Construction method and system of deep vein thrombosis risk assessment model

Abstract

The invention relates to the technical field of data processing, in particular to a deep vein thrombosis risk assessment model construction method and system, and the method comprises the steps: obtaining target clinical data of a tibial plateau fracture patient; based on the target clinical data, grouping the tibial plateau fracture patients to obtain a concurrent deep venous thrombosis group and a non-concurrent deep venous thrombosis group; based on the target clinical data, the concurrent deep venous thrombosis group and the non-concurrent deep venous thrombosis group, determining a plurality of independent prediction factors for predicting deep venous thrombosis in the target clinical data; and based on the plurality of independent predictive factors, constructing a deep vein thrombosis risk assessment model in the form of a column diagram. According to the method, a more accurate evaluation tool can be provided, so that a high-risk patient can be identified in time, individual intervention is guided, and the risk of DVT of a tibial plateau fracture patient is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for constructing a deep vein thrombosis risk assessment model in the technical field of data processing. Background Art

[0002] Tibial plateau fractures are a common type of lower limb fracture, usually caused by high-energy trauma or falls. This type of fracture not only has a significant impact on the patient's functional recovery, but is also accompanied by a high risk of complications. Among them, deep vein thrombosis (DVT) is one of the most common and serious complications. The occurrence of DVT may lead to fatal consequences, such as pulmonary embolism, posing a major threat to the patient's life safety and functional recovery. Therefore, in clinical practice, early identification and assessment of the risk of DVT in patients with tibial plateau fractures is crucial to optimize treatment strategies and prevent adverse prognosis.

[0003] At present, D-dimer testing is usually used to assess the risk of DVT in clinical practice. This method is widely used because of its high sensitivity. However, relevant studies have shown that D-dimer testing has limitations in specific circumstances, especially in elderly patients and patients with significantly elevated C-reactive protein levels, which may produce false positive results or misleading assessments. This deficiency limits its applicability in clinical practice. In addition, the evaluation method of a single biomarker is difficult to fully reflect the patient's individualized risk, making the prediction accuracy of deep vein thrombosis low. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for constructing a deep vein thrombosis risk assessment model. The technical solutions adopted are as follows: In a first aspect, an embodiment of the present invention provides a method for constructing a deep vein thrombosis risk assessment model, the method comprising: Obtain targeted clinical data on patients with tibial plateau fractures; Based on the target clinical data, the patients with tibial plateau fractures were divided into a group complicated with deep vein thrombosis and a group not complicated with deep vein thrombosis; Based on the target clinical data, the complicated deep vein thrombosis group and the uncomplicated deep vein thrombosis group, determining a plurality of independent predictive factors for predicting deep vein thrombosis in the target clinical data; Based on the multiple independent predictive factors, a deep vein thrombosis risk assessment model was constructed in the form of a nomogram.

[0005] In a second aspect, a system for constructing a deep vein thrombosis risk assessment model is provided, the system comprising: an acquisition module for acquiring target clinical data of patients with tibial plateau fractures; A grouping module, for grouping the patients with tibial plateau fractures based on the target clinical data to obtain a group complicated with deep vein thrombosis and a group not complicated with deep vein thrombosis; A determination module, configured to determine a plurality of independent predictive factors for predicting deep vein thrombosis in the target clinical data based on the target clinical data, the complicated deep vein thrombosis group and the uncomplicated deep vein thrombosis group; A building module is used to construct a deep vein thrombosis risk assessment model in the form of a nomogram based on the multiple independent predictors.

[0006] According to a third aspect, a computer program product is provided. The computer program product includes: a computer program code. When the computer program code is executed on a computer, the computer executes the method according to the first aspect.

[0007] According to a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program code. When the computer program code is executed on a computer, the computer executes the method according to the first aspect.

[0008] The present invention has the following beneficial effects: after obtaining the target clinical data of patients with tibial plateau fractures, the patients with tibial plateau fractures are grouped according to the target clinical data to obtain a complicated deep vein thrombosis group and a non-complicated deep vein thrombosis group; then, the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group are combined, and the target clinical data are screened to determine multiple independent predictive factors for predicting deep vein thrombosis; in this way, the target clinical data is quickly obtained through clinical examination, thereby simplifying the process of determining independent predictive factors, making the constructed deep vein thrombosis risk assessment model simple and easy to use, suitable for clinical application. Finally, a deep vein thrombosis risk assessment model is constructed in the form of a nomogram through independent predictive factors; in this way, constructing a deep vein thrombosis risk assessment model in the form of a nomogram helps to intuitively assess the risk of patients with complicated deep vein thrombosis, and is easy to operate and has clear results, and can provide a more accurate assessment tool, so as to timely identify high-risk patients and guide individualized intervention, thereby reducing the risk of DVT in patients with tibial plateau fractures. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0010] Figure 1 It is a schematic diagram of the implementation process of a method for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of an application scenario of a method for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 3 It is a structural schematic diagram of a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 4 It is another implementation flow diagram of a method for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 5 It is a schematic diagram of verifying a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 6 is another verification schematic diagram of a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 7 This is another verification schematic diagram of a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Figure 8 It is another verification schematic diagram of a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Fig. 9 It is a schematic diagram of the composition structure of a system for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention; Fig.10 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the construction method of a deep vein thrombosis risk assessment model proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0012] Among them, in the description of the embodiments of the present invention, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a way to describe the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" refers to two or more than two.

