Method for detecting pathological type of pulmonary nodule and storage medium

Through the preprocessing and classifier construction of pulmonary nodule urine protein expression profile data, non-invasive detection of pulmonary nodule pathological types is achieved, solving the problem of inconsistent physical and psychological burden and diagnosis results of traditional methods, and providing more Accurate and reliable classification of pathological types.

CN118942552AActive Publication Date: 2024-11-12BEIJING CHAOYANG HOSPITAL CAPITAL MEDICAL UNIVERSITY +1
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411020775.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-12
Estimated Expiration
2044-07-29

AI Technical Summary

Technical Problem

Traditional pulmonary nodule pathological type detection methods have problems with physical injury and psychological burden caused by invasive operations, as well as inconsistent diagnosis results.

Method used

By obtaining pulmonary nodule urine protein expression profile data for preprocessing, a pre-classifier is constructed for feature matching and prediction, and the pathological type classification results are output, non-invasive detection is achieved and differences in the dependence of doctors' diagnostic level are reduced.

Benefits of technology

A non-invasive, comfortable and safe detection process is achieved, providing more unified and reliable classification results, reducing diagnostic differences due to doctors' experience level, and is of great significance for early detection and precise treatment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118942552A_ABST
    Figure CN118942552A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a method for detecting pathological types of pulmonary nodules and a storage medium, and belongs to the technical field of medical treatment. The method for detecting the pathological type of the pulmonary nodule comprises the following steps: acquiring pulmonary nodule urine protein expression profile data of a to-be-analyzed individual sample; preprocessing the pulmonary nodule urine protein expression profile data to obtain data after quantitative data standardization based on the sum; the standardized data are input into a pre-constructed classifier; and performing feature matching and prediction on the input data by a pre-constructed classifier, and outputting a pathological type classification result of the individual sample to be analyzed. The method is used for solving the problems that a traditional pulmonary nodule pathology type detection method brings physical injuries and psychological burdens to a patient, and diagnosis results are not uniform.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a method and a storage medium for detecting the pathological type of pulmonary nodules. Background Art

[0002] Lung cancer is a major global health threat, accounting for approximately 11.4% of all cancer diagnoses in 2020, ranking at the forefront of cancer incidence. Moreover, lung cancer is the leading cause of cancer-related deaths, accounting for approximately 18% of total cancer deaths. In my country, the incidence and mortality rates of lung cancer in 2020 ranked first among all cancers. Pulmonary nodules are closely related to lung cancer. For patients with lung nodules, their lung nodules may be benign or malignant. Malignant lung nodules are a manifestation of early lung cancer. Therefore, the detection of the pathological type of lung nodules is of great significance for the early detection, precise treatment and prognosis evaluation of lung cancer.

[0003] The traditional method of diagnosing the pathological type of patients with lung nodules mainly relies on histopathological examinations, such as bronchoscopy, fine needle aspiration biopsy, and open lung biopsy. These traditional histopathological examinations are invasive procedures and may cause complications such as bleeding, infection, and pneumothorax (lung collapse). In particular, open lung biopsy may also cause general anesthesia risks, bleeding risks, and postoperative recovery problems, causing harm to the patient's body. In addition, invasive procedures may cause a greater psychological burden on patients.

[0004] In addition, traditional diagnostic results rely on the individual level of the doctor, and the diagnostic results are easily different due to the doctor's experience level, and cannot provide unified and reliable diagnostic results.

[0005] From the above analysis, it can be seen that the traditional method of diagnosing the pathological type of patients with lung nodules has the defects of causing physical harm and psychological burden to patients and inconsistent diagnostic results. Summary of the invention

[0006] The purpose of the present invention is to provide a method and storage medium for detecting the pathological type of pulmonary nodules, so as to solve the problem that the above-mentioned traditional method for detecting the pathological type of pulmonary nodules brings physical harm and psychological burden to patients and the problem of inconsistent diagnostic results.

[0007] In order to achieve the above object, in a first aspect, the present invention provides a method for detecting the pathological type of pulmonary nodules, comprising the following steps:

[0008] Acquire urine protein expression spectrum data of lung nodules of individual samples to be analyzed; preprocess the urine protein expression spectrum data of lung nodules to obtain data standardized based on the total quantitative data; input the standardized data into a pre-constructed classifier; and the pre-constructed classifier performs feature matching and prediction on the input data, and outputs the pathological type classification result of the individual sample to be analyzed.

