Medical examination sample automatic classification and identification system and method

Abstract

The invention discloses a medical examination sample automatic classification and identification system and method, and the system comprises a data collection module which is used for collecting the related data of a medical examination sample; the data preprocessing module is connected with the data acquisition module; the feature extraction module is connected with the data preprocessing module and is used for performing feature extraction on related data of different types of medical examination samples; the feature selection module is connected with the feature extraction module and is used for screening the extracted features by adopting a method based on correlation analysis and information gain; the classification and recognition model construction module is connected with the feature selection module and selects a deep neural network to construct a classification and recognition model; and the result output module is connected with the classification and identification model construction module, outputs the classification result and the identification information of the sample, and displays the classification result and the identification information in a visual mode. The method is more comprehensive and accurate, the medical examination sample classification and identification efficiency and accuracy can be improved, and the subjectivity and errors of manual operation are reduced.

Description

Technical Field

[0001] The present invention relates to the field of medical testing technology, and in particular to a system and method for automatically classifying and identifying medical test samples. Background Art

[0002] In modern medicine, medical testing plays a key role as an important part of disease diagnosis, treatment monitoring, and prognosis assessment. With the continuous development of medical technology and the rapid growth of medical data, the number and types of medical test samples are increasing, making the accurate and efficient classification and identification of samples increasingly important. Traditionally, the classification and identification of medical test samples relies primarily on manual operations, with testers analyzing and judging samples based on their own professional knowledge and experience. This method not only consumes a great deal of time and effort, but is also easily influenced by subjective factors, making it difficult to ensure the accuracy and consistency of classification and identification results. Furthermore, medical test sample data is diverse and complex, including various types of data, including physical and chemical indicators, image information, and basic patient information.

[0003] Currently, while some advanced medical testing equipment and technologies can provide some sample information, these devices and technologies are often independent and lack effective data integration and analysis capabilities. Furthermore, existing sample classification and identification methods are mostly based on a single data type or feature, failing to fully utilize the sample's multi-source data information, limiting the accuracy and reliability of classification and identification. Furthermore, despite the rapid development of artificial intelligence technology, while some application attempts have been made in the medical field, its application in medical test sample classification and identification is still immature and lacks a systematic and responsive solution. Summary of the Invention

[0004] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides an automatic classification and identification system for medical test samples to solve the problems that traditional medical test sample classification and identification relies on manual labor, is susceptible to subjective influence, has low efficiency, and is difficult to ensure accuracy and consistency.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a system for automatically classifying and identifying medical test samples, comprising: Data collection module, used to collect relevant data of medical test samples from multiple channels; A data preprocessing module, connected to the data acquisition module, preprocesses the relevant data of the medical test samples; A feature extraction module, connected to the data preprocessing module, extracts features from relevant data of different types of medical test samples; A feature selection module, connected to the feature extraction module, screens the extracted features using a method based on correlation analysis and information gain; A classification and recognition model construction module is connected to the feature selection module, selects a deep neural network to construct a classification and recognition model, and inputs the selected features into the model for training and testing; The result output module is connected to the classification and recognition model building module, outputs the classification results and recognition information of the samples, and displays them in a visual manner.

[0007] As a preferred solution of the automatic classification and identification system for medical test samples of the present invention, the data acquisition module is used to collect relevant data of medical test samples from multiple channels, including: Collect physical and chemical index data of samples from medical testing equipment, including routine blood parameters and biochemical indicators of blood samples, and routine urine parameters of urine samples; Obtain image information of samples from imaging devices, including microscope images of pathological sections, X-rays, and CT images; Extract the patient's basic information from the hospital information system, including age, gender, medical history, and allergy history.

