Neurosyphilis detection and prediction model based on clinical diagnosis variables and construction method and application of model

Abstract

The invention relates to the field of medical diagnosis, and discloses a neurosyphilis detection and prediction model based on clinical diagnosis variables and a construction method and application of the model. The model is constructed based on the following five key clinical variables: cerebral ischemia or infarction, serum syphilis specific antibody (anti-TP), TRUST titer, ataxia and hypopsia. Through numerical analysis of the variables, LASSO regression screening variables and multi-factor logistic regression analysis are adopted, and a risk prediction model in a column diagram form is constructed and used for calculating the neurosyphilis risk probability of a patient. The model integrates serology, iconography and clinical manifestation data, and can help doctors identify high-risk patients on the non-invasive premise. The innovation not only fills the blank of the prior art, but also significantly optimizes the clinical diagnosis process, and provides powerful support for early screening and intervention of neurosyphilis.

Description

Technical Field

[0001] The present application relates to the field of medical diagnosis, and more specifically, to a neurosyphilis risk prediction model based on clinical data and mathematical modeling technology, and a method for constructing and applying the model. Background Art

[0002] Neurosyphilis (NS) is a central nervous system complication caused by Treponema pallidum (TP) infection. It may occur at any stage of syphilis infection and has a relatively complicated course. According to clinical manifestations and pathological characteristics, neurosyphilis can be divided into asymptomatic, symptomatic and delayed types. Asymptomatic patients usually have no obvious neurological symptoms and are only diagnosed by cerebrospinal fluid examination; symptomatic patients show various neurological lesions such as meningitis, cerebral vasculitis or spinal cord damage; delayed neurosyphilis usually occurs more than 10 years after infection, with more severe symptoms and often accompanied by irreversible neurological damage. These diverse manifestations make neurosyphilis a major challenge in clinical diagnosis.

[0003] At present, the diagnosis of neurosyphilis depends on cerebrospinal fluid examination, especially abnormal white blood cell count and protein level in the cerebrospinal fluid. Specific tests such as syphilis antibody (anti-TP) positivity and non-specific tests such as TRUST titer are also important reference indicators. However, these methods all require lumbar puncture to obtain samples, and lumbar puncture is an invasive procedure that not only has potential risks but also causes psychological burden to some patients. In addition, there are large differences in the cerebrospinal fluid test results of different patients, resulting in the sensitivity and specificity of the diagnosis cannot fully meet clinical needs.

[0004] The clinical manifestations of neurosyphilis are complex and varied, and are closely related to the time, site and treatment of infection. Some patients may only experience mild symptoms such as headache and nausea, while others may experience ischemic stroke, ataxia or mental disorders caused by cerebrovasculitis. These symptoms overlap greatly with a variety of central nervous system diseases (such as encephalitis, stroke, Alzheimer's disease, etc.), making neurosyphilis easy to be misdiagnosed or missed. At the same time, due to the slow progression of symptoms, many patients are not diagnosed until serious complications occur, missing the best time for early intervention.

[0005] Epidemiological studies have shown that the incidence of neurosyphilis varies significantly among different populations. HIV-positive patients have a significantly higher incidence of neurosyphilis due to impaired immune function, and their condition is often more complicated. In addition, elderly patients are more susceptible to central nervous system infections due to decreased immune system function. In the eastern coastal areas of my country, the number of neurosyphilis cases is relatively high, which may be related to the prevalence of syphilis in the region.

[0006] At present, there are still many deficiencies in the diagnosis technology of neurosyphilis. On the one hand, the traditional diagnostic method is mainly based on cerebrospinal fluid testing, and there is a lack of non-invasive diagnostic tools based on serology, imaging and clinical manifestations; on the other hand, existing studies often focus on the analysis of a single variable (such as TRUST titer or cerebrospinal fluid leukocyte level) and fail to integrate multiple clinical indicators. Although some studies have pointed out that specific variables (such as high TRUST titer) are associated with the risk of neurosyphilis, a standardized risk prediction model has not yet been formed.

[0007] In recent years, the application of machine learning technology in medical diagnosis has provided a new direction for the risk prediction of neurosyphilis. By integrating a variety of clinical data and statistical analysis methods, an efficient and accurate risk assessment tool can be constructed. For example, the LASSO regression method screens key variables in high-dimensional data to avoid overfitting; the logistic regression model can quantitatively predict the patient's probability of disease. At the same time, the visualization of the nomogram makes it easier for clinicians to quickly interpret the risk score, improving the convenience of actual operation.

