A prognosis prediction system and device for bacterial meningitis in children

Abstract

The present invention provides a prognosis prediction system and device for bacterial meningitis in children, which relates to the technical field of prognosis prediction. The system mainly includes: a data receiving module, a data processing module and a visualization module; the data receiving module is used to receive the prediction factor data of Chinese children with bacterial meningitis under test; the data processing module includes a model unit and a prediction unit; the model unit stores a prediction model for predicting the prognosis of bacterial meningitis; the prediction unit calls the prediction model, inputs the prediction factor data, and outputs the stratified scores and prognosis probabilities; the visualization module visualizes the stratified scores and prognosis probabilities, obtains a nomogram and sends it out. This scheme pioneered a prediction model for the prognosis of bacterial meningitis in Chinese children, and has good discrimination, calibration, clinical applicability and stability for the prognosis prediction of bacterial meningitis in Chinese children.

Description

Technical Field

[0001] The present invention relates to the technical field of prognosis prediction, and in particular to a prognosis prediction system and device for children with bacterial meningitis. Background Art

[0002] Worldwide, several prognostic prediction models for bacterial meningitis (BM) have been developed and validated, but they mainly focus on adult patients.

[0003] For example, Weisfelt et al. developed and validated a prediction model that can predict the risk of poor outcome (GOS ≤ 4 points) in adult patients with bacterial meningitis in the Netherlands. The predictors in this risk score include age, heart rate, Glasgow Coma Scale score, cranial nerve palsy, and CSF WBC count <1000 / mm 3 and CSF Gram stain showed G+ cocci. The consistency index of the risk score was 0.84 (95% CI: 0.80-0.87), and the external validation cohort was 0.81 (95% CI: 0.74-0.87). This risk score can assist clinicians in risk stratification of patients with bacterial meningitis and identify high-risk cases with poor prognosis.

[0004] However, the modeling cohorts of existing models do not include Chinese children samples, so the risk scoring results of these models are not accurate for Chinese children; moreover, these models also have problems such as small sample size, low predictive efficiency, low sensitivity and low specificity, resulting in a higher risk of use. Summary of the invention

[0005] The object of the present invention is to provide a system and device for predicting the prognosis of bacterial meningitis in children, so as to solve at least one of the above-mentioned technical problems existing in the prior art.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a prognosis prediction system for bacterial meningitis in children, comprising a data receiving module, a data processing module and a visualization module; The data receiving module is used to receive the predictive factor data of the Chinese pediatric patients with bacterial meningitis; The data processing module includes a model unit and a prediction unit; The model unit stores a prediction model for predicting the prognosis of bacterial meningitis; The prediction unit calls the prediction model, inputs prediction factor data, and outputs stratification scores and prognosis probabilities; The visualization module visualizes the stratified scores and prognostic probabilities to obtain a nomogram and sends it out, which helps to perform prognostic analysis intuitively and quantitatively.

[0007] In a feasible implementation manner, the process of constructing the prediction model includes: Step 1: construct a database based on multi-center case data of bacterial meningitis in Chinese children, such as multi-center BM project case data of the BM collaboration group of Futang Children's Medical Development Research Center, BM case data of Beijing Children's Hospital, etc., so as to enrich the sample space and help to train a reliable prediction model for the prognosis of bacterial meningitis in Chinese children in the subsequent sequence; the database includes multiple inclusion factors; Step 2: randomly divide the case data in the database into a training set and a validation set according to a preset ratio, such as 6:4; and use the prognosis as a stratification label; Step 3: Use data analysis tools such as SPSS 26.0 and Stata SE 15.0 to conduct multiple regression analyses on the included factors and construct the first candidate prediction model, including: in Lasso regression analysis, modeling is based on the minimum extended Bayesian information criterion (EBIC) value and EBIC±standard error (SE) value; in Logistic regression analysis, variables are selected and modeled by stepwise selection, forward selection, backward elimination, forward selection followed by backward elimination, etc.; Step 4: Based on the Akaike information criterion (AIC) and the Bayesian information criterion (BIC), the AIC value and the BIC value of each first candidate prediction model are calculated, and the first candidate prediction model with the lowest AIC value and / or BIC value is selected as the second candidate prediction model; the inclusion factors selected by each second candidate prediction model are used as prediction factors to construct a prediction model; Step 5: Evaluate and verify the prediction model, draw the corresponding receiver operating characteristics (ROC) curve, calculate the corresponding area under the ROC curve (AUC), and evaluate the discrimination of the prediction model; evaluate the calibration of the prediction model through HL (Hosmer-Lemeshow) test; evaluate the clinical applicability of the prediction model through decision curve analysis (DCA); evaluate the stability of the prediction model through ten-fold cross validation (k=10) and bootstrap validation.