[0008] As a preferred solution of the automatic classification and identification system for medical test samples of the present invention, a data preprocessing module is connected to the data acquisition module to preprocess the relevant data of the medical test samples, including: Check for missing values in the physical and chemical index data. If the data is normally distributed, use the mean to fill the missing values; if the data distribution is relatively discrete, use the median to fill the missing values. Use statistical methods to identify and correct outliers, normalize the data, and map the data to the [0, 1] interval. Perform denoising on the image information, enhance the image contrast through histogram equalization, and then normalize the image to unify the image size and format; The basic information of the patient is formatted and coded, and the text information is converted into digital code. As a preferred solution of the automatic classification and identification system of medical test samples of the present invention, the feature extraction module is connected to the data preprocessing module to extract features of the relevant data of different types of medical test samples, including: Use convolutional neural networks to extract texture features, shape features, and color features of images; The principal component analysis method is used to convert high-dimensional physical and chemical index data into a few principal components, and key information that can represent the main characteristics of the data is extracted; For the encoded patient basic information, features related to sample classification are extracted.

[0009] As a preferred embodiment of the automatic classification and identification system for medical test samples of the present invention, a feature selection module is connected to the feature extraction module and uses a method based on correlation analysis and information gain to screen the extracted features, including: Calculate the Pearson correlation coefficient between each feature and the sample classification result to measure the linear correlation between the feature and the classification result; Calculate the information gain of each feature to evaluate the information contribution of the feature to the classification result; Set the correlation threshold and information gain threshold, and remove features whose absolute value of correlation coefficient is less than the correlation threshold and whose information gain is less than the information gain threshold; The remaining features are ranked by importance according to the absolute value of the feature correlation coefficient and the size of the information gain, and the features with higher importance are retained.

[0010] As a preferred embodiment of the automatic classification and recognition system for medical test samples of the present invention, the classification and recognition model construction module is connected to the feature selection module, selects a deep neural network to construct a classification and recognition model, and inputs the selected features into the model for training and testing, including: Build a deep neural network model, determine the number of layers, the number of neurons in each layer, and the activation function, use the filtered features as the input of the input layer, and use the actual classification results of the samples as supervision information; The stochastic gradient descent algorithm is used to update the weight and bias parameters of the model. The cross entropy loss function is used as the loss function to measure the difference between the model prediction results and the actual classification results. The dataset is divided into training set and test set in a ratio of 7:3. The model is iteratively trained multiple times on the training set, and the model parameters are continuously adjusted to gradually reduce the loss function value. After the training is completed, the model is tested on the test set to evaluate the model's classification accuracy, recall rate, and F1 value indicators, and the model is optimized and adjusted based on the test results.

[0011] As a preferred embodiment of the automatic classification and identification system for medical test samples of the present invention, the result output module is connected to the classification and identification model building module, outputs the classification results and identification information of the sample, and displays them in a visual manner, including: After the medical test sample data to be classified and identified has gone through the data collection, preprocessing, feature extraction and selection process, it is input into the trained model to obtain the classification results and identification information of the samples. The basic information of the samples, classification results and identification confidence are displayed in a table format. At the same time, a bar chart or pie chart is generated to intuitively display the proportion of different classification results. For image samples, key features and classification basis are marked on the image to facilitate users' intuitive understanding; Generate a presentation report containing sample feature information and the contribution of each feature in the classification process. Users can further analyze and view the results.

[0012] In a second aspect, the present invention provides a method for an automatic classification and identification system for medical test samples, comprising: Collect relevant data of medical test samples from multiple channels; Preprocessing relevant data of medical test samples; Extract features from relevant data of different types of medical test samples; The extracted features are screened using methods based on correlation analysis and information gain; Use deep neural networks to build classification and recognition models, and input the filtered features into the model for training and testing; Output the classification results and identification information of the sample and display them in a visual way.

[0013] In a third aspect, the present invention provides a computing device, comprising: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the automatic classification and identification system for medical test samples are implemented.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the automatic classification and identification system for medical test samples.

[0015] Compared with the existing technology, the beneficial effects of the present invention are as follows: the present invention uses advanced convolutional neural networks and principal component analysis technology to deeply mine the key features of different types of data. Compared with traditional methods, it is more comprehensive and accurate, can improve the efficiency and accuracy of medical test sample classification and identification, and reduce the subjectivity and errors of manual operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them: Figure 1 Schematic diagram of a system for automatic classification and identification of medical test samples according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0020] The present invention is described in detail with reference to schematic diagrams. For ease of illustration, cross-sectional views of device structures may be partially enlarged and not to scale when describing embodiments of the present invention. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of the present invention. Furthermore, in actual production, the three-dimensional dimensions of length, width, and depth should be included.

