Wound-related acute kidney injury target spot prediction device based on biomarker

Abstract

The invention discloses a trauma-related acute kidney injury target prediction device based on a biomarker, and relates to the technical field of injury prediction, and the device comprises the following steps: obtaining a biomarker data set of trauma-related acute kidney injury; performing treatment analysis on the trauma-related acute kidney injury, and determining treatment target point information; obtaining a plurality of biomarker features, and constructing a feature relationship matrix according to the plurality of biomarker features; constructing a deep learning model, and obtaining prediction information of a plurality of target spots; matching and screening the multiple pieces of target spot prediction information, and determining drug candidate molecular parameters; and obtaining a treatment effect score, associating the treatment effect score with the multiple pieces of target spot prediction information, and generating a target spot prediction report. The technical problems of low accuracy of early diagnosis and treatment target prediction of the trauma-related acute kidney injury and low drug screening efficiency in the prior art are solved, and the technical effects of improving the target prediction precision, optimizing the drug screening process and realizing personalized treatment are achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to injury prediction, and specifically to a device for predicting targets of trauma-related acute kidney injury based on biomarkers. Background Art

[0002] Trauma-Associated Acute Kidney Injury (T-AKI) is one of the most common functional injuries in trauma patients. Acute Kidney Injury (AKI) refers to a sharp decline in kidney function, usually manifested by symptoms such as increased serum creatinine levels and decreased urine output. As the name suggests, T-AKI is an acute kidney injury caused by factors such as trauma or surgical trauma. Its pathogenesis is complex and involves multiple pathological changes in the kidney, including tubular injury, renal hypoperfusion, immune response, etc. With the continuous development of the application of biomarkers in disease diagnosis and treatment, and the rapid development of the biomedical field, targeted therapy has gradually become a new direction for disease treatment. However, the current treatment of trauma-related AKI mainly relies on supportive measures and lacks targeted drug therapy. Although some studies have identified some potential biomarkers and therapeutic targets through bioinformatics methods, further experimental verification and drug development are still needed. Traditional target screening usually relies on experimental data or existing clinical experience, which makes target prediction inaccurate and inefficient, thus affecting the treatment effect.

[0003] Therefore, in the current relevant technologies, there are technical problems such as low accuracy in early diagnosis of trauma-related acute kidney injury, low prediction of treatment targets, and low efficiency in drug screening. Summary of the invention

[0004] This application solves the technical problems of low accuracy in early diagnosis and treatment target prediction and low drug screening efficiency of trauma-related acute kidney injury in the prior art by providing a biomarker-based trauma-related acute kidney injury target prediction device, thereby achieving the technical effects of improving target prediction accuracy, optimizing the drug screening process and realizing personalized treatment.

[0005] The present application provides a device for predicting a target of trauma-related acute kidney injury based on biomarkers, the device comprising: a biomarker data set acquisition module, used for screening based on a public database to obtain a biomarker data set of trauma-related acute kidney injury; a treatment target information determination module, used for calling a literature report library in combination with the public database to perform treatment analysis of trauma-related acute kidney injury and determine treatment target information; a biomarker feature acquisition module, used for performing feature analysis based on the biomarker data set and the treatment target information to obtain multiple biomarker features, and constructing a feature relationship matrix based on the multiple biomarker features; a target prediction information acquisition module, used for constructing a deep learning model based on a self-attention mechanism. type, synchronizing the biomarker data set to the deep learning model according to the feature relationship matrix for prediction, and obtaining multiple target prediction information, wherein the multiple target prediction information includes multiple confidence scores; a verification result generating module, for matching and screening the multiple target prediction information using a drug database in combination with the multiple confidence scores, determining drug candidate molecule parameters, verifying the multiple target prediction information according to the drug candidate molecule parameters, and generating a verification result; a treatment effect score obtaining module, for performing a treatment evaluation on the multiple target prediction information according to the verification result, obtaining a treatment effect score, associating the treatment effect score with the multiple target prediction information, and generating a target prediction report.

[0006] In a possible implementation, the biomarker feature acquisition module also performs the following processing: performing correlation calculation based on the biomarker data set and the therapeutic target information to generate multiple correlation coefficients; performing functional annotation of the therapeutic target information according to the multiple correlation coefficients to determine multiple target function annotation information; performing acute kidney injury analysis on the biomarker data set based on the multiple target function annotation information combined with the multiple correlation coefficients to determine multiple biomarker features; constructing a null-valued two-dimensional matrix, converting the multiple target function annotation information into matrix columns, and converting the multiple biomarker features into matrix rows; filling the null-valued two-dimensional matrix with the matrix columns and the matrix rows according to the multiple correlation coefficients to generate the feature relationship matrix.

[0007] In a possible implementation, the target prediction information acquisition module also performs the following processing: based on the self-attention mechanism, regularization is used for weighted processing to construct a self-attention layer, and the input end of the self-attention layer is communicated with the output end of the input layer in the deep learning model; classification is performed based on the feature relationship matrix to determine multiple subspaces, and the multiple subspaces are connected using the self-attention mechanism to construct a multi-head attention mechanism; the target prediction task is retrieved, and a loss function is introduced according to the target prediction task for classification and regression to construct a fully connected layer; the self-attention layer, the multi-head attention mechanism, and the fully connected layer are evaluated by cross-validation, and the deep learning model is constructed according to the evaluation results; wherein, the output end of the self-attention layer is communicated with the input end of the multi-head attention mechanism, the input end of the fully connected layer is communicated with the output end of the multi-head attention mechanism and the output end of the input layer, and the output end of the fully connected layer is the output end of the deep learning model.

[0008] In a possible implementation, the target prediction information acquisition module also performs the following processing: synchronizing the biomarker data set as the first vector input data to the input layer, and synchronizing the feature relationship matrix as the second vector input data to the input layer; using the self-attention layer to perform weighted calculation on the second vector input data to obtain multiple weight coefficients; using the multi-head attention mechanism to perform multi-dimensional capture of the first vector input data and the second vector input data according to the multiple weight coefficients to generate multi-spatial attention parameters; performing confidence scoring based on the multi-spatial attention parameters to generate multiple confidence scores; activating the fully connected layer based on the multiple confidence scores, and performing classification regression prediction on the first vector input data and the second vector input data according to the fully connected layer to generate classification regression prediction results; matching the multiple confidence scores with the classification regression prediction results to generate the multiple target prediction information.