[0008] In a feasible implementation manner, the specific calculation formula of the AIC value in step 4 may be: ; in, represents the likelihood function value of the first alternative prediction model, Represents the number of variables in the first candidate prediction model.

[0009] In a feasible implementation manner, the specific calculation formula of the BIC value in step 4 may be: ; in, Indicates the sample size.

[0010] In the second aspect, based on the same inventive concept, the present application also provides a device for predicting the prognosis of bacterial meningitis in children, including a processor, a memory and a bus, wherein the memory stores instructions and data that can be read by the processor, and the processor is used to call the instructions and data in the memory to implement the prognosis prediction system for bacterial meningitis in children as described above, and the bus connects the functional components for transmitting information.

[0011] In a feasible implementation, the device further comprises a display for displaying the prognosis prediction results of bacterial meningitis in Chinese children.

[0012] Preferably, the display includes a touch screen for realizing human-computer interaction.

[0013] By adopting the above technical solution, the present invention has the following beneficial effects: The present invention provides a prognosis prediction system and device for bacterial meningitis in children. It is based on multi-center, large-sample case data of bacterial meningitis in Chinese children, covers various risk factors related to poor prognosis and death of bacterial meningitis in Chinese children, and obtains a prediction model containing optimal predictive factors through step-by-step and hierarchical screening. This prediction model is original. It has been verified that the discrimination, calibration, clinical applicability and stability of the prediction model for the prognosis prediction of bacterial meningitis in Chinese children are better than the existing models, and can effectively identify high-risk cases with poor prognosis and reduce the prediction risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 A diagram of a prognosis prediction system for bacterial meningitis in children provided by an embodiment of the present invention; Figure 2 A flowchart of the process of constructing a prediction model provided by an embodiment of the present invention; Figure 3 The training set ROC curve analysis diagram of the prediction model provided in the embodiment of the present invention; Figure 4 The ROC curve analysis diagram of the validation set of the prediction model provided by the embodiment of the present invention; Figure 5 A calibration curve diagram of a training set of a prediction model provided by an embodiment of the present invention; Figure 6 A calibration curve diagram of a validation set of a prediction model provided by an embodiment of the present invention; Figure 7 A training set net benefit rate diagram of the prediction model provided by an embodiment of the present invention; Figure 8 A net benefit rate graph of a validation set of a prediction model provided by an embodiment of the present invention; Fig. 9 A cross-validation diagram of a prediction model provided by an embodiment of the present invention; Fig.10 A diagram illustrating the assignment of values ​​to a nomogram provided by an embodiment of the present invention; Fig.11 This is a diagram of the prediction results of the nomogram provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0018] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0019] The present invention is further explained below in conjunction with specific implementation modes.

[0020] It should also be noted that the following specific embodiments or specific implementations are a series of optimized settings listed in the present invention to further explain the specific content of the invention, and these settings can be used in combination or in association with each other.

[0021] Embodiment 1: like Figure 1 As shown, this embodiment provides a prognosis prediction system for bacterial meningitis in Chinese children, including a data receiving module, a data processing module and a visualization module; The data receiving module is used to receive the predictive factor data of the Chinese pediatric patients with bacterial meningitis; The data processing module includes a model unit and a prediction unit; The model unit stores a prediction model for predicting the prognosis of bacterial meningitis; The prediction unit calls the prediction model, inputs prediction factor data, and outputs stratification scores and prognosis probabilities; The visualization module visualizes the stratification scores and prognosis probabilities to obtain a nomogram and sends it out.