[0021] Furthermore, in the description of the present invention, it should be noted that the terms "upper, lower, inner, and outer" and other references to orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the systems or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first, second, or third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0022] In this disclosure, unless otherwise specified or limited, the terms "mounted," "connected," and "connected" should be interpreted broadly. For example, they may refer to fixed, removable, or integral connections. They may also refer to mechanical, electrical, or direct connections, indirect connections through an intermediary, or internal communication between two components. Those skilled in the art will understand the specific meanings of these terms in this disclosure.

[0023] Example 1 Reference Figure 1 , as one embodiment of the present invention, provides a medical test sample automatic classification and identification system, comprising: Data collection module, used to collect relevant data of medical test samples from multiple channels; Preferably, the physical and chemical index data of the samples are collected from medical testing equipment, including routine blood parameters and biochemical indicators of blood samples, and routine urine parameters of urine samples; Preferably, image information of the sample is obtained from an imaging device, including microscope images of pathological sections, X-rays, and CT images; Preferably, the patient's basic information, including age, gender, medical history, and allergy history, is extracted from the hospital information system.

[0024] The data preprocessing module is connected to the data acquisition module to preprocess the relevant data of the medical test samples; Preferably, the missing values in the physical and chemical index data are checked. If the data is normally distributed, the mean is used to fill the missing values; if the data distribution is relatively discrete, the median is used to fill the missing values. Statistical methods are used to identify and correct outliers, and the data are normalized and mapped to the interval [0, 1]. Preferably, the image information is denoised, the contrast of the image is enhanced by a histogram equalization method, and then the image is normalized to unify the size and format of the image; the basic information of the patient is formatted and encoded, and the text information is converted into digital code.

[0025] The feature extraction module is connected to the data preprocessing module to extract features from relevant data of different types of medical test samples; Preferably, a convolutional neural network is used to extract the texture features, shape features, and color features of the image; a principal component analysis method is used to convert high-dimensional physical and chemical index data into a few principal components to extract key information that can represent the main features of the data; and for the encoded basic patient information, features related to sample classification are extracted.

[0026] The feature selection module is connected to the feature extraction module and uses a method based on correlation analysis and information gain to screen the extracted features; Preferably, the Pearson correlation coefficient between each feature and the sample classification result is calculated to measure the linear correlation between the feature and the classification result; the information gain of each feature is calculated to evaluate the information contribution of the feature to the classification result; Preferably, a correlation threshold and an information gain threshold are set to remove features whose absolute value of the correlation coefficient is less than the correlation threshold and whose information gain is less than the information gain threshold; the remaining features are ranked by importance according to the absolute value of the correlation coefficient and the size of the information gain of the features, and features with higher importance are retained.

[0027] It should be noted that in terms of feature extraction, specialized methods are employed for different data types, using convolutional neural networks to extract texture, shape, and color features from images. This allows for a deep dive into the information inherent in images, capturing image features more comprehensively and accurately than traditional methods, providing a rich and crucial basis for subsequent classification and recognition. Principal component analysis is employed to process high-dimensional physical and chemical index data, converting it into a small number of principal components. This effectively extracts key information representing the data's primary characteristics, reducing data dimensionality, minimizing data redundancy, and improving data processing efficiency. From the encoded patient information, features relevant to sample classification are extracted, enabling patient information to better serve the sample classification task. Regarding feature selection, methods based on correlation analysis and information gain are scientific and effective. By calculating the Pearson correlation coefficient to measure the linear correlation between features and classification results, and by calculating information gain to assess the contribution of features to the classification results, features closely related to the classification results can be accurately selected from the numerous extracted features. Setting correlation and information gain thresholds to remove features with low correlation and information gain prevents interference from irrelevant or weakly correlated features, reducing model complexity and improving training efficiency and generalization. The remaining features are ranked according to their importance based on the absolute value of the feature correlation coefficient and the size of the information gain, and important features are retained, so that subsequent classification and recognition models can focus on key features, further improving the accuracy and reliability of the model.