[0009] In a possible implementation, the verification result generation module further performs the following processing: arranging the multiple target prediction information in descending order according to the multiple confidence scores to generate a target prediction sequence; calculating the mean according to the target prediction sequence and setting a confidence threshold; performing initial screening on the multiple target prediction information according to comparing the multiple confidence scores with the confidence threshold, storing the target prediction information corresponding to the confidence score greater than or equal to the confidence threshold, and generating an initial screening result; using the initial screening result as index information, performing a matching search on the drug database to generate a drug interaction relationship; and traversing the drug database based on the drug interaction relationship to perform screening to obtain the drug candidate molecule parameters.

[0010] In a possible implementation, the verification result generation module also performs the following processing: classifying the drug candidate molecules according to the drug interaction relationship to generate a plurality of data to be tested, wherein the plurality of data to be tested include in vitro data to be tested and in vivo data to be tested; performing in vitro experimental analysis based on the in vitro data to be tested in combination with the drug candidate molecules to obtain in vitro experimental influence effect parameters; performing in vivo experimental analysis based on the in vivo data to be tested in combination with the drug candidate molecules to obtain in vivo experimental influence effect parameters; performing in vitro pharmacodynamic evaluation on the plurality of target prediction information based on the in vitro experimental influence effect parameters to generate in vitro pharmacodynamic scores; performing in vivo pharmacodynamic evaluation on the plurality of target prediction information based on the in vivo experimental influence effect parameters to generate in vivo pharmacodynamic scores; performing safety verification on the plurality of target prediction information based on the in vitro pharmacodynamic scores in combination with the in vivo pharmacodynamic scores to generate a target safety factor, and adding the target safety factor to the verification result.

[0011] In a possible implementation, the treatment effect score acquisition module also performs the following processing: setting a quality assessment indicator based on the target safety factor; traversing the multiple target prediction information to perform treatment scoring according to the quality assessment indicator to generate an in vivo treatment quality score and an in vitro treatment quality score; performing an in vivo efficacy analysis based on the in vivo treatment quality score combined with the in vivo efficacy score to generate an in vivo treatment effect score; performing an in vitro efficacy analysis based on the in vitro treatment quality score combined with the in vitro efficacy score to generate an in vitro treatment effect score; and calculating the in vivo treatment effect score, the in vitro treatment quality score and the multiple confidence scores of the multiple target prediction information respectively to generate the treatment effect score.

[0012] In a possible implementation, the therapeutic effect score acquisition module also performs the following processing: performing a mapping analysis based on the in vivo therapeutic effect score and the multiple target prediction information to generate a first mapping relationship parameter; performing a mapping analysis based on the in vitro therapeutic effect score and the multiple target prediction information to generate a second mapping relationship parameter; constructing a mapping relationship network according to the multiple confidence scores based on the first mapping relationship parameter combined with the second mapping relationship parameter; performing transfer learning on the multiple target prediction information according to the mapping relationship network based on the therapeutic effect score to generate a target learning result; feeding back the target learning result to the drug database for response, generating a treatment recommendation, and adding the treatment recommendation to the target prediction report.

[0013] Through the biomarker-based target prediction device for trauma-related acute kidney injury proposed in this application, it is planned to obtain a biomarker data set for trauma-related acute kidney injury; conduct treatment analysis of trauma-related acute kidney injury and determine treatment target information; obtain multiple biomarker features and construct a feature relationship matrix based on multiple biomarker features; construct a deep learning model to obtain multiple target prediction information; match and screen multiple target prediction information to determine drug candidate molecular parameters; obtain treatment effect scores, associate treatment effect scores with multiple target prediction information, and generate target prediction reports. The technical problems of low accuracy in early diagnosis and treatment target prediction of trauma-related acute kidney injury and low efficiency in drug screening in the prior art are solved, and the technical effects of improving target prediction accuracy, optimizing the drug screening process, and realizing personalized treatment are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the device according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic diagram of the structure of a biomarker-based target prediction device for trauma-related acute kidney injury provided in an embodiment of the present application;

[0016] Figure 2 A schematic diagram of the execution process of a biomarker feature acquisition module in a biomarker-based trauma-related acute kidney injury target prediction device provided in an embodiment of the present application.

[0017] Explanation of the reference numerals: biomarker data set acquisition module 10, treatment target information determination module 20, biomarker feature acquisition module 30, target prediction information acquisition module 40, verification result generation module 50, treatment effect score acquisition module 60. DETAILED DESCRIPTION

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0019] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0020] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, device, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0021] The present application embodiment provides a device for predicting targets of trauma-related acute kidney injury based on biomarkers, such as Figure 1 As shown, the device comprises:

[0022] The biomarker data set acquisition module 10 is used to perform screening based on a public database to obtain a biomarker data set for trauma-related acute kidney injury.

[0023] Preferably, biomarker information related to trauma-related acute kidney injury (T-AKI) is extracted from a public database (a public, verified resource database), which usually contains a large amount of biological data, including biological information at different levels such as gene expression, protein level, and metabolite changes. Specifically, relevant biological data resources are screened from a variety of public databases. For example, Gene Expression Omnibus (GEO) is a public database containing a large amount of gene expression data, covering many disease-related gene expression data sets; KEGG (Kyoto Encyclopedia of Genes and Genomes) is an information library that integrates genes, proteins, and metabolic pathways, which can provide a theoretical basis for the screening of biomarkers; Human Protein Atlas provides human protein expression data, which can reveal the relationship between proteins and diseases, etc. In public databases, through keyword retrieval, biomarkers related to acute kidney injury or trauma are located, such as looking for genes, proteins or metabolites that are significantly changed in patients with trauma or acute kidney injury. These screened related biomarker data are then integrated to form a comprehensive data set containing trauma-related acute kidney injury markers, namely a biomarker data set, which includes not only the types of markers (such as certain genes or proteins), but also their expression levels under different pathological conditions, related biological information, etc.

[0024] The treatment target information determination module 20 is used to call the literature report library and combine it with the public database to perform treatment analysis on trauma-related acute kidney injury and determine the treatment target information.