[0009] Furthermore, the lung nodule urine protein expression spectrum data is preprocessed to obtain the data normalized based on the total quantitative data, including:

[0010] Proteins whose occurrence frequency in the lung nodule urine protein expression spectrum data is lower than the preset occurrence frequency are eliminated, or proteins whose occurrence frequency is higher than the preset occurrence frequency are retained, and redundant proteins are removed to obtain an alternative protein set; proteins whose iFOT values ​​rank before the preset positions in the alternative protein set are selected to obtain a protein set for screening, wherein the iFOT value of each protein is the ratio of the iBAQ value of the protein to the sum of the iBAQ values ​​of all proteins in the alternative protein set; and the sum of iFOT of the protein set for screening is calculated, and then the iFOT of each protein in the protein set for screening is divided by the sum of iFOT to obtain data after quantitative data standardization based on the sum.

[0011] Furthermore, the pre-constructed classifier is a pulmonary nodule pathology type classifier, which includes a pulmonary nodule pathology benign and malignant classifier or a pulmonary nodule pathology subtype classifier.

[0012] Furthermore, the construction of the pulmonary nodule pathological type classifier includes the following steps:

[0013] A lung nodule urine protein expression spectrum data set is obtained, which is marked with a lung nodule pathological type label; a portion of data is selected from the lung nodule urine protein expression spectrum data set as a training set; the training set is preprocessed to obtain the data of the training set after the quantitative data is standardized based on the sum; each of the standardized data is a feature, the features are sorted in a certain way, and k features are selected therefrom; k is a preset number of feature selections; and a classifier is constructed based on the k features and the lung nodule pathological type label to obtain the lung nodule pathological type classifier.

[0014] Furthermore, when the constructed classifier is a pulmonary nodule pathology benign and malignant classifier, the pulmonary nodule pathology type label includes benign pulmonary nodules and malignant pulmonary nodules; when the constructed classifier is a pulmonary nodule pathology subtype classifier, the pulmonary nodule pathology type label includes high-risk subtypes and medium- and low-risk subtypes.

[0015] Furthermore, the features are sorted in a certain way, and k features are selected therefrom, including:

[0016] The features are selected using the SelectKBest class, and the F value of each feature, that is, the variance ratio, is obtained through the f_classif analysis and calculation of ANOVA; and after obtaining the F value of each feature, the features are sorted from high to low according to the F value of the feature, and then the top k features with the highest F value are selected.

[0017] Furthermore, the lung nodule urine protein expression spectrum dataset includes a multicenter cohort sample dataset and an independent external cohort sample dataset; to prevent model overfitting, the classifier adopts a five-fold cross-validation, that is, 4 / 5 of the multicenter cohort sample dataset is selected as a training set for classifier training, the remaining 1 / 5 of the multicenter cohort sample dataset is selected for internal validation of the classifier, and an independent external cohort sample dataset is selected for external validation of the classifier.

[0018] Furthermore, when constructing a classifier, an artificial intelligence model is used, preferably a random forest or a support vector machine.

[0019] Furthermore, the construction of the classifier also includes evaluation of the classifier, and the evaluation indicators include accuracy, specificity and AUC, and the best classifier is retained based on the indicators.

[0020] Further, the k features include at least one of the following proteins: CBR1, CDH15, DSC3, FRK, GSTP1, HSPD1, LGALS3, LGALS3BP, MYH9, NEU1, PAH, PPIA, SCNN1G, SRC, TKT, TUBA3C, TXN, TUBA1A, VAMP8, MGAM, WASL, KL, HDAC6, TOM1, NAMPT, TUBA1B, ST3GAL6, OLFM4, RAB21, CHMP2B, ATP6V1D, TUBA8, MIOX, ALDH8A1, AGMAT, TUBA1C, TUBA3E, TUBA3D, CLRN3 and LRRK2.

[0021] Further, the k features include at least one of the following proteins: APOC2, APOD, ATP6V1A, B2M, CALML3, CAT, CD9, CDH13, CRABP2, CST3, GLO1, IGFBP6, LTA4H, PAH, PCCB, SERPINI1, PRKACB, PSAP, PTMA, CXCL12, SLC4A1, TSN, UMOD, STK24, ATRN, PTER, NAPSA, ADAMTS1, DPP3, TSPAN9, FRMPD1, RAB21, AHCYL2, QPCT, CHMP2A, CD248, SLC44A2, ITFG1, TMEM192 and UBE2NL.

