CN115240859A - System and method for predicting late schistosomiasis - Google Patents

Abstract

The invention relates to a system and a method for predicting late schistosomiasis. The system for predicting late schistosomiasis comprises: a data collector and a processor. After the data acquisition unit acquires the biomarkers of the blood of the patient to be predicted, the biomarkers are implanted into the SPSS statistical analysis software package in the processor, and the acquired biomarkers of the blood of the patient to be predicted are input into the judgment and diagnosis model, so that the prediction result of whether the patient belongs to the late schistosomiasis can be obtained, and the purposes of early discovery and early intervention and effective reduction of new cases of the late schistosomiasis can be achieved.

Description

System and method for predicting late schistosomiasis

Technical Field

The invention relates to the technical field of medical systems, in particular to a system and a method for predicting late schistosomiasis.

Background

Schistosomiasis in the late stage is mainly characterized by hepatosplenopathy, such as symptoms of hepatic fibrosis around portal vein, portal hypertension, splenomegaly and congestion, severe growth and development disorder or marked granulomatous proliferation of colon. Fibrosis of egg deposition tissue is the most serious outcome of schistosome infection. Due to the factors of long course of disease, high treatment cost, poor healing and the like, the patients and families thereof are burdened with huge psychology and economy. Schistosomiasis belongs to immune diseases in terms of disease occurrence and pathogenic mechanism, and the pathological basis is the immune response of a host to schistosome and egg. Eggs laid by adults parasitizing in a portal vein system enter the liver along with blood flow, because the eggs with larger diameters are retained in front of hepatic sinuses to block blood vessels, egg antigens are released to sensitize T lymphocytes with delayed allergy, the sensitized T lymphocytes with delayed allergy generate a series of lymphokines, and various lymphokines attract inflammatory cells (including macrophages, lymphocytes, eosinophilic granulocytes and the like) to gather around the eggs to cause a series of inflammatory reactions to form granuloma of the eggs. Macrophages in granulomas in eggs release profibrotic cytokines that activate Hepatic Stellate Cells (HSCs) to transform into Myofibroblasts (Myofibroblasts), which produce large amounts of extracellular matrix (ECM) and secrete profibrotic factors, resulting in an imbalance between fibrogenesis and degradation. If no effective intervention is performed, fibrogenesis develops further, eventually leading to late stage schistosomiasis. The late schistosomiasis is difficult to completely recover, so with the increasing proportion of late schistosomiasis cases in the whole schistosomiasis disease spectrum, an effective prevention strategy of early detection, early diagnosis and early treatment is necessary to be provided, which has important significance for effectively controlling schistosomiasis.

At present, the diagnosis method of schistosomiasis liver fibrosis mainly comprises the following steps: (1) histological examination is a reliable method for diagnosing hepatic fibrosis, but has great trauma to tissues, and the whole condition of the liver of a patient cannot be reflected by the material-taking part; (2) the imaging method, the ultrasonic examination is the most widely used method at present, and is applied to diagnosis of schistosomiasis liver disease, liver fibrosis evaluation grading, curative effect observation and the like; (3) body fluid (serum or plasma) markers, including extracellular matrix components, degradation products, enzymes and cytokines involved in their metabolism, etc., are easily measured and non-invasive, but specificity remains to be assessed. The liver is the main place for synthesizing blood coagulation factors, anticoagulant substances (such as AT-III, PC and PS) and fibrinolytic substances (such as PLG), and liver function and liver parenchyma damage caused by schistosomiasis, chronic hepatitis B and liver cirrhosis can cause abnormal synthesis of the coagulation factors, the anticoagulant substances and the fibrinolytic substances.

However, the existing method has the problems of low judgment accuracy, low efficiency and the like in the judgment process of development of the schistosomiasis japonica at a late stage.

Disclosure of Invention

Based on the problems in the prior art, the invention provides a system and a method for predicting the late schistosomiasis so as to achieve the purposes of early discovery and early intervention and effectively reducing new cases of the late schistosomiasis.

In order to achieve the purpose, the invention provides the following scheme:

a system for predicting late stage schistosomiasis, comprising:

the data acquisition unit is used for acquiring the biomarkers of the blood of the patient to be predicted;

the processor is connected with the data acquisition unit and is implanted with an SPSS statistical analysis software package; the SPSS statistical analysis software package is used for executing the following operations:

and inputting the biomarker of the blood of the patient to be predicted into a discrimination diagnosis model to obtain a prediction result.

