CN115732078A

CN115732078A - Pain disease distinguishing and classifying method and device based on multivariate decision tree model

Info

Publication number: CN115732078A
Application number: CN202211442891.7A
Authority: CN
Inventors: 张翔; 王云晓; 张宝杰
Original assignee: Wuzheng Intelligent Technology Beijing Co ltd
Current assignee: Wuzheng Intelligent Technology Beijing Co ltd
Priority date: 2022-11-17
Filing date: 2022-11-17
Publication date: 2023-03-03

Abstract

The application discloses a pain disease distinguishing and classifying method and device based on a multivariate decision tree model, wherein a sample data set is obtained, and associated characteristic information of pain diseases in the sample data set is collected; normalization processing is carried out on the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and a disease knowledge base associated with the characteristic information set is established; inputting semantic feature information corresponding to a disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of associated feature information, and training the multivariate decision tree model through the semantic feature information subset to obtain a pain disease discrimination classification model; and receiving the pain symptom information and the characteristic information input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output the pain disease distinguishing and classifying result of the patient. The technical problem that diagnosis of painful diseases is difficult in the related art is solved, and intelligent judgment and classification of the painful diseases are achieved.

Description

Pain disease distinguishing and classifying method and device based on multivariate decision tree model

Technical Field

The application belongs to the technical field of computers, and particularly relates to a pain disease distinguishing and classifying method and device based on a multivariate decision tree model, electronic equipment and a storage medium.

Background

Pain, known as the fifth vital sign of the human body, is an important cue for potential or existing damage in humans, but at the same time, the long-term severe pain also seriously affects the quality of life of people. In fact most diseases are accompanied by symptoms of pain. Statistically, more than half of the patients at the outpatient clinic have complaints of pain. The causes of painful diseases are complex, the expressed symptoms are different, the diagnosis of most painful diseases is difficult, and the primary diseases are often covered by the pain symptoms, so that misdiagnosis and missed diagnosis delay the disease condition. For example, the mild person may give pain and economic burden to the patient, and the severe person may be disabled or bereaved by losing the diagnosis and treatment timing.

Aiming at the problem that the diagnosis of painful diseases in the related technology is difficult, an effective solution is not provided at present.

Disclosure of Invention

Accordingly, embodiments of the present invention are directed to a method, an apparatus, an electronic device and a storage medium for pain disease identification and classification based on a multivariate decision tree model, which are used to solve at least one of the problems of the prior art.

In order to achieve the above object, in a first aspect, the present application provides a method for pain disease discrimination and classification based on a multivariate decision tree model, comprising:

acquiring a sample data set, and collecting associated characteristic information of pain diseases in the sample data set;

carrying out normalization processing on the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and establishing a disease knowledge base associated with the characteristic information set;

inputting semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multivariate decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model;

and receiving the symptom information and the characteristic information of the pain input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output a pain disease distinguishing and classifying result of the patient.

In one embodiment, the associated characteristic information includes at least one of: the name of the pain disorder, the cause of the disease, the site of occurrence, the typical symptoms, the accompanying symptoms, the concurrent symptoms, the degree of pain, the exacerbation or alleviation factor of the pain, and the time and duration of the occurrence.

In one embodiment, the normalizing the associated feature information to obtain the feature information set of the associated feature information includes: and performing normalization processing on the associated characteristic information based on a word frequency-inverse document frequency algorithm, extracting key characteristic words of the associated characteristic information, and obtaining a characteristic information set of the associated characteristic information.

In one embodiment, the normalizing the associated feature information based on the word frequency-inverse document frequency algorithm, and extracting the key feature words of the associated feature information includes: and calculating a TF-IDF value of each word in the text information and data samples through a word frequency-inverse document frequency algorithm, arranging each word in a descending order based on the TF-IDF value, and extracting N words arranged at the top as the key feature words, wherein N is a positive integer.

In one embodiment, before the semantic feature information corresponding to the disease knowledge base is input into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, the method further includes: and converting the disease knowledge base into corresponding semantic feature information through a Bayesian network technology.

