Early screening information management system for lung cancer

Abstract

The invention discloses a lung cancer early screening information management system, and relates to the technical field of medical treatment, the system can accurately calculate the probability that a feature value belongs to a rare lung cancer mode, construct a probability vector P and complete feature extraction by combining with a random forest algorithm, which is like accurately screening features highly related to rare lung cancer from a large amount of data. And a key basis is provided for subsequent diagnosis. And the feature substitution subsystem locates a target patient according to the feature set, and sensitively discovers a clustering group which may represent a rare lung cancer mode through clustering similarity analysis and abnormal score YF calculation. For example, patients with similar characteristics are classified into a group, and clustering groups with high abnormal scores are highly likely to be related to rare lung cancer. On the basis, combined condition testing is executed, the probability corresponding to the rare lung cancer mode is accurately recognized, and the screening capacity of the rare lung cancer is further improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to an early screening information management system for lung cancer. Background Art

[0002] In the broad and vital field of medical informationization, the lung cancer early screening information management system occupies an extremely critical position. As a malignant tumor with extremely high morbidity and mortality worldwide, early screening of lung cancer is of decisive significance for improving the survival rate and cure rate of patients. By calling the data of storage devices in the hospital and carefully combing the characteristics of patients, it has become the core support tool in the early diagnosis and intervention process of lung cancer, and plays an indispensable role in the formulation of medical decisions and the planning of patient treatment plans.

[0003] Currently, early screening for lung cancer faces many challenges. In terms of diagnostic models, ordinary screening methods are seriously inadequate in recognizing rare lung cancer patterns, and misdiagnosis and missed diagnosis occur frequently. For example, many rare lung cancer features differ slightly from common types of lung cancer or normal physiological conditions. Conventional diagnosis cannot accurately capture the information contained in these feature combinations, making it difficult for patients to receive timely and accurate diagnoses in the early stages, delaying treatment opportunities. This will reduce the patient's survival rate, impose a heavy burden on the patient's family and society, and seriously affect the overall effectiveness of early lung cancer screening and the improvement of medical service quality. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present application provides an early screening information management system for lung cancer.

[0005] The disclosed embodiment provides an early screening information management system for lung cancer, including: a prediction subsystem, a feature substitution subsystem, and an information aggregation subsystem;

[0006] The prediction subsystem is used to collect relevant data of early-stage lung cancer patients in the hospital using the storage devices in the hospital, generate a lung cancer database, and combine the random forest algorithm to analyze the probability that the corresponding feature values ​​belong to rare lung cancer patterns, so as to complete feature extraction and construct a feature set;

[0007] The feature substitution subsystem is used to extract the target early-stage lung cancer patients in the hospital, and extract the feature set of the target early-stage lung cancer patients. After clustering similarity analysis, the abnormal score YF is traversed to obtain the abnormal score YF. Based on the abnormal score YF, the combined condition test operation is performed to identify the probability of the corresponding rare lung cancer pattern under the corresponding conditions.

[0008] The information summary subsystem is used to summarize the information after the test to generate an information evaluation form and upload it to the hospital information system.

[0009] Optionally, the prediction subsystem includes a data collection unit, a soft classification unit, and a feature extraction unit;

[0010] The data collection unit is used to collect the basic information, diagnosis information, examination information and medication records of each early lung cancer patient according to the hospital information system within the hospital, and share and interact with the image storage and transmission system, laboratory information system and electronic medical record system according to the calling function of the hospital information system, so as to fill in the medical images, pathology and laboratory data and electronic medical records of each early lung cancer patient, and generate a lung cancer database after statistics.

[0011] Optionally, the soft classification unit is used to construct a random forest model based on the random forest algorithm, divide the data in the lung cancer database into a training set, a test set and a validation set, and pre-set an initial range of decision trees. The training set is used to train the random forest model, and then the validation set is used to perform an accuracy A calculation on the random forest model to evaluate the performance of the random forest model under different decision tree quantities. By comparing the accuracy A values ​​under different decision tree quantities, the number of decision trees corresponding to the maximum accuracy A value is selected as the optimal number of decision trees M in the random forest.

[0012] According to the optimal number of decision trees, each feature in the lung cancer database is input into the root node of each decision tree, and the direction of each feature is determined through the feature judgment condition of each internal node of the decision tree, and the leaf node of the corresponding decision tree is reached after traversal. After statistics, the number of features N of the training set contained in the leaf node of the corresponding decision tree is obtained, and according to the number of features N of the training set contained in the leaf node of the corresponding decision tree, the number of features n of the training set contained in the leaf node of the corresponding decision tree belonging to each rare lung cancer pattern is determined. According to the number of features N of the training set contained in the leaf node of the corresponding decision tree and the number of features n of the training set contained in the leaf node of the corresponding decision tree belonging to each rare lung cancer pattern, the probability of the corresponding feature value belonging to the rare lung cancer pattern is analyzed, and the probability vector P is constructed, which is specifically: ; ;

[0013] In the formula, Represents a decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern The probability estimate of ; Represented in the decision tree The leaf nodes of The number of features; Represents a decision tree The probability vector of feature x in the lung cancer database belonging to different rare lung cancer patterns; , and Represents decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern Probability estimate, decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern Probability estimate and decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern The probability estimate of ; m represents the number of types of rare lung cancer patterns; i represents the decision tree number; j represents the rare lung cancer pattern number.