[0022] In a second aspect, the present invention provides a machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute any of the above-mentioned methods for detecting the pathological type of pulmonary nodules of the present application.

[0023] In a third aspect, the present invention provides a computer program product, which includes instructions for causing a machine to execute any of the above methods for detecting the pathological type of pulmonary nodules in the present application.

[0024] In a fourth aspect, the present invention provides a device for detecting the pathological type of lung nodules, the device comprising a processor and a memory, the memory storing instructions for enabling the processor to execute any one of the above-mentioned methods for detecting the pathological type of lung nodules in the present application.

[0025] Through the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0026] 1. The present invention uses urine protein to detect the pathological type of lung nodules. The patient only needs to provide a urine sample, thus achieving true non-invasive detection. Compared with traditional invasive operations, this method avoids physical harm and psychological burden on patients, and the detection process is more comfortable and safe.

[0027] 2. The present invention classifies the pathological types of individual samples to be analyzed through a pre-constructed classifier, thereby avoiding dependence on the doctor's diagnostic level, helping to reduce diagnostic differences caused by the doctor's experience level, and providing more unified and reliable classification results.

[0028] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0030] Figure 1 It is a flowchart of a method for detecting pathological types of pulmonary nodules provided in an embodiment of the present invention.

[0031] Figure 2 It is a schematic diagram of the process of constructing a pulmonary nodule pathological type classifier provided in an embodiment of the present invention.

[0032] Figure 3 It is a schematic diagram of ROC evaluation of the internal validation set of the benign and malignant pathology classifier of pulmonary nodules provided in an embodiment of the present invention.

[0033] Figure 4 It is a schematic diagram of ROC evaluation of the external validation set of the benign and malignant pathology classifier of pulmonary nodules provided in an embodiment of the present invention.

[0034] Figure 5 It is a schematic diagram of ROC evaluation of the internal validation set of the pulmonary nodule pathological subtype classifier provided in an embodiment of the present invention.

[0035] Figure 6 It is a schematic diagram of ROC evaluation of the external validation set of the pulmonary nodule pathological subtype classifier provided in an embodiment of the present invention.

[0036] Figure 7 It is a schematic diagram of the accuracy evaluation of the internal validation set of the benign and malignant pathology classifier of pulmonary nodules provided in an embodiment of the present invention.

[0037] Figure 8 It is a schematic diagram of the accuracy evaluation of the internal validation set of the pulmonary nodule pathological subtype classifier provided in an embodiment of the present invention.

[0038] Fig. 9 It is a schematic diagram of the external validation set parameter evaluation of the benign and malignant pathology classifier of pulmonary nodules provided in an embodiment of the present invention.

[0039] Fig.10 It is a schematic diagram of external validation set parameter evaluation of the pulmonary nodule pathological subtype classifier provided in an embodiment of the present invention.

[0040] Fig.11 It is a schematic diagram of the confusion matrix of the external validation set of the benign and malignant lung nodule pathology classifier provided in an embodiment of the present invention.

[0041] Fig.12 It is a schematic diagram of the confusion matrix of the external validation set of the pulmonary nodule pathological subtype classifier provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0043] Figure 1 Figure 2 is a flow chart of a method for detecting the pathological type of pulmonary nodules. Figure 1 As shown, the present invention provides a method for detecting the pathological type of pulmonary nodules, comprising the following steps:

[0044] S101: Obtaining lung nodule urine protein expression profile data of the individual sample to be analyzed;

[0045] S102: preprocessing the lung nodule urine protein expression spectrum data to obtain data normalized based on the total quantitative data;

[0046] S103: inputting the standardized data into a pre-built classifier; and

[0047] S104: The pre-built classifier performs feature matching and prediction on the input data, and outputs the classification result of the pathological type of the individual sample to be analyzed.

[0048] It should be noted that when performing feature matching, logarithmic transformation of the features is also included.

[0049] There are more than 8,000 proteins in human urine, about 40% of which come from plasma proteins. The urine of healthy individuals contains more than 1,800 proteins that are significantly expressed in the lungs. Therefore, the present invention detects the pathological type of lung nodules in patients with lung nodules through the proteins present in the urine. Compared with other detection methods, urine collection provides a non-invasive and easy-to-obtain method to obtain a large number of samples, which causes less physical harm to the patient, and the detection process is more comfortable and safe, avoiding physical harm and psychological burden on the patient.