Preferably, the discriminant diagnosis model is:

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

wherein CAS is a late schistosomiasis prediction value, CCS is a chronic schistosomiasis prediction value, X1 is a value of a biomarker hemoglobin, X2 is a value of a biomarker monocyte, X3 is a value of a biomarker biosphere, X4 is a value of a biomarker gamma glutamyltransferase, X5 is a value of a biomarker activated partial thromboplastin time, X6 is a value of a biomarker VIII factor activity, and X7 is a value of a biomarker fibrinogen.

Preferably, the processor further comprises:

the curve generation module is connected with the data acquisition unit and is used for obtaining an ROC curve by taking the advanced schistosomiasis patient and the chronic schistosomiasis patient as state variables and taking the detection value of the biomarker of the blood of the advanced schistosomiasis patient and the detection value of the biomarker of the blood of the chronic schistosomiasis patient as test variables;

the index determining module is connected with the curve generating module and is used for acquiring the sensitivity and the specificity corresponding to the ROC curve coordinate and determining a Youden index based on the sensitivity and the specificity;

the critical value determining module is connected with the index determining module and is used for acquiring the numerical value of the Youden index and taking the acquired numerical value as a diagnosis critical value;

the marker screening module is connected with the data collector and is used for screening the biomarker of the blood of the patient with the advanced schistosomiasis and the biomarker of the blood of the patient with the chronic schistosomiasis to obtain a screened biomarker;

and the model construction module is respectively connected with the marker screening module and the critical value determination module and is used for constructing a discriminant diagnosis model based on the screened biomarkers and the diagnosis critical value.

Preferably, the marker screening module comprises:

and the marker screening unit is used for screening the biomarkers of the blood of the late stage schistosomiasis patient and the biomarkers of the blood of the chronic schistosomiasis patient based on single factors and multiple factors to obtain screened biomarkers.

Preferably, the processor further comprises:

and the storage module is used for storing the SPSS statistical analysis software package.

Preferably, the storage module is a computer-readable storage medium.

Preferably, the processor is a computer.

According to the specific embodiment provided by the invention, the invention discloses the following technical effects:

the invention provides a system for predicting late schistosomiasis, which comprises the following steps: a data collector and a processor. After the data acquisition unit acquires the biomarkers of the blood of the patient to be predicted, the SPSS statistical analysis software package is implanted into the processor, and the acquired biomarkers of the blood of the patient to be predicted are input into the judgment and diagnosis model, so that the prediction result of whether the patient belongs to the late schistosomiasis can be obtained, and the purposes of early discovery and early intervention and effective reduction of new cases of the late schistosomiasis can be achieved.

Corresponding to the specific functions implemented by the system for predicting late schistosomiasis provided above, the present invention also provides a method for predicting late schistosomiasis, comprising:

obtaining a biomarker of blood of a patient to be predicted;

acquiring a discrimination diagnosis model;

and inputting the biomarker of the blood of the patient to be predicted into the discriminant diagnosis model to obtain a prediction result.

Preferably, the discriminant diagnosis model is:

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

wherein CAS is a late schistosomiasis prediction value, CCS is a chronic schistosomiasis prediction value, X1 is a value of a biomarker hemoglobin, X2 is a value of a biomarker monocyte, X3 is a value of a biomarker biosphere, X4 is a value of a biomarker gamma glutamyl transferase, X5 is a value of a biomarker activated partial thromboplastin time, X6 is a value of a biomarker VIII factor activity, and X7 is a value of a biomarker fibrinogen.

Preferably, the construction process of the discriminant diagnosis model includes:

taking a late stage schistosomiasis patient and a chronic schistosomiasis patient as state variables, and taking a detection value of a biomarker of blood of the late stage schistosomiasis patient and a detection value of a biomarker of blood of the chronic schistosomiasis patient as test variables to obtain an ROC curve;

acquiring sensitivity and specificity corresponding to the ROC curve coordinate, and determining a Youden index based on the sensitivity and the specificity;

obtaining a numerical value of said Youden index and using said obtained numerical value as a diagnostic threshold;

screening the biomarkers of the blood of the patients with the advanced schistosomiasis and the biomarkers of the blood of the patients with the chronic schistosomiasis based on single factors and multiple factors to obtain screened biomarkers;

and constructing a discriminant diagnosis model based on the screening biomarkers.