[0014] Optionally, based on the probability vector P, a majority voting method is used to predict the prediction result corresponding to the corresponding feature value, specifically: ;

[0015] In the formula, Represents the prediction result corresponding to feature x; represents a collection of rare lung cancer patterns, = ; Indicates that all decision trees give the conclusion that feature x belongs to a rare lung cancer pattern The average probability of

[0016] According to the predicted results and actual results corresponding to the corresponding feature values, the features in the test set are shuffled to construct the feature importance score ZP and analyze the importance of each feature for rare lung cancer screening, specifically: ;

[0017] In the formula, represents the importance score of feature x; Represents the classification error of the decision tree without disrupting the value of feature x; Represents the classification error of the decision tree when the value of feature x is shuffled.

[0018] Optionally, the feature extraction unit is used to set an importance threshold based on the feature importance score ZP of different features, and compare the feature importance score ZP of the corresponding feature with the importance threshold; if the feature importance score ZP of the corresponding feature exceeds the importance threshold, the corresponding feature is included in the feature set; if the feature importance score ZP of the corresponding feature does not exceed the importance threshold, the corresponding feature is not included in the feature set.

[0019] Optionally, the feature substitution subsystem includes a clustering unit, an anomaly analysis unit, and a recombination analysis unit;

[0020] The clustering unit is used to extract the early-stage lung cancer patients corresponding to the relevant features in the feature set in the hospital according to the feature set in the feature extraction unit, and mark the extracted early-stage lung cancer patients as target early-stage lung cancer patients. Based on the target early-stage lung cancer patients, the relevant data of the target early-stage lung cancer patients are extracted from the lung cancer database. According to the DBSCAN algorithm, the relevant data of the target early-stage lung cancer patients are clustered to obtain several cluster groups. According to the corresponding cluster groups, clustering information is generated, and the clustering information includes the features in cluster group z and the number of features in cluster group z. , the average number of features across all clusters , the average distance from cluster group z to its nearest neighbor cluster and the average distance of all clusters to the nearest neighbor cluster .

[0021] Optionally, the anomaly analysis unit is used to traverse all cluster groups according to the cluster information to calculate and obtain an anomaly score YF. The anomaly score YF is specifically obtained by the following formula: ;

[0022] In the formula, represents the anomaly score of cluster group z; represents the average number of features across all clusters; represents the number of features in cluster group z; Represents the average distance from cluster group z to its nearest neighbor cluster; Represents the average distance of all clusters to the nearest neighbor cluster; and Represents the weight coefficient, which is used to adjust the importance of feature quantity and distance factors in anomaly scoring.

[0023] Optionally, the recombination analysis unit is used to obtain the abnormality scores YF of the corresponding cluster groups according to the acquisition method of the abnormality scores YF, and extract the cluster group corresponding to the maximum abnormality score YF as the target cluster group, randomly combine the features in the target cluster group, and test the combined conditions after random combination to identify the probability of the corresponding rare lung cancer pattern under the corresponding conditions, specifically: ;

[0024] In the formula, Indicates that in the combination condition Cases, rare lung cancer pattern The probability of In a known rare lung cancer pattern In the case of The probability of Indicates rare lung cancer patterns in historical data The proportion of total cases; Indicates combination conditions probability.

[0025] Optionally, the information summary subsystem is used to organize the test results according to the recombinant analysis units to generate an information evaluation aid table, and store it in the form of an electronic document in the hospital information system to assist doctors in analyzing early-stage lung cancer patients.

[0026] The present invention provides an early screening information management system for lung cancer, which has the following beneficial effects:

[0027] (1) By combining the random forest algorithm, the system can accurately calculate the probability that the feature value belongs to a rare lung cancer pattern, construct a probability vector P and complete feature extraction, just like accurately screening out features that are highly related to rare lung cancer from a large amount of data, providing key basis for subsequent diagnosis. The feature substitution subsystem locates the target patient based on the feature set, and through cluster similarity analysis and abnormality score YF calculation, it keenly discovers cluster groups that may represent rare lung cancer patterns. For example, patients with similar features are grouped together, and cluster groups with high abnormality scores are very likely to be related to rare lung cancer. On this basis, combined conditional tests are performed to accurately identify the probability of corresponding rare lung cancer patterns, further improving the screening ability of rare lung cancer. The information aggregation subsystem integrates the information after the test to generate an evaluation table and upload it, so that doctors can easily obtain comprehensive information. For example, when a doctor opens the system, he can clearly see the comprehensive evaluation of the patient's various test results, quickly make diagnosis and treatment decisions, improve the efficiency and quality of medical services, and help the efficient implementation of early lung cancer screening.