[0025] Preferably, a comprehensive analysis is performed by combining the treatment research results in the literature and the biomarker data in the public database to identify potential therapeutic targets for trauma-related acute kidney injury (T-AKI). The literature report library refers to the data that organizes and summarizes the research results of trauma-related acute kidney injury, usually including a large number of scientific literature, clinical research, experimental research, etc. For example, databases such as PubMed and Scholar contain various research results on trauma, acute kidney injury and their treatment. Specifically, public databases can not only find biomarker information of trauma-related acute kidney injury, but also obtain targeted treatment research results related to these markers. For example, a certain biomarker (such as KIM-1, IL-18, etc.) is closely related to acute kidney injury, and the literature report library is a library of literature reports. There may be studies in the literature showing specific drugs or treatments for these markers. By calling these literature reports and combining the biomarker information in the public database with the treatment strategies in the literature, potential drug targets or treatment targets can be identified. These targets may be molecules or proteins directly related to the pathogenesis of trauma-related acute kidney injury, and then the treatment target information can be determined. Among them, the treatment target refers to the key molecule or signaling pathway that can be intervened by drugs or other treatment methods during the disease process. For example, the occurrence of trauma-related acute kidney injury is usually accompanied by kidney oxidative stress, inflammatory response and cell death. Targets related to these processes, such as specific receptors of renal tubular cells, key enzymes or inflammatory factors (such as TNF-α, IL-1β, etc.), may be treatment targets.

[0026] The biomarker feature acquisition module 30 is used to perform feature analysis based on the biomarker data set and the therapeutic target information to obtain multiple biomarker features and construct a feature relationship matrix based on the multiple biomarker features.

[0027] Preferably, by in-depth analysis of the biomarker dataset and therapeutic target information, relevant features are extracted, and then a matrix describing the relationship between these features is constructed, that is, a feature relationship matrix is ​​constructed, which helps to provide effective input features for subsequent deep learning models to optimize prediction and screening of therapeutic targets. Specifically, by comparing and analyzing the biomarker dataset and therapeutic targets, key features are identified, that is, multiple biomarker features are obtained, wherein biomarker features refer to specific attributes extracted that can characterize pathological states or are related to therapeutic targets, such as certain biomarkers may be upregulated in trauma-related acute kidney injury. Others may change after targeted therapy. Common biomarker features may include differences in the expression of markers in healthy and injured states, dynamic changes of markers over time, and correlations between markers and other markers or therapeutic targets. Multiple biomarker features are then used to construct a feature relationship matrix. Specifically, the rows and columns of the feature relationship matrix represent different biomarker features, and each element in the matrix represents the strength of the relationship between the two features, such as correlation or synergistic effect. The feature relationship matrix can capture the complex relationships between different biomarker features, thereby improving the accuracy and reliability of therapeutic target prediction.

[0028] The target prediction information acquisition module 40 is used to build a deep learning model based on the self-attention mechanism, synchronize the biomarker data set to the deep learning model according to the feature relationship matrix for prediction, and obtain multiple target prediction information, wherein the multiple target prediction information includes multiple confidence scores.

[0029] Preferably, the self-attention mechanism is used to process the biomarker data set and the feature relationship matrix in the deep learning model, and then the therapeutic targets of trauma-related acute kidney injury (T-AKI) are predicted, and a corresponding confidence score is provided for each predicted target. The self-attention mechanism is an important technology in deep learning, especially when processing sequence data (such as text or time series data). Each marker in the biomarker data set can be regarded as a feature or data at a time point. The relationship between these features is modeled through the self-attention mechanism, and their correlation is automatically learned to obtain a deep learning model (for example, a deep learning model is constructed using a large model architecture such as Transformer or its variants, such as BERT, GPT, etc.) to better capture the role of different markers in trauma-related acute kidney injury. Specifically, the biomarker data set and the feature relationship matrix are synchronized to the deep learning model, and the independent information of each marker is processed by the deep learning model, and the complex correlation between them can also be captured, thereby improving the accuracy of target prediction.

[0030] Preferably, the deep learning model is then used to start predicting the therapeutic targets of trauma-related acute kidney injury. Specifically, the deep learning model predicts multiple potential therapeutic targets based on the input biomarker data and its characteristic relationship matrix, which are usually molecules (such as genes, proteins, receptors, etc.) closely related to the occurrence, development or treatment of the disease. Each therapeutic target may contain multiple information, such as the predicted target name, the type of target (such as genes, proteins, enzymes, etc.), the biological process that the target may involve, etc., and each therapeutic target is often accompanied by a confidence score, which indicates the reliability of the model's prediction of the target. In the prediction of targets for trauma-related acute kidney injury, the confidence score is usually between 0 and 1. The closer it is to 1, the more confident the model is that the target is an effective therapeutic target. For example, when predicting targets for trauma-related acute kidney injury, the model may predict target 1 (such as KIM-1) with a confidence score of 0.85, indicating that the model has 85% confidence in predicting KIM-1 as a target; target 2 (such as IL-18), with a confidence score of 0.92, indicating that the model has 92% confidence in predicting IL-18 as a target; target 3 (such as NGAL), with a confidence score of 0.75, indicating that the model has 75% confidence in predicting NGAL as a target. Through deep learning and self-attention mechanisms, biomarker data and their feature relationships are analyzed to predict potential therapeutic targets for trauma-related acute kidney injury, and a confidence score is provided for each predicted target to help improve the accuracy and reliability of target prediction.

[0031] The verification result generation module 50 is used to use the drug database in combination with the multiple confidence scores to match and screen the multiple target prediction information, determine the drug candidate molecule parameters, verify the multiple target prediction information according to the drug candidate molecule parameters, and generate a verification result.