In a second aspect, the present application further provides a pain disease distinguishing and classifying device based on a multivariate decision tree model, including:

the characteristic acquisition module is used for acquiring a sample data set and acquiring associated characteristic information of pain diseases in the sample data set;

the characteristic extraction module is used for carrying out normalization processing on the associated characteristic information to obtain a characteristic information set of the associated characteristic information and establishing a disease knowledge base associated with the characteristic information set;

the model construction module is used for inputting semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multivariate decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model;

and the cognitive recognition module is used for receiving the pain symptom information and the characteristic information input by the patient, carrying out similarity calculation based on the pain disease distinguishing and classifying model, obtaining the pain disease distinguishing and classifying result of the patient and outputting the pain disease distinguishing and classifying result.

In one embodiment, the normalizing the associated feature information to obtain the feature information set of the associated feature information includes: and performing normalization processing on the associated characteristic information based on a word frequency-inverse document frequency algorithm, and extracting key characteristic words of the associated characteristic information to obtain a characteristic information set of the associated characteristic information.

In one embodiment, the normalizing the associated feature information based on the word frequency-inverse document frequency algorithm, and extracting the key feature words of the associated feature information includes: and calculating the TF-IDF value of each word in the text information and data samples through a word frequency-inverse document frequency algorithm, arranging each word in a descending order based on the TF-IDF value, and extracting N words arranged at the top as the key feature words, wherein N is a positive integer.

In a third aspect, the present application further provides an electronic device comprising a memory and a processor, the memory having stored therein a computer program that, when executed by the processor, causes the processor to perform the steps of the multivariate decision tree model based pain disorder discrimination and classification method.

In a fourth aspect, the present application further provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, causes the processor to perform the steps of the method for pain disease discrimination and classification based on multivariate decision tree models.

According to the pain disease distinguishing and classifying method and device based on the multivariate decision tree model, the electronic equipment and the storage medium, the associated characteristic information of pain diseases in a sample data set is collected by obtaining the sample data set; carrying out normalization processing on the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and establishing a disease knowledge base associated with the characteristic information set; inputting semantic feature information corresponding to the disease knowledge base into a multi-variable decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multi-variable decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model; and receiving the symptom information and the characteristic information of the pain input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output a pain disease distinguishing and classifying result of the patient. The problem that diagnosis of painful diseases is difficult in the related technology is solved, and the following beneficial effects are achieved: through the application of the multivariate decision tree model, the pain diseases are intelligently distinguished and classified, corresponding conditioning and treatment schemes are provided, doctors are assisted to improve the diagnosis and treatment efficiency, and the misdiagnosis rate of the pain diseases is reduced.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, are included to provide a further understanding of the application and to enable other features, objects, and advantages of the application to be more apparent. The drawings and their description illustrate the embodiments of the invention and do not limit it. In the drawings:

fig. 1 is a flowchart of implementing a pain disease distinguishing and classifying method based on a multivariate decision tree model according to an embodiment of the present application;

FIG. 2 is a schematic diagram of the main modules of the pain diagnosis and classification apparatus based on multivariate decision tree model according to the embodiment of the present application;

FIG. 3 is a diagram of an exemplary system architecture that may be used with embodiments of the present application;

fig. 4 is a schematic block diagram of a computer system suitable for implementing a terminal device or a server according to an embodiment of the present application.

Detailed Description

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without making any creative effort shall fall within the protection scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances in order to facilitate the description of the embodiments of the application herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In this application, the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings. These terms are used primarily to better describe the present application and its embodiments, and are not used to limit the indicated devices, elements or components to a particular orientation or to be constructed and operated in a particular orientation.

Moreover, some of the above terms may be used to indicate other meanings besides the orientation or positional relationship, for example, the term "on" may also be used to indicate some kind of attachment or connection relationship in some cases. The specific meaning of these terms in this application will be understood by those of ordinary skill in the art as the case may be.

In addition, the term "plurality" shall mean two as well as more than two.

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Fig. 1 shows a flow of implementing a pain disease distinguishing and classifying method based on a multivariate decision tree model provided by an embodiment of the present application, and for convenience of illustration, only the relevant parts of the embodiment of the present application are shown, which are detailed as follows:

a pain disease distinguishing and classifying method based on a multivariate decision tree model comprises the following steps:

s101: acquiring a sample data set, and acquiring associated characteristic information of pain diseases in the sample data set;

s102: normalizing the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and establishing a disease knowledge base associated with the characteristic information set;

s103: inputting semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multivariate decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model;

s104: and receiving the pain symptom information and the characteristic information input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output a pain disease distinguishing and classifying result of the patient.