[0050] When classifying and diagnosing the pathological types of pulmonary nodules, the traditional diagnostic method depends on the doctor's diagnostic level, and the accuracy of diagnosis by doctors with less experience will be significantly reduced. The present invention can effectively identify and utilize proteins in urine by constructing a pulmonary nodule pathological type classifier, improve the accuracy of pulmonary nodule pathological type classification, help reduce diagnostic differences caused by the doctor's experience level, and provide more unified and reliable classification results, which is of great significance for early detection, precision treatment and prognosis evaluation.

[0051] By classifying the patient's lung nodule pathological types, doctors can better formulate treatment plans based on the classification results, such as deciding whether surgery is needed, assisting in planning the scope of surgery before surgery, identifying people with a high risk of postoperative recurrence, and assisting in formulating postoperative management plans, thereby achieving personalized treatment.

[0052] Preferably, the lung nodule urine protein expression spectrum data is preprocessed to obtain the data normalized based on the total quantitative data, including:

[0053] Eliminate proteins whose occurrence frequency is lower than a preset occurrence frequency in the lung nodule urine protein expression spectrum data or retain proteins whose occurrence frequency is higher than a preset occurrence frequency, remove redundant proteins, and obtain an alternative protein set;

[0054] Selecting proteins whose iFOT values ​​rank before a preset position in the candidate protein set to obtain a protein set for screening, wherein the iFOT value of each protein is the ratio of the iBAQ value of the protein to the sum of the iBAQ values ​​of all proteins in the candidate protein set; and

[0055] The sum of iFOTs of the protein set for screening is calculated, and then the iFOT of each protein in the protein set for screening is divided by the sum of iFOTs to obtain data normalized based on the sum of quantitative data.

[0056] It should be noted that the iBAQ value of a protein is the sum of the peak areas of all corresponding peptide segments of the protein / theoretical number of peptide segments, and is calculated using the label-free quantitative iBAQ method based on peak area.

[0057] It is understood that the number of data contained in the data normalized based on the total quantitative data is the same as the number of proteins in the protein set for screening.

[0058] In this embodiment, when removing proteins, proteins with an occurrence frequency of less than 10% are preferentially removed.

[0059] Preferably, the pre-constructed classifier is a pulmonary nodule pathology type classifier, which includes a pulmonary nodule pathology benign and malignant classifier or a pulmonary nodule pathology subtype classifier.

[0060] Figure 2 Figure 1 is a flowchart of the construction of a pulmonary nodule pathological type classifier. Figure 2 As shown, the construction of the pulmonary nodule pathological type classifier includes the following steps:

[0061] S201: Obtain a dataset of urine protein expression profiles of lung nodules, which is marked with labels of lung nodule pathological types;

[0062] S202: Selecting part of the data from the lung nodule urine protein expression spectrum dataset as a training set;

[0063] S203: preprocessing the training set to obtain data of the training set after quantitative data standardization based on the sum;

[0064] S204: Each of the standardized data is a feature, the features are sorted in a certain way, and k features are selected therefrom; k is a preset number of feature selections; and

[0065] S205: Based on the k features and the pulmonary nodule pathology type label, a classifier is constructed to obtain the pulmonary nodule pathology type classifier.

[0066] Each of the standardized data is a feature, each feature corresponds to a protein, and the proteins corresponding to the features are also called markers.

[0067] Preferably, when the constructed classifier is a benign or malignant lung nodule pathology classifier, the lung nodule pathology type label includes benign lung nodules and malignant lung nodules;

[0068] When the constructed classifier is a pulmonary nodule pathological subtype classifier, the pulmonary nodule pathological type label includes a high-risk subtype and a medium- and low-risk subtype.

[0069] In one embodiment of the present invention, when the classifier is constructed as a benign and malignant classifier for pulmonary nodule pathology, the selected k features include at least one of the following proteins: CBR1, CDH15, DSC3, FRK, GSTP1, HSPD1, LGALS3, LGALS3BP, MYH9, NEU1, PAH, PPIA, SCNN1G, SRC, TKT, TUBA3C, TXN, TUBA1A, VAMP8, MGAM, WASL, KL, HDAC6, TOM1, NAMPT, TUBA1B, ST3GAL6, OLFM4, RAB21, CHMP2B, ATP6V1D, TUBA8, MIOX, ALDH8A1, AGMAT, TUBA1C, TUBA3E, TUBA3D, CLRN3 and LRRK2.

