戶沒有接受高血壓治療，則將收縮壓數值取自然對數乘以第一收縮壓因子；若用戶有接受高血壓治療，則將收縮壓數值取自然對數乘以第二收縮壓因子；加上抽菸布林值乘以抽菸係數；加上糖尿病布林值乘以糖尿病係數；加上高敏感性發炎指標取自然對數乘以高敏感性發炎指標係數；及減去偏移值。 Further, in the risk management system of the present creation, the operator is configured to sequentially perform the following steps: multiplying the age value by the natural logarithm by the age coefficient; plus adding the total cholesterol value to the natural logarithm multiplied by the total cholesterol coefficient; The high-density cholesterol value is taken as the natural logarithm multiplied by the high-density cholesterol coefficient; plus the systolic blood pressure value is taken as the natural logarithm multiplied by the first systolic blood pressure factor or the second systolic blood pressure factor, wherein if the user does not receive hypertension treatment, the contraction will be The pressure value is taken by the natural logarithm multiplied by the first systolic blood pressure factor; if the user has received hypertension treatment, the systolic blood pressure value is taken as the natural logarithm multiplied by the second systolic blood pressure factor; plus the smoking boolean value multiplied by the smoking coefficient; Plus the diabetes Bollinger value multiplied by the diabetes coefficient; plus the high-sensitivity inflammatory index multiplies the natural logarithm by the high-sensitivity inflammatory index coefficient; and subtracts the offset value.
進一步地，在本創作的風險管理系統中，若性別是男性，則年齡係數是3.06117；總膽固醇係數是1.12370；高密度膽固醇係數是0.93263；第一收縮壓因子是1.93303；第二收縮壓因子是1.99881；抽菸係數是0.65451；糖尿病係數是0.57367；高敏感性發炎指標係數是0.03；且偏移值是23.9802；若性別是女性，則年齡係數是2.32888；總膽固醇係數是1.20904；高密度膽固醇係數是0.70833；第一收縮壓因子是2.76157；第二收縮壓因子是2.82263；抽菸係數是0.52873；糖尿病係數是0.69154；高敏感性發炎指標係數是0.03；且偏移值是26.1931。 Further, in the risk management system of the present creation, if the gender is male, the age coefficient is 3.06117; the total cholesterol coefficient is 1.12370; the high density cholesterol coefficient is 0.93263; the first systolic blood pressure factor is 1.93303; the second systolic blood pressure factor is 1.99881; smoking coefficient is 0.65451; diabetes coefficient is 0.57367; high sensitivity inflammatory index coefficient is 0.03; and offset value is 23.9802; if gender is female, age coefficient is 2.32888; total cholesterol coefficient is 1.20904; high density cholesterol coefficient Is 0.70833; the first systolic blood pressure factor is 2.76157; the second systolic blood pressure factor is 2.82263; the smoking coefficient is 0.52873; the diabetes coefficient is 0.69154; the high sensitivity inflammatory index coefficient is 0.03; and the offset value is 26.1611.
進一步地，在本創作的風險管理系統中，轉換器是組態成以下列公式將風險因子riskfactor轉換成風險百分比risk(%)：若性別是男性，則risk(%)=100×(1-0.88936×exp(riskfactor)) Further, in the risk management system of the present creation, the converter is configured to convert the risk factor riskfactor into a risk percentage risk (%) by the following formula: if the gender is male, then risk(%)=100×(1- 0.88936×exp(riskfactor))
若性別是女性，則risk(%)=100×(1-0.95012×exp(riskfactor))其中，exp()表示取自然指數。 If the gender is female, then risk(%)=100×(1-0.95012×exp(riskfactor)) where exp() represents the natural index.