[0032] Preferably, the target prediction results are combined with the information in the drug database, and through screening and verification, the drug candidate molecules that may have therapeutic potential for trauma-related acute kidney injury (T-AKI) are finally determined, and their correlation with the predicted targets and efficacy are evaluated. Specifically, the drug database is used in combination with the confidence score to match and screen the target prediction information. The drug database usually contains a large amount of drug molecule information, including approved drugs, drug molecules in clinical trials, and some potential new drug candidate molecules. Common drug databases include DrugBank, PubChem, ChEMBL, etc. In the drug database, drug candidate molecules related to these predicted targets are searched and screened. For example, the drug database It may provide known inhibitors, agonists or other related molecules for a certain target. By comparing the molecules in the drug database with the target prediction information, possible drug candidate molecules are screened out, and then the parameters of the drug candidate molecules are determined, which usually refers to the relationship between the drug molecule and the target and its potential therapeutic effect, which may include the structural characteristics of the drug molecule, such as the chemical structure, molecular weight, lipophilicity (such as LogP), solubility, etc. of the drug, drug-target interaction information, the binding affinity of the drug candidate molecule with the target (such as IC50, Ki value, etc.), binding site, etc., and the biological effect of the drug; each drug molecule in the database will be matched with the target prediction results according to these parameters to evaluate its therapeutic effect and possibility as a target.

[0033] Preferably, verifying multiple target prediction information according to the drug candidate molecule parameters means that after the drug candidate molecule parameters are determined, the target prediction information is verified to ensure that these candidate molecules can effectively act on the predicted targets, thereby verifying their potential as a treatment for T-AKI. The verification process may include molecular docking, which evaluates the affinity of the drug molecule and the target by calculating and simulating the binding of the drug molecule to the target, and further confirms whether the drug can effectively bind to and inhibit the target; biological verification, verifying the effectiveness of the drug candidate molecule in vitro (such as a cell model) or in vivo (such as an animal model) in the laboratory, such as detecting the binding effect of the drug on the target, the relief effect on kidney damage, etc.; based on the results of the above verification process, generating the verification results of the drug candidate molecule, including drug target verification, verifying whether the drug can effectively bind to the predicted target and inhibit its function; therapeutic effect evaluation, how is the therapeutic effect of the drug on trauma-related acute kidney injury, whether it can reduce kidney damage and improve kidney function; finally confirming the effectiveness of the drug on the target, thereby providing medical data support and decision-making.

[0034] The treatment effect score obtaining module 60 is used to perform treatment evaluation on the multiple target prediction information according to the verification results, obtain a treatment effect score, associate the treatment effect score with the multiple target prediction information, and generate a target prediction report.

[0035] Preferably, based on the validation results of the drug candidate molecules, the therapeutic effect of each target is further evaluated, and a detailed target prediction report is generated based on these evaluation results, and the therapeutic potential of each target and the effect of its corresponding drug candidate molecules are displayed. Specifically, the therapeutic evaluation of multiple target prediction information based on the validation results usually includes whether the drug candidate molecule can effectively bind to the target and exert its effect, whether it can alleviate the pathological process (such as tubular injury, inflammatory response, etc.) in trauma-related acute kidney injury (T-AKI), whether the target plays an important role in the pathological process, whether the drug improves kidney function by regulating the target, and whether the drug candidate molecule can reduce damage and improve renal function by acting on the target. Renal function or improvement of clinical symptoms, and then obtain a treatment effect score, that is, a quantitative expression of the treatment effect of each target, such as scoring according to the effect of the drug on the target, such as the drug may be rated as effective, partially effective or ineffective, or the adaptability of the drug in clinical applications, including patient acceptance, drug metabolism characteristics, etc.; the treatment effect score is then associated with the prediction information of each target as a target prediction report, which lists in detail the treatment effect, prediction information and possible drug candidate molecules of each target, which is used to continuously optimize the prediction model, help researchers understand the therapeutic potential of each target, and select the most promising targets and drugs based on the treatment effect score of the target, thereby providing a scientific basis for targeted therapy.

[0036] According to an embodiment of the present invention, a target prediction device for trauma-related acute kidney injury based on biomarkers is used to solve the technical problems of low accuracy of target prediction and low efficiency of drug screening in the early diagnosis and treatment of trauma-related acute kidney injury in the prior art, and achieves the technical effects of improving target prediction accuracy, optimizing the drug screening process and realizing personalized treatment. A target prediction device for trauma-related acute kidney injury based on biomarkers includes: a biomarker data set acquisition module 10, a treatment target information determination module 20, a biomarker feature acquisition module 30, a target prediction information acquisition module 40, a verification result generation module 50, and a treatment effect score acquisition module 60.

[0037] The specific configuration of the biomarker feature acquisition module 30 will be described in detail below. The biomarker feature acquisition module 30 may further include: performing correlation calculation based on the biomarker data set and the treatment target information to generate multiple correlation coefficients; performing functional annotation on the treatment target information according to the multiple correlation coefficients to determine multiple target function annotation information; performing acute kidney injury analysis on the biomarker data set based on the multiple target function annotation information combined with the multiple correlation coefficients to determine multiple biomarker features; constructing a null value two-dimensional matrix, converting the multiple target function annotation information into matrix columns, and converting the multiple biomarker features into matrix rows; filling the matrix columns and the matrix rows into the null value two-dimensional matrix according to the multiple correlation coefficients to generate the feature relationship matrix.

[0038] Preferably, a statistical method (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) is used to calculate the correlation between the biomarker and the therapeutic target, and multiple correlation coefficients are generated, indicating the linear relationship between each biomarker and the specific therapeutic target, wherein the correlation coefficient is a value between -1 and 1, indicating the strength and direction of the relationship between the two (positive correlation or negative correlation). For example, if the correlation coefficient between a biomarker and the target is close to 1, it indicates that there is a strong positive correlation between them; if it is close to -1, it indicates a negative correlation; if it is close to 0, it indicates that there is no obvious relationship between the two. According to the calculated correlation coefficient, the function of each target is annotated, that is, according to the biological function of the target (such as participation in inflammatory response, immune regulation, cell apoptosis, etc.), the functional characteristics of the target are extracted, including the role of the target in the biological process, related signal pathways, potential disease associations, etc., to determine multiple target function annotation information, and then based on the target function annotation information and the correlation coefficient, the biomarker data set is analyzed for acute kidney injury, that is, which biomarkers are closely related to acute kidney injury, and further screen out marker features closely related to the therapeutic target or the kidney injury process.

