Rapid detection and clinical diagnosis support system for autoimmune disease antibody

Abstract

The invention provides an autoimmune disease antibody rapid detection and clinical diagnosis support system. The system comprises a rapid detection module, a data processing module, an intelligent diagnosis module and a remote management module, the rapid detection module is used for completing rapid detection of autoimmune diseases of a detected person and outputting detection data; the data processing module is used for performing standardization processing on the detection data to generate standardized detection data; the intelligent diagnosis module is used for providing diagnosis support based on standardized detection data in combination with artificial intelligence; the remote management module is used for managing and storing standardized detection data and optimizing subsequent diagnosis support; according to the invention, through rapid detection, standardized data processing, personalized intelligent diagnosis and remote management optimization, efficient, accurate and adaptive optimization autoimmune disease antibody detection and diagnosis support is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of medical detection and intelligent diagnosis systems, and particularly to a rapid detection and clinical diagnosis support system for autoimmune disease antibodies. Background Art

[0002] Autoimmune diseases are a class of chronic diseases caused by the abnormal attack of the immune system on its own tissues, including systemic lupus erythematosus, rheumatoid arthritis, Sjogren's syndrome, etc. Their etiology is complex, early symptoms are atypical, and there are large individual differences among patients, which brings great challenges to clinical diagnosis and precise treatment.

[0003] Currently, the laboratory detection of autoimmune diseases mainly relies on the detection of autoantibodies. Common methods include enzyme-linked immunosorbent assay, immunoblotting, and fluorescence immunoassay, etc. However, existing detection methods usually take a long time, the equipment is complex, and the requirements for the laboratory environment are relatively high, making it difficult to meet the needs of rapid screening and primary medical applications.

[0004] On the other hand, during the clinical diagnosis process, doctors usually rely on the patient's symptoms, laboratory test results, and medical experience for comprehensive judgment. However, due to individual differences, the traditional diagnostic method based on fixed reference standards may lead to misdiagnosis or missed diagnosis, especially in patients with early-stage diseases or atypical symptoms. Therefore, how to combine rapid detection technology and artificial intelligence to improve the accuracy of detection and the level of personalized diagnosis has become a key issue in the current field of medical detection and diagnosis. Summary of the Invention

[0005] The purpose of the present invention is to propose a rapid detection and clinical diagnosis support system for autoimmune disease antibodies in view of the current deficiencies.

[0006] The present invention adopts the following technical solutions: A rapid detection and clinical diagnosis support system for autoimmune disease antibodies, the system includes a rapid detection module, a data processing module, an intelligent diagnosis module, and a remote management module; the rapid detection module is used to complete the rapid detection of autoimmune diseases of the detector and output detection data; the data processing module is used to perform standardized processing on the detection data to generate standardized detection data; the intelligent diagnosis module is used to provide diagnosis support based on the standardized detection data combined with artificial intelligence; the remote management module is used to manage and store the standardized detection data and optimize subsequent diagnosis support.

[0007] The rapid detection module includes a sample processing unit, a fluorescence detection unit, and a data output unit; the sample processing unit is used to obtain a physiological sample of the detector, and the physiological sample includes blood, serum or urine; and introduce the physiological sample into an immunochromatographic test strip to complete the antigen-antibody reaction; the fluorescence detection unit is used to excite the fluorescent label on the immunochromatographic test strip and detect the fluorescence signal intensity on the immunochromatographic test strip through an optical sensor; the data output unit is used to receive and output the fluorescence signal intensity of the fluorescence detection unit as detection data.

[0008] Furthermore, the data processing module includes a data correction unit, a result determination unit, a concentration analysis unit, and a standardization output unit; the data correction unit is used to complete the background noise correction of the detection data; the result determination unit is used to combine the detection data to output a negative or positive determination of the detection result of the detector's physiological sample, and determine whether the detector's physiological sample contains autoimmunity disease-related antibodies; the concentration analysis unit is used to compare the fluorescence signal intensity with the standard fluorescence curve and calculate the concentration of autoimmunity disease-related antibodies in the detector's physiological sample; the standardization output unit is used to receive the determination result of the result determination unit and the concentration data of the concentration analysis unit and integrate them to generate standardized detection data.