The technical effect of the method for predicting the late schistosomiasis is the same as that of the system for predicting the late schistosomiasis, so the method is not repeated herein.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

FIG. 1 is a schematic diagram of the structure of the system for predicting late schistosomiasis provided by the present invention;

FIG. 2 is a flow chart of the method for predicting late schistosomiasis provided by the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The invention aims to provide a system and a method for predicting late schistosomiasis, which can achieve the purposes of early discovery and early intervention and effectively reducing new cases of late schistosomiasis.

In order to make the aforementioned objects, features and advantages of the present invention more comprehensible, the present invention is described in detail with reference to the accompanying drawings and the detailed description thereof.

As shown in fig. 1, the system for predicting late schistosomiasis provided by the present invention comprises: a data collector and a processor. Wherein, the processor is connected with the data acquisition unit. The data collector is used for obtaining the biomarkers of the blood of the patient to be predicted. The processor is embedded with an SPSS statistical analysis software package. The SPSS statistical analysis software package is used for inputting the biomarkers of the blood of the patient to be predicted into the discriminant diagnosis model to obtain a prediction result. The processor employed in the present invention may be a computer or other device having software execution and processing capabilities.

The discriminant diagnosis model adopted by the invention is as follows:

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

in the formula, CAS is a late schistosomiasis prediction value, CCS is a chronic schistosomiasis prediction value, X1 is a value of biomarker Hemoglobin (HGB), X2 is a value of biomarker Monocyte (MON), X3 is a value of biomarker Globulin (GLB), X4 is a value of biomarker Gamma Glutamyltransferase (GGT), X5 is a value of biomarker Activated Partial Thromboplastin Time (APTT), X6 is a value of biomarker VIII factor activity (VIII), and X7 is a value of biomarker fibrinogen (Fbg).

In the invention, the data collector can be a detection instrument used for obtaining blood biomarkers in a hospital, and is connected to the processor in a wireless mode, and the processor crawls data generated in the detection instrument in a data crawling mode. Of course, the data collector may also be a computer, and the obtained blood biomarkers may be input manually.

Further, in order to improve the accuracy of the prediction result, the processor adopted by the invention may further include: the system comprises a curve generation module, an index determination module, a critical value determination module, a marker screening module and a model construction module. The curve generation module is connected with the data acquisition unit. The index determining module is connected with the curve generating module. The critical value determining module is connected with the index determining module. The marker screening module is connected with the data acquisition unit. The model building module is respectively connected with the marker screening module and the critical value determining module.

After the data acquisition unit obtains the biomarkers of blood of the patients with advanced schistosomiasis and the biomarkers of blood of the patients with chronic schistosomiasis, the curve generation module obtains the ROC curve by using the patients with advanced schistosomiasis and the patients with chronic schistosomiasis as state variables and using the detection values of the biomarkers of blood of the patients with advanced schistosomiasis and the detection values of the biomarkers of blood of the patients with chronic schistosomiasis as test variables. And the index determination module acquires the sensitivity and specificity corresponding to the ROC curve coordinate and calculates the maximum Youden index through [ (sensitivity + specificity) -1 ]. The critical value determining module obtains the numerical value of the Youden index, and takes the obtained numerical value as a diagnosis critical value, wherein the diagnosis critical value is the optimal diagnosis critical value. The marker screening module screens the biomarkers of blood of patients with advanced schistosomiasis and the biomarkers of blood of patients with chronic schistosomiasis to obtain screened biomarkers. The model construction module constructs a discriminant diagnosis model based on the screening biomarkers and the diagnosis critical value.

For example, the marker screening module may be provided with a marker screening unit for screening the biomarkers of blood of late stage schistosomiasis patient and the biomarkers of blood of chronic schistosomiasis patient based on single factor and multifactorial to obtain screening biomarkers. Wherein, the single factor: after the included quantitative values of the 22 blood biomarkers are converted into qualitative 0 s and 1 s by taking the optimal diagnosis critical value as a standard, chi-square verification is carried out together with the 6 biomarker data such as SJ, five hepatitis B and the like, and the result shows that: except that 7 biomarkers such AS SJ, RBC, HBsAg, HBsAb, HBeAg, HBeAb and HBcAb have no statistical significance, other biomarkers such AS AS group and CS group have significant difference, but because the P value of RBC and HBcAb is less than 0.2, the invention still incorporates the biomarkers into multi-factor analysis according to the initial statistical constraint.