[0013] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0014] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0015] The following is a detailed description of a method for constructing a deep vein thrombosis risk assessment model provided by the present invention in conjunction with the accompanying drawings. Figure 1 , which shows a schematic diagram of an implementation flow of a method for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention, the method comprising: 101, to obtain targeted clinical data for patients with tibial plateau fractures.

[0016] Here, the target clinical data is obtained by screening the acquired patients with tibial plateau fractures. The target clinical data includes: medical quantitative data, the medical qualitative data and the medical image feature data. Among them, the medical quantitative data includes: age, body mass index (BMI), white blood cell count, neutrophil count, lymphocyte count, red blood cell count, hemoglobin level, platelet count, electrolyte indicators (such as potassium, sodium, calcium), blood sugar level, alanine aminotransferase (ALT), aspartate aminotransferase (AST), uric acid, urea, creatinine, fibrinogen, D-dimer, and parameters related to coagulation function, including activated partial thromboplastin time (APTT), thrombin time (TT) and international normalized ratio (INR), etc.

[0017] In some possible implementations, the above step 101 may be implemented by the following steps 111 to 113 (not shown): 111, according to the pre-specified inclusion criteria, the initial clinical data of patients with multiple tibial plateau fractures were obtained.

[0018] Here, the preset inclusion criteria can be custom set, for example, the preset inclusion criteria include: (a) age > 18 years old; (b) confirmed as tibial plateau fracture by clinical symptoms or imaging examination; (c) complete clinical and imaging data; (d) receiving complete surgical treatment and hospitalization management. Based on the preset inclusion criteria, patients with tibial plateau fractures are screened to obtain patients with tibial plateau fractures that meet the preset inclusion criteria, and then the initial clinical data of the patients with tibial plateau fractures can be obtained by clinically examining the patients with tibial plateau fractures. The initial clinical data include: age, body mass index, white blood cell count, neutrophil count, lymphocyte count, red blood cell count, hemoglobin level, platelet count, electrolyte indicators (such as potassium, sodium, calcium), blood glucose level, alanine aminotransferase, aspartate aminotransferase, uric acid, urea, creatinine, fibrinogen, D-dimer, and parameters related to coagulation function. Among them, the parameters related to coagulation function include: activated partial thromboplastin time, thrombin time, international normalized ratio, whether the patient suffers from hypertension and diabetes, Schatzker fracture classification, Doppler ultrasound imaging data and X-ray film imaging data, etc.

[0019] 112. Exclude the initial clinical data that meet a preset exclusion criterion from the initial clinical data of the multiple patients with tibial plateau fractures to obtain candidate clinical data.

[0020] Here, the preset exclusion criteria include: (a) old or pathological tibial plateau fractures; (b) the presence of autoimmune diseases or blood system diseases; (c) a history of anticoagulant or antiplatelet drug use; (d) combined vascular injury or multiple fractures; (e) a history of cardiovascular disease, liver or kidney dysfunction; (f) a diagnosis of gout or other diseases related to uric acid metabolism. Based on this, the data that meet the preset exclusion criteria in the initial clinical data are excluded to obtain candidate clinical data.

[0021] 113. Process the clinical data to obtain medical quantitative data, medical qualitative data, and medical image feature data of the patient with tibial plateau fracture.

[0022] The target clinical data includes: the medical quantitative data, the medical qualitative data and the medical image feature data. The medical quantitative data covers a number of physiological and biochemical indicators, such as age, body mass index (BMI), white blood cell count, neutrophil count, lymphocyte count, red blood cell count, hemoglobin level, platelet count, electrolyte indicators (such as potassium, sodium, calcium), blood sugar level, alanine aminotransferase (ALT), aspartate aminotransferase (AST), uric acid, urea, creatinine, fibrinogen, D-dimer, and parameters related to coagulation function, including activated partial thromboplastin time (APTT), thrombin time (TT) and international normalized ratio (INR).