[0070] In one embodiment of the present invention, when the classifier is constructed as a pulmonary nodule pathological subtype classifier, particularly an invasive lung adenocarcinoma pathological subtype classifier, the selected k features include at least one of the following proteins: APOC2, APOD, ATP6V1A, B2M, CALML3, CAT, CD9, CDH13, CRABP2, CST3, GLO1, IGFBP6, LTA4H, PAH, PCCB, SERPINI1, PRKACB, PSAP, PTMA, CXCL12, SLC4A1, TSN, UMOD, STK24, ATRN, PTER, NAPSA, ADAMTS1, DPP3, TSPAN9, FRMPD1, RAB21, AHCYL2, QPCT, CHMP2A, CD248, SLC44A2, ITFG1, TMEM192 and UBE2NL.

[0071] It can be understood that the pulmonary nodule pathological benign and malignant classifier and the pulmonary nodule pathological subtype classifier are both binary classifiers. The main difference lies in the different pathological type labels and the different k features selected, but the construction steps of the two are the same.

[0072] Preferably, the features are sorted in a certain manner, and k features are selected therefrom, including:

[0073] SelectKBest class is used to select the features, and the F value of each feature, i.e., the variance ratio, is obtained by f_classif analysis and calculation of ANOVA; and

[0074] After obtaining the F value of each feature, sort the features from high to low according to their F value, and then select the top k features with the highest F value.

[0075] It can be understood that the F value can reflect the importance of each feature, that is, the statistical significance of each feature for the pathological label. In this embodiment, the F value of each feature is obtained by the f_classif analysis and calculation of ANOVA. Each F value corresponds to an index, and the F values ​​are sorted, that is, the indexes of the F values ​​are sorted, and then the F values ​​are searched according to the index. The value range of the pre-set feature selection number k is [40,1500]. In this embodiment, k=40 is selected.

[0076] Preferably, the lung nodule urine protein expression profile dataset includes a multi-center cohort sample dataset and an independent external cohort sample dataset;

[0077] To prevent model overfitting, the classifier uses a five-fold cross-validation, that is, 4 / 5 of the multi-center cohort sample data set is selected as the training set for classifier training, the remaining 1 / 5 of the multi-center cohort sample data set is selected for internal validation of the classifier, and an independent external cohort sample data set is selected for external validation of the classifier.

[0078] It should be noted that in actual operation, the multi-center cohort sample data set was divided into five subsets of the same size, 4 / 5 of the multi-center cohort sample data set, that is, four subsets were selected for classifier training, and then the remaining subset was selected for classifier internal verification. The training process was repeated five times, and a different subset was selected each time for classifier internal verification, and finally five evaluation results were obtained for classifier evaluation.

[0079] Preferably, when constructing a classifier, an artificial intelligence model is used, preferably a random forest or a support vector machine.

[0080] Preferably, the construction of the classifier further includes evaluating the classifier, and the evaluation indicators include accuracy, specificity and AUC, and the optimal classifier is retained based on the indicators.

[0081] When constructing the classifier, a total of 7 types are constructed in this embodiment, including support vector machine (SVM), multi-layer perceptron (MLP), logistic regression, decision tree, random forest, gradient boosting and K-nearest neighbors.

[0082] Figure 3 This is a schematic diagram of the ROC evaluation of the internal validation set of the benign and malignant lung nodule pathology classifier. Figure 4 This is a schematic diagram of ROC evaluation of the external validation set of the benign and malignant lung nodule pathology classifier. Figure 5 This is a schematic diagram of the ROC evaluation of the internal validation set of the pulmonary nodule pathological subtype classifier. Figure 6 This is a schematic diagram of the ROC evaluation of the external validation set of the pulmonary nodule pathological subtype classifier. When selecting the optimal classifier, the AUC is mainly considered. The AUC is the area under the ROC curve and the coordinate axis. The larger the area, the higher the AUC score, indicating that the classifier has a better classification effect. Figure 3 , Figure 4 , Figure 5 and Figure 6 It can be seen that when using two validation sets for validation, the AUC of the support vector machine showed the highest score for both the pulmonary nodule pathology benign and malignant classifier and the pulmonary nodule pathology subtype classifier, and the classification effect was better than other classifiers.