[0008] Therefore, if an efficient neurosyphilis detection and prediction model can be proposed, it will be able to optimize the clinical diagnostic process of neurosyphilis and provide strong support for the early screening and intervention of neurosyphilis. Summary of the invention

[0009] In view of the shortcomings of the above-mentioned background technology, the present invention proposes a neurosyphilis risk prediction model based on retrospective data and multivariate analysis, which can efficiently assess the risk of patients and provide decision support for clinicians.

[0010] In order to achieve the above objectives, this application adopts the following technical solutions:

[0011] The present invention discloses a neurosyphilis risk prediction model based on multivariate analysis, which aims to quickly quantify the risk probability of patients suffering from neurosyphilis by constructing a set of non-invasive prediction tools. This technical solution completes model construction and evaluation by screening multiple clinical variables significantly associated with neurosyphilis, combining machine learning and statistical modeling techniques, and ultimately realizes a clinically operable risk assessment system.

[0012] In the first aspect, the present application discloses a neurosyphilis detection and prediction model based on clinical diagnostic variables, which is constructed based on the following five key clinical variables: cerebral ischemia or infarction, serum syphilis-specific antibodies (anti-TP), TRUST titer, ataxia and visual impairment; through numerical analysis of the above variables, LASSO regression screening variables and multivariate logistic regression analysis are used to construct a risk prediction model in the form of a nomogram, which is used to calculate the patient's neurosyphilis risk probability.

[0013] It should be noted that the core variables of the model were screened based on clinical practice and statistical analysis results. The risk factors included cerebral ischemic infarction, serum syphilis-specific antibodies, serum TRUST titer, ataxia, and visual impairment.

[0014] The "cerebral ischemia or infarction" mentioned in the present invention is the same as "cerebral ischemia or infarction", which is a sign of central nervous system damage determined by cranial MRI examination. This variable is a common imaging abnormality in patients with neurosyphilis and is closely related to central nervous system infection and cerebral vasculitis. Treponema pallidum infection can cause meningeal vasculitis, leading to intimal fibrosis, lumen stenosis, and reduced blood flow, thereby causing cerebral insufficiency and ischemic lesions. This variable has a high detection rate in patients with neurosyphilis and is an important indicator for evaluating the involvement of the patient's central nervous system.

[0015] The serum syphilis-specific antibody described in the present invention, the same as "Serum anti-TP (S / CO)", is a specific serological marker for syphilis infection and can indicate the status of syphilis infection. Although its lifelong positivity may interfere with the dynamic evaluation of some patients, its numerical changes are still an important reference for judging the involvement of the central nervous system. As a highly specific indicator, anti-TP provides basic support for the risk assessment of neurosyphilis.

[0016] The "serum TRUST titer" described in the present invention is the same as the "Serum TRUST titer", i.e. the toluidine red unheated serum test, which is a non-specific antibody test for detecting the activity of syphilis. The result is expressed in the form of titer, and the titer value reflects the activity of the disease and the degree of infection. In the risk assessment of neurosyphilis, the TRUST titer is an important parameter for measuring the disease activity of patients, which can significantly improve the predictive ability of the model.

[0017] The "ataxia" described in the present invention is the same as "Ataxia", which is an abnormal manifestation of the nervous system, including unsteady walking, dysarthria, limb weakness, dizziness, etc. In neurosyphilis, damage to the cerebellum or spinal cord often causes ataxia, which manifests as uncoordinated movements or impaired balance. This symptom may gradually worsen, and in severe cases affect the patient's daily activities. The manifestations of ataxia are diverse, and some patients may only show mild unsteady gait, while others are accompanied by significant dizziness and limb weakness. The presence of ataxia usually indicates deep involvement of the nervous system and is an important clinical feature for evaluating neurosyphilis.

[0018] The "visual impairment" mentioned in the present invention is the same as "Decreased vision", which refers to blurred vision, narrowed visual field or even blindness of patients, which is usually related to optic nerve damage. In patients with neurosyphilis, visual impairment is a common neurological manifestation, usually caused by optic neuritis or cerebrovascular disease. Patients may describe blurred vision, obstruction of shadows or visual field defects, which may lead to unilateral or bilateral vision loss in severe cases. Visual impairment not only affects the quality of life of patients, but may also indicate the progression of neurosyphilis. Early identification and intervention are crucial to improving patient prognosis.