根據本創作的另一種觀點，是提出一種風險管理方法，包括： 步驟A：取得用戶的年齡數值Age、總膽固醇數值Totalcholesterol、高密度膽固醇數值HDLcholesterol、收縮壓數值SystolicBP、抽菸布林值Cig、糖尿病布林值DM及高敏感性發炎指標hsCRP；步驟B：運算出風險因子riskfactors：若用戶是男性，則：riskfactor=ln(Age)×3.06117+ln(Totalcholesterol)×1.12370-ln(HDLcholesterol)×0.93263+ln(SystolicBP)×SysBPFactor+Cig×0.65451+DM×0.57367+ln(hsCRP)×0.03-23.9802 According to another aspect of this creation, a risk management method is proposed, including: Step A: Obtain the user's age value Age, total cholesterol value Totalcholesterol, high-density cholesterol value HDLcholesterol, systolic blood pressure value SystolicBP, smoking Bollinger value Cig, diabetes Bollinger value DM, and high-sensitivity inflammatory index hsCRP; Step B: Operation Risk factor riskfactors: If the user is male, then: riskfactor=ln(Age)×3.06117+ln(Totalcholesterol)×1.12370-ln(HDLcholesterol)×0.93263+ln(SystolicBP)×SysBPFactor+Cig×0.65451+DM×0.57367+ Ln(hsCRP)×0.03-23.9802
若用戶是女性，則：riskfactor=ln(Age)×2.32888+ln(Totalcholesterol)×1.20904-ln(HDLcholesterol)×0.70833+ln(SystolicBP)×SysBPFactor+Cig×0.52873+DM×0.69154+ln(hsCRP)×0.03-26.1931 If the user is female, then: riskfactor=ln(Age)×2.32888+ln(Totalcholesterol)×1.20904-ln(HDLcholesterol)×0.70833+ln(SystolicBP)×SysBPFactor+Cig×0.52873+DM×0.69154+ln(hsCRP)× 0.03-26.1931
其中，ln表示取自然對數；且若用戶是男性且沒有接受高血壓治療，則SysBPFactor為1.93303；若用戶是男性且有接受高血壓治療，則SysBPFactor為1.99881；若用戶是女性且沒有接受高血壓治療，則SysBPFactor為2.76157；若用戶是女性且有接受高血壓治療，則SysBPFactor為2.82263；及步驟C：將風險因子轉換成風險百分比risk(%)：若用戶是男性，則risk(%)=100×(1-0.88936×exp(riskfactor)) Where ln means taking the natural logarithm; and if the user is male and does not receive hypertension treatment, the SysBPFactor is 1.93303; if the user is male and has hypertension treatment, the SysBPFactor is 1.99881; if the user is female and does not receive hypertension For treatment, the SysBPFactor is 2.76157; if the user is female and has hypertension treatment, the SysBPFactor is 2.82263; and step C: converts the risk factor into a risk percentage risk (%): if the user is male, then risk(%)= 100×(1-0.88936×exp(riskfactor))
若用戶是女性，則risk(%)=100×(1-0.95012×exp(riskfactor)) If the user is female, then risk(%)=100×(1-0.95012×exp(riskfactor))
其中，exp()表示取自然指數。 Among them, exp() means taking the natural index.
進一步地，本創作的風險管理方法更包括：步驟D：根據風險百分比進行第一次分類：若風險百分比大於20%，則將用戶分類至高度風險族群；若風險百分比介於10%至20%，則將用戶分類至中度風險族群；且若風險百分比小於10%，則將用戶分類至低度風險族群；及步驟E：根據動脈硬化指數進行第二次分類：若動脈硬化指數小於0.9，則將用戶自低度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至高度風險族群；若動脈硬化指數介於1.1至1.4，則將用戶自高度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至低度風險族群。 Further, the risk management method of the present invention further includes: Step D: performing the first classification according to the percentage of risk: if the risk percentage is greater than 20%, classifying the user into a high risk group; if the risk percentage is between 10% and 20% , the user is classified into a moderate risk group; and if the risk percentage is less than 10%, the user is classified into a low risk group; and step E: the second classification according to the arteriosclerosis index: if the arteriosclerosis index is less than 0.9, Reclassify users from low risk groups to moderate risk groups; or reclassify from moderate risk groups to highly risk groups; if the arteriosclerosis index is between 1.1 and 1.4, reclassify users from high risk groups to Degree risk group; or reclassify from a moderate risk group to a low risk group.
綜上所述，根據本創作的風險管理系統及方法，可基於用戶的多個生理特徵完整地評估其在十年間罹患心血管疾病的風險，而將其分類至正確的風險族群，以達成有效的管理。醫療人員可根據本創作的評估結果給予用戶保健或治療方案。 In summary, according to the risk management system and method of the present invention, the risk of cardiovascular disease in a decade can be completely evaluated based on multiple physiological characteristics of the user, and classified into the correct risk group to achieve effective Management. The medical staff can give the user a health care or treatment plan based on the evaluation results of this creation.
此外，本創作的風險管理系統及方法的各項係數來自完整的臨床實驗及充分的統計數據，並可透過回授進行動態更新，以促進評估的精準度。 In addition, the various factors of the risk management system and method of this creation come from complete clinical trials and sufficient statistical data, and can be dynamically updated through feedback to promote the accuracy of the assessment.