[0023] The patient's physiological data is used to analyze whether the patient suffers from diabetes, hypertension, etc., and the patient's imaging analysis indicators (including Doppler ultrasound imaging data and X-ray film imaging data, etc.) are used to determine the patient's Schatzker fracture classification, whether the patient suffers from deep vein thrombosis, etc., thereby obtaining the patient's medical qualitative data; among which, the medical qualitative data include: gender, whether the patient suffers from hypertension, whether the patient suffers from diabetes, Schatzker fracture classification, and whether the patient suffers from deep vein thrombosis.

[0024] By extracting features from the patient's imaging analysis indicators (including Doppler ultrasound imaging data and X-ray film imaging data, etc.), the patient's medical image feature data is obtained. In this way, the initial clinical data is screened through the preset inclusion criteria and the preset exclusion criteria, so that the target clinical data that is convenient for building a deep vein thrombosis risk assessment model can be obtained.

[0025] Here, patients with tibial plateau fractures are divided into two groups according to the medical image feature data, namely: a group with complicated deep vein thrombosis and a group without complicated deep vein thrombosis; among them, the diagnostic criteria for deep vein thrombosis are in line with the diagnostic criteria for deep vein thrombosis in the "Guidelines for the Diagnosis and Treatment of Deep Vein Thrombosis", that is, patients show symptoms such as swelling and pain in the affected limb, increased soft tissue tension, and increased skin temperature, and deep vein thrombosis is confirmed by imaging techniques such as color Doppler ultrasound and CT.

[0026] 102. Based on the target clinical data, the patients with tibial plateau fractures are grouped into a group with complicated deep vein thrombosis and a group without complicated deep vein thrombosis.

[0027] Here, since the target clinical data includes: medical quantitative data, medical qualitative data and medical image feature data, multiple tibial plateau fracture patients are grouped according to the image feature data to obtain a group with complicated deep vein thrombosis and a group without complicated deep vein thrombosis.

[0028] 103. Based on the target clinical data, the group with complicated deep vein thrombosis and the group without complicated deep vein thrombosis, determine a plurality of independent predictive factors for predicting deep vein thrombosis in the target clinical data.

[0029] Here, the target clinical data are grouped to obtain a training data set and a validation data set, thereby screening variables with greater contributions from the training data set according to a confidence threshold, that is, obtaining multiple independent predictors.

[0030] In some possible implementations, the above step 103 may be implemented by the following steps 131 to 133 (not shown): 131. Based on the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group, divide the target clinical data into a training data set and a validation data set.

[0031] Here, the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group were input into R software and randomly divided into training data set and validation data set in a ratio of 7:3 using the caret package.

[0032] 132. Performing logistic regression on the training data set to obtain the contribution of each training data in the training data set to the deep vein thrombosis.

[0033] Here, the glmnet package is called in R software to use the least absolute shrinkage and selection operator (LASSO) regression on the training data set to screen the variables that contribute most to the outcome indicator deep vein thrombosis. Among them, the cross-validation curve of LASSO regression is as follows: Figure 2 As shown in curve 21 in the figure, the vertical axis represents the binomial deviation, which is an error measure used to measure the performance of the model in cross-validation. The smaller the deviation, the better the performance of the model.

[0034] 133. Screen the training data set according to the confidence threshold and the contribution of each training data to the deep vein thrombosis to obtain the multiple independent prediction factors.

[0035] Here, five variables were selected from age, sex, body mass index, history of hypertension, history of diabetes, white blood cell count, neutrophil, lymphocyte, red blood cell count, hemoglobin, platelet count, potassium ion, sodium ion, calcium ion, glucose, alanine aminotransferase, aspartate aminotransferase, uric acid, urea, creatinine, fibrinogen, partially activated thromboplastin time, thrombin time, international normalized ratio, D-dimer, Schatzker fracture classification, etc. The generalized linear model (GLM) function was called in R software, and multivariate logistic regression was used to confirm that the five variables selected in the LASSO regression of the training data set were independent predictors of DVT. In this way, the training data set was screened according to the contribution of each training data to deep vein thrombosis, so as to screen out multiple independent predictors with greater contribution, thereby making the constructed deep vein thrombosis risk assessment model more accurate.

[0036] 104. Based on the independent predictive factors, a deep vein thrombosis risk assessment model is constructed in the form of a nomogram.

[0037] In some possible implementations, the above step 104 may be implemented by the following steps 141 to 143 (not shown): 141, determining the influence value of each of the multiple independent predictors on the occurrence of deep vein thrombosis in the patient with tibial plateau fracture.