[0019] Furthermore, the present invention presents the prediction model in the form of a nomogram, and the model includes five key risk factors: cerebral ischemic infarction, serum syphilis-specific antibodies, TRUST titer, ataxia and visual impairment.

[0020] Furthermore, the prediction model generates a total risk score for the patient by accumulating the individual scores of each variable, and predicts the risk probability of the patient suffering from neurosyphilis based on the total score.

[0021] Preferably, when the total neurosyphilis risk prediction score is less than 42, the patient is assessed as low risk;

[0022] Preferably, when the total score is greater than 46, the patient is assessed as high risk;

[0023] Preferably, for patients with a total score between 42 and 46, the risk level cannot be clearly determined and needs to be comprehensively evaluated in combination with other clinical factors or further examinations.

[0024] In a second aspect, the present application provides a method for constructing the above model, comprising the following steps:

[0025] Step 1: Collect patients infected with syphilis, screen patients with serum anti-TP and TRUST double positive and include them in the study, and collect clinical diagnostic variable data;

[0026] Step 2: The patients were randomly divided into a training set and a test set according to the proportion. The training set was used for model development, and the validation set was used to evaluate the model's discrimination ability, prediction consistency, and clinical practicality.

[0027] Step 3: Perform univariate analysis and LASSO regression screening on the clinical diagnosis variable data in the training set to identify key variables significantly associated with neurosyphilis;

[0028] Step 4: Use a multivariate logistic regression model to model the key variables screened, calculate the regression coefficient of each variable, and generate a risk prediction model;

[0029] Step 5: Model validation.

[0030] Preferably, the construction method has at least one of the following characteristics:

[0031] In step 1, clinical diagnostic variable data included age, sex, TRUST titer, anti-TP, cerebral ischemic infarction (MRI), ataxia, visual impairment, blood pressure, blood sugar, hearing loss, syncope, headache, and history of stroke;

[0032] In step 2, patients were randomly divided into training and test sets in a ratio of 8:2;

[0033] In step 3, the key variables screened included cerebral ischemic infarction, anti-TP, TRUST titer, ataxia, and visual impairment;

[0034] The process of step 5 is as follows: first, the receiver operating characteristic (ROC) curve is used to evaluate the discrimination ability of the model, and the area under the curve (AUC) of the training set and the validation set is calculated respectively; secondly, the consistency between the predicted value of the model and the actual observed value is evaluated by calibration curve analysis; and the net benefit of the model at different risk thresholds is evaluated by clinical decision curve (DCA) analysis.

[0035] The performance of the model was evaluated by ROC curve, with an AUC of 0.816 for the training set and 0.843 for the validation set.

[0036] In a third aspect, the present application provides a computer device for constructing the above-mentioned model, wherein the computer device comprises a memory and a processor, wherein the memory stores a program, and the processor implements the above-mentioned construction method when executing the program.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program;

[0038] Wherein, when the computer program is running, the computer-readable storage medium is controlled to implement the above-mentioned construction method.

[0039] In a fifth aspect, the present application provides the application of the above-mentioned model and the above-mentioned construction method in the preparation of neurosyphilis detection and prediction products.

[0040] In a sixth aspect, the present application provides a system for detecting and predicting neurosyphilis, comprising:

[0041] A data collection module is used to collect clinical diagnostic variable data of patients;

[0042] The data processing module is used to randomly divide the patients into training sets and test sets in proportion, perform univariate analysis and LASSO regression screening on the clinical diagnostic variable data of the training set, determine the key variables significantly associated with neurosyphilis, use the multivariate logistic regression model to model the screened key variables, calculate the regression coefficient of each variable, generate a risk prediction model, and verify it in the test set;

[0043] The risk prediction module is used to perform risk stratification based on the patient's total score and output the results;

[0044] Among them, patients with a total score less than 42 are at low risk; patients with a total score greater than 46 are at high risk; and patients with a total score between 42 and 46 are at medium risk.

[0045] In a seventh aspect, the present application provides a clinical application method of the above-mentioned model and the above-mentioned construction method, comprising the following steps:

[0046] a) Collect clinical data of patients, including cerebral ischemia or infarction, serum anti-TP, TRUST titer, ataxia and visual impairment;

[0047] b) using a nomogram to calculate a total risk score based on patient variable values;

[0048] c) To predict the probability of neurosyphilis in patients based on the total score and to perform risk stratification;

[0049] d) Provide further CSF examination and treatment recommendations for patients at intermediate and high risk.