[0009] Furthermore, the intelligent diagnosis module includes a multi-model unit, an information input unit, a model selection unit, and a result output unit; the multi-model unit includes multiple diagnosis models set based on different individual characteristic labels; the information input unit is used to input the individual characteristic labels of the detector; the model selection unit is used to match the individual characteristic labels of the detector with the individual characteristic labels corresponding to each diagnosis model in the multi-model unit and select the most suitable diagnosis model for the detector; the result output unit is used to input the standardized detection data of the detector into the selected diagnosis model to complete the diagnosis support for the detector.

[0010] Furthermore, in the multi-model unit, the establishment method of each diagnosis model is as follows: S11: Set multiple different individual characteristic labels according to clinical medical experience and patient population data analysis, and each individual characteristic label represents a specific patient group; S12: For each individual characteristic label, establish a corresponding diagnosis model framework, the diagnosis model framework includes input features, target outputs, and model architectures, and set initial model parameters; S13: Obtain historical diagnosis data from a medical database, and each sample data in the historical diagnosis data includes individual characteristic label data, standardized detection data, and diagnosis support results, and divide the historical diagnosis data into a training set and a validation set; S14: Use the training set to train each diagnostic model respectively to optimize the model parameters; during the training process, the loss function of each training model satisfies: ; Where, is the total loss function of a certain training model, is the total number of samples in the training set, is the classification weight factor of the th sample; is the cross-entropy loss of the th sample, is the th sample's true class label, obtained from the diagnostic support results in the training set; is the predicted class probability distribution of the th sample; For satisfies: ; Where, is the similarity control parameter, used to control the influence of the similarity degree of individual feature labels on the classification weight factor, set through preliminary experiments; is the similarity measurement function between the individual feature label of the th sample and the individual feature label corresponding to the current diagnostic model, satisfying: ; Where, is the individual feature label vector of the th sample, is the individual feature label vector corresponding to the current diagnostic model; S15: Use the validation set to evaluate the performance of each diagnostic model and complete the deployment.

[0011] Furthermore, the model selection unit calculates the similarity measurement function between the individual feature label of the detector and the individual feature label corresponding to each diagnostic model, and selects the diagnostic model corresponding to the highest similarity measurement function value as the most suitable diagnostic model for the detector.

[0012] Further, the remote management module includes a data storage management unit, a data access unit, and a model optimization unit; the data storage management unit is used to manage and store standardized detection data and diagnostic support results, and classify the standardized detection data and diagnostic support results according to different individual characteristic tags set in advance; the data access unit is used to support doctors and testers to remotely access the standardized detection data and diagnostic support results of the tester to assist in clinical decision-making; the model optimization unit is used to dynamically optimize the diagnostic model in the intelligent diagnosis module to improve the diagnostic accuracy and adaptability of the diagnostic model.

[0013] Beneficial effects achieved by the present invention: By integrating four major modules of rapid detection, data processing, intelligent diagnosis, and remote management, the present invention realizes efficient, accurate, personalized, and sustainable optimization of autoimmune disease antibody detection and diagnostic support; through a multi-model matching mechanism based on individual characteristic tags, testers can match the diagnostic model most suitable for their characteristics, ensuring high accuracy of diagnostic results and improving the level of personalized medicine; and by dynamically optimizing the diagnostic model, it is ensured that the diagnostic model can maintain the best performance for a long time, improving the stability and generalization ability of the overall diagnostic support system. Description of the Drawings

[0014] The present invention can be further understood from the following description in conjunction with the drawings. The components in the drawings are not necessarily drawn to scale, but the emphasis is placed on showing the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0015] Figure 1 It is a schematic diagram of the overall module of the present invention.

[0016] Figure 2 It is a schematic diagram of the establishment process of each diagnostic model in the multi-model unit of the present invention.