In step S101: acquiring a sample data set, and collecting associated characteristic information of pain diseases in the sample data set. By collecting the associated characteristic information of the pain diseases in the sample data set, a plurality of pieces of historical information of different types of pain diseases can be known, and subsequent data analysis and model training are facilitated. The sample data set is a diagnosis report of a plurality of pain disease patients collected in advance, and can be stored in a text form or a data table form and directly accessed and obtained by calling a data interface. Here, the related characteristic information includes text information of the sample data and information of the data class, for example, the text information may be a disease name, a symptom, and the like of the pain disease of the patient, and the information of the data class may be a disease onset time, a duration, and the like of the pain disease of the patient.

In one embodiment, the associated characteristic information includes at least one of: the name of the pain disorder, the cause of the disease, the site of occurrence, the typical symptoms, the accompanying symptoms, the concurrent symptoms, the degree of pain, the exacerbation or remission factor of the pain, and the time and duration of the occurrence. Through the associated characteristic information of the pain diseases, the associated characteristic information can be associated, so that the construction of a knowledge base and the training of a model are facilitated, and the pain diseases of different types can be identified more accurately. Here, the type of the sample data set needs to make samples of different types of pain diseases as comprehensive as possible, so that a more comprehensive disease knowledge base is constructed, and a more accurate prediction model is trained.

Illustratively, the name of the pain disorder, the cause of the disease, the site of occurrence, the typical symptoms, the accompanying symptoms, the concurrent symptoms, the degree of pain, the exacerbation or alleviation factors of the pain, and the time, duration of the occurrence, etc. may include the following relevant information: 1) The pain part is mostly the pathological change part, for example, precordial and poststernal pain are mostly angina or myocardial infarction, pleurisy is mostly on chest side, right upper abdominal pain can be cholelithiasis, cholecystitis and liver abscess, and the pain part of central nerve injury is dispersive and wide in range; 2) The nature and degree of pain are etiologically related, for example, dull or dull pain is visceral pain, myocardial infarction and arterial dissection are often manifested as severe laceration-like pain; 3) The duration of pain, which may be manifested as paroxysmal, persistent, spontaneous, etc.; 4) With symptoms: nervous system diseases, such as pain with emesis, vertigo, epilepsy, and visual disturbance, such as cerebral hemorrhage, vertebrobasilar ischemia, cerebrovascular malformation, and glaucoma; disorders of the locomotor system, manifested by pain associated with abnormal limb movement such as fracture, arthralgia associated with erythema, photosensitivity, systemic lupus erythematosus, etc.; cardiovascular system diseases, manifested by chest pain with pale, profuse sweating, blood pressure drop or shock, such as myocardial infarction, and dissecting aneurysm; respiratory system diseases, manifested as chest pain with cough, expectoration, and hemoptysis, such as lobar pneumonia and lung cancer; digestive and urinary system diseases, manifested by abdominal pain with acid regurgitation, emesis, diarrhea, jaundice, such as gastrointestinal obstruction, gastroduodenal ulcer, intestinal inflammation, and liver, gallbladder and pancreas diseases; lumbago is accompanied by frequent micturition, urgent micturition, incomplete urination, and hematuria, such as urinary tract infection and ureteral calculus. The above is merely an example description of the associated features of a pain disorder to facilitate understanding of the present solution. In the practical embodiment, it is not limited thereto.

In step S102: and carrying out normalization processing on the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and establishing a disease knowledge base associated with the characteristic information set. The characteristic information set is obtained by carrying out normalization processing on the associated characteristic information, the associated characteristics of various pain diseases can be more conveniently counted and analyzed, and the disease knowledge base associated with the characteristic information set is established, so that the disease knowledge base can be continuously updated, namely, sample data is continuously collected in the above mode, and then the disease knowledge base is continuously enriched so as to train a more accurate prediction model.

In one embodiment, the normalizing the associated feature information to obtain the feature information set of the associated feature information includes: and performing normalization processing on the associated characteristic information based on a word frequency-inverse document frequency algorithm, extracting key characteristic words of the associated characteristic information, and obtaining a characteristic information set of the associated characteristic information. Here, the correlation characteristic information is normalized through a TF-IDF (word frequency-inverse document frequency) technology, common words in the correlation characteristic information can be filtered out, and important words are retained, so that the formed key characteristic words can accurately express the characteristics of the pain disease.