[0028] (2) In terms of prediction results, the majority voting method combined with the probability vector P is used to predict the results corresponding to the feature values, which can make full use of the information of multiple decision trees in the random forest. By calculating the average probability given by all decision trees that feature x belongs to a rare lung cancer pattern, and then taking the rare lung cancer pattern corresponding to the maximum value as the prediction result, the bias that may be generated by a single decision tree is effectively avoided, making the prediction results more accurate and reliable. For example, when faced with complex patient feature data, multiple decision trees analyze from different angles, and the comprehensive prediction results can better reflect the actual situation and provide doctors with more valuable diagnostic references. In feature importance analysis, by constructing the feature importance score ZP and comparing the classification errors of the decision trees before and after the feature values ​​are disrupted, the impact of each feature on rare lung cancer screening can be accurately measured. If a feature has a high importance score, it means that it plays a key role in the screening results. For example, among many patient features, some gene features have a high importance score, so in subsequent diagnosis and research, these features can be focused on to improve the targetedness of screening.

[0029] (3) The clustering unit is guided by the feature set provided by the feature extraction unit. It can efficiently and accurately extract early-stage lung cancer patients with relevant features from a large number of patients in the hospital, mark them as target early-stage lung cancer patients, and obtain their relevant data from the lung cancer database. After clustering these data using the DBSCAN algorithm, the patients are grouped according to feature similarity to generate comprehensive clustering information. This process helps medical personnel quickly sort out the feature distribution of rare lung cancer patients and identify groups of patients with similar features. For example, in actual applications, it may be found that a group of patients present similar features in terms of genetic test results, imaging features, and lifestyle habits, which provides a clear direction for subsequent in-depth analysis.

[0030] (4) The reorganization analysis unit obtains the abnormal score YF of each cluster group and accurately locks the target cluster group corresponding to the maximum abnormal score. This is like finding the key path to the clue of rare lung cancer in a complex patient data maze. The features within the target cluster group are randomly combined and tested, and the probability of suffering from a rare lung cancer pattern under the corresponding conditions is calculated, which can deeply explore the potential connection between feature combinations and rare lung cancer. For example, by combining the patient's genetic test data, imaging features, living habits and other features in a variety of ways, it may be found that the probability of association between certain unique combinations and rare lung cancer is far higher than expected, providing a valuable reference direction for clinical diagnosis, and further improving the pertinence and accuracy of rare lung cancer diagnosis. With the help of the generation of a table, the information evaluation clearly presents key data such as the probability of rare lung cancer under different feature combinations and patient clustering. Whether it is a preliminary diagnosis for patients in the outpatient clinic or an in-depth discussion of the condition in the ward, doctors can quickly make scientific and reasonable judgments based on this information evaluation with the help of the table. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application.

[0032] Figure 1 This is a block diagram of an early lung cancer screening information management system of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] Example 1

[0035] like Figure 1 As shown, the present invention provides an early screening information management system for lung cancer, including a prediction subsystem, a feature substitution subsystem and an information aggregation subsystem;

[0036] The prediction subsystem is used to collect relevant data of early-stage lung cancer patients in the hospital using the storage devices in the hospital, generate a lung cancer database, and combine the random forest algorithm to analyze the probability that the corresponding feature values ​​belong to rare lung cancer patterns, construct a probability vector P, complete feature extraction, and construct a feature set;

[0037] The feature substitution subsystem is used to extract the target early-stage lung cancer patients in the hospital based on the feature set, and extract the feature set of the target early-stage lung cancer patients, traverse to obtain the abnormal score YF through clustering similarity analysis, and perform the combined condition test operation based on the abnormal score YF to identify the probability of the corresponding rare lung cancer pattern under the corresponding conditions;

[0038] The information summary subsystem is used to summarize the information after the test to generate an information evaluation form and upload it to the hospital information system.

[0039] It should be noted that, for people who have not yet been diagnosed with lung cancer, the present invention can identify the probability of corresponding rare lung cancer patterns occurring under corresponding conditions. It can also be used to assess the risk of people who have not yet been diagnosed with lung cancer suffering from different types of lung cancer, conduct stratified management according to the level of risk, strengthen monitoring and preventive measures for high-risk groups, such as increasing the frequency of screening and providing more stringent lifestyle interventions, etc. For low-risk groups, the intensity of screening can be appropriately reduced to achieve a rational allocation of medical resources.

[0040] In an embodiment of the present invention, in terms of data utilization, the prediction subsystem calls HIS to collect data to generate a lung cancer database, so that the originally scattered and messy data can be centralized and integrated. For example, it can gather the patient's imaging data, genetic test results, medical history and other information, which is convenient for subsequent in-depth analysis and improves the availability and value of the data. The probability vector P and feature set are constructed through the random forest algorithm, which further improves the accuracy and pertinence of feature extraction. For example, after analyzing a large amount of patient data, the features closely related to the rare lung cancer pattern are accurately screened out to provide strong support for subsequent diagnosis. The feature substitution subsystem locates the target patient based on the feature set, and through cluster similarity analysis and abnormal score calculation, it can keenly identify cluster groups that may represent rare lung cancer patterns. For example, patients with similar characteristics are grouped together. If a cluster group has a high abnormal score, it means that the feature combination of this group of patients may be related to rare lung cancer, and then a combined condition test is performed to accurately identify the probability of suffering from rare lung cancer under the corresponding conditions, which greatly improves the screening efficiency and accuracy of rare lung cancer. The information summary subsystem generates an information evaluation form and uploads it to the hospital information system for doctors to review at any time. For example, when a doctor opens the system, he or she can see a comprehensive evaluation of the patient's test results, quickly understand the condition, make more reasonable diagnosis and treatment decisions, improve the efficiency and quality of medical services, and provide convenience for early screening and treatment of lung cancer.