[0017] Figure 3 It is a schematic diagram of the working process of the model optimization unit of the present invention. Detailed Embodiments

[0018] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the following further details the present invention in conjunction with its embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention; for those skilled in the art, after referring to the following detailed description, other systems, methods, and / or features of this embodiment will become obvious; it is intended that all such additional systems, methods, features, and advantages are included in this specification; included within the scope of the present invention and protected by the appended claims; additional features of the disclosed embodiments are described in the following detailed description, and these features will be obvious according to the following detailed description.

[0019] In the drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", etc. indicating the orientation or positional relationship, it is based on the orientation or positional relationship shown in the drawings. This is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and cannot be construed as a limitation of this patent. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances. Embodiment

[0020] As Figure 1 As shown, this embodiment provides a rapid detection and clinical diagnosis support system for autoimmune disease antibodies. The system includes a rapid detection module, a data processing module, an intelligent diagnosis module, and a remote management module; the rapid detection module is used to complete the rapid detection of autoimmune diseases for the detector and output detection data; the data processing module is used to perform standardized processing on the detection data to generate standardized detection data; the intelligent diagnosis module is used to provide diagnosis support based on the standardized detection data combined with artificial intelligence; the remote management module is used to manage and store the standardized detection data and optimize the subsequent diagnosis support; The rapid detection module includes a sample processing unit, a fluorescence detection unit, and a data output unit; the sample processing unit is used to obtain the physiological sample of the detector, and the physiological sample includes blood, serum or urine; and introduce the physiological sample into the immunochromatographic test strip to complete the antigen-antibody reaction; the fluorescence detection unit is used to excite the fluorescent label on the immunochromatographic test strip and detect the fluorescence signal intensity on the immunochromatographic test strip through an optical sensor; the data output unit is used to receive and output the fluorescence signal intensity of the fluorescence detection unit as detection data; The data processing module includes a data correction unit, a result determination unit, a concentration analysis unit, and a standardized output unit; the data correction unit is used to complete the background noise correction of the detection data; the result determination unit is used to output a negative or positive determination of the detection result of the detector's physiological sample in combination with the detection data, and judge whether the detector's physiological sample contains antibodies related to autoimmune diseases; the concentration analysis unit is used to compare the fluorescence signal intensity with the standard fluorescence curve and calculate the concentration of antibodies related to autoimmune diseases in the detector's physiological sample; the standardized output unit is used to receive the determination result of the result determination unit and the concentration data of the concentration analysis unit and integrate them to generate standardized detection data; Further, the intelligent diagnosis module includes a multi-model unit, an information input unit, a model selection unit, and a result output unit; the multi-model unit includes multiple diagnosis models set based on different individual characteristic labels; the information input unit is used to input the individual characteristic labels of the detector; the model selection unit is used to match the individual characteristic labels of the detector with the individual characteristic labels corresponding to each diagnosis model in the multi-model unit, and select the most suitable diagnosis model for the detector; the result output unit is used to input the standardized detection data of the detector into the selected diagnosis model to complete the diagnosis support for the detector; Specifically, the content in the individual characteristic labels includes but is not limited to age, gender, medical history, and genetic background information; Further, as Figure 2 shown, in the multi-model unit, the establishment methods of each diagnosis model are as follows: S11: Set multiple different individual characteristic labels according to clinical medical experience and patient population data analysis, and each individual characteristic label represents a specific patient group; S12: For each individual characteristic label, establish a corresponding diagnosis model framework, the diagnosis model framework includes input features, target outputs, and model architectures, and set initial model parameters; S13: Obtain historical diagnosis data from the medical database, each sample data in the historical diagnosis data includes individual characteristic label data, standardized detection data, and diagnosis support results, and divide the historical diagnosis data into a training set and a validation set; S14: Use the training set to train each diagnosis model to optimize the model parameters; during the training process, the loss function of each training model satisfies: ; Among them, is the total loss function of a certain training model, is the total number of samples in the training set, is the th sample's classification weight factor; is the th sample's cross-entropy loss, that is, the difference between the sample's true class label and the probability distribution predicted by the model; is the th sample's true class label, obtained from the diagnosis support results in the training set; is the th sample's predicted class probability distribution, that is, the predicted output of the diagnosis model; For it satisfies: ; Among them, is a similarity control parameter, used to control the influence of the similarity degree of individual feature labels on the classification weight factor, and is set through pre-experiments; is the similarity measurement function between the individual feature label of the th sample and the individual feature label corresponding to the current diagnostic model, and satisfies: where is the individual feature label vector of the th sample, and is the individual feature label vector corresponding to the current diagnostic model; S15: Use the validation set to evaluate the performance of each diagnostic model and complete the deployment; Specifically, for an intuitive understanding of the construction and role of the diagnostic model, taking the diagnostic model corresponding to an individual feature label where the disease type is rheumatoid arthritis as an example, in the step S11, the specific examples of the individual feature labels corresponding to this diagnostic model are as follows: Age: 40 - 50 years old; Gender: Female; Medical history: Continuously ill for 6 - 12 months; Genetic background: HLA - DR4 genotype positive; In the step S12, the input features of this diagnostic model framework are specifically the content in the standardized test data, including but not limited to the RF concentration and anti - CCP concentration indicators; the target output is specifically the content in the diagnostic support result, including: Diagnostic classification: Determine whether the tester has rheumatoid arthritis; Disease severity assessment result; Evaluate the disease severity of the patient with rheumatoid arthritis and quantitatively output it as a disease activity parameter to guide clinical treatment decisions; import numpy as np import torch import torch.nn as nn import torch.optim as optim from sklearn.model_selection import train_test_split from scipy.spatial.distance import cosine # --- S11: Set multiple individual feature labels --- def define_patient_groups(): """ Define individual characteristic labels based on clinical experience and patient group data.