[0039] Preferably, an empty two-dimensional matrix is ​​constructed, the target function annotation information is converted into matrix columns, and the biomarker features are converted into matrix rows, that is, the columns of the matrix represent the functional information of the therapeutic target, and the rows represent the features of the biomarkers. The correlation coefficient is used as the element of the matrix to represent the strength of the relationship between the target and the marker. Then, according to the correlation coefficient (the strength of the relationship between the biomarker and the target), the elements in the two-dimensional matrix are filled to determine the feature relationship matrix, which contains the relationship information between the biomarker features and the target functions. The value of each matrix element (correlation coefficient) represents the similarity between the biomarker features and the target function annotation. For example, the feature relationship matrix may be as shown in the following table:

[0040]

[0041] The specific configuration of the target prediction information acquisition module 40 will be described in detail below. The target prediction information acquisition module 40 may further include: using regularization for weighted processing based on the self-attention mechanism to construct a self-attention layer, the input end of the self-attention layer is connected to the output end of the input layer in the deep learning model for communication; classifying based on the feature relationship matrix, determining multiple subspaces, connecting the multiple subspaces using the self-attention mechanism, and constructing a multi-head attention mechanism; calling the target prediction task, introducing the loss function according to the target prediction task for classification and regression, and constructing a fully connected layer; cross-validating the self-attention layer, the multi-head attention mechanism, and the fully connected layer for evaluation, and constructing the deep learning model according to the evaluation results; wherein the output end of the self-attention layer is connected to the input end of the multi-head attention mechanism for communication, the input end of the fully connected layer is connected to the output end of the multi-head attention mechanism and the output end of the input layer for communication, and the output end of the fully connected layer is the output end of the deep learning model.

[0042] Preferably, the self-attention mechanism dynamically adjusts the weights of input data (such as features, marker data, etc.) by calculating the relative importance between them, thereby helping the model to better focus on important features and calculating the relationship between different input features (such as biomarker features or target information). Regularized weighted processing reduces the network's dependence on training data by controlling the weights, so that the model has better performance on new data and helps the model to generalize better. A self-attention layer is constructed through the self-attention mechanism to process the features in the input data and weight them according to the relationship between them. The input end of the self-attention layer is connected to the input layer of the deep learning model to receive and process the input data. Based on the feature relationship matrix, the model divides the input data (i.e., biomarker features) into multiple subspaces, each of which contains a set of specific features. The self-attention mechanism is used to connect multiple subspaces to construct a multi-head attention mechanism, that is, multiple self-attention mechanisms are executed in parallel and their results are combined to capture more patterns and information.

[0043] Preferably, the target prediction task is retrieved, that is, predicting therapeutic targets related to acute kidney injury (AKI). Depending on the target prediction task, a suitable loss function can be selected for classification regression. For example, for classification tasks, a cross entropy loss function is used; for regression tasks, a mean square error (MSE) loss function is used to evaluate the accuracy of target prediction. Depending on the task, it may be a classification task (predicting target category) or a regression task (predicting the value or confidence of the target). A fully connected layer is constructed, which is usually used to integrate the outputs of different layers and make a final prediction, including the output of the self-attention layer and the multi-head attention mechanism, and finally generate a prediction result; cross-validation is used for evaluation, and a deep learning model is constructed based on the evaluation results, that is, the data set is divided into multiple subsets (for example, 5-fold cross-validation), and the model is trained and evaluated on each subset to ensure that the model is stable on different data subsets and has good generalization ability.

[0044] Preferably, the output of the self-attention layer is connected to the input of the multi-head attention mechanism; the input of the fully connected layer is connected to the output of the multi-head attention mechanism and the output of the input layer. The model can use the information extracted from the input layer to the self-attention layer and the multi-head attention mechanism to further perform feature fusion and generate the final prediction result; the output of the fully connected layer is the output of the deep learning model, which generates the final prediction result of the model, that is, the prediction of the target, and can accurately and efficiently predict the target.

[0045] The specific configuration of the target prediction information acquisition module 40 will be described in detail below. The target prediction information acquisition module 40 may further include: synchronizing the biomarker data set as the first vector input data to the input layer, synchronizing the feature relationship matrix as the second vector input data to the input layer; using the self-attention layer to perform weighted calculation on the second vector input data to obtain multiple weight coefficients; using the multi-head attention mechanism to perform multi-dimensional capture of the first vector input data and the second vector input data according to the multiple weight coefficients to generate multi-space attention parameters; performing confidence scoring according to the multi-space attention parameters to generate multiple confidence scores; activating the fully connected layer according to the multiple confidence scores, and performing classification regression prediction on the first vector input data and the second vector input data according to the fully connected layer to generate classification regression prediction results; matching the multiple confidence scores with the classification regression prediction results to generate the multiple target prediction information.

[0046] Preferably, the biomarker data set and the feature relationship matrix are synchronously input into the input layer as the first vector input data and the second vector input data, respectively. The second vector input data (feature relationship matrix) is weighted by using the self-attention layer to calculate the weight coefficient of each feature (such as the correlation between the target and the marker) to obtain multiple weight coefficients. Then, the multi-head attention mechanism processes different parts of the input data in parallel through multiple independent heads. Each head can focus on different aspects or subspaces of the input data. Through the parallel calculation of multiple heads, the model can simultaneously focus on multiple dimensions in the biomarker data and the feature relationship matrix, capture the interrelationships and complexity between different dimensions, and then generate multi-space attention parameters, which can describe the multi-dimensional relationship between biomarkers and different target functions.

[0047] Preferably, a confidence score is performed according to the multi-spatial attention parameters, that is, a confidence score is calculated to reflect the confidence of the model in the prediction results of each target. The confidence score is a value between 0 and 1. The closer it is to 1, the higher the confidence of the model in the prediction of the target. Each target will generate a corresponding confidence score, which indicates the model's confidence in the prediction of each target as a therapeutic target. Then, the fully connected layer is activated according to multiple confidence scores, and the fully connected layer performs classification regression prediction, that is, the therapeutic potential of each target is predicted based on the input biomarker features (first vector) and target function information (second vector). The goal of regression-like prediction is to output whether a target is an effective therapeutic target, or the confidence score of the target, and then generate prediction results, including classification results, such as whether a target is an effective therapeutic target; regression results, the therapeutic effect score or confidence score of a target; finally, multiple confidence scores are matched and combined with classification regression prediction results to generate prediction information of the target. The prediction information of each target includes the target name and prediction category (for example, whether it is an effective therapeutic target), the confidence score of the target, that is, the confidence of the model in the therapeutic effect or effectiveness of the target, etc., as the final output result.