[0022] Furthermore, if Figure 2 As shown, the construction process of the prediction model includes: Step 1, based on the multi-center case data of bacterial meningitis in children in China, specifically including the case data of the multi-center BM project of the BM collaboration group of the Futang Children's Medical Development Research Center and the case data of the BM of Beijing Children's Hospital between January 1, 2010 and December 31, 2022, a database is constructed; the database includes a variety of inclusion factors, and the inclusion factors are divided into four categories, including demographic information, basic diseases / inducing factors, clinical symptoms and signs, and laboratory examinations, as shown in Table 1;

[0023] Step 2: randomly divide the case data in the database into a training set and a validation set in a ratio of 6:4: the training set contains 1109 cases, with a poor prognosis rate of 15.15% (168 / 1109); the validation set contains 739 cases, with a poor prognosis rate of 13.80% (102 / 739); the prognosis is used as a stratification label; Step 3. Using the data analysis tools SPSS 26.0 and Stata SE 15.0, multiple regression analyses were performed on the included factors and the first alternative prediction model was constructed, including: in the Lasso regression analysis, the lambda penalty coefficient was applied from large to small, and all the included factors were sorted according to their correlation with the poor prognosis of bacterial meningitis in Chinese children, and the models were built based on the minimum extended Bayesian information criterion (EBIC) value and EBIC±standard error (SE) value, respectively, to obtain Model M1 and Model M2; in the Logistic regression analysis, variables were selected and built by the stepwise selection method, forward selection method, backward elimination method, and forward selection followed by backward elimination method, respectively, to obtain Model M3, Model M4, Model M5, and Model M6; The Bayesian information criterion is a statistical criterion for model selection, which belongs to the prior art and is used to evaluate the model. It fully considers the goodness of fit and model complexity of the model, so that the optimal model can be selected; Step 4: Based on the Akaike information criterion (AIC) and the Bayesian information criterion (BIC), the AIC values ​​and BIC values ​​of the six first candidate prediction models are calculated, as shown in Table 2; the Akaike information criterion belongs to the prior art and can be used to find a model that can best explain the data and contains the least free parameters (variables);

[0024] As shown in Table 2, the model with the lowest AIC value (744.077) is M6, and the models with the lowest BIC value (790.177) are M1, M4 and M5. At the same time, it is found that the number of variables of M1, M4 and M5 (9, including 8 included factors and 1 constant) is less than that of M6 (10), so the first candidate prediction model with the lowest BIC value is selected as the second candidate prediction model, namely M1, M4 and M5, so that the preferred model with fewer variables can be obtained, which is conducive to reducing the calculation time of the model and improving the prediction efficiency; the 8 included factors selected by the three second candidate prediction models are used as predictive factors, including convulsion, consciousness change, cranial nerve palsy, muscle weakness, increased muscle tension, first CSF WBC count, first CSF protein value and first CSF glucose value, to construct a prediction model, as shown in Table 3;

[0025] Step 5: Evaluate and verify the prediction model, draw the corresponding receiver operating characteristics (ROC) curve, calculate the corresponding area under the ROC curve (AUC), and evaluate the discrimination of the prediction model. The result is: training set AUC = 0.822 (95% CI: 0.786-0.858). Figure 3 As shown; validation set AUC = 0.815 (95% CI: 0.770-0.859), as Figure 4 As shown; it shows that the prediction model has good discrimination; The calibration of the prediction model was evaluated by HL (Hosmer-Lemeshow) test, and the results were: training set HL test P = 0.4819, such as Figure 5 As shown; validation set HL test P = 0.1168, as Figure 6 As shown, it shows that the prediction model is well calibrated. The clinical applicability of the prediction model was evaluated by decision curve analysis (DCA). The results showed that when the threshold probability was 0.5, the net benefit rates of the training set and the validation set were both 60%-80%. Figure 7-8 As shown, the clinical applicability of the prediction model is good; The stability of the prediction model was evaluated through ten-fold cross validation (k=10), such as Fig. 9 As shown, the results are: AUC of the complete data set = 0.822, and AUC of the test set = 0.8125, which means that the prediction model has good stability; The stability of the prediction model was evaluated through 1000 bootstrap validations, and the result was: 1000 bootstrap validations P < 0.001, that is, the stability of the prediction model was good.