[0038] Here, after obtaining multiple independent predictors, the influence of each independent predictor on deep vein thrombosis in patients with tibial plateau fractures can be obtained by calling the generalized linear model function in R software. For example, age (odds ratio [OR] 1.035, 95% confidence interval [95% CI] 1.003-1.068), hypertension (OR 2.918, 95% CI 1.206-7.064), Schatzker fracture classification (OR 4.401, 95% CI 2.051-9.444), D-dimer (OR 1.053, 95% CI 1.002-1.110), and uric acid (OR 0.990, 95% CI 0.985-0.995).

[0039] 142. Determine a risk assessment value for deep vein thrombosis in the patient with tibial plateau fracture based on the influence value of each independent predictor on the deep vein thrombosis and the weight corresponding to each independent predictor.

[0040] In some possible implementations, the total impact value of the multiple independent predictive factors on the tibial plateau fracture patient's risk of deep vein thrombosis is determined by using the impact values ​​corresponding to the multiple independent predictive factors; and based on the total impact value and the weight, the risk assessment value of the tibial plateau fracture patient's risk of deep vein thrombosis is determined.

[0041] Here, multiple independent predictors were identified, including age, history of hypertension, uric acid, D-dimer, and Schatzker fracture classification.

[0042] In some possible implementations, the weight corresponding to each independent prediction factor is multiplied by the corresponding influence value to obtain the multiplication result corresponding to each independent prediction factor; and the risk assessment value is determined based on the proportion of the multiplication result corresponding to each independent prediction factor in the total influence value.

[0043] Among them, the multiplication results corresponding to each independent predictor include: 0.035×age, 1.489×Schatzker fracture classification, 1.075×hypertension, 0.052×D-dimer, -0.010×uric acid, etc. These multiple multiplication results are summed to obtain the total impact value; the risk assessment value can be obtained by calculating the proportion of the multiplication results corresponding to each independent predictor in the total impact value. Here, the total impact value is: 0.035×age+1.489×Schatzker fracture classification+1.075×hypertension+0.052×D-dimer-0.010×uric acid-1.285. Based on this, the risk assessment value obtained is, Figure 3 In this way, by multiplying the influence value of each independent predictor on the deep vein thrombosis and the weight corresponding to each independent predictor, and calculating the percentage of the multiplication result, a more accurate risk assessment value can be obtained.

[0044] 143. Constructing the deep vein thrombosis risk assessment model based on the multiple independent predictors, the impact values ​​and the risk assessment values.

[0045] In some possible implementations, the ordinate of the deep vein thrombosis risk assessment model is constructed by the names corresponding to each independent prediction factor, the total impact value, and the risk assessment value; and the abscissa of the deep vein thrombosis risk assessment model is constructed by the impact value corresponding to each independent prediction factor, the total impact value, and the numerical value corresponding to the risk assessment; finally, the deep vein thrombosis risk assessment model is constructed by the ordinate and abscissa.

[0046] Here, the Regression Modeling Strategies (RMS) package was called in the R software to construct a Nomogram risk assessment model using the independent predictors confirmed by multivariate logistic regression: age, history of hypertension, uric acid, D-dimer, and Schatzker fracture classification. Figure 3 As shown in the figure, the risk model includes: the horizontal axis shows the risk relationship between deep vein thrombosis and age, Schatzker fracture classification, hypertension, D-dimer, and uric acid in patients with tibial plateau fractures; the vertical axis shows the corresponding "score" value of each risk factor vertically; the "total score" is the sum of the "score" values ​​corresponding to all predictive factors, that is, the total impact value, which vertically corresponds to the "risk of suffering from deep vein thrombosis", that is, the risk assessment value of deep vein thrombosis in patients with tibial plateau fractures. In this model, the risk relationship between age, Schatzker fracture classification, hypertension, D-dimer, and uric acid is combined to establish a risk assessment model for deep vein thrombosis in patients with tibial plateau fractures. In this way, by combining multiple independent factors, corresponding impact values, and risk assessment values, a high-precision deep vein thrombosis risk assessment model can be created.

[0047] In some possible implementations, after building a deep vein thrombosis risk assessment model, the performance of the model is verified using a training data set and a validation data set, which can be achieved in the following ways: Method 1: First, based on the verification data set and the training data set, the deep vein thrombosis risk assessment model is respectively verified by ROC curve to obtain a first verification result and a second verification result; then, based on the first verification result and the second verification result, the clinical application value of the deep vein thrombosis risk assessment model is determined.