The method comprises the following steps: and carrying out multifactorial analysis on the 23 blood biomarkers with P <0.2 in the monokine analysis, and finally screening to obtain 9 blood biomarkers such as HGB, LYM, MON, DBiL, GLB, GGT, APTT, fbg, VIII and the like.

Further, the curve generation module, the index determination module, the critical value determination module, the marker screening module and the model construction module provided above may be software functional units.

Further, in order to increase the software running speed, the processor further comprises: and the storage module is used for storing the SPSS statistical analysis software package. In the present invention, the storage module may be a software functional unit or a computer readable storage medium. When a computer-readable storage medium, it may be sold or used as a stand-alone product.

The following provides a specific validation example to illustrate the technical effects that the system for predicting late schistosomiasis provided by the invention can achieve.

Example one

In this example, the proportion of male and female patients with Advanced Schistosomiasis (AS) and Chronic Schistosomiasis (CS) involved in the study was 1. The mean ages of the As and CS patients in men were 64.1 and 61.8 years old, and those in women were 57.8 and 59.9 years old, and the mean ages in both groups were not statistically significant (t) _Male ＝2.401，t _Female ＝1.484，Pall>0.05). AS and CS groups were matched for gender, age, AS shown in table 1.

TABLE 1 basic information Table of patients

And (3) taking AS and CS patients AS state variables, taking the detected value of each blood biomarker AS a test variable, calculating the maximum Youden index according to the sensitivity and specificity corresponding to the ROC curve coordinate, and acquiring the numerical value of the maximum Youden index, wherein the numerical value is the optimal diagnosis critical value of the biomarker values of the two groups of patients. The analysis results show that: 8 of the 30 blood biomarkers failed to calculate optimal diagnostic cutoff values, so these 8 blood biomarkers were discarded as shown in table 2.

TABLE 2 optimal diagnostic cutoff tables for blood biomarkers by ROC screening of AS and CS patients

After the included quantitative values of the 22 blood biomarkers are converted into qualitative 0 s and 1 s by taking the optimal diagnosis critical value as a standard, chi-square verification is carried out together with the 6 biomarker data such as SJ, five hepatitis B and the like, and the result shows that: except that 7 biomarkers such AS SJ, RBC, HBsAg, HBsAb, HBeAg, HBeAb and HBcAb have no statistical significance, other biomarkers such AS AS group and CS group have significant difference (Table 4), but because the P value of RBC and HBcAb is less than 0.2, the invention still incorporates the biomarkers into multi-factor analysis according to the initial statistical constraint, AS shown in Table 3.

TABLE 3 Single-factor analysis of blood biomarker qualitative data for AS and CS groups

The 23 blood biomarkers with P <0.2 in the single factor analysis were subjected to multifactorial analysis, and finally 9 blood biomarkers, such as HGB, LYM, MON, DBiL, GLB, GGT, APTT, fbg and VIII, were obtained as shown in table 4.

TABLE 4 Multi-factor analysis of blood biomarker qualitative data for AS and CS groups

The AS group and the CS group were distinguished according to the blood biomarkers, and the FDA model was constructed from 9 biomarkers screened based on multi-factor analysis. According to the results of the statistical analysis, 7 variables with statistical significance were selected: HGB (X1), MON (X2), GLB (X3), GGT (X4), APTT (X5), VIII (X6), and Fbg (X7), thus yielding the following discriminant functions (Wilks' lambda =0.624, χ 2=125.033, df =7, p = 0.000):

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

the accuracy of the FDA model was self-assessed by a cross-validation method. The classification results showed that 86.7% of the participants were correctly classified into the AS group and the CS group, AS shown in table 5.

TABLE 5 Classification results Table of Primary and Cross-validation methods

Further, the blood biomarker data for the remaining 109 CS patients were replaced with discriminant functions and the CAS and CCS values were calculated, respectively. CS patients who did not incorporate statistical analysis were classified by comparison of CAS and CCS values according to the following principles: if CAS > CCS, then it is judged AS likely AS patient, otherwise CS patient is judged. And revisit work was performed in 2020 to determine whether these patients had developed AS, for a total of 7 CS patients who were missed. The results show that: of 29 patients judged AS by the discriminant function, 18 patients finally developed AS with a coincidence rate of 62.1%, while of 73 patients judged AS CS, 8 patients finally developed AS with a total coincidence rate of 81.4% and accounting for 11.0% of visitors, AS shown in table 6.