[0026] Furthermore, the specific calculation formula of the AIC value in step 4 can be: ; in, represents the likelihood function value of the first alternative prediction model, Represents the number of variables in the first candidate prediction model.

[0027] Furthermore, the specific calculation formula of the BIC value in step 4 can be: ; in, Indicates the sample size.

[0028] Furthermore, if Fig.10 As shown, the nomogram can be drawn using the nomolog command, and the specific assignment of stratified scores is as follows: Initial CSF WBC count: >1500×10 6 , 1 point; >500~≤1500×10 6 , 2 points; >150~≤500×10 6 , 3 points; >15~≤150×10 6 , 4 points; ≤15×10 6 , 5 points; This can highlight the stratified scores of the first CSF WBC count and reflect the inverse relationship between the first CSF WBC count and the probability of poor prognosis in a gradient manner.

[0029] First CSF glucose value: ≥2.8mmol / L, 1 point; ≥2.1~<2.8mmol / L, 2.5 points; ≥0.9~<2.1mmol / L, 4 points; ≥0.25~<0.9mmol / L, 5 points; <0.25mmol / L, 6 points; This can highlight the higher weight of the first CSF glucose value in the stratification score and reflect the inverse relationship between the first CSF glucose value and the probability of poor prognosis in a gradient manner.

[0030] First CSF protein value: ≤0.45 g / L, 2 points; >0.45~≤1.2g / L, 4 points; >1.2~≤2.4g / L, 6 points; >2.4~≤4.0g / L, 8 points; >4.0g / L, 10 points; This can highlight the highest weight of the first CSF protein value in the stratification score and reflect the proportionality between the first CSF protein value and the probability of poor prognosis in a gradient manner.

[0031] Altered consciousness: If the patient has altered consciousness, 2.5 points are recorded; If the patient has no change in consciousness, a score of 0 is recorded; This can take into account the rule that changes in consciousness have a certain influence weight on the probability of poor prognosis.

[0032] twitch: If the patient has convulsions, score 3.5 points; If the patient has no convulsion, the score is 0; This can take into account the rule that convulsions have a greater influence weight on the probability of poor prognosis.

[0033] There is a decrease in muscle strength: If the patient has decreased muscle strength, it is recorded as 2.5 points; If the patient has no muscle strength loss, the score is 0; This can take into account the rule that decreased muscle strength has a certain influence weight on the probability of poor prognosis.

[0034] Cranial nerve palsy: The patient has cranial nerve palsy, which is scored as 3.5 points; If the patient has no cranial nerve palsy, a score of 0 is recorded; This can take into account the rule that cranial nerve palsy has a greater influence weight on the probability of poor prognosis.

[0035] Increased muscle tone: If the patient has increased muscle tone, score 4 points; If the patient has no increased muscle tension, a score of 0 is recorded; This can take into account the rule that increased muscle tension has a greater influence weight on the probability of poor prognosis.

[0036] After summing up the scores of each stratification, the total stratification score is obtained, such as Fig.11 The specific prognostic probabilities are as follows: 4 to 10 points, the probability of poor prognosis is 1% to 5%; 10.5-13 points, the probability of poor prognosis is 5%-10%; 13.5-16.5 points, the probability of poor prognosis is 10%-20%; 17-19 points, the probability of poor prognosis is 20%-30%; 19.5-20.5 points, the probability of poor prognosis is 30%-40%; 21-22.5 points, the probability of poor prognosis is 40%-50%; 23-24 points, the probability of poor prognosis is 50%-60%; 24.5-26 points, the probability of poor prognosis is 60%-70%; 26.5-28 points, the probability of poor prognosis is 70%-80%; 28.5-31.5 points, the probability of poor prognosis is 80%-90%; 32-34.5 points, the probability of poor prognosis is 90%-95%; If the score is between 34.5 and 37, the probability of poor prognosis is >95%.