1‧‧‧風險管理系統 1‧‧‧Risk Management System
2‧‧‧風險管理系統 2‧‧‧Risk Management System
3‧‧‧風險管理系統 3‧‧‧ Risk Management System
10‧‧‧用戶資料庫 10‧‧‧User database
20‧‧‧統計資料庫 20‧‧‧Statistical database
30‧‧‧運算器 30‧‧‧Operator
40‧‧‧轉換器 40‧‧‧ converter
50‧‧‧第一分類器 50‧‧‧First classifier
60‧‧‧第二分類器 60‧‧‧Second classifier
70‧‧‧回授器 70‧‧‧receiver
圖1顯示本創作一實施例的風險管理系統1。 Figure 1 shows a risk management system 1 of an embodiment of the present invention.
圖2顯示本創作另一實施例的風險管理系統2。 Figure 2 shows a risk management system 2 of another embodiment of the present creation.
圖3顯示本創作一實施例的第二分類器的重新分類流程圖。 FIG. 3 shows a reclassification flowchart of the second classifier of an embodiment of the present invention.
圖4顯示本創作再一實施例的風險管理系統3。 Fig. 4 shows a risk management system 3 of still another embodiment of the present creation.
圖5顯示本創作一實施例的風險管理方法。 FIG. 5 shows a risk management method of an embodiment of the present creation.
以下將提供本創作不同實施例。可理解的是，這些實施例並非用以限制。本創作的技術特徵可加以修飾、置換、組合、分離及設計以應用於其他實施例。 Different embodiments of the present work will be provided below. It is to be understood that these embodiments are not intended to be limiting. The technical features of the present invention can be modified, replaced, combined, separated, and designed to be applied to other embodiments.
圖1顯示本創作一實施例的風險管理系統1。 Figure 1 shows a risk management system 1 of an embodiment of the present invention.
本創作的風險管理系統1主要包括用戶資料庫10、統計資料庫20、運算器30、轉換器40及第一分類器50。 The risk management system 1 of the present invention mainly includes a user database 10, a statistical database 20, an arithmetic unit 30, a converter 40, and a first classifier 50.
用戶資料庫10儲存至少一個用戶的多個生理特徵，例如：1.年齡數值(Age)，單位為歲；2.血液總膽固醇數值(Totalcholesterol)，單位為毫克/分升(mg/dL)；3.血液高密度膽固醇數值(HDLcholesterol)，單位為毫克/分升(mg/dL)；4.血管收縮壓數值(SystolicBP)，單位為毫米汞柱(mmHg)；5.抽菸布林值(Cig)，其中，抽菸布林值是0或1，若用戶沒有抽菸，則抽菸布林值是0，若用戶有抽菸，則抽菸布林值是1；6.糖尿病布林值(DM)，其中，糖尿病布林值是0或1，若用戶沒有糖尿病，則糖尿病布林值是0，若用戶有糖尿病，則糖尿病布林值是1；及7.高敏感性發炎指標(hsCRP)，單位為毫克/分升(mg/dL)。 The user database 10 stores a plurality of physiological characteristics of at least one user, for example: 1. Age value (Age), the unit is years old; 2. Blood total cholesterol value (Totalcholesterol), the unit is mg/dl (mg/dL); 3. Blood high-density cholesterol (HDLcholesterol) in milligrams per deciliter (mg/dL); 4. Vasostrictive pressure value (SystolicBP) in millimeters of mercury (mmHg); 5. smoking boolean value ( Cig), wherein the smoking Bollinger value is 0 or 1. If the user does not smoke, the smoking Bollinger value is 0. If the user has smoking, the smoking Bollinger value is 1; 6. Diabetic Brin Value (DM), wherein the diabetes boll value is 0 or 1. If the user does not have diabetes, the diabetes boolean value is 0. If the user has diabetes, the diabetes boolean value is 1; and 7. The high sensitivity inflammation index (hsCRP) in milligrams per deciliter (mg/dL).