In one embodiment, the normalizing the associated feature information based on the word frequency-inverse document frequency algorithm, and extracting the key feature words of the associated feature information includes: and calculating a TF-IDF value of each word in the text information and data samples through a word frequency-inverse document frequency algorithm, arranging each word in a descending order based on the TF-IDF value, and extracting N words arranged at the top as the key feature words, wherein N is a positive integer. Here, N may be set as needed. For example, if the associated feature information includes the following words: vomit, labor pain, diarrhea, hot pain, deformity and the like, and when extracting the key words, the sequence according to TF-IDF values is as follows: vomit, labor pain, diarrhea, heat pain and deformity, wherein N is set to be 3, and the extracted disease description keywords of the sample case are the vomit, the labor pain and the diarrhea.

It should be noted that the main idea of TF-IDF is: if a word appears in an article with a high frequency TF and rarely appears in other articles, the word or phrase is considered to have a good classification capability and is suitable for classification. The Term Frequency (TF) represents the frequency with which terms (keywords) appear in text. This number is typically normalized (typically word frequency divided by the total word count of the article) to prevent it from being biased towards long documents.

In step S103: and inputting semantic feature information corresponding to the disease knowledge base into a multi-variable decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multi-variable decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model.

In one embodiment, before inputting the semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, the method further includes: and converting the disease knowledge base into corresponding semantic feature information through a Bayesian network technology.

Here, a bayesian network is a causal association model of uncertainty. The Bayesian network is formed by plotting the random variables involved in a study system independently in a directed graph G (V, E) depending on whether conditions exist. Generally speaking, the bayesian formulation is the simplest bayesian network consisting of two nodes, a and b, and a directed arc pointing from a to b. The network can be used to infer the occurrence probability of various states of b from the state of a, or vice versa. The method comprises the following steps: the goal of a bayesian network is to infer the conditional probability of each state of some nodes based on the existing state of other nodes.

It should be noted that the classification belongs to a prediction task, that is, an objective function f (model) is obtained through learning of an existing data set (training set), each attribute set x is mapped to an objective attribute y (class), and y must be discrete (or belongs to a regression algorithm if y is continuous). In fact, most machine learning classification algorithms consider data with n attributes as a point in an n-dimensional space, and the classification process finds a hyperplane in the n-dimensional space or a higher-dimensional space and divides the points. Decision tree learning is one of the most extensive inductive reasoning algorithms in data mining. Generally, a decision tree includes a root node, a plurality of internal nodes, and a plurality of leaf nodes. Leaf nodes correspond to decision results; the internal nodes are classified corresponding to the attribute values, and a sample set contained in each internal node is divided into sub-nodes according to the result (value) of the attribute test; the root node contains a sample complete set, and a path from the root node to each leaf node corresponds to one judgment test sequence. The purpose of decision tree learning is to generate a decision tree which has strong generalization capability and can judge that no example classification result is seen, and the basic flow follows a simple and intuitive divide-and-conquer strategy. With the progress of the partitioning process, it is desirable that the branch nodes of the decision tree contain samples belonging to the same class as much as possible, i.e., the purity (purity) of the nodes is higher and higher. The purity of the node is used for measuring the optimal selection of attribute division, and the information entropy (informationcopy) index is used for measuring the purity of the sample set. Assuming that the ratio of the kth type sample in the current sample set D is pk (k =1,2,. Gtii Y |), the information entropy of D is defined as:

the smaller the value of Ent (D), the higher the purity of D. | Y | represents the number of categories of the sample.

Suppose that there are V possible values of the discrete attribute a { a } ¹ ,a ² ,...,a ^V And if a is used to divide the sample set D, V branch nodes are generated, where the V branch node includes all the values a of attribute a in D ^v Sample of (2), denoted as D ^v . We can calculate D according to the formula of information entropy ^v Considering the different number of samples contained in different branch nodes, the entropy of the information in (1) gives a weight | D to the branch node ^v I/| D i.e. the more samples, the greater the influence of the branch nodes, and thus the "information gain" obtained by dividing the sample set D by the attribute a can be calculated "

Generally, the higher the information gain, the greater the purity improvement obtained by partitioning using attribute a, and thus the information gain magnitude is used as a choice for decision tree partition attributes. On the decision tree generation algorithm, the attribute a × = argmax Gain (D, a) is selected as the basis for partitioning.