[0028] The classification and recognition model construction module is connected to the feature selection module, and a deep neural network is selected to build a classification and recognition model. The selected features are input into the model for training and testing; Preferably, a deep neural network model is constructed, the number of network layers, the number of neurons in each layer, and the activation function are determined, the filtered features are used as the input of the input layer, and the actual classification results of the samples are used as supervision information; the weights and bias parameters of the model are updated using the stochastic gradient descent algorithm, and the cross entropy loss function is selected as the loss function to measure the difference between the model prediction results and the actual classification results; Preferably, the data set is divided into a training set and a test set in a ratio of 7:3. The model is iteratively trained multiple times on the training set, and the model parameters are continuously adjusted to gradually reduce the loss function value. After the training is completed, the model is tested on the test set to evaluate the model's classification accuracy, recall rate, and F1 value indicators, and the model is optimized and adjusted according to the test results.

[0029] The result output module is connected to the classification and recognition model building module to output the classification results and recognition information of the samples and display them in a visual way.

[0030] Preferably, the medical test sample data to be classified and identified is input into the trained model after the data collection, preprocessing, feature extraction and selection process to obtain the classification results and identification information of the samples, and the basic information of the samples, classification results and identification confidence are displayed in a table form, and a bar chart or pie chart is generated to intuitively display the proportion of different classification results; Preferably, for image samples, key features and classification basis are marked on the image to facilitate user intuitive understanding; a display report is generated, which includes the feature information of the sample and the contribution of each feature in the classification process, and the user can further analyze and view the results.

[0031] It should be noted that the present invention collects multi-source sample data through multiple channels and performs targeted preprocessing, effectively integrating complex and diverse data and improving data quality and availability. The feature extraction link uses advanced convolutional neural networks and principal component analysis technologies to deeply explore the key features of different types of data, which is more comprehensive and accurate than traditional methods. Feature selection uses correlation analysis and information gain methods to accurately screen key features, reduce data dimensions, avoid overfitting, and improve model training efficiency. The classification and recognition model construction uses deep neural networks, combined with scientific training and optimization strategies, so that the model has high accuracy and strong generalization ability, and can accurately classify and identify samples. The result output module displays detailed results in a visual way, including classification results, recognition information, feature contributions, etc., which is convenient for users to intuitively understand and further analyze. Overall, the system significantly improves the efficiency and accuracy of medical test sample classification and identification, and reduces the subjectivity and errors of manual operations.

[0032] The method for automatically classifying and identifying medical test samples in this embodiment includes: Collect relevant data of medical test samples from multiple channels; Preprocessing relevant data of medical test samples; Extract features from relevant data of different types of medical test samples; The extracted features are screened using methods based on correlation analysis and information gain; Use deep neural networks to build classification and recognition models, and input the filtered features into the model for training and testing; Output the classification results and identification information of the sample and display them in a visual way.

[0033] This embodiment further provides a computing device applicable to a medical test sample automatic classification and identification system, including: Memory and processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the automatic classification and identification system for medical test samples proposed in the above embodiment.

[0034] This embodiment further provides a storage medium storing a computer program, which, when executed by a processor, implements the automatic classification and identification system for medical test samples as proposed in the above embodiment.

[0035] The storage medium proposed in this embodiment and the system for realizing automatic classification and identification of medical test samples proposed in the above embodiment belong to the same inventive concept. For technical details not described in detail in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0036] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0037] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A medical test sample automatic classification and identification system, characterized by: include: Data collection module, used to collect relevant data of medical test samples from multiple channels; A data preprocessing module, connected to the data acquisition module, preprocesses the relevant data of the medical test samples; A feature extraction module, connected to the data preprocessing module, extracts features from relevant data of different types of medical test samples; A feature selection module, connected to the feature extraction module, screens the extracted features using a method based on correlation analysis and information gain; A classification and recognition model construction module is connected to the feature selection module, selects a deep neural network to construct a classification and recognition model, and inputs the selected features into the model for training and testing; The result output module is connected to the classification and recognition model building module, outputs the classification results and recognition information of the samples, and displays them in a visual manner.

2. The automatic classification and identification system for medical test samples according to claim 1, characterized in that: The data collection module is used to collect relevant data of medical test samples from multiple channels, including: Collect physical and chemical index data of samples from medical testing equipment, including routine blood parameters and biochemical indicators of blood samples, and routine urine parameters of urine samples; Obtain image information of samples from imaging devices, including microscope images of pathological sections, X-rays, and CT images; Extract the patient's basic information from the hospital information system, including age, gender, medical history, and allergy history.