統計資料庫20儲存多個統計值，對應於多個生理特徵。多個統計值來自完整的臨床實驗及充分的統計數據，可促進評估的精準度。多個統計值是例如：1.年齡係數是對應於年齡數值(Age)； 2.總膽固醇係數是對應於總膽固醇數值(Totalcholesterol)；3.高密度膽固醇係數是對應於高密度膽固醇數值(HDLcholesterol)；4.第一收縮壓因子及第二收縮壓因子是對應於收縮壓數值(SystolicBP)，其中，第一收縮壓因子及第二收縮壓因子統稱為收縮壓因子(SysBPfactor)，若用戶沒有接受高血壓治療，則將收縮壓數值取自然對數乘以第一收縮壓因子；若用戶有接受高血壓治療，則將收縮壓數值取自然對數乘以第二收縮壓因子；5.抽菸係數是對應於抽菸布林值(Cig)；6.糖尿病係數是對應於糖尿病布林值(DM)；7.高敏感性發炎指標係數是對應於高敏感性發炎指標(hsCRP)；及8.偏移值是對應至多個臨床資料。 The statistical repository 20 stores a plurality of statistical values corresponding to a plurality of physiological features. Multiple statistical values come from complete clinical trials and sufficient statistical data to facilitate the accuracy of the assessment. The plurality of statistical values are for example: 1. The age coefficient corresponds to the age value (Age); 2. The total cholesterol coefficient corresponds to the total cholesterol value (Totalcholesterol); 3. The high density cholesterol coefficient corresponds to the high density cholesterol value (HDLcholesterol); 4. The first systolic blood pressure factor and the second systolic blood pressure factor correspond to the systolic blood pressure The value (SystolicBP), wherein the first systolic pressure factor and the second systolic blood pressure factor are collectively referred to as a systolic blood pressure factor (SysBPfactor). If the user does not receive hypertension treatment, the systolic blood pressure value is taken as the natural logarithm multiplied by the first systolic blood pressure factor. If the user has received hypertension treatment, the systolic blood pressure value is taken as the natural logarithm multiplied by the second systolic blood pressure factor; 5. The smoking coefficient corresponds to the smoking Bollinger value (Cig); 6. The diabetes coefficient corresponds to diabetes Boolean value (DM); 7. High sensitivity inflammatory index coefficient corresponds to high sensitivity inflammatory index (hsCRP); and 8. Offset value corresponds to multiple clinical data.
運算器30是連結至用戶資料庫10及統計資料庫20，並結合多個生理特徵及多個統計值以運算出風險因子(riskfactors)。進一步地，運算器30是組態成依序執行下列步驟：1.將年齡數值(Age)取自然對數乘以年齡係數，其中，年齡係數男性例如是3.06117，女性例如是2.32888；2.加上總膽固醇值(Totalcholesterol)取自然對數乘以總膽固醇係數，其中，總膽固醇係數例如男性是1.12370，女性例如是1.20904；3.減去高密度膽固醇數值(HDLcholesterol)取自然對數乘以高密度膽固醇係數，其中，高密度膽固醇係數男性例如是0.93263，女性例如是0.70833；4.加上收縮壓數值(SystolicBP)取自然對數乘以第一收縮壓因子或第二收縮壓因子，其中，第一收縮壓因子男性例如是1.93303，女性例如是2.76157；第二收縮壓因子男性例如是1.99881，女性例如是2.82263； 5.加上抽菸布林值(Cig)乘以抽菸係數，其中，抽菸係數男性例如是0.65451，女性例如是0.52873；6.加上糖尿病布林值(DM)乘以糖尿病係數，其中，糖尿病係數男性例如是0.57367，女性例如是0.69154；7.加上高敏感性發炎指標(hsCRP)取自然對數乘以高敏感性發炎指標係數，其中，高敏感性發炎指標係數(男性或女性)例如是0.03；8.減去偏移值，其中，偏移值男性例如是23.9802，女性例如是26.1931；及9.輸出執行上列步驟的運算結果，作為風險因子。 The computing unit 30 is coupled to the user database 10 and the statistical database 20, and combines a plurality of physiological features and a plurality of statistical values to calculate risk factors. Further, the operator 30 is configured to perform the following steps in sequence: 1. The age value (Age) is taken as the natural logarithm multiplied by the age coefficient, wherein the age coefficient male is, for example, 3.06117, and the female is, for example, 2.32888; 2. plus The total cholesterol value (Totalcholesterol) is taken as the natural logarithm multiplied by the total cholesterol coefficient, wherein the total cholesterol coefficient is 1.12370 for men and 1.209940 for women; 3. minus high-density cholesterol (HDLcholesterol) is taken by natural logarithm multiplied by high-density cholesterol coefficient. Wherein, the high density cholesterol coefficient male is, for example, 0.93263, and the female is, for example, 0.70833; 4. The systolic blood pressure value (SystolicBP) is taken as the natural logarithm multiplied by the first systolic blood pressure factor or the second systolic blood pressure factor, wherein the first systolic blood pressure The factor male is, for example, 1.93303, the female is, for example, 2.76157; the second systolic blood pressure factor is, for example, 1.99881, and the female is, for example, 2.82263; 5. Add the smoking Bollinger value (Cig) multiplied by the smoking coefficient, wherein the smoking coefficient male is, for example, 0.65451, and the female is, for example, 0.52873; 6. Plus the Diabetes Boolean value (DM) multiplied by the diabetes coefficient, wherein The coefficient of diabetes is, for example, 0.57367 for men and 0.691514 for women. 7. Add high-sensitivity inflammatory index (hsCRP) by natural logarithm multiplied by high-sensitivity inflammatory index coefficient, among which high-sensitivity inflammatory index coefficient (male or female) For example, 0.03; 8. The offset value is subtracted, wherein the offset value is, for example, 23.9802 for men and 26.1931 for women, and 9. The output is the result of performing the above steps as a risk factor.