[0083] Figure 7 This is a schematic diagram of the accuracy evaluation of the internal validation set of the lung nodule pathology benign and malignant classifier. Figure 8 This is a schematic diagram of the accuracy evaluation of the internal validation set of the pulmonary nodule pathological subtype classifier. Figure 7 and Figure 8 All of them are box plots. In this embodiment, the judgment is mainly based on the median. The higher the median, the higher the accuracy of the classifier. It can be understood that the median is the median of the five evaluation results obtained by the five-fold cross validation. Figure 7 and Figure 8 As can be seen from the figure, the support vector machine still shows the highest accuracy.

[0084] It should be noted that when other indicators perform poorly, even if the AUC is excellent, it should be considered abandoned. Fig. 9 This is a schematic diagram of the external validation set parameter evaluation for the benign and malignant lung nodule pathology classifier. Fig.10 Figure 1 is a schematic diagram of the external validation set parameter evaluation for the pulmonary nodule pathological subtype classifier. Fig. 9 and Fig.10 It can be seen that the support vector machine has achieved high scores in all indicators. Therefore, in this embodiment, the support vector machine is selected as the optimal classifier to be retained. However, the present invention is not limited to this. If a classifier with better performance in various indicators is constructed in other embodiments, other types of classifiers can also be retained.

[0085] Fig.11 This is a diagram of the confusion matrix of the external validation set for the benign and malignant classifier of pulmonary nodules. Fig.12 This is a schematic diagram of the confusion matrix of the external validation set of the pulmonary nodule pathological subtype classifier. Fig.11 and Fig.12 The classifiers in are all support vector machines, where malignant tumors are malignant pulmonary nodules, benign nodules are benign pulmonary nodules, moderately differentiated pulmonary nodules are moderately low-risk subtypes, and poorly differentiated pulmonary nodules are high-risk subtypes. From the two confusion matrices, it can be calculated that the retained optimal pulmonary nodule pathology benign and malignant classifier can identify malignant pulmonary nodules with 85% sensitivity while maintaining a specificity of 75%, and the pulmonary nodule pathology subtype classifier can identify high-risk subtypes with a sensitivity of 66.67% while maintaining a specificity of 79.17%, showing excellent classification performance.

[0086] The present invention selects the optimal classifier by evaluating the classifier, improves the accuracy of the constructed classifier, and can effectively identify and utilize proteins in urine, which is of great significance for early detection, precision treatment and prognosis evaluation. Accurate pathological type detection can help doctors choose the most appropriate treatment method, avoid unnecessary surgery and treatment, reduce medical costs, and improve treatment effects and patient survival rates.

[0087] An embodiment of the present invention further provides a machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the method for detecting the pathological type of pulmonary nodules described in the present application.

[0088] Those skilled in the art should understand that the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0089] The present application is described with reference to the flowcharts and / or block diagrams of the methods according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of the processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0090] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0092] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0093] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0094] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0095] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity, or apparatus.

[0096] The acquisition, transmission, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0097] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0098] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for detecting the pathological type of pulmonary nodules, characterized in that: The following steps are involved: Obtaining lung nodule urine protein expression profile data of individual samples to be analyzed; Preprocessing the lung nodule urine protein expression spectrum data to obtain data normalized based on the total quantitative data; Inputting the standardized data into a pre-built classifier; as well as The pre-built classifier performs feature matching and prediction on the input data and outputs the classification results of the pathological type of the individual sample to be analyzed.

2. The method according to claim 1, characterized in that The lung nodule urine protein expression spectrum data is preprocessed to obtain the data based on the total quantitative data normalization, including: Eliminate proteins whose occurrence frequency is lower than a preset occurrence frequency in the lung nodule urine protein expression spectrum data or retain proteins whose occurrence frequency is higher than a preset occurrence frequency, remove redundant proteins, and obtain an alternative protein set; Selecting proteins whose iFOT values ​​rank before a preset position in the candidate protein set to obtain a protein set for screening, wherein the iFOT value of each protein is the ratio of the iBAQ value of the protein to the sum of the iBAQ values ​​of all proteins in the candidate protein set; and The sum of iFOTs of the protein set for screening is calculated, and then the iFOT of each protein in the protein set for screening is divided by the sum of iFOTs to obtain data normalized based on the sum of quantitative data.