[0050] The present invention uses advanced machine learning algorithms and statistical methods to construct a neurosyphilis risk prediction model, which specifically includes the following steps:

[0051] a) Variable screening. Univariate logistic regression analysis was used to screen out variables significantly associated with neurosyphilis. To avoid model overfitting and redundancy, LASSO regression was further used to screen key variables, and five core variables were finally determined: cerebral ischemic infarction, anti-TP, TRUST titer, ataxia, and visual impairment.

[0052] b) Multivariate logistic regression modeling. The five key variables screened were introduced into the multivariate logistic regression model. By calculating the variable regression coefficients, the contribution of each variable to the risk of neurosyphilis was quantified to form a complete mathematical model. Based on the logistic regression model, an intuitive nomogram tool was generated.

[0053] c) Nomogram construction: The nomogram quickly calculates the patient's total risk score by summing up the risk weight scores of each variable and predicts the probability of neurosyphilis.

[0054] d) Model performance verification. The model’s ability to distinguish was evaluated by ROC curve analysis, and the AUC value reflected the accuracy of the model’s prediction. The calibration curve was used to verify the consistency between the model’s predicted value and the actual observed value, and the clinical decision curve (DCA) was used to evaluate the model’s net benefit at different thresholds, thereby ensuring the model’s practicality and reliability.

[0055] e) Clinical application. The model of the present invention can achieve risk assessment of neurosyphilis without invasive operation. Low-risk patients can avoid unnecessary cerebrospinal fluid examination, and high-risk patients can obtain further diagnosis and timely intervention.

[0056] Due to the adoption of the above technical solution, the beneficial effects of this application are:

[0057] The present invention integrates multi-dimensional information from imaging, serology, and neurological symptoms, uses machine learning algorithms to optimize variable selection, and constructs a non-invasive, accurate, and efficient neurosyphilis risk prediction model. The model is visualized through a nomogram, providing clinicians with an easy-to-use and highly reliable diagnostic support tool. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flowchart used to recruit patients with neurosyphilis and non-neurosyphilis in the study, showing in detail the specific steps for patient enrollment and the inclusion and exclusion criteria. This flowchart clearly illustrates the case selection and data processing process of the study, and is the basis for model development and analysis.

[0059] Figure 2 This is the process of LASSO feature selection, showing the detailed steps of screening variables through LASSO regression. Figure 2 A: LASSO regression coefficient path diagram. Each curve represents the regression coefficient trajectory of a variable. As the lambda value increases, the regression coefficients of some variables gradually decrease to zero, and finally the five key variables that contribute most to the model are screened out: cerebral ischemia or infarction, ataxia, visual impairment, serum anti-TP, and TRUST titer. Figure 2 B: Binomial deviation graph, with binomial deviation on the vertical axis and log(λ) on the horizontal axis. The optimal lambda value (lambda.1se) is selected through cross-validation, which corresponds to the minimum model complexity and the best model performance.

[0060] Figure 3 This is an example of a nomogram for the neurosyphilis risk prediction model, showing the weight of each variable and its contribution to the total risk score. The nomogram assigns different weights to each variable, and can calculate the total risk score and predict the probability of disease based on the variable values ​​of a specific patient.

[0061] Figure 4It is the ROC curve, which is used to evaluate the discrimination ability of the prediction model and shows the AUC values ​​of the training set and the validation set. Figure 4 A: ROC curve of the training set. The AUC of the model in the training set was 0.816, indicating that the model had good ability to distinguish between patients with neurosyphilis and non-neurosyphilis. Figure 4 B: ROC curve of the validation set. The AUC of the model in the validation set is 0.843, which further verifies the model's ability to distinguish and proves the stability and reliability of the model.

[0062] Figure 5 is the calibration curve, which evaluates the consistency between the model predictions and the actual observed values. Figure 5 A: Calibration curve of the training set. The model prediction value and the actual observation value are close to the diagonal line, indicating that the prediction accuracy of the model in the training set is high and the calibration performance is good. Figure 5 B: Calibration curve of the validation set. The calibration curve of the model in the validation set is also close to the ideal line, indicating that the model's prediction value in the independent data set has good calibration, further proving the reliability and practicality of the model.