Considering a special case that the value v of the attribute a has exactly m values (the same as the number of the D sample sets), that is, each sample has a different value on the attribute a, and there are m values in total, this branch will generate m, each branch node contains only one sample, and the purity is the maximum, and it is obvious that it is natural to select a as the basis for attribute classification. However, the decision tree generated by such division does not have generalization capability, and cannot effectively predict new samples. Generally, the information gain criterion favors attributes with a large number of attribute values, and in order to reduce the possible adverse effect caused by the preference, the information gain is not directly used, but the gain rate is used to select the optimal partition attribute.

The gain ratio is defined as:

wherein, the first and the second end of the pipe are connected with each other,

referred to as the "eigenvalue" of attribute a, the larger the number of possible values of attribute a (i.e., the larger V), the larger the value of IV (a) will typically be. However, the gain rate criterion favors the attribute with less number of possible values, so the C4.5 algorithm does not directly select the partition attribute candidate with the highest gain, but uses a heuristic: the attribute that the information gain is higher than the average level is found out from the candidate partition attributes, and then the attribute with the highest gain rate is selected.

The CART decision tree [ Breiman et al, 1984] uses the "Gini index" (Gini index) to select partition attributes. The purity of data set D can be measured by a kini value:

the Gini value Gini (D) reflects the probability that two samples, whose class labels are inconsistent, are randomly drawn from the data set D. Thus, the smaller the Gini value Gini (D), the higher the purity of the data set D. For a specific attribute a, its kini index is defined as:

in the candidate attribute set a, an attribute with the smallest post-partitioning kini index is selected as the optimal partitioning attribute, i.e., a × = arg minGain _ index (D, a).

In addition, the structure of the multivariate decision tree is a binary tree, which is actually a combination of multiple linear models: 1. from "tree" to "rule": a decision tree corresponds to a "rule set", and each branch path from a root node to a leaf node corresponds to a rule; 2. dividing the axes in parallel: if we regard each attribute as a coordinate axis in a coordinate space, the samples described by d attributes correspond to a data point in a d-dimensional space, and classifying the samples means finding classification boundaries between different types of samples in the coordinate space, and the decision boundary formed by the decision tree has an obvious characteristic: the axes are parallel (axis-parallel), i.e. its classification boundary is composed of several segments parallel to the coordinate axes.

1) Univariate decision tree: only one partition attribute is considered in each non-leaf node, an axis parallel classification surface is generated, and a classification boundary is searched. When the classification boundary corresponding to the learning task is complex, a good approximation can be obtained by very multi-segment division. If an oblique partition model can be used, the decision tree model is greatly simplified. This introduces a "multivariate decision tree".

2) Multivariate decision tree: each non-leaf node not only considers one attribute, such as an 'oblique decision tree' (object decision tree), but also establishes a linear classifier instead of finding the optimal partition attribute for each non-leaf node. Each non-leaf node of the multivariate decision tree is a judgment of a combination of multiple attributes. The combination mode of the attributes can be simple linear combination or complex combination, for example, a neural network is arranged for each non-leaf node to judge the complex combination of the attributes. Multivariate decision trees enable "skewed partitions" or even more complex partitioned decision trees. Instead of testing only a certain attribute, the non-leaf nodes are each a linear combination of \ sum _ { i =1} { d } w _ ia _ i = t.

Specifically, using the above data set, each node is examined, starting with the root node. When looking at node n, a linear regression model linear (the classification function that the regression model actually implements here, but other models, such as logistic regression, may also be used) is fitted with the data set D that node n owns (the root node owns all the training data). And dividing the data set D, and classifying the data set D into two classes (a positive class and a negative class, and the output can be represented as less than 0 or more than 0) according to the predicted output of the linear. Two sets D are obtained ^- And D ⁺ . Survey set D ^- Evaluating the accuracy of the evaluation by using linear, and if the accuracy is greater than or equal to a preset threshold value or D ^- If the node is empty, setting the left sub-node of the node nn as a leaf node, and marking the category as a negative category; if the precision is less than the threshold value, D is set ^- Copy to the left child node of node nnn, recursively explore the left child node. In the same way, consider set D ⁺ Evaluating the accuracy of the image by using linear, and if the accuracy is greater than or equal to a preset threshold value or D ⁺ If the node n is empty, setting the right sub-node of the node n as a leaf node, and marking the category as a positive category; if the precision is less than the threshold value, D is set ⁺ Copied to the right child of node n, recursively exploring the right child.