[0041] Example 2

[0042] like Figure 1 As shown, specifically: the prediction subsystem includes a data collection unit, a soft classification unit and a feature extraction unit;

[0043] The data collection unit is used to collect the basic information, diagnosis information, examination information and medication records of each early lung cancer patient according to the hospital information system within the hospital, and share and interact with the picture archiving and transmission system (PACS), laboratory information system (LIS) and electronic medical record system (EMR) according to the calling function of the hospital information system, so as to fill in the medical images, pathology and laboratory data and electronic medical records of each early lung cancer patient, and generate a lung cancer database after statistics.

[0044] The soft classification unit is used to construct a random forest model based on the random forest algorithm, divide the data in the lung cancer database into a training set, a test set, and a validation set, and pre-set the initial range of decision trees. The training set is used to train the random forest model, and then the validation set is used to perform an accuracy A calculation on the random forest model to evaluate the performance of the random forest model under different decision tree quantities. By comparing the accuracy A values ​​under different decision tree quantities, the number of decision trees corresponding to the maximum accuracy A value is selected as the optimal number of decision trees M in the random forest.

[0045] A decision tree is a model that makes decisions based on a tree structure. Each internal node is a test on a feature, each branch is a test output, and each leaf node is a category. In a classification task, a series of tests are performed on the input features to eventually classify the features into a certain category.

[0046] According to the optimal number of decision trees, each feature in the lung cancer database is input into the root node of each decision tree, and the direction of each feature is determined through the feature judgment condition of each internal node of the decision tree, and the leaf node of the corresponding decision tree is reached after traversal (the leaf node is a node without child nodes in the decision tree, which represents the final decision result). After statistics, the number of features N of the training set contained in the leaf node of the corresponding decision tree is obtained, and according to the number of features N of the training set contained in the leaf node of the corresponding decision tree, the number of features n of the training set contained in the leaf node of the corresponding decision tree belonging to each rare lung cancer pattern is determined. According to the number of features N of the training set contained in the leaf node of the corresponding decision tree and the number of features n of the training set contained in the leaf node of the corresponding decision tree belonging to each rare lung cancer pattern, the probability of the corresponding feature value belonging to the rare lung cancer pattern is analyzed, and the probability vector P is constructed, which is specifically: ;

[0047] The meaning of this formula is that under the feature combination represented by the leaf node, the feature belongs to the rare lung cancer pattern The probability is equal to the rare lung cancer pattern in the leaf node The ratio of the number of features to the total number of features.

[0048] ;

[0049] Specifically, by combining the probabilities corresponding to each category, we get the entire probability vector , this probability vector reflects the decision tree Estimate the likelihood that feature x belongs to each category.

[0050] In the formula, Represents a decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern The probability estimate of ; Represented in the decision tree The leaf nodes of The number of features; Represents a decision tree The probability vector of feature x in the lung cancer database belonging to different rare lung cancer patterns; , and Represents decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern Probability estimate, decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern Probability estimate and decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern The probability estimate of ; m represents the number of types of rare lung cancer patterns; i represents the decision tree number; j represents the rare lung cancer pattern number.

[0051] Among them, rare lung cancer patterns refer to lung malignancies with lower incidence and special pathological types compared to common lung cancers (such as non-small cell lung cancer and small cell lung cancer). These rare types usually have unique histological features, gene mutations, treatment options and prognosis. They are less common in clinical practice and may not respond well to conventional treatments. Rare lung cancer patterns include but are not limited to adenoid cystic carcinoma, clear cell lung carcinoma, alveolar soft tissue sarcoma and pulmonary hemangiopericytoma.

[0052] In the embodiment of the present invention, the data collection unit relies on the powerful calling function of the hospital information system to achieve multi-source data integration. Obtaining the patient's lung images from the image storage and transmission system can intuitively show lung lesions; collecting pathological and laboratory data from the laboratory information system, such as tumor marker values, to provide a quantitative basis for diagnosis; integrating past medical history and medication records from the electronic medical record system to build a comprehensive patient portrait. Through data sharing and interaction, fill the data gap and generate a complete lung cancer database. For example, for an early lung cancer patient, the information such as lung CT images, blood tumor marker test results, and past treatment medications are systematically integrated, providing a rich data basis for subsequent analysis. The soft classification unit divides the training set, test set, and validation set, pre-sets the decision tree quantity range, and performs a multi-dimensional evaluation of the random forest model. After the accuracy calculation under different decision tree quantities, the optimal decision tree number M is screened out to ensure the optimal model performance. Afterwards, according to the optimal decision tree number, the features in the lung cancer database are input into the decision tree, and the node feature judgment is accurately traversed to the leaf node.