[0048] The specific configuration of the verification result generation module 50 will be described in detail below. The verification result generation module 50 may further include: arranging the multiple target prediction information in descending order according to the multiple confidence scores to generate a target prediction sequence; calculating the mean according to the target prediction sequence and setting a confidence threshold; performing an initial screening of the multiple target prediction information according to the multiple confidence scores and the confidence threshold, storing the target prediction information corresponding to the confidence score greater than or equal to the confidence threshold, and generating an initial screening result; using the initial screening result as index information, performing a matching search on the drug database to generate a drug interaction relationship; traversing the drug database based on the drug interaction relationship to screen and obtain the drug candidate molecule parameters.

[0049] Preferably, multiple target prediction information is arranged in descending order according to multiple confidence scores, which means that the target prediction information is sorted in descending order according to the confidence score of each target (usually between 0 and 1), that is, the target prediction information with a higher confidence score will be ranked in front, indicating that the relationship between the target in the biomarker and the therapeutic target is stronger, and the accuracy of these target model predictions is higher, and the target prediction sequence is obtained, and then the average confidence level of all target predictions is reflected by calculating the average confidence score of all targets, and it is set as the confidence threshold to help remove those targets with lower confidence and poor prediction accuracy; the confidence score of each target is compared with the set confidence threshold, and the targets with a confidence score greater than or equal to the threshold are screened out to ensure that the target information with a higher prediction confidence is retained, and the targets with low confidence are eliminated as the initial screening results.

[0050] Preferably, the initial screening results are input as index information into the drug database for matching retrieval. By retrieving drugs related to the targets in the initial screening results, the model will be able to find drug molecules with potential interactions with these targets, and then generate a table of drug-target interaction relationships, i.e., drug interactions, listing drug molecules related to each target, such as target inhibitors, agonists, or drugs that can regulate biological processes related to the target; then based on the drug interaction relationships, further traverse the relevant drug molecules in the drug database to screen out the drugs most relevant to the target. Specifically, the screened high-confidence targets are matched with the targets in the drug database, and through the target-drug interaction information of the drug, drug candidate molecules that can bind to the target and may have a therapeutic effect on TR-AKI are screened, and then the drug candidate molecule parameters are obtained, which may include chemical structure information, pharmacological parameters, and drug-target interactions.

[0051] The specific configuration of the verification result generation module 50 will be described in detail below. The verification result generation module 50 may further include: classifying the drug candidate molecules according to the drug interaction relationship to generate a plurality of data to be tested, wherein the plurality of data to be tested include in vitro data to be tested and in vivo data to be tested; performing in vitro experimental analysis based on the in vitro data to be tested in combination with the drug candidate molecules to obtain in vitro experimental effect parameters; performing in vivo experimental analysis based on the in vivo data to be tested in combination with the drug candidate molecules to obtain in vivo experimental effect parameters; performing in vitro efficacy evaluation on the plurality of target prediction information based on the in vitro experimental effect parameters to generate in vitro efficacy scores; performing in vivo efficacy evaluation on the plurality of target prediction information based on the in vivo experimental effect parameters to generate in vivo efficacy scores; performing safety verification on the plurality of target prediction information based on the in vitro efficacy scores in combination with the in vivo efficacy scores to generate target safety factors, and adding the target safety factors to the verification results.

[0052] Preferably, the drug interaction relationship indicates the potential effect (such as inhibition, activation, etc.) between the drug and the target. According to the drug interaction relationship, the drug candidate molecules are classified into multiple experimental data (in vitro experimental data, in vivo experimental data), wherein the in vitro experimental data refers to the experiment that the drug candidate molecules will be subjected to, usually the experiment conducted in cells, tissues or molecular models (for example, cell proliferation inhibition experiment, enzyme activity test, etc.), and the in vivo experimental data refers to the experiment that the drug candidate molecules will be subjected to in animals or humans, usually the drug metabolism, efficacy evaluation, toxicity test conducted in mice, rats or other animal models. Experiments such as in vitro experiments; in vitro experimental analysis is performed based on the in vitro experimental data combined with the drug candidate molecules, that is, in vitro experiments are performed on the screened drug candidate molecules, and the binding effect of the drug and the target is verified using a cell culture system, and the effect of the drug on TR-AKI-related cell types (such as renal tubular epithelial cells, endothelial cells, etc.) is evaluated. For example, drug candidate molecules may be tested for their effects on cell proliferation, cell death, and specific protein activity, etc., and ultimately in vitro experimental effect parameters are obtained, such as IC50 values ​​(half-maximal inhibitory concentration of the drug), EC50 values ​​(effective concentration), etc., which indicate the activity and effect of the drug in an in vitro environment.

[0053] Preferably, in vivo experimental analysis is performed based on the in vivo experimental data combined with the drug candidate molecule, usually in an animal model, that is, the efficacy of the drug is verified in the animal model, its improvement effect on trauma-related acute kidney injury is tested, and the effectiveness, safety and side effects of the drug are evaluated through drug administration trials to evaluate the pharmacokinetics (ADMET, such as absorption, distribution, metabolism, excretion, toxicity) and pharmacodynamics (such as anti-inflammatory, anti-tumor, antioxidant, etc.) of the drug, and then the in vivo experimental effect parameters are obtained, such as plasma concentration curve, maximum drug concentration (Cmax), half-life (T1 / 2), bioavailability of the drug, etc., and then multiple target prediction information is in vivo based on the in vivo experimental effect parameters. Internal efficacy evaluation, that is, evaluating the repair effect of the drug on acute kidney injury by detecting biomarkers related to kidney function (such as urea nitrogen, etc.), paying attention to the distinction between efficacy and therapeutic effect here. Efficacy is the ideal effect of medication, and therapeutic effect is the actual effect after use; then based on the in vitro efficacy score combined with the in vivo efficacy score, multiple target prediction information is verified for safety, that is, the in vitro efficacy score and the in vivo efficacy score are combined to evaluate the therapeutic effect and safety of drug candidate molecules on the target, and then obtain the target safety factor, which represents the safety of the drug on the target. Finally, the target safety factor is added to the verification result of the drug, and a comprehensive evaluation is performed together with other experimental data of the drug (such as efficacy, drug-target interaction, etc.).