TABLE 6 results of revisit of CS patients

In summary, the feature framework is a reliable predictive model developed in conjunction with easily clinically achievable indices. Has important guiding function for effectively controlling the occurrence of the schistosomiasis japonica at the late stage, and lays a firm foundation for achieving the aim of eliminating the schistosomiasis japonica.

In the embodiment, 36 body fluid (serum/plasma) markers and related clinical data of 271 schistosomiasis patients (CS 132 cases and AS139 cases) are retrospectively and contrastively analyzed, and HGB, MON, GLB, GGT, APTT, VIII and Fbg are found to be related to fibrosis.

There were 3 markers reflecting the coagulation function in the FDA model finally constructed from 7 markers from this example. The life history of adult schistosomes, including parasitism and migration in the venous system, and deposition of eggs in liver tissue, cause specific pathological reactions in the host. Theoretically, vascular injury first causes a local inflammatory response, followed by an imbalance in the coagulation and fibrinolytic systems, which interact with each other, ultimately leading to a systemic pathological response in the host. Systemic coagulation plays an important role in the compensation of parasitic immunity as a subsequent response of inflammation to schistosome parasitism. However, to maintain the homeostasis of the blood system, fibrin excessively secreted in blood coagulation needs to be further degraded by fibrinolytic factors such as plasminogen and fibrinolytic protein. Previous studies report blood coagulation dysfunction in schistosomiasis patients, especially abnormal blood coagulation status in late stage schistosomiasis patients. If the coagulation and anticoagulation systems and the fibrinolysis and antifibrinolysis systems are balanced, a hypercoagulable state or bleeding tendency may occur, wherein the increase of VIII is mainly seen in the hypercoagulable state and thrombotic diseases, which may be caused by the deposition of a large amount of eggs in mesenteric blood vessels after schistosome infection, and the egg antigens stimulate the blood vessel wall to activate the related coagulation system. At present, no serum biomarker or detection method with clinical application value is available for evaluating fibrosis of patients with advanced schistosomiasis, and partial thromboplastin time activated by APTT, fbg fibrinogen and factor VIII activity may be good candidate indexes.

In conclusion, the present invention finds that the discriminant function established by using the body fluid (serum/plasma) marker can effectively perform risk early warning on the occurrence of AS through the establishment and verification of the FDA model, which is a reliable prediction model developed by combining with indexes that are easy to be realized clinically, and has the characteristic of strong practicability. Has important guiding function for effectively controlling the generation of AS and lays a firm foundation for achieving the aim of eliminating schistosomiasis. By adopting the system provided by the invention, the advance treatment intervention can be further carried out on the CS patient which is possibly developed into AS, the optimal treatment strategy is established for the chronic schistosomiasis patient, and the development of AS is effectively prevented.

Corresponding to the specific functions implemented by the system for predicting late schistosomiasis provided above, the present invention also provides a method for predicting late schistosomiasis, as shown in fig. 2, the method comprising:

step 100: a biomarker of the blood of a patient to be predicted is obtained.

Step 101: and obtaining a discriminant diagnosis model.

Step 102: and inputting the biomarker of the blood of the patient to be predicted into the discriminant diagnosis model to obtain a prediction result.

Preferably, the discriminant diagnosis model is:

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

wherein CAS is the late schistosomiasis prediction value, CCS is the chronic schistosomiasis prediction value, X1 is the value of biomarker HGB, X2 is the value of biomarker MON, X3 is the value of biomarker GLB, X4 is the value of biomarker GGT, X5 is the value of biomarker APTT, X6 is the value of biomarker VIII, and X7 is the value of biomarker Fbg.

Preferably, the construction process of the discriminant diagnosis model includes:

taking the late stage schistosomiasis patient and the chronic schistosomiasis patient as state variables, and taking the detection value of the biomarker of the blood of the late stage schistosomiasis patient and the detection value of the biomarker of the blood of the chronic schistosomiasis patient as test variables to obtain the ROC curve.