[0053] Taking the rare lung adenocarcinoma pattern as an example, in a decision tree leaf node, the total number of features N is found to be 100, of which the number of features n belonging to the rare lung adenocarcinoma pattern is 30. According to the formula, the probability that the leaf node feature belongs to the rare lung adenocarcinoma pattern is 30%. Combining the probabilities of each category into a probability vector P provides a quantitative basis for the accurate judgment of rare lung cancer patterns, further improving the recognition ability and accuracy of rare lung cancer patterns in early lung cancer screening.

[0054] Example 3

[0055] like Figure 1As shown, specifically: based on the probability vector P, the majority voting method is used to predict the prediction results corresponding to the corresponding feature values, specifically: ;

[0056] In the formula, Represents the prediction result corresponding to feature x; represents a collection of rare lung cancer patterns, = ; Indicates that all decision trees give the conclusion that feature x belongs to a rare lung cancer pattern The average probability of; argmax (*) represents the value of the independent variable that makes the function reach the maximum value, which is a mathematical term;

[0057] Among them, all decision trees give the probability that feature x belongs to a rare lung cancer pattern. The average probability The specific formula is as follows: ;

[0058] In the formula, M represents the optimal number of decision trees;

[0059] It should be noted that by averaging the probabilities given by all decision trees regarding feature x belonging to each lung cancer type, the results of multiple decision trees can be comprehensively considered to more comprehensively analyze the likelihood that feature x corresponds to different lung cancer types.

[0060] According to the predicted results and actual results corresponding to the corresponding feature values, the features in the test set are shuffled to construct the feature importance score ZP and analyze the importance of each feature for rare lung cancer screening, specifically: ;

[0061] In the formula, represents the importance score of feature x; Represents the classification error of the decision tree without disrupting the value of feature x; Represents the classification error of the decision tree when the value of feature x is shuffled;

[0062] It should be noted that without disrupting the value of feature x, the classification error of the decision tree The steps to obtain are: traverse each decision tree in the random forest, use the test set data to make predictions, and get the prediction results (the prediction results corresponding to feature x). ), calculate the error between the predicted result and the true label (true result), that is, the proportion of samples with classification errors, and take the average of the classification errors of each decision tree as the classification error of the decision tree of the entire random forest without disrupting the value of feature x. .

[0063] The classification error of the decision tree when the value of feature x is shuffled The steps for obtaining are as follows: first select the feature x to be evaluated, specify it through the feature index, then traverse each decision tree in the random forest, randomly shuffle the value of feature x in the test set data, obtain the shuffled feature matrix, and use the shuffled feature matrix for prediction to obtain the prediction result, calculate the error between the prediction result and the true label (true result), that is, the proportion of samples with classification errors, take the average of the classification errors of each decision tree after shuffling the feature x, and use it as the classification error of the decision tree of the entire random forest when the value of feature x is shuffled .

[0064] The specific process of the shuffled feature matrix is ​​as follows: clarify the index position of the feature x to be shuffled in the feature matrix, extract the column vector corresponding to the feature x from the feature matrix, and perform a random shuffle operation on the extracted column vector. Here, a random number generator can be used to generate a random permutation index sequence with the same length as the column vector, and then rearrange the elements in the column vector according to the index sequence. Here, it is assumed that the length of the column vector is Y, and the generated random permutation index sequence is , ,..., , then the ath element of the shuffled column vector is the ath element of the original column vector elements, and replace the shuffled column vector back to the column position corresponding to the feature x in the original feature matrix to obtain the shuffled feature matrix.

[0065] Specifically, the higher the feature importance score, the greater the impact of the feature on rare lung cancer screening. During the model training process, the most important features can be selected according to the feature importance score to reduce the complexity of the model and improve the performance of the model.

[0066] Among them, by introducing sample data of rare lung cancer types for model training and optimization, we prepare for improving the system's sensitivity to rare lung cancer types, so that it can more accurately identify rare lung cancer patients, reduce missed diagnoses and misdiagnoses of rare lung cancer patients, avoid unnecessary further examinations and treatments, and improve the utilization efficiency of medical resources.

[0067] The feature extraction unit is used to set an importance threshold according to the feature importance score ZP of different features, and compare the feature importance score ZP of the corresponding feature with the importance threshold. If the feature importance score ZP of the corresponding feature exceeds the importance threshold, the corresponding feature is included in the feature set; if the feature importance score ZP of the corresponding feature does not exceed the importance threshold, the corresponding feature is not included in the feature set.