[0063] Figure 6 is the clinical decision curve (DCA), showing the net benefit of the model at different decision thresholds. Figure 6 A: Decision curve of the training set. It shows that the model can bring higher net benefits in most risk threshold ranges, which is better than the strategy of "all patients receive intervention" or "all patients do not receive intervention". Figure 6 B: Decision curve of the validation set. Consistent with the results of the training set, the model also showed a high net benefit in the validation set, indicating that the model has good clinical practicality and can provide effective decision support in actual diagnosis and treatment. DETAILED DESCRIPTION

[0064] The present invention is described in detail below in conjunction with the accompanying drawings and examples. It should be noted that the following description is only a preferred embodiment of the present invention, which is used to better understand and implement the technical solution of the present invention. For those of ordinary skill in the art, various improvements and supplements may be made to it without departing from the core method of the present invention, and these improvements and supplements should also be regarded as the protection scope of the present invention.

[0065] Example: Construction of a clinical diagnostic model for neurosyphilis based on a retrospective study.

[0066] 1. Data source and queue information.

[0067] This study was based on patients with syphilis who were admitted to the Shanghai Public Health Clinical Center from January 2021 to December 2023. A total of 702 patients with serum anti-TP and TRUST double positive were screened out from 1078 suspected syphilis cases and included in the study (see Appendix for details). Figure 1 ). Inclusion criteria included: age ≥ 18 years, positive serum Treponema pallidum specific antibody (anti-TP), and complete clinical and imaging examination data, including TRUST titer, cranial MRI, and neurological examination results. Exclusion criteria were patients with other neurological diseases (such as encephalitis, brain tumors) or missing key test data. All patients' data were desensitized before enrollment and approved by the ethics committee.

[0068] The baseline characteristics of the patient cohort were as follows: 586 males (83.5%) and 116 females (16.5%); the median age was 50 years (IQR: 34-62). In terms of age distribution, the largest number of patients were over 60 years old (199 patients, accounting for 28.3%), 147 patients were 51-60 years old (20.9%), 101 patients were 41-50 years old (14.4%), 146 patients were 31-40 years old (20.8%), and 109 patients were under 30 years old (15.5%). According to serum TRUST titer, 397 patients (56.6%) had low titer (≤1:16), 167 patients (23.8%) had medium titer (1:32 and 1:64), and 136 patients (19.4%) had high titer (≥1:128). Among the neurological manifestations, 531 patients (75.6%) were normal, and 171 patients (24.4%) had neurological abnormalities. Specifically, 67 patients (9.5%) had ataxia, including speech disorders, dysarthria, unsteady walking, limb weakness, disorientation, and dizziness and headache; 48 patients (6.8%) showed mental and behavioral abnormalities, including delirium, depression, dementia, and apathy; 31 patients (4.4%) had visual impairment; 22 patients (3.1%) had memory impairment; and 3 patients (0.4%) had hearing loss. Imaging examinations showed that the most common abnormality in cranial MRI was cerebral ischemia and infarction (120 cases, accounting for 17.1%), and other abnormalities included white matter degeneration (13 cases, accounting for 1.9%), brain atrophy (11 cases, accounting for 1.6%), and demyelinating lesions (2 cases, accounting for 0.3%) (see Table 1 for details).

[0069] Table 1 Baseline characteristics of the 702 syphilis-infected patients included in the study

[0070]

[0071]

[0072] According to the results of cerebrospinal fluid tests, the patients were divided into neurosyphilis group and non-neurosyphilis group. The neurosyphilis group (NS) included 246 patients who showed typical neurosyphilis features in cerebrospinal fluid tests, including elevated white blood cell counts, abnormal protein levels, and positive cerebrospinal fluid TRUST. The non-neurosyphilis group (NNS) included 456 patients whose cerebrospinal fluid test results were normal. Comparison of the clinical and laboratory characteristics of the two groups of patients showed that there was no statistical difference in age, gender, and whether there was concurrent HIV infection between the non-neurosyphilis group and the neurosyphilis group. However, there were significant differences between the two groups in the following indicators: cerebral ischemic infarction (P<0.0001), memory loss (P=0.0221), visual impairment (P<0.0001), ataxia (P<0.0001), mental and behavioral abnormalities (P=0.0071), serum specific antibodies (anti-TP, P<0.0001), serum TRUST titer (P<0.0001), complete blood red blood cell count (P=0.0002), platelet count (P=0.0222) and prothrombin time (PT, P=0.0474) (see Table 2 for details).