[0048] Here, by averaging the first verification result and the second verification result, the clinical application value can be obtained. The program package was called in the R software to perform ROC curve analysis on the prediction model, and the curve areas corresponding to the training data set and the verification data set (that is, the first verification result and the second verification result) were 0.838, (95% CI: 0.786-0.894) and 0.801, (95% CI: 0.675-0.923), respectively; where the curve area is the area under the ROC curve and the coordinate axis. The results show that there are higher curve areas in the training data set and the verification data set, which means that the deep vein thrombosis risk assessment model has a higher clinical application value. Figure 5 As shown, Figure 5 Figure A in Figure 1 shows the first verification result. Figure 5 Figure B in FIG. 1 represents the second verification result.

[0049] Method 2: Based on the verification data set and the training data set, a statistical test is performed on the deep vein thrombosis risk assessment model to obtain a statistical test result.

[0050] like Figure 6 As shown, the ResourceSelection package was called in the R software, and the Hosmer-Lemeshow test was performed on the risk assessment models applied to the training data set and the validation data set, where Figure 6 Figure A in the figure shows the statistical test results of the Hosmer-Lemeshow test on the training data set. Figure 6 Figure B in Figure 2 shows the statistical test results of the Hosmer-Lemeshow test on the validation data set. Figure 6 It can be seen that the P values ​​are all greater than 0.05, with no statistical difference. The results show that the estimated risk of the model is basically consistent with the actual risk.

[0051] Method three: Determine the decision curve and clinical impact curve of the deep vein thrombosis risk assessment model based on the validation data set and the training data set.

[0052] Based on the validation data set and the training data set, a decision curve of the deep vein thrombosis risk assessment model is determined, such as Figure 7 As shown, the package is called in the R software to perform decision curve analysis on the training data set and the validation data set, where: Figure 7 Figure A in is the decision curve corresponding to the training data set. Figure 7 Figure B in Figure 2 is the decision curve corresponding to the validation data set. Figure 7 The results shown indicate that, over a wide range, intervention for patients at high risk of DVT assessed by the risk assessment model is more beneficial than all treatment or no treatment, and the model has a high clinical applicability.

[0053] Based on the validation data set and the training data set, the clinical impact curve of the deep vein thrombosis risk assessment model is determined as follows: Figure 8 As shown, the clinical impact curve was drawn by calling the package in R software to evaluate the probability stratification in the 1000-person population; Figure 8 Figure A in is the clinical impact curve corresponding to the training data set. Figure 8 Figure B in Figure 1 is the clinical impact curve corresponding to the validation dataset. Figure 8 The clinical impact curves shown demonstrate high agreement between the model's estimated risk and the actual risk in both the training and validation datasets.

[0054] After obtaining the statistical test results, the decision curve and the clinical impact curve in the above manner, the performance of the deep vein thrombosis risk assessment model is verified based on at least one of the statistical test results, the decision curve and the clinical impact curve. In an embodiment of the present invention, after obtaining the target clinical data of patients with tibial plateau fractures, the target clinical data is screened to determine independent predictive factors for predicting deep vein thrombosis; in this way, the target clinical data is quickly obtained through clinical examinations, thereby simplifying the process of determining independent predictive factors, making the constructed deep vein thrombosis risk assessment model simple and easy to use, suitable for clinical application. Finally, through independent predictive factors, a deep vein thrombosis risk assessment model is constructed in the form of a nomogram; in this way, constructing a deep vein thrombosis risk assessment model in the form of a nomogram helps to intuitively assess the risk of patients with deep vein thrombosis, and is easy to operate and has clear results, and can provide a more accurate assessment tool to facilitate timely identification of high-risk patients and guide individualized intervention, thereby reducing the risk of DVT in patients with tibial plateau fractures.

[0055] In some possible implementations, the method for constructing a deep vein thrombosis risk assessment model provided by the embodiments of the present invention can be implemented by the following process: In the first step, patients with tibial plateau fractures who met the inclusion and exclusion criteria were screened.

[0056] like Figure 4 As shown, 41 patients with tibial plateau fractures who met the inclusion criteria were identified.

[0057] In the second step, the clinical data of several patients with tibial plateau fractures selected in the first step were collected and processed to obtain quantitative data, qualitative data and image feature data. According to the image feature data, the patients were divided into a group with complicated deep vein thrombosis and a group without complicated deep vein thrombosis.

[0058] In the third step, the data set consisting of the quantitative data, qualitative data, image feature data and grouping information obtained in the second step was imported into R software and randomly grouped into a training data set and a validation data set in a ratio of 7:3 using the caret package; Figure 4 As shown, a validation set 42 and a training set 43 are obtained.