And acquiring the sensitivity and specificity corresponding to the ROC curve coordinate, and determining the Youden index based on the sensitivity and specificity.

The numerical value of the Youden index is obtained and the obtained numerical value is used as a diagnostic threshold value.

Screening the biomarker of blood of patients with advanced schistosomiasis and the biomarker of blood of patients with chronic schistosomiasis based on single factor and multifactorial to obtain the screened biomarkers.

And constructing a discriminant diagnosis model based on the screening biomarkers.

Example two

Based on the technical scheme provided by the embodiment, in the embodiment, the case contrast study comprises two groups of cases from 8 counties (Nanchang, new construction, income virtues, stars, duchang, yongzhen, poyang and the left trunk) of schistosomiasis reissue area in Poyang lake area in Jiangxi province, and 271 cases are recruited from 2013 in 2-3 months, wherein 139 cases of AS patients and 132 cases of CS patients have diagnosis standards according to schistosomiasis diagnosis standards (WS 261-2006) released by Ministry of health of the people's republic of China. These cases do not include diseases such as metabolic inherited liver disease, other parasitic infections, tumors, cardiovascular system, kidney disease, respiratory system, digestive system, diabetes, infection and tissue necrosis, bacteremia, and systemic lupus erythematosus, while minimizing confounding effects of other liver diseases (except hepatitis b). In addition, the embodiment also reserves whether 109 CS patients have been developed into AS when observed in 2020, and the AS is used for evaluating the accuracy of discriminant function early warning.

Morning and post-fasting cubital venous blood was collected under sterile conditions and biomarker testing was completed within 2 hours for all subjects. The biomarkers comprise 36 blood routine, liver function, fibrin degradation product D-dimer, blood coagulation indexes, HBV, alpha fetoprotein, four examinations of liver fibrosis and the like (see Table 7).

TABLE 7 detection methods of blood biomarkers and product provider Table

30 of the 36 blood biomarkers are continuous variables, and the variables are subjected to an orthomorphism test, so that the data of not all the variables are in an orthomorphism distribution, and parts of the data are not in the orthomorphism distribution although logarithmically corrected. Thus, to facilitate uniform analysis of the data, the present embodiment converts these continuous variables into categorical variables. Namely, the method is to evaluate various biomarkers of AS and CS patients by applying Receiver Operating Characteristics (ROC) curve, find out the optimal clinical diagnosis critical point and finish qualitative classification, namely, when the value is smaller than the critical point, the value is assigned to be 0, otherwise, the value is 1. Simultaneously, the area understrate ROC curve (AUC) is less than or equal to 0.5, or P is more than 0.05.

Statistical analysis was performed using the SPSS Statistics 22.0 software (SPSS inc., chicago, IL, USA), test level α =0.05. The data analysis method comprises the following parts:

first, subjects participating in the study were generally characterized by gender and age to ensure consistency between samples. Secondly, single factor analysis is carried out on the difference between the biomarker classified variables of the AS group and the CS group through chi-square check, and variables which can be used for the next step are screened out.

Meanwhile, a certain relation possibly exists between the variable without significant difference and other confounding variables during single-factor analysis, so that in order to avoid the fact that the real effect of the variable is covered by the effect of other confounding indexes, all variables with P <0.2 in the single-factor analysis are included in the multi-factor analysis, and then after the variable P <0.10, the variable P is gradually selected backwards and is kept in the multivariate model. The final report is presented as the Odds Ratio (ORs) and 95% confidence interval (95% CI) and significance level value (P-value).

The Fisher Discriminant Analysis (FDA) is a commonly used multivariate statistical method, which uses projection technique to perform dimension reduction for determining a linear function of a variable to maximize the difference between multiple class samples and minimize the difference between the same class samples. Therefore, in this embodiment, the discrimination function is established by using the FDA model selection stepping method for the variables screened by the multi-factor analysis, and the established discrimination function is self-checked.

In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims (10)

1. A system for predicting late stage schistosomiasis, comprising:

the data acquisition unit is used for acquiring the biomarkers of the blood of the patient to be predicted;

the processor is connected with the data acquisition unit and is implanted with an SPSS statistical analysis software package; the SPSS statistical analysis software package is used for executing the following operations:

and inputting the biomarker of the blood of the patient to be predicted into a discrimination diagnosis model to obtain a prediction result.