[0037] The above-mentioned poor prognosis probability distribution facilitates detailed and accurate prognosis estimation for Chinese children with bacterial meningitis, thereby facilitating accurate identification of high-risk cases in the subsequent period.

[0038] Embodiment 2: The present embodiment provides a device for predicting the prognosis of bacterial meningitis in Chinese children, including a processor, a memory and a bus. The memory stores instructions and data that can be read by the processor. The processor is used to call the instructions and data in the memory to implement the above-mentioned system for predicting the prognosis of bacterial meningitis in Chinese children. The bus connects the functional components for transmitting information.

[0039] Furthermore, the device also includes a display, which is electrically connected to the processor and is used to display the prognosis prediction results of bacterial meningitis in Chinese children.

[0040] Preferably, the display includes a touch screen for realizing human-computer interaction.

[0041] In another embodiment, the present solution can be implemented by a device, which may include a corresponding module for performing each or several steps in the above-mentioned embodiments. The module may be one or more hardware modules specially configured to perform the corresponding steps, or implemented by a processor configured to perform the corresponding steps, or stored in a computer-readable medium for implementation by a processor, or implemented by some combination.

[0042] The processor performs the various methods and processes described above. For example, the method implementation in the present solution can be implemented as a software program, which is tangibly contained in a machine-readable medium, such as a memory. In some embodiments, part or all of the software program can be loaded and / or installed via a memory and / or a communication interface. When the software program is loaded into the memory and executed by the processor, one or more steps in the method described above can be executed. Alternatively, in other embodiments, the processor can be configured to perform one of the above methods by any other appropriate means (for example, by means of firmware).

[0043] The device can be implemented using a bus architecture. The bus architecture can include any number of interconnecting buses and bridges, depending on the specific application and overall design constraints of the hardware. The bus connects various circuits including one or more processors, memories, and / or hardware modules together. The bus can also connect various other circuits such as peripherals, voltage regulators, power management circuits, external antennas, etc.

[0044] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Component (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A prognosis prediction system for bacterial meningitis in Chinese children, characterized in that: It includes a data receiving module, a data processing module and a visualization module; The data receiving module is used to receive the predictive factor data of the Chinese pediatric patients with bacterial meningitis; The data processing module includes a model unit and a prediction unit; The model unit stores a prediction model for predicting the prognosis of bacterial meningitis; The prediction unit calls the prediction model, inputs prediction factor data, and outputs stratification scores and prognosis probabilities; The visualization module visualizes the stratification scores and prognosis probabilities to obtain a nomogram and sends it out.

2. The system according to claim 1, characterized in that The construction process of the prediction model includes: Step 1: construct a database based on multi-center case data of bacterial meningitis in children in China; the database includes multiple inclusion factors; Step 2: randomly divide the case data in the database into a training set and a validation set according to a preset ratio; and use the prognosis as a stratification label; Step 3: Use data analysis tools to conduct multiple regression analyses on the included factors and construct the first candidate prediction model, including: in Lasso regression analysis, modeling based on the minimum extended Bayesian information criterion value and its standard error value; in Logistic regression analysis, variables are selected and modeled through stepwise selection, forward selection, backward elimination, and forward selection followed by backward elimination; Step 4: Based on the Akaike information criterion and the Bayesian information criterion, the AIC value and the BIC value of each first candidate prediction model are calculated, and the first candidate prediction model with the lowest AIC value and / or BIC value is selected as the second candidate prediction model; the inclusion factors selected by each second candidate prediction model are used as prediction factors to construct a prediction model; the AIC value represents the Akaike information criterion value; the BIC value represents the Bayesian information criterion value; Step 5: Evaluate and verify the prediction model, draw the corresponding receiver operating characteristic curve, calculate the area under the receiver operating characteristic curve, and evaluate the discrimination of the prediction model; evaluate the calibration of the prediction model through HL test; evaluate the clinical applicability of the prediction model through decision curve analysis; evaluate the stability of the prediction model through ten-fold cross validation and bootstrap validation.