[0059] The fourth step is to use LASSO regression on the training data set selected in the third step (such as Figure 4 LASSO regression as shown in 44) was used to screen for variables that contribute significantly to the outcome indicator, deep vein thrombosis.

[0060] In the fifth step, based on the variables selected in the fourth step, multivariate logistic regression (such as Figure 4 Logistic regression analysis (45) was performed to confirm whether it was an independent predictor of deep vein thrombosis.

[0061] In the sixth step, the nomogram risk assessment model was constructed using the independent predictors established in the fifth step (e.g. Figure 4 Nomogram model 401 is shown).

[0062] In the seventh step, the ROC curve, clinical decision curve and clinical impact curve (e.g. Figure 4 Receiver operating characteristic (ROC) curve 46 , calibration and decision curves 47 , and clinical impact curve 48 ) are shown.

[0063] The clinical application of the risk assessment model was analyzed by using the receiver operating characteristic (ROC) curve46, calibration curve, decision curve47 and clinical impact curve48. Figure 4 Clinical application shown 49. Included: Patient number: Sample-1, DVT risk 80%; Patient number: Sample-2, DVT risk 60%.

[0064] In an embodiment of the present invention, the deep vein thrombosis risk assessment model is used to assess the risk of deep vein thrombosis in patients with tibial plateau fractures. Based on parameters that can be quickly obtained through clinical examination, it is easy to operate and convenient for clinical promotion and application, and can intuitively assess the possibility of deep vein thrombosis in patients. The implementation method of the model is simple and the results are clear, which provides important support for clinicians to optimize treatment strategies.

[0065] The present invention provides a system for constructing a deep vein thrombosis risk assessment model. Fig. 9 , which shows a schematic diagram of the composition structure of a system for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention, the system 900 includes: An acquisition module 901 is used to acquire target clinical data of patients with tibial plateau fractures; A grouping module 902 is used to group the patients with tibial plateau fractures based on the target clinical data to obtain a group complicated with deep vein thrombosis and a group not complicated with deep vein thrombosis; A determination module 903, configured to determine a plurality of independent predictive factors for predicting deep vein thrombosis in the target clinical data based on the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group; The construction module 904 is used to construct a deep vein thrombosis risk assessment model in the form of a nomogram based on the multiple independent predictors.

[0066] In some possible implementations, the construction module 904 is further used to determine the influence value of each independent predictor among the multiple independent predictors on the deep vein thrombosis complicated by the patient with tibial plateau fracture; determine the risk assessment value of the patient with tibial plateau fracture suffering from deep vein thrombosis based on the influence value of each independent predictor on the deep vein thrombosis and the weight corresponding to each independent predictor; and construct the deep vein thrombosis risk assessment model based on the multiple independent predictors, the influence values ​​and the risk assessment values.

[0067] In some possible implementations, the construction module 904 is further used to determine the total impact value of the multiple independent predictive factors on the tibial plateau fracture patient suffering from deep vein thrombosis based on the impact values ​​corresponding to the multiple independent predictive factors; and determine the risk assessment value of the tibial plateau fracture patient suffering from deep vein thrombosis based on the total impact value and the weight.

[0068] In some possible implementations, the construction module 904 is further used to construct the ordinate of the deep vein thrombosis risk assessment model based on the name corresponding to each independent prediction factor, the total impact value, and the risk assessment value; to construct the abscissa of the deep vein thrombosis risk assessment model based on the impact value corresponding to each independent prediction factor, the total impact value, and the numerical value corresponding to the risk assessment; and to obtain the deep vein thrombosis risk assessment model based on the ordinate and the abscissa.

[0069] In some possible implementations, the construction module 904 is further used to multiply the weight corresponding to each independent prediction factor by the corresponding influence value to obtain the multiplication result corresponding to each independent prediction factor; and determine the risk assessment value based on the proportion of the multiplication result corresponding to each independent prediction factor in the total influence value.

[0070] In some possible implementations, the acquisition module 901 is further used to acquire initial clinical data of multiple patients with tibial plateau fractures according to preset inclusion criteria; among the initial clinical data of the multiple patients with tibial plateau fractures, the initial clinical data that meet the preset exclusion criteria are excluded to obtain candidate clinical data; and the candidate clinical data are processed to obtain medical quantitative data, medical qualitative data, and medical image feature data of the patients with tibial plateau fractures; wherein the target clinical data includes: the medical quantitative data, the medical qualitative data, and the medical image feature data.