2. The system for predicting late schistosomiasis according to claim 1, wherein the discriminant diagnosis model is:

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

wherein CAS is a late schistosomiasis prediction value, CCS is a chronic schistosomiasis prediction value, X1 is a value of a biomarker hemoglobin, X2 is a value of a biomarker monocyte, X3 is a value of a biomarker biosphere, X4 is a value of a biomarker gamma glutamyl transferase, X5 is a value of a biomarker activated partial thromboplastin time, X6 is a value of a biomarker VIII factor activity, and X7 is a value of a biomarker fibrinogen.

3. The system for predicting late schistosomiasis as in claim 1, wherein the processor further comprises:

the curve generation module is connected with the data acquisition unit and is used for obtaining an ROC curve by taking the advanced schistosomiasis patient and the chronic schistosomiasis patient as state variables and taking the detection value of the biomarker of the blood of the advanced schistosomiasis patient and the detection value of the biomarker of the blood of the chronic schistosomiasis patient as test variables;

the index determining module is connected with the curve generating module and used for acquiring the sensitivity and the specificity corresponding to the ROC curve coordinate and determining a Youden index based on the sensitivity and the specificity;

the critical value determining module is connected with the index determining module and is used for acquiring the numerical value of the Youden index and taking the acquired numerical value as a diagnosis critical value;

the marker screening module is connected with the data collector and is used for screening the biomarkers of the blood of the patient with the advanced schistosomiasis and the biomarkers of the blood of the patient with the chronic schistosomiasis to obtain screened biomarkers;

and the model construction module is respectively connected with the marker screening module and the critical value determination module and is used for constructing a discriminant diagnosis model based on the screened biomarkers and the diagnosis critical value.

4. The system for predicting late stage schistosomiasis according to claim 3, wherein the marker screening module comprises:

and the marker screening unit is used for screening the biomarkers of the blood of the late stage schistosomiasis patient and the biomarkers of the blood of the chronic schistosomiasis patient based on single factors and multiple factors to obtain screened biomarkers.

5. The system for predicting late stage schistosomiasis according to claim 1, wherein the processor further comprises:

and the storage module is used for storing the SPSS statistical analysis software package.

6. The system for predicting late schistosomiasis as in claim 5, wherein the storage module is a computer readable storage medium.

7. The system for predicting late schistosomiasis as in claim 1, wherein the processor is a computer.

8. A method for predicting late stage schistosomiasis, comprising:

obtaining a biomarker of blood of a patient to be predicted;

acquiring a discriminant diagnosis model;

and inputting the biomarker of the blood of the patient to be predicted into the discriminant diagnosis model to obtain a prediction result.

9. The method for predicting late schistosomiasis according to claim 8, wherein the discriminant diagnosis model is:

CAS＝0.923X1+3.058X2+2.672X3+2.694X4+4.364X5+2.226X6+7.744X7-8.211

CCS＝1.843X1+1.930X2+1.002X3+1.586X4+2.893X5+2.863X6+8.875X7-7.621

wherein CAS is a late schistosomiasis prediction value, CCS is a chronic schistosomiasis prediction value, X1 is a value of a biomarker hemoglobin, X2 is a value of a biomarker monocyte, X3 is a value of a biomarker biosphere, X4 is a value of a biomarker gamma glutamyl transferase, X5 is a value of a biomarker activated partial thromboplastin time, X6 is a value of a biomarker VIII factor activity, and X7 is a value of a biomarker fibrinogen.

10. The method for predicting late schistosomiasis according to claim 8, wherein the discriminant diagnosis model is constructed by the steps of:

taking a late stage schistosomiasis patient and a chronic schistosomiasis patient as state variables, and taking a detection value of a biomarker of blood of the late stage schistosomiasis patient and a detection value of a biomarker of blood of the chronic schistosomiasis patient as test variables to obtain an ROC curve;

acquiring sensitivity and specificity corresponding to the ROC curve coordinate, and determining a Youden index based on the sensitivity and the specificity;

acquiring a numerical value of the Youden index, and taking the acquired numerical value as a diagnosis critical value;

screening the biomarkers of the blood of the patients with the advanced schistosomiasis and the biomarkers of the blood of the patients with the chronic schistosomiasis based on single factors and multiple factors to obtain screened biomarkers;

and constructing a discriminant diagnosis model based on the screening biomarkers.

Images