執行上述步驟將獲得下列運算結果：若用戶是男性，則風險因子riskfactor=ln(Age)×3.06117+ln(Totalcholesterol)×1.12370-ln(HDLcholesterol)×0.93263+ln(SystolicBP)×SysBPFactor+Cig×0.65451+DM×0.57367+ln(hsCRP)×0.03-23.9802 Performing the above steps will obtain the following results: If the user is male, the risk factor riskfactor=ln(Age)×3.06117+ln(Totalcholesterol)×1.12370-ln(HDLcholesterol)×0.93263+ln(SystolicBP)×SysBPFactor+Cig×0.65451 +DM×0.57367+ln(hsCRP)×0.03-23.9802
若用戶是女性，則風險因子riskfactor=ln(Age)×2.32888+ln(Totalcholesterol)×1.20904-ln(HDLcholesterol)×0.70833+ln(SystolicBP)×SysBPFactor+Cig×0.52873+DM×0.69154+ln(hsCRP)×0.03-26.1931 If the user is female, the risk factor riskfactor=ln(Age)×2.32888+ln(Totalcholesterol)×1.20904-ln(HDLcholesterol)×0.70833+ln(SystolicBP)×SysBPFactor+Cig×0.52873+DM×0.69154+ln(hsCRP) ×0.03-26.1931
其中，ln表示取自然對數。 Where ln means taking the natural logarithm.
轉換器40是連結至運算器30，並將風險因子(riskfactor)轉換成風險百分比(risk(%))。進一步地，轉換器40是組態成以下列公式將風險因子(riskfactor)轉換成風險百分比(risk(%))： 若用戶是男性，則risk(%)=100×(1-0.88936×exp(riskfactor)) The converter 40 is coupled to the operator 30 and converts the risk factor into a risk percentage (risk (%)). Further, converter 40 is configured to convert a risk factor into a risk percentage (risk (%)) by the following formula: If the user is male, then risk(%)=100×(1-0.88936×exp(riskfactor))
若用戶是女性，則risk(%)=100×(1-0.95012×exp(riskfactor))其中，exp表示取自然指數。 If the user is a female, risk(%)=100×(1-0.95012×exp(riskfactor)) where exp represents the natural index.
在上述公式中，0.88936及0.95012分別是基於臨床資料，使用Kaplan-Meier預估量(estimator)所得的男性及女性的(Baseline Survival Function)基準存活函數的係數，可用於精準預測用戶罹患心血管疾病的風險。 In the above formula, 0.88936 and 0.95012 are the coefficients of the baseline survival function of the male and female (Baseline Survival Function) obtained by Kaplan-Meier estimator based on clinical data, respectively, which can be used to accurately predict the cardiovascular disease of users. risks of.
第一分類器50是連結至轉換器40，並根據風險百分比(risk(%))將用戶分類至多個族群的一個族群，並輸出分類標籤。進一步地，第一分類器50是組態成：若風險百分比(risk(%))大於20%，則將用戶分類至高度風險族群；若風險百分比(risk(%))介於10%至20%，則將用戶分類至中度風險族群；且若風險百分比(risk(%))小於10%，則將用戶分類至低度風險族群。 The first classifier 50 is coupled to the converter 40 and classifies the user into a group of a plurality of ethnic groups according to the risk percentage (risk (%)), and outputs the classification tag. Further, the first classifier 50 is configured to classify the user into a high risk group if the risk percentage (risk (%)) is greater than 20%; if the risk percentage (risk (%)) is between 10% and 20 %, the user is classified into a moderate risk group; and if the risk percentage (risk (%)) is less than 10%, the user is classified into a low risk group.
圖2顯示本創作另一實施例的風險管理系統2。 Figure 2 shows a risk management system 2 of another embodiment of the present creation.