[0054] The specific configuration of the treatment effect score acquisition module 60 will be described in detail below. The treatment effect score acquisition module 60 may further include: setting a quality assessment index based on the target safety factor; traversing the multiple target prediction information and performing treatment scoring according to the quality assessment index to generate an in vivo treatment quality score and an in vitro treatment quality score; performing in vivo efficacy analysis based on the in vivo treatment quality score combined with the in vivo efficacy score to generate an in vivo treatment effect score; performing in vitro efficacy analysis based on the in vitro treatment quality score combined with the in vitro efficacy score to generate an in vitro treatment effect score; respectively calculating the in vivo treatment effect score, the in vitro treatment quality score and the multiple confidence scores of the multiple target prediction information to generate the treatment effect score.

[0055] Preferably, a quality assessment index is set according to the target safety factor to measure the predicted therapeutic potential of the target, and then multiple target prediction information is traversed, and the therapeutic potential of each target is scored according to the set quality assessment index, which may include a comprehensive evaluation of the target's safety, in vivo and in vitro efficacy, toxicity, etc., and then an in vivo treatment quality score and an in vitro treatment quality score are generated, wherein the in vivo treatment quality score considers factors such as the drug's effectiveness, safety, and pharmacokinetics in vivo, and the in vitro treatment quality score is scored for in vitro experimental data to evaluate the drug's in vitro effect and toxicity; then, an in vivo efficacy analysis is performed based on the in vivo treatment quality score combined with the in vivo efficacy score to generate an in vivo treatment effect score, such as using a five-point system (0-5) or a percentage scoring system to represent the efficacy to generate an in vivo treatment effect score, For example, 0 indicates ineffectiveness, 1-2 indicates mild efficacy, 3-4 indicates moderate efficacy, and 5 indicates complete cure; the in vitro efficacy analysis is then performed based on the in vitro treatment quality score combined with the in vitro efficacy score to generate an in vitro therapeutic efficacy score, which reflects the effect of the drug in the in vitro experiment; the in vivo therapeutic efficacy score (since the in vivo experimental results are closer to clinical reality, a higher weight can be given), the in vitro treatment quality score (in vitro experiments provide preliminary data on the action of the drug, but due to their limitations, a moderate weight can be given) and multiple confidence scores of multiple target prediction information (according to the training results of the deep learning model, the confidence score has a greater impact on the credibility of the target, so a higher weight is given) are calculated to obtain a therapeutic efficacy score, which represents the overall effect and reliability of the drug in treating the target.

[0056] The specific configuration of the treatment effect score obtaining module 60 will be described in detail below. The treatment effect score obtaining module 60 may further include:

[0057] A mapping analysis is performed based on the in vivo therapeutic effect score and the prediction information of multiple targets, that is, through statistical or mathematical methods (such as regression analysis, cluster analysis, correlation analysis, etc.), the performance of each target in vivo is determined according to the in vivo efficacy score, and then the first mapping relationship parameter is determined, which reflects the relationship between the in vivo therapeutic effect score and each target, and indicates the therapeutic potential or effect strength of the target in vivo; similarly, a mapping analysis is performed based on the in vitro therapeutic effect score and the prediction information of multiple targets to generate a second mapping relationship parameter, which reflects the performance of the target in the in vitro experiment, and may describe the effectiveness of the target, the response to the drug, etc.; then the in vivo and in vitro mapping relationship parameters (i.e., the first and second mapping relationship parameters) are combined with the confidence score of the target to construct a mapping relationship network, which represents the relationship between the therapeutic effects of different targets (in vivo and in vitro data) and the model's prediction information for the target. In this way, the model can comprehensively evaluate the therapeutic potential of the target from different data sources (in vivo and in vitro experimental data and prediction information).

[0058] Preferably, transfer learning is performed on multiple target prediction information according to the mapping relationship network based on the treatment effect score, which means using the treatment effect score (in vivo and in vitro score) and the information in the mapping relationship network to optimize the target learning, that is, to transfer the in vivo and in vitro experimental data (structure constructed by the mapping relationship network) to the target learning, and update the understanding of the target. Among them, transfer learning is a machine learning method that uses the knowledge gained in one task to help improve the learning efficiency in another related task, outputs the target learning results, and represents the therapeutic effect, potential, relevance and confidence of each target. The results are fed back to the drug database for response, that is, personalized treatment recommendations are made based on the therapeutic effects of the target and the drug recommendation results, such as drug screening, recommending which drugs are most likely to effectively treat the target, and then helping doctors choose appropriate treatment plans, such as recommending a certain treatment method or combination of drugs based on the characteristics and therapeutic effects of the target. These treatment recommendations are added to the target prediction report, which contains detailed information about each target, therapeutic effects, recommended drugs and their candidate molecules, experimental data support, etc., so that effective therapeutic targets can be efficiently screened out, providing a scientific basis for drug development, clinical treatment plan formulation and personalized treatment.

[0059] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0060] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A biomarker-based target prediction device for trauma-related acute kidney injury, characterized in that: The device comprises: A biomarker dataset acquisition module, which is used to obtain a biomarker dataset of trauma-related acute kidney injury by screening based on public databases; A treatment target information determination module, used to call the literature report library and combine it with the public database to perform treatment analysis on trauma-related acute kidney injury and determine the treatment target information; A biomarker feature acquisition module, configured to perform feature analysis based on the biomarker data set and the therapeutic target information to obtain a plurality of biomarker features, and to construct a feature relationship matrix based on the plurality of biomarker features; A target prediction information acquisition module, which is used to build a deep learning model based on a self-attention mechanism, synchronize the biomarker data set to the deep learning model according to the feature relationship matrix for prediction, and obtain multiple target prediction information, wherein the multiple target prediction information includes multiple confidence scores; A verification result generating module, used to match and screen the multiple target prediction information using a drug database in combination with the multiple confidence scores, determine drug candidate molecule parameters, verify the multiple target prediction information according to the drug candidate molecule parameters, and generate a verification result; The treatment effect score acquisition module is used to perform treatment evaluation on the multiple target prediction information according to the verification results, obtain a treatment effect score, associate the treatment effect score with the multiple target prediction information, and generate a target prediction report.