[0021] """ patient_groups = { "Group_1": np.array([0.2, 0.7, 0.1]), # Example: Feature vector such as age, gender, medical history, etc. "Group_2": np.array([0.8, 0.1, 0.6]), "Group_3": np.array([0.3, 0.5, 0.9]) } return patient_groups # --- S12: Diagnostic model framework --- class DiagnosisModel(nn.Module): """ The structure of the diagnostic model, which can adjust input features, hidden layers, etc. as needed.

[0022] """ def __init__(self, input_size, output_size): super(DiagnosisModel, self).__init__() self.fc1 = nn.Linear(input_size, 32) self.fc2 = nn.Linear(32, output_size) self.relu = nn.ReLU() self.softmax = nn.Softmax(dim=1) def forward(self, x): x = self.relu(self.fc1(x)) x = self.softmax(self.fc2(x)) return x def build_model_framework(patient_groups): """ Create a corresponding diagnostic model for each individual feature label.

[0023] """ models = {} input_size = 3 # Assume feature dimension output_size = 2 # Binary classification (0 = negative, 1 = positive) for group in patient_groups: models[group] = DiagnosisModel(input_size, output_size) return models # --- S13: Load medical data --- def load_medical_data(): """ Load historical diagnostic data from the database and divide it into training and validation sets.

[0024] """ num_samples = 1000 feature_size = 3 # Generate simulated data X = np.random.rand(num_samples, feature_size) # Standardized test data y = np.random.randint(0, 2, num_samples) # True diagnosis class labels patient_tags = np.random.rand(num_samples, feature_size) # Individual feature labels X_train, X_val, y_train, y_val, tags_train, tags_val = train_test_split(X, y, patient_tags, test_size = 0.2) return X_train, X_val, y_train, y_val, tags_train, tags_val # --- S14: Calculate the similarity of individual features --- def compute_similarity(patient_vector, model_vector): """ Calculate the similarity (cosine similarity) between the individual feature labels and the model feature labels.

[0025] """ return 1 - cosine(patient_vector, model_vector) # Cosine similarity, the closer the value is to 1, the more similar # --- S14: Train the model --- def train_model(model, X_train, y_train, tags_train, group_vector, lr=0.01, epochs=100): """ Train the diagnostic model and adjust the sample weights according to the individual feature similarity.