本創作的風險管理系統2主要包括用戶資料庫10、統計資料庫20、運算器30、轉換器40、第一分類器50及第二分類器60。用戶資料庫10、統計資料庫20、運算器30、轉換器40、第一分類器50皆可採用前述實施例的風險管理系統1所述者。 The risk management system 2 of the present invention mainly includes a user database 10, a statistical database 20, an arithmetic unit 30, a converter 40, a first classifier 50, and a second classifier 60. The user database 10, the statistical database 20, the arithmetic unit 30, the converter 40, and the first classifier 50 can all be as described in the risk management system 1 of the foregoing embodiment.
第二分類器60是連結至第一分類器50，並根據第一分類器50所輸出的分類標籤及調整指數將用戶重新分類至多個族群的一個族群。進一步地，調整指數例如是動脈硬化指數(ABI)，無單位，具體而言，動脈硬化指數(ABI)是腳踝血壓與上臂血壓的比值。 The second classifier 60 is coupled to the first classifier 50 and reclassifies the user to a group of a plurality of ethnic groups based on the classification tag and the adjustment index output by the first classifier 50. Further, the adjustment index is, for example, an arteriosclerosis index (ABI), no unit, specifically, an arteriosclerosis index (ABI) is a ratio of ankle blood pressure to upper arm blood pressure.
圖3顯示本創作一實施例的第二分類器60的重新分類流程圖。 FIG. 3 shows a reclassification flowchart of the second classifier 60 of an embodiment of the present invention.
如圖3所示，第二分類器60是組態成若動脈硬化指數(ABI)小於0.9，則將用戶自低度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至高度風險族群，也就是提高至更高風險之族群；若動脈硬化指數(ABI)介於1.1至1.4，則將用戶自高度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至低度風險族群，也就是降低至更低風險之族群。其他情形，無須重新分類。 As shown in FIG. 3, the second classifier 60 is configured to reclassify the user from a low risk group to a moderate risk group if the arteriosclerosis index (ABI) is less than 0.9; or reclassify from a moderate risk group to Highly risked populations, ie groups that are raised to higher risk; if the arteriosclerosis index (ABI) is between 1.1 and 1.4, the user is reclassified from a high risk group to a moderate risk group; or reclassified from a moderate risk group To the low risk group, that is, to the lower risk group. In other cases, there is no need to reclassify.
圖4顯示本創作再一實施例的風險管理系統3。 Fig. 4 shows a risk management system 3 of still another embodiment of the present creation.
本創作的風險管理系統3主要包括用戶資料庫10、統計資料庫20、運算器30、轉換器40、第一分類器50及回授器70。用戶資料庫10、統計資料庫20、運算器30、轉換器40、第一分類器50皆可採用前述實施例的風險管理系統1所述者。 The risk management system 3 of the present invention mainly includes a user database 10, a statistical database 20, an arithmetic unit 30, a converter 40, a first classifier 50, and a feedback device 70. The user database 10, the statistical database 20, the arithmetic unit 30, the converter 40, and the first classifier 50 can all be as described in the risk management system 1 of the foregoing embodiment.
回授器70是連結至統計資料庫20，並根據多個用戶的多個臨床資料更新多個統計值。進一步地，回授器70可連接至第一分類器50，以比對用戶的分類標籤及其實際患病的情況，以判斷第一分類器50的分類是否正確，進而更新多個統計值，使得根據新的多個統計值運算而得的新的分類標籤符合用戶實際患病的情況。 The retriever 70 is linked to the statistical database 20 and updates a plurality of statistical values based on a plurality of clinical data of a plurality of users. Further, the feedback device 70 can be connected to the first classifier 50 to compare the classification label of the user and the actual condition of the user to determine whether the classification of the first classifier 50 is correct, thereby updating a plurality of statistical values. The new classification label calculated according to the new plurality of statistical values is made to conform to the actual illness of the user.
可理解的是，回授器70亦適用於本創作的風險管理系統2，在本例中，回授器可進一步連接至第二分類器60。 It will be appreciated that the retriever 70 is also applicable to the risk management system 2 of the present author, in which case the backhaul may be further coupled to the second classifier 60.
圖5顯示本創作一實施例的風險管理方法。 FIG. 5 shows a risk management method of an embodiment of the present creation.
本創作的風險管理方法適於由電腦程式產品執行，以供電腦系統使用。其可實現為有形或非暫態媒體的一系列電腦可讀指令，而存在於電腦可讀媒體，例如磁碟、CD-ROM、ROM、快閃記憶體或硬碟。 The risk management method of this creation is suitable for execution by a computer program product for use by a computer system. It can be implemented as a series of computer readable instructions of tangible or non-transitory media, but on a computer readable medium such as a magnetic disk, CD-ROM, ROM, flash memory or hard disk.