2. The biomarker-based trauma-related acute kidney injury target prediction device according to claim 1, characterized in that: Based on the biomarker data set and the therapeutic target information, feature analysis is performed to obtain multiple biomarker features, and a feature relationship matrix is ​​constructed according to the multiple biomarker features. The device includes: performing correlation calculation based on the biomarker data set and the therapeutic target information to generate a plurality of correlation coefficients; Functionally annotating the therapeutic target information according to the multiple correlation coefficients to determine multiple target function annotation information; Performing acute kidney injury analysis on the biomarker dataset based on the multiple target function annotation information combined with the multiple correlation coefficients to determine multiple biomarker features; Constructing a two-dimensional matrix of null values, converting the plurality of target function annotation information into matrix columns, and converting the plurality of biomarker features into matrix rows; According to the multiple correlation coefficients, the matrix columns and the matrix rows are filled into the null-value two-dimensional matrix to generate the characteristic relationship matrix.

3. The biomarker-based target prediction device for trauma-related acute kidney injury according to claim 1, characterized in that: A deep learning model is constructed based on the self-attention mechanism. The device includes: Based on the self-attention mechanism, regularization is used for weighted processing to construct a self-attention layer, and the input end of the self-attention layer is connected to the output end of the input layer in the deep learning model for communication; Classify based on the feature relationship matrix, determine multiple subspaces, connect the multiple subspaces using the self-attention mechanism, and construct a multi-head attention mechanism; Retrieve the target prediction task, introduce the loss function according to the target prediction task to perform classification regression, and construct a fully connected layer; The self-attention layer, the multi-head attention mechanism, and the fully connected layer are evaluated by cross-validation, and the deep learning model is constructed according to the evaluation results; Among them, the output end of the self-attention layer is communicatively connected to the input end of the multi-head attention mechanism, the input end of the fully connected layer is communicatively connected to the output end of the multi-head attention mechanism and the output end of the input layer, and the output end of the fully connected layer is the output end of the deep learning model.

4. The biomarker-based trauma-related acute kidney injury target prediction device according to claim 3, characterized in that: The device for synchronizing the biomarker data set to the deep learning model according to the feature relationship matrix for prediction to obtain multiple target prediction information includes: Synchronizing the biomarker dataset as first vector input data to the input layer, and synchronizing the feature relationship matrix as second vector input data to the input layer; Using the self-attention layer to perform weighted calculation on the second vector input data to obtain a plurality of weight coefficients; The first vector input data and the second vector input data are multi-dimensionally captured according to the multiple weight coefficients by the multi-head attention mechanism to generate multi-spatial attention parameters; Performing confidence scoring according to the multiple spatial attention parameters to generate multiple confidence scores; activating a fully connected layer according to the multiple confidence scores, performing classification regression prediction on the first vector input data and the second vector input data according to the fully connected layer, and generating a classification regression prediction result; The multiple confidence scores are matched with the classification regression prediction results to generate the multiple target prediction information.

5. The biomarker-based target prediction device for trauma-related acute kidney injury according to claim 1, characterized in that: The device uses a drug database combined with the multiple confidence scores to match and screen the multiple target prediction information to determine the parameters of drug candidate molecules, and includes: Arranging the multiple target prediction information in descending order according to the multiple confidence scores to generate a target prediction sequence; Calculate the mean according to the target prediction sequence and set the confidence threshold; According to the comparison of the multiple confidence scores with the confidence threshold, the multiple target prediction information are initially screened, the target prediction information corresponding to the confidence score greater than or equal to the confidence threshold is stored, and an initial screening result is generated; Using the initial screening results as index information, performing matching search on the drug database to generate drug interaction relationships; The drug database is traversed and screened based on the drug interaction relationship to obtain the drug candidate molecule parameters.

6. The biomarker-based target prediction device for trauma-related acute kidney injury according to claim 5, characterized in that: The device verifies the plurality of target prediction information according to the drug candidate molecule parameters and generates a verification result, comprising: Classifying the drug candidate molecules according to the drug interaction relationship to generate a plurality of data to be tested, wherein the plurality of data to be tested includes in vitro data to be tested and in vivo data to be tested; Based on the in vitro experimental data and the drug candidate molecule, an in vitro experimental analysis is performed to obtain in vitro experimental effect parameters; Based on the in vivo experimental data and the candidate drug molecule, an in vivo experimental analysis is performed to obtain in vivo experimental effect parameters; Performing in vitro efficacy evaluation on the plurality of target prediction information based on the in vitro experimental effect parameters to generate an in vitro efficacy score; Performing in vivo efficacy evaluation on the plurality of target prediction information based on the in vivo experimental effect parameters to generate an in vivo efficacy score; The safety of the plurality of target prediction information is verified according to the in vitro efficacy score combined with the in vivo efficacy score, a target safety factor is generated, and the target safety factor is added to the verification result.

7. The biomarker-based trauma-related acute kidney injury target prediction device according to claim 6, characterized in that: The multiple target prediction information is evaluated for treatment according to the verification result to obtain a treatment effect score, and the device includes: Setting a quality assessment index based on the target safety factor; Traversing the plurality of target prediction information, performing treatment scoring according to the quality assessment index, and generating an in vivo treatment quality score and an in vitro treatment quality score; Performing in vivo efficacy analysis according to the in vivo treatment quality score combined with the in vivo efficacy score to generate an in vivo treatment efficacy score; Performing in vitro efficacy analysis according to the in vitro treatment quality score combined with the in vitro efficacy score to generate an in vitro treatment efficacy score; The in vivo treatment effect score, the in vitro treatment quality score and the multiple confidence scores of the multiple target prediction information are calculated respectively to generate the treatment effect score.

8. The biomarker-based target prediction device for trauma-related acute kidney injury according to claim 7, characterized in that: The treatment effect score is associated with the plurality of target prediction information to generate a target prediction report, the device comprising: Performing a mapping analysis based on the in vivo treatment effect score and the plurality of target prediction information to generate a first mapping relationship parameter; Performing a mapping analysis based on the in vitro treatment effect score and the plurality of target prediction information to generate a second mapping relationship parameter; constructing a mapping relationship network according to the first mapping relationship parameter in combination with the second mapping relationship parameter and the multiple confidence scores; Based on the treatment effect score, transfer learning is performed on the plurality of target prediction information according to the mapping relationship network to generate a target learning result; The target learning result is fed back to the drug database for response, a treatment suggestion is generated, and the treatment suggestion is added to the target prediction report.

Images