[0073] Table 2 Baseline characteristics and laboratory parameters of the non-neurosyphilis group and the neurosyphilis group

[0074]

[0075]

[0076]

[0077] 2. Variable screening.

[0078] After data collection, 702 patients were randomly divided into training set and validation set in a ratio of 8:2. Among them, the training set contained 577 patients for model development; the validation set contained 125 patients for evaluating the performance of the model. Before modeling, 13 candidate variables were listed from the patient data, including age, gender, TRUST titer, anti-TP, cerebral ischemic infarction (MRI), ataxia, visual impairment, blood pressure, blood sugar, hearing loss, syncope, headache and history of stroke. Through univariate analysis and LASSO regression screening of candidate variables, 5 key variables significantly associated with neurosyphilis were finally identified, including cerebral ischemic infarction, anti-TP, TRUST titer, ataxia and visual impairment (see Table 3 for details). The screening process and results of these variables are shown in the attached. Figure 2 shown.

[0079] Table 3. Logistic regression analysis of clinical diagnosis model

[0080]

[0081]

[0082] 3. Model construction.

[0083] After completing the variable screening, the multivariate logistic regression model was used to model the screened variables, calculate the regression coefficient of each variable, and generate a risk prediction formula. The core of the model is to quantify the contribution weight of each variable to the risk of neurosyphilis, accumulate the scores of each variable, generate the patient's total score, and further calculate the probability of suffering from neurosyphilis based on the total score. For the convenience of clinicians, the model is visualized in the form of a nomogram, which clearly shows the weight of each variable and its contribution to the total risk score. A specific example of the nomogram is shown in the attached figure. Figure 3 Clinicians can determine the corresponding variable scores in the nomogram one by one according to the patient's specific data, add up the scores of each variable to get the total score, and predict the patient's risk level based on the total score.

[0084] 4. Model validation.

[0085] The performance of the model was comprehensively evaluated using multiple indicators. First, the receiver operating characteristic (ROC) curve was used to evaluate the discriminative ability of the model, and the area under the curve (AUC) of the training set and the validation set was calculated. The AUC of the training set was 0.816, and the AUC of the validation set was 0.843 (see Table 4 for details), indicating that the model has a high discriminative ability. The ROC curve is shown in the attached figure. Figure 4 Secondly, the consistency between the model's predicted value and the actual observed value was evaluated by calibration curve analysis. The results showed that the calibration curves of the training set and the validation set were close to the ideal line, indicating that the model has good predictive stability. The calibration curves are shown in the attached figure. Figure 5 In addition, the net benefit of the model at different risk thresholds was evaluated through clinical decision curve (DCA) analysis. The results showed that within the medium and high risk thresholds (0.2 to 0.6), the net benefit of the model was significantly higher than the "full intervention" or "no intervention" strategy, which verified the clinical practicality of the model. Figure 6 .

[0086] Table 4 ROC curve analysis results of the model in the training set and validation set

[0087]

[0088] 5. Specific application scenarios.

[0089] In practical applications, the present invention proposes a set of stratification schemes based on risk scores to guide clinical decision-making. According to the model scores, patients are divided into three risk levels. Patients with a total score of less than 42 are assessed as low risk, and it is recommended to continue follow-up observation without immediate cerebrospinal fluid examination; patients with a total score greater than 46 are assessed as high risk, and it is recommended to conduct cerebrospinal fluid examination in time for diagnosis and take intervention measures as soon as possible; patients with a total score between 42 and 46 are intermediate risk, and their risk level cannot be clearly judged, and a comprehensive assessment is required in combination with other clinical factors or supplementary examinations (see Table 5 for details). This stratification scheme can effectively optimize the allocation of clinical resources, reduce unnecessary invasive examinations, and ensure that high-risk patients receive timely diagnosis and treatment.

[0090] Table 5 Neurosyphilis risk score stratification and its clinical significance

[0091] Risk score range Risk Stratification Clinical Recommendations Total score < 42 Low risk Follow-up observation, no cerebrospinal fluid examination required Total score 42-46 Intermediate Risk Combined with other clinical manifestations or additional examinations, further evaluation Total score > 46 High risk It is recommended to immediately perform cerebrospinal fluid examination to confirm the diagnosis and timely intervention treatment

[0092] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed, but such modifications are protected by the patent law as long as they are within the scope of the claims of the present application.