3. The method according to claim 1, characterized in that The pre-constructed classifier is a pulmonary nodule pathology type classifier, which includes a pulmonary nodule pathology benign and malignant classifier or a pulmonary nodule pathology subtype classifier.

4. The method according to claim 3, characterized in that The construction of the pulmonary nodule pathological type classifier comprises the following steps: Obtain a dataset of urine protein expression profiles of lung nodules, which is labeled with the pathological type of lung nodules; Selecting part of the data from the lung nodule urine protein expression spectrum data set as a training set; Preprocessing the training set to obtain the data of the training set after quantitative data standardization based on the sum; Each of the standardized data is a feature, the features are sorted in a certain way, and k features are selected therefrom; wherein k is a preset number of feature selections; and Based on the k features and the pulmonary nodule pathological type label, a classifier is constructed to obtain the pulmonary nodule pathological type classifier.

5. The method according to claim 4, characterized in that When the constructed classifier is a pulmonary nodule pathology benign or malignant classifier, the pulmonary nodule pathology type labels include benign pulmonary nodules and malignant pulmonary nodules; When the constructed classifier is a pulmonary nodule pathological subtype classifier, the pulmonary nodule pathological type label includes a high-risk subtype and a medium- and low-risk subtype.

6. The method according to claim 4, characterized in that The features are sorted in a certain way, and k features are selected from them, including: SelectKBest class is used to select the features, and the F value of each feature, i.e., the variance ratio, is obtained by f_classif analysis and calculation of ANOVA; and After obtaining the F value of each feature, sort the features from high to low according to their F value, and then select the top k features with the highest F value.

7. The method according to claim 4, characterized in that The lung nodule urine protein expression profile dataset includes a multi-center cohort sample dataset and an independent external cohort sample dataset; To prevent model overfitting, the classifier uses a five-fold cross-validation, that is, 4 / 5 of the multi-center cohort sample data set is selected as the training set for classifier training, the remaining 1 / 5 of the multi-center cohort sample data set is selected for internal validation of the classifier, and an independent external cohort sample data set is selected for external validation of the classifier.

8. The method according to claim 4, characterized in that When building a classifier, use an artificial intelligence model, preferably a random forest or support vector machine.

9. The method according to claim 4, characterized in that The construction of the classifier also includes evaluation of the classifier, and the evaluation indicators include accuracy, specificity and AUC, and the best classifier is retained based on the indicators.

10. The method according to claim 4, characterized in that The k features include at least one of the following proteins: CBR1, CDH15, DSC3, FRK, GSTP1, HSPD1, LGALS3, LGALS3BP, MYH9, NEU1, PAH, PPIA, SCNN1G, SRC, TKT, TUBA3C, TXN, TUBA1A, VAMP8, MGAM, WASL, KL, HDAC6, TOM1, NAMPT, TUBA1B, ST3GAL6, OLFM4, RAB21, CHMP2B, ATP6V1D, TUBA8, MIOX, ALDH8A1, AGMAT, TUBA1C, TUBA3E, TUBA3D, CLRN3 and LRRK2.

11. The method according to claim 4, characterized in that The k features include at least one of the following proteins: APOC2, APOD, ATP6V1A, B2M, CALML3, CAT, CD9, CDH13, CRABP2, CST3, GLO1, IGFBP6, LTA4H, PAH, PCCB, SERPINI1, PRKACB, PSAP, PTMA, CXCL12, SLC4A1, TSN, UMOD, STK24, ATRN, PTER, NAPSA, ADAMTS1, DPP3, TSPAN9, FRMPD1, RAB21, AHCYL2, QPCT, CHMP2A, CD248, SLC44A2, ITFG1, TMEM192 and UBE2NL.

12. A machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the method described in any one of claims 1 to 11 of the present application.

Citation Information

Patent Citations

  • Gastric mucosa lesion protein molecular typing, lesion progression and gastric cancer related protein marker and method for predicting lesion progression risk

    CN112071363A

  • Pancreaticobiliary ampulla carcinoma classification model generation method and image classification method

    CN113762395A

  • Application of urine protein marker in preparation of kit for detecting diabetic nephropathy

    CN113917152A

  • Application of urine protein marker in preparation of kit for detecting clinical stages of diabetic nephropathy

    CN113917153A

  • Application of urine protein marker in preparation of kit for clinical early detection of diabetic nephropathy

    CN113917154A