In step S104: and receiving the pain symptom information and the characteristic information input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output a pain disease distinguishing and classifying result of the patient. The patient can input the symptom information and the characteristic information of the pain on a terminal user interface and then submit the information, the system responds based on the information submitted by the patient, similarity calculation is carried out based on the pain disease distinguishing and classifying model, and the pain disease distinguishing and classifying result of the patient is obtained and output. For example, the pain disease discrimination and classification model deduces the pain disease discrimination and classification result, and proposes a corresponding medical conditioning and treatment scheme to be sent to the patient terminal, where the patient terminal may also be a doctor terminal.

For example, heart problems, mainly manifested as precordial pain, and also manifested as pain on the medial side of the left arm, occur. But often accompanied by other symptoms such as chest distress, dyspnea, purple lips, cold limbs, etc.; problems with the liver, manifested by discomfort in the right upper abdomen and right upper shoulder, may accompany yellowing of the skin, yellow staining of the sclera of the eye, yellow color in urine, sweat, tears; problems of the gallbladder are manifested as pain in the right upper abdomen and right shoulder, but also manifested as pain in the scapular region of the back; problems occur in the lung, pain sensation in the chest can occur, and common symptoms are as follows: cough, expectoration, hemoptysis, bloody sputum and dyspnea, and lung diseases without position are different in expression, but chronic cough, dyspnea and incapability of activities are caused in the early stage; the major problem of appendicitis is right lower abdominal pain. Nausea, vomiting, diarrhea, etc. may also be accompanied; problems in the intestines and stomach are manifested by pain in the abdomen, and besides, symptoms such as halitosis, flatulence in the lower abdomen, frequent diarrhea, inappetence and vomiting are also manifested; problems in ureters are mainly manifested by pain in the groin area and on both sides of the lower abdomen. Urinary calculi not only feel pain, but also have the condition of hematuria; problems with other internal organs such as the pancreas, mainly manifested in left upper abdominal pain; the ovaries are problematic and mainly manifested as pain in the left, right, middle, lower abdomen, etc.

It should be noted that, in the conventional C4.5 algorithm, the attribute with the largest information gain rate is selected as the classification attribute at each node, but when the information gain rate of the attribute with the largest information gain rate is very close to the information gain rate of some other attributes, it is not necessarily optimal to separately take the attribute with the largest information gain rate as the classification attribute. The algorithm of the invention combines the attributes by utilizing some heuristic information to form a conjunction expression which is used as the classification attribute of the current node to obtain a decision tree with the same classification effect but smaller scale.

Therefore, according to the pain disease distinguishing and classifying method based on the multivariate decision tree model, the sample data set is obtained, and the associated characteristic information of the pain disease in the sample data set is collected; carrying out normalization processing on the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and establishing a disease knowledge base associated with the characteristic information set; inputting semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multivariate decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model; and receiving the symptom information and the characteristic information of the pain input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output a pain disease distinguishing and classifying result of the patient. The problem that diagnosis of painful diseases is difficult in the related technology is solved, and the following beneficial effects are achieved: through the application of the multivariate decision tree model, the pain diseases are intelligently distinguished and classified, corresponding conditioning and treatment schemes are provided, doctors are assisted to improve the diagnosis and treatment efficiency, and the misdiagnosis rate of the pain diseases is reduced.