[0026] """ optimizer = optim.Adam(model.parameters(), lr=lr) criterion = nn.CrossEntropyLoss() X_train_tensor = torch.tensor(X_train, dtype=torch.float32) y_train_tensor = torch.tensor(y_train, dtype=torch.long) for epoch in range(epochs): optimizer.zero_grad() # Calculate the similarity α_i similarity_weights = np.array([np.exp(1.0 * compute_similarity(t, group_vector)) for t in tags_train]) similarity_weights = similarity_weights / np.sum(similarity_weights) # Forward propagation outputs = model(X_train_tensor) # Calculate weighted loss loss = sum(similarity_weights[i] * criterion(outputs[i].unsqueeze(0), y_train_tensor[i].unsqueeze(0)) for i in range(len(X_train))) # Backward propagation loss.backward() optimizer.step() return model # --- S15: Evaluate the model --- def evaluate_model(model, X_val, y_val): """ Evaluate the model accuracy """ X_val_tensor = torch.tensor(X_val, dtype=torch.float32) y_val_tensor = torch.tensor(y_val, dtype=torch.long) with torch.no_grad(): outputs = model(X_val_tensor) predictions = torch.argmax(outputs, dim=1) accuracy = (predictions == y_val_tensor).sum().item() / len(y_val) return accuracy # --- Run the entire process --- def main(): # 1. Set individual feature labels patient_groups = define_patient_groups() # 2. Build a multi-model framework models = build_model_framework(patient_groups) # 3. Load medical data X_train, X_val, y_train, y_val, tags_train, tags_val = load_medical_data() # 4. Train the model for group, model in models.items(): print(f"Training model for {group}...") models[group] = train_model(model, X_train, y_train, tags_train, patient_groups[group]) # 5. Evaluate the model for group, model in models.items(): acc = evaluate_model(model, X_val, y_val) print(f"Model for {group} - Accuracy: {acc:.4f}") if __name__ == "__main__": main().

[0027] Furthermore, the model selection unit calculates the similarity metric function between the individual feature tags of the detector and the individual feature tags corresponding to each diagnostic model, and selects the diagnostic model corresponding to the highest similarity metric function value as the diagnostic model most suitable for the detector; the similarity metric function is as follows: ; wherein, is the similarity metric function between the detector and the th diagnostic model, is the individual feature tag vector corresponding to the th diagnostic model, satisfying ; The individual characteristic label vector of the detector; In this solution, multiple individual characteristic labels are preset in advance, and a dedicated diagnostic model is trained for each individual characteristic label, so as to ensure that patients in different characteristic groups can be matched with the most suitable diagnostic method, improving the pertinence and adaptability; by calculating the similarity between the individual characteristic label of the detector and the individual characteristic labels of each diagnostic model, the most suitable diagnostic model for the detector is selected, so as to ensure that the corresponding diagnostic model can provide accurate diagnostic support results for the detector, improving the accuracy and reliability of personalized diagnosis. Embodiment