Claims

1. A neurosyphilis detection and prediction model based on clinical diagnostic variables, characterized in that: The model was constructed based on the following five key clinical variables: cerebral ischemia or infarction, serum syphilis-specific antibodies (anti-TP), TRUST titer, ataxia and visual impairment. Through numerical analysis of the above variables, LASSO regression was used to screen variables and multivariate logistic regression analysis was used to construct a risk prediction model in the form of a nomogram to calculate the patient's risk probability of neurosyphilis.

2. The neurosyphilis detection and prediction model according to claim 1, characterized in that: The nomogram assigns a corresponding weight to each variable based on the variable regression coefficient, generates a total score by accumulating the scores of each variable, and predicts the probability of neurosyphilis in the patient based on the total score.

3. The neurosyphilis detection and prediction model according to claim 2, characterized in that: The model stratifies patients into risk groups based on their total score, including: Patients with a total score of less than 42 were considered low risk; Patients with a total score greater than 46 were considered high risk; Patients with a total score between 42 and 46 are considered intermediate risk.

4. The method for constructing a model according to any one of claims 1 to 3, characterized in that: The following steps are involved: Step 1: Collect patients infected with syphilis, screen patients with serum anti-TP and TRUST double positive and include them in the study, and collect clinical diagnostic variable data; Step 2: The patients were randomly divided into a training set and a test set according to the proportion. The training set was used for model development, and the validation set was used to evaluate the model's discrimination ability, prediction consistency, and clinical practicality. Step 3: Perform univariate analysis and LASSO regression screening on the clinical diagnosis variable data in the training set to identify key variables significantly associated with neurosyphilis; Step 4: Use a multivariate logistic regression model to model the key variables screened, calculate the regression coefficient of each variable, and generate a risk prediction model; Step 5: Model validation.

5. The construction method according to claim 4, characterized in that: Have at least one of the following characteristics: In step 1, clinical diagnostic variable data included age, sex, TRUST titer, anti-TP, cerebral ischemic infarction (MRI), ataxia, visual impairment, blood pressure, blood sugar, hearing loss, syncope, headache, and history of stroke; In step 2, patients were randomly divided into training and test sets in a ratio of 8:2; In step 3, the key variables screened included cerebral ischemic infarction, anti-TP, TRUST titer, ataxia, and visual impairment; The process of step 5 is as follows: first, the receiver operating characteristic (ROC) curve is used to evaluate the discrimination ability of the model, and the area under the curve (AUC) of the training set and the validation set is calculated respectively; secondly, the consistency between the predicted value of the model and the actual observed value is evaluated by calibration curve analysis; and the net benefit of the model at different risk thresholds is evaluated by clinical decision curve (DCA) analysis. The performance of the model was evaluated by ROC curve, with an AUC of 0.816 for the training set and 0.843 for the validation set.

6. A computer device for constructing a model according to any one of claims 1 to 3, characterized in that: The computer device includes a memory and a processor, the memory stores a program, and the processor implements the construction method described in claim 4 or 5 when executing the program.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program; Wherein, when the computer program is running, the computer-readable storage medium is controlled to implement the construction method described in claim 4 or 5.

8. Use of the model according to any one of claims 1 to 3 and the construction method according to claim 4 or 5 in the preparation of neurosyphilis detection and prediction products.

9. A system for detecting and predicting neurosyphilis, characterized in that: include: A data collection module is used to collect clinical diagnostic variable data of patients; The data processing module is used to randomly divide the patients into training sets and test sets in proportion, perform univariate analysis and LASSO regression screening on the clinical diagnostic variable data of the training set, determine the key variables significantly associated with neurosyphilis, use the multivariate logistic regression model to model the screened key variables, calculate the regression coefficient of each variable, generate a risk prediction model, and verify it in the test set; The risk prediction module is used to perform risk stratification based on the patient's total score and output the results; Among them, patients with a total score less than 42 are at low risk; patients with a total score greater than 46 are at high risk; Patients with a total score between 42 and 46 are considered intermediate risk.

10. A clinical application method of the model according to any one of claims 1 to 3 and the construction method according to claim 4 or 5, characterized in that: The following steps are involved: a) Collect clinical data of patients, including cerebral ischemia or infarction, serum anti-TP, TRUST titer, ataxia and visual impairment; b) using a nomogram to calculate a total risk score based on patient variable values; c) To predict the probability of neurosyphilis in patients based on the total score and to perform risk stratification; d) Provide further CSF examination and treatment recommendations for patients at intermediate and high risk.

Images