3. The automatic classification and identification system for medical test samples according to claim 1 or 2, characterized in that: The data preprocessing module is connected to the data acquisition module and preprocesses the relevant data of the medical test samples, including: Check for missing values in the physical and chemical index data. If the data is normally distributed, use the mean to fill the missing values; if the data distribution is relatively discrete, use the median to fill the missing values. Use statistical methods to identify and correct outliers, normalize the data, and map the data to the [0, 1] interval. Perform denoising on the image information, enhance the image contrast through histogram equalization, and then normalize the image to unify the image size and format; The patient's basic information is formatted and coded, and text information is converted into digital codes.

4. The automatic classification and identification system for medical test samples according to claim 3, characterized in that: The feature extraction module is connected to the data preprocessing module and performs feature extraction on the relevant data of different types of medical test samples, including: Use convolutional neural networks to extract texture features, shape features, and color features of images; The principal component analysis method is used to convert high-dimensional physical and chemical index data into a few principal components, and key information that can represent the main characteristics of the data is extracted; For the encoded patient basic information, features related to sample classification are extracted.

5. The automatic classification and identification system for medical test samples according to claim 4, characterized in that: The feature selection module is connected to the feature extraction module and uses a method based on correlation analysis and information gain to screen the extracted features, including: Calculate the Pearson correlation coefficient between each feature and the sample classification result to measure the linear correlation between the feature and the classification result; Calculate the information gain of each feature to evaluate the information contribution of the feature to the classification result; Set the correlation threshold and information gain threshold, and remove features whose absolute value of correlation coefficient is less than the correlation threshold and whose information gain is less than the information gain threshold; The remaining features are ranked by importance according to the absolute value of the feature correlation coefficient and the size of the information gain, and the features with higher importance are retained.

6. The automatic classification and identification system for medical test samples according to claim 5, characterized in that: The classification and recognition model construction module is connected to the feature selection module, selects a deep neural network to construct a classification and recognition model, and inputs the selected features into the model for training and testing, including: Build a deep neural network model, determine the number of layers, the number of neurons in each layer, and the activation function, use the filtered features as the input of the input layer, and use the actual classification results of the samples as supervision information; The stochastic gradient descent algorithm is used to update the weight and bias parameters of the model. The cross entropy loss function is used as the loss function to measure the difference between the model prediction results and the actual classification results. The dataset is divided into training set and test set in a ratio of 7:

3. The model is iteratively trained multiple times on the training set, and the model parameters are continuously adjusted to gradually reduce the loss function value. After the training is completed, the model is tested on the test set to evaluate the model's classification accuracy, recall rate, and F1 value indicators, and the model is optimized and adjusted based on the test results.

7. The automatic classification and identification system for medical test samples according to claim 6, characterized in that: The result output module is connected to the classification and recognition model building module to output the classification results and recognition information of the sample and display them in a visual way, including: After the medical test sample data to be classified and identified has gone through the data collection, preprocessing, feature extraction and selection process, it is input into the trained model to obtain the classification results and identification information of the samples. The basic information of the samples, classification results and identification confidence are displayed in a table format. At the same time, a bar chart or pie chart is generated to intuitively display the proportion of different classification results. For image samples, key features and classification basis are marked on the image to facilitate users' intuitive understanding; Generate a presentation report containing sample feature information and the contribution of each feature in the classification process. Users can further analyze and view the results.

8. A method for automatic classification and identification of medical test samples, characterized in that: include, Collect relevant data of medical test samples from multiple channels; Preprocessing relevant data of medical test samples; Extract features from relevant data of different types of medical test samples; The extracted features are screened using methods based on correlation analysis and information gain; Use deep neural networks to build classification and recognition models, and input the filtered features into the model for training and testing; Output the classification results and identification information of the sample and display them in a visual way.

9. An electronic device comprising: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the automatic classification and identification system for medical test samples according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the steps of the automatic classification and identification system for medical test samples according to any one of claims 1 to 7.

Images