[0028] This embodiment should be understood as including at least all the features of any one of the foregoing embodiments and being further improved on this basis; This embodiment provides a rapid detection and clinical diagnosis support system for autoimmune disease antibodies. The system includes a rapid detection module, a data processing module, an intelligent diagnosis module, and a remote management module; the rapid detection module is used to complete the rapid detection of autoimmune diseases of the detector and output detection data; the data processing module is used to perform standardized processing on the detection data to generate standardized detection data; the intelligent diagnosis module is used to provide diagnostic support based on the standardized detection data combined with artificial intelligence; the remote management module is used to manage and store the standardized detection data and optimize subsequent diagnostic support. Furthermore, the remote management module includes a data storage management unit, a data access unit, and a model optimization unit; the data storage management unit is used to manage and store the standardized detection data and diagnostic support results, and classify the standardized detection data and diagnostic support results according to different preset individual characteristic labels; the data access unit is used to support doctors and detectors to remotely access the standardized detection data and diagnostic support results of the detector to assist clinical decision-making; the model optimization unit is used to dynamically optimize the diagnostic model in the intelligent diagnosis module to improve the diagnostic accuracy and adaptability of the diagnostic model. Furthermore, the data storage management unit calculates the similarity metric function between each sample data and each preset individual characteristic label, and classifies according to the set similarity threshold. The sample data includes standardized detection data, diagnostic support results, and the individual characteristic label of the sample; if the similarity metric function value between a certain sample data and a certain individual characteristic label is higher than the preset similarity threshold, it is classified into the corresponding individual characteristic label category; if the similarity metric function values between a certain sample data and all individual characteristic labels are lower than the similarity threshold, the nearest neighbor classification method is used for allocation. Furthermore, as Figure 3 shown, the model optimization unit completes the dynamic optimization of the diagnostic model through the following methods: S21: Set consecutive evaluation periods, and evaluate the accuracy of the diagnostic support results output by each diagnostic model within each evaluation period; S22: If the accuracy of a certain diagnostic model is lower than the set accuracy threshold, trigger model update, retrain the diagnostic model using newly collected sample data, and readjust the composition ratio of the training data to optimize the adaptability of the diagnostic model; S23: During the training process, mix two types of data as the training data for the diagnostic model. The first type of data is the sample data belonging to the individual feature label of this diagnostic model, and the second type of data is all sample data that does not belong to the individual feature label of this diagnostic model; S24: Determine the composition ratio of the first type of data and the second type of data in the training data through the following method: ; ; where, is the composition ratio of the first type of data, is the composition ratio of the second type of data; is the accuracy of the current diagnostic model in the current evaluation period, is the preset accuracy threshold, is the mixing ratio control coefficient, which is used to control the change of the data mixing ratio and is set through preliminary experiments; S25: After obtaining the composition ratio of the two types of data, in the newly collected sample dataset, sample in the newly collected sample dataset at the composition ratio of the two types of data as the sampling frequencies of the two types of data respectively until the total number of sample data obtained from the two types of data reaches the requirements of the preset training dataset; S26: Use the training dataset obtained in step S25 as the training data for the diagnostic model, and execute and complete the training optimization of the diagnostic model; This solution intelligently classifies and manages the stored data through the similarity calculation based on the individual feature label, thereby improving the data retrieval efficiency and providing a data basis for subsequent model training; by dynamically evaluating the accuracy of the diagnostic model and adjusting the composition ratio of the training data, it ensures that the model maintains the generalization ability while adapting to a specific individual feature group, thereby improving the accuracy and long-term stability of the diagnostic model.

[0029] The content disclosed above is only the preferred feasible embodiment of the present invention, and does not limit the protection scope of the present invention. Therefore, all equivalent technical changes made by using the content of the specification and drawings of the present invention are included in the protection scope of the present invention. In addition, the elements therein can be updated with the development of technology.

Claims

1. A rapid detection and clinical diagnosis support system for autoimmune disease antibodies, characterized in that, The system includes a rapid detection module, a data processing module, an intelligent diagnosis module, and a remote management module; the rapid detection module is used to rapidly detect the autoimmune diseases of the detector and output detection data; the data processing module is used to perform standardized processing on the detection data to generate standardized detection data; the intelligent diagnosis module is used to provide diagnostic support based on the standardized detection data combined with artificial intelligence; the remote management module is used to manage and store the standardized detection data and optimize subsequent diagnostic support; The rapid detection module includes a sample processing unit, a fluorescence detection unit, and a data output unit; the sample processing unit is used to obtain the physiological samples of the detector, and the physiological samples include blood, serum or urine; and introduce the physiological samples into the immunochromatographic test strip to complete the antigen-antibody reaction; the fluorescence detection unit is used to excite the fluorescent label on the immunochromatographic test strip and detect the fluorescence signal intensity on the immunochromatographic test strip through an optical sensor; the data output unit is used to receive and output the fluorescence signal intensity of the fluorescence detection unit as detection data.