3. The system according to claim 2, characterized in that In step 3: In the Lasso regression analysis, a λ penalty coefficient was applied from large to small, and all included factors were ranked according to their correlation with the poor prognosis of bacterial meningitis in Chinese children. Models were built based on the EBIC value and the EBIC±standard error value, respectively, to obtain Model M1 and Model M2; in the Logistic regression analysis, variables were selected and modeled by the stepwise selection method, forward selection method, backward elimination method, and forward selection followed by backward elimination method, respectively, to obtain Model M3, Model M4, Model M5, and Model M6; the EBIC represents the minimum extended Bayesian information criterion.

4. The system according to claim 2, characterized in that The specific calculation formula of the AIC value in step 4 is: ； in, represents the likelihood function value of the first alternative prediction model, Represents the number of variables in the first candidate prediction model.

5. The system according to claim 2, characterized in that The specific calculation formula of the BIC value in step 4 is: ； in, represents the sample size; represents the likelihood function value of the first alternative prediction model, Represents the number of variables in the first candidate prediction model.

6. The system according to claim 2, characterized in that Predictors in step 4 included seizures, altered consciousness, cranial nerve palsies, decreased muscle strength, increased muscle tone, first CSF WBC count, first CSF protein value, and first CSF glucose value.

7. The system according to claim 6, characterized in that The specific assignment instructions for stratified scores include: Initial CSF WBC count: >1500×10 6 , 1 point; >500~≤1500×10 6 , 2 points; >150~≤500×10 6 , 3 points; >15~≤150×10 6 , 4 points; ≤15×10 6 , 5 points; First CSF glucose value: ≥2.8mmol / L, 1 point; ≥2.1~<2.8mmol / L, 2.5 points; ≥0.9~<2.1mmol / L, 4 points; ≥0.25~<0.9mmol / L, 5 points; <0.25mmol / L, 6 points; First CSF protein value: ≤0.45 g / L, 2 points; >0.45~≤1.2g / L, 4 points; >1.2~≤2.4g / L, 6 points; >2.4~≤4.0g / L, 8 points; >4.0g / L, 10 points; Altered consciousness: If the patient has altered consciousness, 2.5 points are recorded; If the patient has no change in consciousness, a score of 0 is recorded; twitch: If the patient has convulsions, score 3.5 points; If the patient has no convulsion, the score is 0; There is a decrease in muscle strength: If the patient has decreased muscle strength, it is recorded as 2.5 points; If the patient has no muscle strength loss, the score is 0; Cranial nerve palsy: The patient has cranial nerve palsy, which is scored as 3.5 points; If the patient has no cranial nerve palsy, a score of 0 is recorded; Increased muscle tone: If the patient has increased muscle tone, score 4 points; Patients with no increased muscle tension were scored as 0 points.

8. The system according to claim 7, characterized in that After summing up the scores of each stratification, the total stratification score is obtained. The specific prognostic probabilities include: 4 to 10 points, the probability of poor prognosis is 1% to 5%; 10.5-13 points, the probability of poor prognosis is 5%-10%; 13.5-16.5 points, the probability of poor prognosis is 10%-20%; 17-19 points, the probability of poor prognosis is 20%-30%; 19.5-20.5 points, the probability of poor prognosis is 30%-40%; 21-22.5 points, the probability of poor prognosis is 40%-50%; 23-24 points, the probability of poor prognosis is 50%-60%; 24.5-26 points, the probability of poor prognosis is 60%-70%; 26.5-28 points, the probability of poor prognosis is 70%-80%; 28.5-31.5 points, the probability of poor prognosis is 80%-90%; 32-34.5 points, the probability of poor prognosis is 90%-95%; If the score is between 34.5 and 37, the probability of poor prognosis is >95%.

9. A prognosis prediction device for bacterial meningitis in Chinese children, characterized in that: It includes a processor, a memory and a bus, wherein the memory stores instructions and data read by the processor, the processor is used to call the instructions and data in the memory to implement the system as described in any one of claims 1-8, and the bus connects the functional components for transmitting information.

10. The device according to claim 9, characterized in that A display is also included, and the display is electrically connected to the processor.

Images