[0071] In some possible implementations, the determination module 903 is further used to divide the target clinical data into a training data set and a validation data set based on the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group; perform logistic regression on the training data set to obtain the contribution of each training data in the training data set to the deep vein thrombosis; and screen the training data set according to the confidence threshold and the contribution of each training data to the deep vein thrombosis to obtain the multiple independent predictors.

[0072] In some possible implementations, the construction module 904 is also used to perform ROC curve verification on the deep vein thrombosis risk assessment model based on the verification data set and the training data set, respectively, to obtain a first verification result and a second verification result; based on the first verification result and the second verification result, determine the clinical application value of the deep vein thrombosis risk assessment model.

[0073] In some possible implementations, the construction module 904 is further used to perform a statistical test on the deep vein thrombosis risk assessment model based on the validation data set and the training data set to obtain a statistical test result; determine a decision curve and a clinical impact curve of the deep vein thrombosis risk assessment model based on the validation data set and the training data set; and perform performance verification on the deep vein thrombosis risk assessment model based on at least one of the statistical test result, the decision curve and the clinical impact curve.

[0074] Optionally, the transmission medium can be a wired link (such as but not limited to coaxial cable, optical fiber and digital subscriber line (DSL), etc.) or a wireless link (such as but not limited to wireless Fidelity (WIFI), Bluetooth and mobile device network, etc.). It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the method embodiments provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0075] Fig.10 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Fig.10As shown, the computer device 1000 includes: a memory 1001, a processor 1002, and a computer program 1003 stored in the memory 1001 and running on the processor 1002, wherein when the processor 1002 executes the computer program 1003, the computer device can execute any one of the methods for constructing a deep vein thrombosis risk assessment model introduced above.

[0076] In addition, an embodiment of the present invention also protects a system, which may include a memory and a processor, wherein an executable program code is stored in the memory, and the processor is used to call and execute the executable program code to execute a method for constructing a deep vein thrombosis risk assessment model provided by an embodiment of the present invention. This embodiment can divide the system into functional modules according to the above method example. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic, which is only a logical function division, and there may be other division methods in actual implementation. It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.

[0077] It should be understood that the system provided in this embodiment is used to execute the above-mentioned method for constructing a deep vein thrombosis risk assessment model, so the same effect as the above-mentioned implementation method can be achieved. In the case of an integrated unit, the system may include a processing module and a storage module. Among them, when the system is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device to execute mutual program codes, etc. Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logic boxes, modules and circuits described in combination with the contents disclosed in the present invention. The processor can also be a combination that implements a computing function, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module can be a memory.

[0078] In addition, the system provided by the embodiment of the present invention may be specifically a chip, a component or a module, and the chip may include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a method for constructing a deep vein thrombosis risk assessment model provided in the above embodiment. This embodiment also provides a computer-readable storage medium, in which a computer program code is stored, and when the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement a method for constructing a deep vein thrombosis risk assessment model provided in the above embodiment.

[0079] This embodiment also provides a computer program product. When the computer program product is run on a computer, the computer executes the above-mentioned related steps to implement a method for constructing a deep vein thrombosis risk assessment model provided in the above embodiment. Among them, the system, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be repeated here. Through the description of the above implementation mode, the technicians in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional modules as needed, that is, the internal structure of the system is divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, system or unit, which may be electrical, mechanical or other forms.

[0080] It should be noted that the sequence of the above-mentioned embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous. The various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments. The above content is only a specific implementation method of the present invention, but the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered within the protection scope of the present invention.

Claims

1. A method for constructing a deep vein thrombosis risk assessment model, characterized in that: The method for constructing the deep vein thrombosis risk assessment model includes: Obtain targeted clinical data on patients with tibial plateau fractures; Based on the target clinical data, the patients with tibial plateau fractures were divided into a group complicated with deep vein thrombosis and a group not complicated with deep vein thrombosis; Based on the target clinical data, the complicated deep vein thrombosis group and the uncomplicated deep vein thrombosis group, determining a plurality of independent predictive factors for predicting deep vein thrombosis in the target clinical data; Based on the multiple independent predictive factors, a deep vein thrombosis risk assessment model was constructed in the form of a nomogram.

2. The method for constructing a deep vein thrombosis risk assessment model according to claim 1, characterized in that: The deep vein thrombosis risk assessment model is constructed in the form of a nomogram based on the multiple independent predictors, including: Determining the influence value of each of the multiple independent predictors on the occurrence of deep vein thrombosis in the patient with tibial plateau fracture; Determine a risk assessment value of deep vein thrombosis in the patient with tibial plateau fracture based on the influence value of each independent predictor on the deep vein thrombosis and the weight corresponding to each independent predictor; The deep vein thrombosis risk assessment model is constructed based on the multiple independent predictors, the influence values ​​and the risk assessment values.