[0068] In an embodiment of the present invention, the majority voting method is used to predict the result corresponding to the feature value based on the probability vector P, which can effectively integrate the judgment of multiple decision trees. Different decision trees classify and predict feature x from their own perspectives, and combine multiple judgment bases by finding the average probability. For example, in a set of lung cancer patient feature data, multiple decision trees respectively give the probability that feature x belongs to different rare lung cancer patterns. After average calculation, the argmax function is used to select the rare lung cancer pattern with the largest average probability as the prediction result, which makes the prediction result more reliable and reduces the possible deviation of a single decision tree. The process of constructing the feature importance score ZP can accurately evaluate the role of each feature in rare lung cancer screening. By comparing the classification error of the decision tree before and after the feature value is disrupted, the feature importance is quantified. If the classification error of a feature x increases significantly after being disrupted, it indicates that the feature has a large impact on the classification and its importance score is high. For example, after the feature "a certain gene mutation situation" is disrupted, the classification error of the decision tree for the rare lung cancer pattern increases from 5% to 30%, indicating that this feature is crucial in rare lung cancer screening. The feature extraction unit screens features based on the feature importance score ZP, sets the importance threshold, and includes features that exceed the threshold in the feature set, which effectively reduces the number of features and removes features that have little impact on screening. When building a screening model, unnecessary interference factors are reduced, the complexity of the model is reduced, and operational efficiency is improved. At the same time, retaining key features helps improve the accuracy of the model for screening rare lung cancer, allowing the model to focus more on core influencing factors, providing doctors with more targeted diagnostic basis, and helping to carry out efficient early screening of lung cancer.

[0069] Example 4

[0070] like Figure 1 As shown, specifically: the feature substitution subsystem includes a clustering unit, an abnormality analysis unit and a recombination analysis unit;

[0071] The clustering unit is used to extract the early-stage lung cancer patients corresponding to the relevant features in the feature set in the hospital according to the feature set in the feature extraction unit, and mark the extracted early-stage lung cancer patients as target early-stage lung cancer patients. Based on the target early-stage lung cancer patients, the relevant data of the target early-stage lung cancer patients are extracted from the lung cancer database. According to the DBSCAN algorithm, the relevant data of the target early-stage lung cancer patients are clustered to obtain several cluster groups. According to the corresponding cluster groups, clustering information is generated, and the clustering information includes the features in cluster group z and the number of features in cluster group z. , the average number of features across all clusters , the average distance from cluster group z to its nearest neighbor cluster and the average distance of all clusters to the nearest neighbor cluster .

[0072] The anomaly analysis unit is used to traverse all cluster groups according to the cluster information to calculate the anomaly score YF. The anomaly score YF is obtained by the following formula: ;

[0073] In the formula, represents the anomaly score of cluster group z; represents the average number of features across all clusters; Represents the number of features in cluster group z. Clusters with fewer features may be more likely to be abnormal, which reflects their rarity. In lung cancer screening, the features corresponding to rare lung cancer patterns are often a minority in the population. It represents the average distance from cluster group z to its nearest neighbor cluster. A long distance means that the cluster is very different from other clusters. Represents the average distance of all clusters to the nearest neighbor cluster; and Represents the weight coefficient, which is used to adjust the importance of the number of features and distance factors in the anomaly score. When the value of is large, it indicates that the cluster is highly separated from other clusters and its characteristics are quite different from those of other clusters. Rare lung cancer patterns usually differ significantly from common lung cancer patterns or normal conditions in their characteristic performance;

[0074] Anomaly score for cluster group z The abnormality score of cluster group z relative to other cluster groups is measured. The higher the abnormality score, the more likely the cluster group is to represent a rare lung cancer pattern.

[0075] In an embodiment of the present invention, the clustering unit accurately locates the target early-stage lung cancer patients in the hospital based on the feature set obtained by the feature extraction unit, and uses the DBSCAN algorithm to perform clustering processing after extracting relevant data. This method can cluster patients with similar features together to form different cluster groups and generate comprehensive cluster information, which helps doctors quickly identify patient groups with similar features and improves the understanding of the distribution of lung cancer patient characteristics. For example, through cluster analysis, it may be found that patients with certain specific gene features and similar living habits are clustered together, providing a basis for subsequent targeted research and treatment. The abnormal analysis unit calculates the abnormal score YF based on the cluster information. The score comprehensively considers the number of features in the cluster group and the average distance to the nearest neighbor cluster, and adjusts the importance of both through the weight coefficient. Clusters with a small number of features and clusters far away from other clusters will obtain higher abnormal scores, which enables the system to keenly capture cluster groups that may represent rare lung cancer patterns. For example, when the number of features of a cluster group is significantly less than the average level and the distance from other clusters is far, its abnormal score will be very high, indicating that the patients in the cluster group may suffer from rare lung cancer. Doctors can conduct in-depth research on these cluster groups with high abnormality scores, carry out more detailed examinations and diagnoses, thereby increasing the early detection rate of rare lung cancer, striving for more timely treatment for patients, and improving their prognosis. At the same time, this precise abnormality screening also helps to rationally allocate medical resources and invest more resources in patients who may have rare lung cancer.

[0076] Example 5

[0077] like Figure 1 As shown, specifically: the recombinant analysis unit is used to obtain the abnormal score YF of the corresponding cluster group according to the acquisition method of the abnormal score YF, and extract the cluster group corresponding to the maximum abnormal score YF as the target cluster group, randomly combine the features in the target cluster group, and test the combined conditions after random combination to identify the probability of the corresponding rare lung cancer pattern under the corresponding conditions, specifically: ;

[0078] In the formula, Indicates that in the combination condition Cases, rare lung cancer pattern The probability of In a known rare lung cancer pattern In the case of This probability can be obtained by statistical analysis of existing lung cancer case data, for example, The frequency of occurrence of the situation where the statistical feature x1 takes the value V1, x2 takes the value V2, etc. in patients with type 2 lung cancer.