本創作的風險管理方法，主要包括下列步驟A、步驟B及步驟C。此外，可包括下列步驟D及步驟E。 The risk management method of the present invention mainly includes the following steps A, B and C. In addition, the following steps D and E can be included.
步驟A是取得用戶的多個生理特徵，包括：年齡數值Age、總膽固醇數值Totalcholesterol、高密度膽固醇數值HDLcholesterol、收縮壓數值SystolicBP、抽菸布林值Cig、糖尿病布林值DM及高敏感性發炎指標hsCRP。 Step A is to obtain a plurality of physiological characteristics of the user, including: age value Age, total cholesterol value Totalcholesterol, high density cholesterol value HDLcholesterol, systolic blood pressure value SystolicBP, smoking Bollinger value Cig, diabetes Bollinger value DM, and high sensitivity inflammation Indicator hsCRP.
步驟B是運算出風險因子riskfactors：若用戶是男性，則：riskfactor=ln(Age)×3.06117+In(Totalcholesterol)×1.12370-ln(HDLcholesterol)×0.93263+ln(SystolicBP)×SysBPFactor+Cig×0.65451+DM×0.57367+ln(hsCRP)×0.03-23.9802 Step B is to calculate the risk factor riskfactors: if the user is male, then: riskfactor=ln(Age)×3.06117+In(Totalcholesterol)×1.12370-ln(HDLcholesterol)×0.93263+ln(SystolicBP)×SysBPFactor+Cig×0.65451+ DM×0.57367+ln(hsCRP)×0.03-23.9802
若用戶是女性，則：riskfactor=ln(Age)×2.32888+In(Totalcholesterol)×1.20904-ln(HDLcholesterol)×0.70833+ln(SystolicBP)×SysBPFactor+Cig×0.52873+DM×0.69154+ln(hsCRP)×0.03-26.1931 If the user is female, then: riskfactor=ln(Age)×2.32888+In(Totalcholesterol)×1.20904-ln(HDLcholesterol)×0.70833+ln(SystolicBP)×SysBPFactor+Cig×0.52873+DM×0.69154+ln(hsCRP)× 0.03-26.1931
其中，ln表示取自然對數；且SysBPFactor的數值如下：
步驟C是將風險因子轉換成風險百分比risk(%)：若用戶是男性，則risk(%)=100×(1-0.88936×exp(riskfactor)) Step C is to convert the risk factor into risk percentage risk (%): if the user is male, then risk (%) = 100 × (1-0.88936 × exp (riskfactor))
若用戶是女性，則risk(%)=100×(1-0.95012×exp(riskfactor)) If the user is female, then risk(%)=100×(1-0.95012×exp(riskfactor))
其中，exp()表示取自然指數。 Among them, exp() means taking the natural index.
步驟D是根據風險百分比進行第一次分類：若風險百分比大於20%，則將用戶分類至高度風險族群；若風險百分比介於10%至20%，則將用戶分類至中度風險族群；且若風險百分比小於10%，則將用戶分類至低度風險族群；及 Step D is to classify the first time according to the percentage of risk: if the risk percentage is greater than 20%, the user is classified into a high risk group; if the risk percentage is between 10% and 20%, the user is classified into a moderate risk group; If the risk percentage is less than 10%, the user is classified into a low risk group; and
步驟E是根據動脈硬化指數進行第二次分類：若動脈硬化指數小於0.9，則將用戶自低度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至高度風險族群；若動脈硬化指數介於1.1至1.4，則將用戶自高度風險族群重新分類至中度風險族群；或自中度風險族群重新分類至低度風險族群。 Step E is a second classification based on the arteriosclerosis index: if the arteriosclerosis index is less than 0.9, the user is reclassified from a low risk group to a moderate risk group; or from a moderate risk group to a highly risk group; The arteriosclerosis index ranged from 1.1 to 1.4, reclassifying users from high-risk groups to moderate risk groups; or reclassifying them from moderate risk groups to low-risk groups.
綜上所述，根據本創作的風險管理系統及方法，可基於用戶的多個生理特徵完整地評估其在十年間罹患心血管疾病的風險，而將其分類至正確的風險族群，以達成有效的管理。醫療人員可根據本創作的評估結果給予用戶保健或治療方案。 In summary, according to the risk management system and method of the present invention, the risk of cardiovascular disease in a decade can be completely evaluated based on multiple physiological characteristics of the user, and classified into the correct risk group to achieve effective Management. The medical staff can give the user a health care or treatment plan based on the evaluation results of this creation.