3. The method for constructing a deep vein thrombosis risk assessment model according to claim 2, characterized in that: Determining the risk assessment value of deep vein thrombosis in the patient with tibial plateau fracture based on the influence value of each independent predictor on the deep vein thrombosis and the weight corresponding to each independent predictor includes: Determining, based on the influence values ​​corresponding to the multiple independent predictors, the total influence values ​​of the multiple independent predictors on the risk of deep vein thrombosis in the tibial plateau fracture patient; Based on the total impact value and the weight, a risk assessment value for deep vein thrombosis in the patient with tibial plateau fracture is determined.

4. The method for constructing a deep vein thrombosis risk assessment model according to claim 3, characterized in that: The step of constructing the deep vein thrombosis risk assessment model based on the multiple independent predictors, the impact values ​​and the risk assessment values ​​comprises: Based on the names corresponding to each independent predictor, the total impact value and the risk assessment value, constructing the ordinate of the deep vein thrombosis risk assessment model; Constructing the abscissa of the deep vein thrombosis risk assessment model based on the impact value corresponding to each independent predictor, the total impact value and the numerical value corresponding to the risk assessment; Based on the ordinate and the abscissa, the deep vein thrombosis risk assessment model is obtained.

5. The method for constructing a deep vein thrombosis risk assessment model according to claim 3, characterized in that: Determining the risk assessment value of deep vein thrombosis in the tibial plateau fracture patient based on the total impact value and the weight includes: Multiplying the weight corresponding to each independent prediction factor by the corresponding influence value to obtain a multiplication result corresponding to each independent prediction factor; The risk assessment value is determined based on the proportion of the multiplication result corresponding to each independent prediction factor in the total impact value.

6. The method for constructing a deep vein thrombosis risk assessment model according to claim 1, characterized in that: The objective clinical data of patients with tibial plateau fractures are obtained, including: According to the pre-set inclusion criteria, the initial clinical data of multiple patients with tibial plateau fractures were obtained; Excluding the initial clinical data that meet a preset exclusion criterion from the initial clinical data of the plurality of patients with tibial plateau fractures, to obtain candidate clinical data; The candidate clinical data are processed to obtain medical quantitative data, medical qualitative data and medical image feature data of the patient with tibial plateau fracture; wherein the target clinical data includes: the medical quantitative data, the medical qualitative data and the medical image feature data.

7. The method for constructing a deep vein thrombosis risk assessment model according to claim 1, characterized in that: Based on the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group, a plurality of independent predictive factors for predicting deep vein thrombosis are determined in the target clinical data, including: Based on the target clinical data, the complicated deep vein thrombosis group and the non-complicated deep vein thrombosis group, dividing the target clinical data into a training data set and a validation data set; Performing logistic regression on the training data set to obtain the contribution of each training data in the training data set to the deep vein thrombosis; The training data set is screened according to a confidence threshold and a contribution of each training data to the deep vein thrombosis to obtain the multiple independent predictors.

8. The method for constructing a deep vein thrombosis risk assessment model according to claim 7, characterized in that: The method further comprises: Based on the validation data set and the training data set, respectively, performing ROC curve validation on the deep vein thrombosis risk assessment model to obtain a first validation result and a second validation result; Based on the first verification result and the second verification result, the clinical application value of the deep vein thrombosis risk assessment model is determined.

9. The method for constructing a deep vein thrombosis risk assessment model according to claim 8, characterized in that: The method further comprises: Based on the validation data set and the training data set, statistically test the deep vein thrombosis risk assessment model to obtain statistical test results; Determining a decision curve and a clinical impact curve of the deep vein thrombosis risk assessment model based on the validation data set and the training data set; Based on at least one of the statistical test result, the decision curve and the clinical impact curve, the performance of the deep vein thrombosis risk assessment model is verified.

10. A system for constructing a deep vein thrombosis risk assessment model, characterized in that: The system comprises: an acquisition module for acquiring target clinical data of patients with tibial plateau fractures; A grouping module, for grouping the patients with tibial plateau fractures based on the target clinical data to obtain a group complicated with deep vein thrombosis and a group not complicated with deep vein thrombosis; A determination module, configured to determine a plurality of independent predictive factors for predicting deep vein thrombosis in the target clinical data based on the target clinical data, the complicated deep vein thrombosis group and the uncomplicated deep vein thrombosis group; A building module is used to construct a deep vein thrombosis risk assessment model in the form of a nomogram based on the multiple independent predictors.

Images