[0079] Indicates rare lung cancer patterns in historical data The proportion of total cases; Indicates combination conditions For example, the probability that features x1, x2, x3, ..., xF take the values ​​V1, V2, V3, ..., VF respectively;

[0080] The information summary subsystem is used to organize the test results according to the recombinant analysis units to generate an information evaluation aid table, which is stored in the hospital information system in the form of electronic documents (such as Excel tables, PDF reports, etc.) to assist doctors in analyzing early lung cancer patients; the information evaluation aid table can be printed or displayed in the system interface as needed.

[0081] Among them, different types of lung cancer may require different treatment plans. Clarifying the probability of a patient suffering from a certain type of lung cancer under a specific combination of features can help doctors plan personalized treatment strategies in advance. For example, if a patient has a higher probability of suffering from a specific type of lung cancer, doctors can more specifically choose a treatment method that is relatively suitable for this type of lung cancer among surgery, radiotherapy, chemotherapy or targeted therapy to improve the treatment effect. In addition, the probability relationship between certain feature combinations and specific types of lung cancer may also be related to the patient's response to specific treatment methods. By analyzing the probability value, doctors can predict the patient's possible response to different treatment options to a certain extent, adjust the treatment plan in advance, and avoid the harm and burden of ineffective treatment on patients.

[0082] In the embodiment of the present invention, the recombinant analysis unit focuses on starting from the abnormal score YF, and by extracting the cluster group corresponding to the maximum abnormal score as the target cluster group, accurately locates the patient group feature set that is most likely to be associated with the rare lung cancer pattern. The features in the target cluster group are randomly combined and tested, and the formula is used to calculate the probability of suffering from the rare lung cancer pattern under the corresponding conditions. This process can deeply explore the potential connection between different feature combinations and rare lung cancer. For example, by testing a variety of combinations of features such as age, specific gene mutations, and tumor marker levels, it may be found that some special feature combinations have a higher probability of association with specific rare lung cancer patterns, providing a more targeted reference for clinical diagnosis. The information summary subsystem receives the test results of the recombinant analysis unit, organizes them to generate an information evaluation table, and stores them in the hospital information system in the form of electronic documents, which enables doctors to easily obtain comprehensive and systematic screening information. Doctors do not need to perform tedious searches in a large amount of scattered data. They can open the system or view electronic documents to clearly and intuitively see various test data and analysis results, such as the probability of rare lung cancer under different feature combinations, patient clustering information, etc. Information evaluation can be printed or displayed on the system interface with the help of a table, meeting the doctor's usage needs in different work scenarios. When making rounds in the ward, doctors can print tables to analyze the actual situation of the patient; in the office, they can view and study various data in detail through the system interface. This efficient information presentation method greatly improves the efficiency and accuracy of doctors' analysis of early lung cancer patients, helping doctors to make scientific and reasonable diagnosis and treatment decisions more quickly, thereby improving patients' medical experience and treatment effects, and also helping hospitals optimize the allocation of medical resources and improve the overall quality of medical services.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A lung cancer early screening information management system, characterized by: It includes prediction subsystem, feature substitution subsystem and information aggregation subsystem; The prediction subsystem is used to collect relevant data of early-stage lung cancer patients in the hospital using the storage devices in the hospital, generate a lung cancer database, and combine the random forest algorithm to analyze the probability that the corresponding feature values ​​belong to rare lung cancer patterns, so as to complete feature extraction and construct a feature set; The feature substitution subsystem is used to extract the target early-stage lung cancer patients in the hospital, and extract the feature set of the target early-stage lung cancer patients. After clustering similarity analysis, the abnormal score YF is traversed to obtain the abnormal score YF. Based on the abnormal score YF, the combined condition test operation is performed to identify the probability of the corresponding rare lung cancer pattern under the corresponding conditions. The information summary subsystem is used to summarize the information after the test to generate an information evaluation form and upload it to the hospital information system.

2. The early screening information management system for lung cancer according to claim 1, characterized in that: The prediction subsystem includes a data collection unit, a soft classification unit, and a feature extraction unit; The data collection unit is used to collect the basic information, diagnosis information, examination information and medication records of each early lung cancer patient according to the hospital information system within the hospital, and share and interact with the image storage and transmission system, laboratory information system and electronic medical record system according to the calling function of the hospital information system, so as to fill in the medical images, pathology and laboratory data and electronic medical records of each early lung cancer patient, and generate a lung cancer database after statistics.