此外，本創作的風險管理系統及方法的各項係數來自完整的臨床實驗及充分的統計數據，並可透過回授進行動態更新，以促進評估的精準度。 In addition, the various factors of the risk management system and method of this creation come from complete clinical trials and sufficient statistical data, and can be dynamically updated through feedback to promote the accuracy of the assessment.
儘管本創作已透過上述實施例說明，可理解的是，在不悖離本創作精神及申請專利範圍之下，可進行許多其他修飾及變化。 Although the present invention has been described in the above embodiments, it is understood that many other modifications and changes can be made without departing from the spirit of the invention.
1‧‧‧風險管理系統 1‧‧‧Risk Management System
10‧‧‧用戶資料庫 10‧‧‧User database
20‧‧‧統計資料庫 20‧‧‧Statistical database
30‧‧‧運算器 30‧‧‧Operator
40‧‧‧轉換器 40‧‧‧ converter
50‧‧‧第一分類器 50‧‧‧First classifier
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW106209139U TWM552639U (en) | 2017-06-23 | 2017-06-23 | Risk management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW106209139U TWM552639U (en) | 2017-06-23 | 2017-06-23 | Risk management system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TWM552639U true TWM552639U (en) | 2017-12-01 |
Family
ID=61228704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW106209139U TWM552639U (en) | 2017-06-23 | 2017-06-23 | Risk management system |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWM552639U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI715250B (en) * | 2019-10-17 | 2021-01-01 | 宏碁股份有限公司 | Feature identifying method and electronic device |
-
2017
- 2017-06-23 TW TW106209139U patent/TWM552639U/en not_active IP Right Cessation
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI715250B (en) * | 2019-10-17 | 2021-01-01 | 宏碁股份有限公司 | Feature identifying method and electronic device |
| US11844633B2 (en) | 2019-10-17 | 2023-12-19 | Acer Incorporated | Feature identifying method and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Collins et al. | An independent external validation and evaluation of QRISK cardiovascular risk prediction: a prospective open cohort study | |
| US20110202486A1 (en) | Healthcare Information Technology System for Predicting Development of Cardiovascular Conditions | |
| JP2019509101A (en) | System and method for determining a hemodynamic instability risk score for pediatric subjects | |
| US20170329918A1 (en) | Internet of things based monitoring and assessment platform | |
| Jørgensen et al. | Electrocardiographic changes improve risk prediction in asymptomatic persons age 65 years or above without cardiovascular disease | |
| Pollard et al. | Perianesthetic and anesthesia-related mortality in a southeastern United States population: a longitudinal review of a prospectively collected quality assurance data base | |
| WO2020248847A1 (en) | Smart heart disease detection method and device, and computer readable storage medium | |
| Leiter et al. | Deep natural language processing to identify symptom documentation in clinical notes for patients with heart failure undergoing cardiac resynchronization therapy | |
| CN114023441A (en) | Severe AKI early risk assessment model and device based on interpretable machine learning model and development method thereof | |
| CN112967803A (en) | Early mortality prediction method and system for emergency patients based on integrated model | |
| CN115862850B (en) | Modeling method and device of hepatocellular carcinoma monitoring model based on longitudinal multidimensional data | |
| Manemann et al. | Longitudinal cohorts for harnessing the electronic health record for disease prediction in a US population | |
| Amutha et al. | A novel approach for the prediction of treadmill test in cardiology using data mining algorithms implemented as a mobile application | |
| AU2016247433A1 (en) | A probabilistic context free grammar for food intake | |
| Kaur et al. | Artificial Intelligence approaches for Predicting Hypertension Diseases: Open Challenges and Research Issues | |
| Chong et al. | Development of automated triage system for emergency medical service | |
| Mamun | Cuff-less blood pressure measurement based on hybrid feature selection algorithm and multi-penalty regularized regression technique | |
| TWM552639U (en) | Risk management system | |
| Reisner et al. | Prehospital heart rate and blood pressure increase the positive predictive value of the Glasgow Coma Scale for high-mortality traumatic brain injury | |
| TWI628591B (en) | Risk management system and method | |
| Ma et al. | Birth outcome measures and infant mortality | |
| Xu et al. | Predicting recurrence for patients with ischemic cerebrovascular events based on process discovery and transfer learning | |
| TW202333618A (en) | Method, electronic system, and computer program product for establishing decision tree for disease prediction | |
| Hsieh et al. | Simplified radius, ulna, and short bone‐age assessment procedure using grouped‐Tanner‐Whitehouse method | |
| CN111968747A (en) | VTE intelligent prevention and control management system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MM4K | Annulment or lapse of a utility model due to non-payment of fees |