2. The rapid detection and clinical diagnosis support system for autoimmune disease antibodies according to claim 1, wherein The data processing module includes a data correction unit, a result determination unit, a concentration analysis unit, and a standardized output unit; the data correction unit is used to complete the background noise correction of the detection data; The result determination unit is used to output the negative or positive determination of the detection result of the detector's physiological sample in combination with the detection data, and judge whether the detector's physiological sample contains antibodies related to autoimmune diseases; the concentration analysis unit is used to compare the fluorescence signal intensity with the standard fluorescence curve and calculate the concentration of antibodies related to autoimmune diseases in the detector's physiological sample; the standardized output unit is used to receive the determination result of the result determination unit and the concentration data of the concentration analysis unit and integrate them to generate standardized detection data.

3. The rapid detection and clinical diagnosis support system for autoimmune disease antibodies according to claim 1, wherein The intelligent diagnosis module includes a multi-model unit, an information input unit, a model selection unit, and a result output unit; the multi-model unit includes multiple diagnosis models set based on different individual characteristic tags; the information input unit is used to input the individual characteristic tags of the detector; the model selection unit is used to match the individual characteristic tags of the detector with the individual characteristic tags corresponding to each diagnosis model in the multi-model unit and select the most suitable diagnosis model for the detector; the result output unit is used to input the standardized detection data of the detector into the selected diagnosis model to complete the diagnostic support for the detector.

4. The rapid detection and clinical diagnosis support system for autoimmune disease antibodies according to claim 3, characterized in that In the multi-model unit, the establishment method of each diagnosis model is as follows: S11: Set multiple different individual characteristic tags according to clinical medical experience and patient population data analysis, and each individual characteristic tag represents a specific patient group; S12: For each individual characteristic tag, establish a corresponding diagnosis model framework, and the diagnosis model framework includes input features, target outputs, and model architectures, and set initial model parameters; S13: Obtain historical diagnosis data from the medical database, and each sample data in the historical diagnosis data includes individual characteristic tag data, standardized detection data, and diagnostic support results, and divide the historical diagnosis data into a training set and a validation set; S14: Use the training set to train each diagnostic model respectively to optimize the model parameters; during the training process, the loss functions of each training model satisfy: ； Among them, is the total loss function of a certain training model, is the total number of samples in the training set, is the classification weight factor of the th sample; is the cross-entropy loss of the th sample, is the true class label of the th sample, obtained from the diagnostic support results in the training set; is the predicted class probability distribution of the For Satisfying: ； Among them, is a similarity control parameter, used to control the influence of the similarity degree of individual feature labels on the classification weight factor, and is set through preliminary experiments; is the similarity metric function between the individual feature label of the th sample and the individual feature label corresponding to the current diagnostic model, and satisfies: ； Among them, is the individual feature label vector of the th sample, and is the individual feature label vector corresponding to the current diagnostic model; S15: Use the validation set to evaluate the performance of each diagnostic model and complete the deployment.

5. The rapid detection and clinical diagnosis support system for autoimmune disease antibodies according to claim 3, characterized in that, The model selection unit calculates the similarity metric function between the individual feature labels of the detector and the individual feature labels corresponding to each diagnostic model, and selects the diagnostic model corresponding to the highest similarity metric function value as the most suitable diagnostic model for the detector.

6. The rapid detection and clinical diagnosis support system for autoimmune disease antibodies according to claim 1, wherein The remote management module includes a data storage management unit, a data access unit, and a model optimization unit; the data storage management unit is used to manage and store the standardized detection data and diagnostic support results, and classify the standardized detection data and diagnostic support results according to different individual feature labels set in advance; the data access unit is used to support doctors and detectors to remotely access the standardized detection data and diagnostic support results of the detector to assist clinical decision-making; the model optimization unit is used to dynamically optimize the diagnostic models in the intelligent diagnosis module to improve the diagnostic accuracy and adaptability of the diagnostic models.

Images