3. The early screening information management system for lung cancer according to claim 2, characterized in that: The soft classification unit is used to construct a random forest model based on the random forest algorithm, divide the data in the lung cancer database into a training set, a test set, and a validation set, and pre-set the initial range of decision trees. The training set is used to train the random forest model, and then the validation set is used to perform an accuracy A calculation on the random forest model to evaluate the performance of the random forest model under different decision tree quantities. By comparing the accuracy A values ​​under different decision tree quantities, the number of decision trees corresponding to the maximum accuracy A value is selected as the optimal number of decision trees M in the random forest. According to the optimal number of decision trees, each feature in the lung cancer database is input into the root node of each decision tree, and the direction of each feature is determined through the feature judgment condition of each internal node of the decision tree, and the leaf node of the corresponding decision tree is reached after traversal. After statistics, the number of features N of the training set contained in the leaf node of the corresponding decision tree is obtained, and according to the number of features N of the training set contained in the leaf node of the corresponding decision tree, the number of features n of the training set contained in the leaf node of the corresponding decision tree belonging to each rare lung cancer pattern is determined. According to the number of features N of the training set contained in the leaf node of the corresponding decision tree and the number of features n of the training set contained in the leaf node of the corresponding decision tree belonging to each rare lung cancer pattern, the probability of the corresponding feature value belonging to the rare lung cancer pattern is analyzed, and the probability vector P is constructed, which is specifically: ； ； In the formula, Represents a decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern The probability estimate of ; Represented in the decision tree The leaf nodes of The number of features; Represents a decision tree The probability vector of feature x in the lung cancer database belonging to different rare lung cancer patterns; , and Represents decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern Probability estimate, decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern Probability estimate and decision tree The feature x in the lung cancer database belongs to a rare lung cancer pattern The probability estimate of ; m represents the number of types of rare lung cancer patterns; i represents the decision tree number; j represents the rare lung cancer pattern number.

4. The early screening information management system for lung cancer according to claim 3, characterized in that: Based on the probability vector P, the majority voting method is used to predict the prediction results corresponding to the corresponding feature values, specifically: ； In the formula, Represents the prediction result corresponding to feature x; represents a collection of rare lung cancer patterns, = ; Indicates that all decision trees give the conclusion that feature x belongs to a rare lung cancer pattern The average probability of According to the predicted results and actual results corresponding to the corresponding feature values, the features in the test set are shuffled to construct the feature importance score ZP and analyze the importance of each feature for rare lung cancer screening, specifically: ； In the formula, represents the importance score of feature x; Represents the classification error of the decision tree without disrupting the value of feature x; Represents the classification error of the decision tree when the value of feature x is shuffled.

5. The early screening information management system for lung cancer according to claim 4, characterized in that: The feature extraction unit is used to set an importance threshold according to the feature importance score ZP of different features, and compare the feature importance score ZP of the corresponding feature with the importance threshold. If the feature importance score ZP of the corresponding feature exceeds the importance threshold, the corresponding feature is included in the feature set; if the feature importance score ZP of the corresponding feature does not exceed the importance threshold, the corresponding feature is not included in the feature set.

6. The early screening information management system for lung cancer according to claim 5, characterized in that: The feature substitution subsystem includes a clustering unit, an anomaly analysis unit, and a recombination analysis unit; The clustering unit is used to extract the early-stage lung cancer patients corresponding to the relevant features in the feature set in the hospital according to the feature set in the feature extraction unit, and mark the extracted early-stage lung cancer patients as target early-stage lung cancer patients. Based on the target early-stage lung cancer patients, the relevant data of the target early-stage lung cancer patients are extracted from the lung cancer database. According to the DBSCAN algorithm, the relevant data of the target early-stage lung cancer patients are clustered to obtain several cluster groups. According to the corresponding cluster groups, clustering information is generated, and the clustering information includes the features in cluster group z and the number of features in cluster group z. , the average number of features across all clusters , the average distance from cluster group z to its nearest neighbor cluster and the average distance of all clusters to the nearest neighbor cluster .

7. The early screening information management system for lung cancer according to claim 6, characterized in that: The anomaly analysis unit is used to traverse all cluster groups according to the cluster information to calculate the anomaly score YF. The anomaly score YF is obtained by the following formula: ； In the formula, represents the anomaly score of cluster group z; represents the average number of features across all clusters; represents the number of features in cluster group z; Represents the average distance from cluster group z to its nearest neighbor cluster; Represents the average distance of all clusters to the nearest neighbor cluster; and Represents the weight coefficient, which is used to adjust the importance of feature quantity and distance factors in anomaly scoring.

8. The early screening information management system for lung cancer according to claim 7, characterized in that: The recombination analysis unit is used to obtain the abnormality scores YF of the corresponding cluster groups according to the acquisition method of the abnormality scores YF, and extract the cluster group corresponding to the maximum abnormality score YF as the target cluster group, randomly combine the features in the target cluster group, and test the combined conditions after random combination to identify the probability of the corresponding rare lung cancer pattern under the corresponding conditions, specifically: ； In the formula, Indicates that in the combination condition Cases, rare lung cancer pattern probability; In a known rare lung cancer pattern In the case of probability; Indicates rare lung cancer patterns in historical data The proportion of total cases; Indicates combination conditions probability.

9. The early screening information management system for lung cancer according to claim 8, characterized in that: The information summary subsystem is used to organize the test results according to the recombinant analysis units to generate information evaluation aid tables, which are stored in the hospital information system in the form of electronic documents to assist doctors in analyzing early-stage lung cancer patients.

Images