Fig. 3 shows a schematic diagram of main modules of the pain disease distinguishing and classifying device based on the multivariate decision tree model provided in the embodiment of the present application, and for convenience of illustration, only the parts related to the embodiment of the present application are shown, which are detailed as follows:

a pain disease discriminating and classifying device 200 based on multivariate decision tree model, comprising:

the feature acquisition module 201: the system comprises a data acquisition module, a data acquisition module and a data processing module, wherein the data acquisition module is used for acquiring a sample data set and acquiring associated characteristic information of pain diseases in the sample data set;

the feature extraction module 202: the system is used for carrying out normalization processing on the associated characteristic information to obtain a characteristic information set of the associated characteristic information and establishing a disease knowledge base associated with the characteristic information set;

the model building module 203: the semantic feature information subset is used for inputting semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model to perform pre-training to obtain semantic feature information subsets of the associated feature information, and the multivariate decision tree model is trained through the semantic feature information subsets to obtain a pain disease distinguishing and classifying model;

the cognitive recognition module 204: and the system is used for receiving the pain symptom information and the characteristic information input by the patient, carrying out similarity calculation based on the pain disease distinguishing and classifying model, obtaining and outputting a pain disease distinguishing and classifying result of the patient.

The characteristic acquisition module 201 is configured to acquire a sample data set and acquire associated characteristic information of pain and disease in the sample data set. By collecting the associated characteristic information of the pain diseases in the sample data set, a plurality of pieces of historical information of different types of pain diseases can be known, and subsequent data analysis and model training are facilitated. The sample data set is a diagnosis report of a plurality of pain disease patients collected in advance, and can be stored in a text form or a data table form and directly accessed and obtained by calling a data interface. Here, the related characteristic information includes text information of the sample data and information of the data class, for example, the text information may be a disease name, a symptom, and the like of the pain disease of the patient, and the information of the data class may be a disease onset time, a duration, and the like of the pain disease of the patient.

Illustratively, the name of the pain disorder, the cause of the disease, the site of occurrence, the typical symptoms, the accompanying symptoms, the concurrent symptoms, the degree of pain, the exacerbation or alleviation factors of the pain, and the time, duration of the occurrence, etc. may include the following relevant information: 1) Pain parts are mostly affected, such as precordial region and poststernal pain, which are mainly angina pectoris or myocardial infarction, pleurisy, right epigastric pain which is cholelithiasis, cholecystitis, liver abscess, and central nerve injury pain parts are dispersive and wide in range; 2) The nature and degree of pain are etiologically related, for example, dull or dull pain is visceral pain, myocardial infarction and arterial dissection are often manifested as severe laceration-like pain; 3) The duration of pain, which may be manifested as paroxysmal, persistent, spontaneous, etc.; 4) With symptoms: nervous system diseases, such as pain accompanied by emesis, vertigo, epilepsy, and visual disturbance, such as cerebral hemorrhage, vertebrobasilar artery insufficiency, cerebrovascular malformation, and glaucoma; disorders of the locomotor system, manifested by pain associated with abnormal limb movement such as fracture, arthralgia associated with erythema, photosensitivity, systemic lupus erythematosus, etc.; cardiovascular system diseases, manifested by chest pain with pale, profuse sweating, blood pressure drop or shock, such as myocardial infarction, and dissecting aneurysm; respiratory system diseases, manifested as chest pain with cough, expectoration, and hemoptysis, such as lobar pneumonia and lung cancer; digestive and urinary system diseases, manifested by abdominal pain with acid regurgitation, emesis, diarrhea, jaundice, such as gastrointestinal obstruction, gastroduodenal ulcer, intestinal inflammation, and liver, gallbladder and pancreas diseases; lumbago with frequent micturition, urgent micturition, incomplete urination, hematuria, such as urinary tract infection and ureteral calculus. The above is merely an example description of the associated features of a pain disorder to facilitate understanding of the present solution. In the practical embodiment, it is not limited thereto.

The feature extraction module 202 is configured to perform normalization processing on the associated feature information to obtain a feature information set of the associated feature information, and establish a disease knowledge base associated with the feature information set. The characteristic information set is obtained by carrying out normalization processing on the associated characteristic information, the associated characteristics of various pain diseases can be more conveniently subjected to statistical analysis, and the disease knowledge base associated with the characteristic information set is established, so that the disease knowledge base can be continuously updated, namely, sample data is continuously collected by the above mode, and then the disease knowledge base is continuously enriched so as to train a more accurate prediction model.

In one embodiment, the normalizing the associated feature information to obtain the feature information set of the associated feature information includes: and performing normalization processing on the associated characteristic information based on a word frequency-inverse document frequency algorithm, and extracting key characteristic words of the associated characteristic information to obtain a characteristic information set of the associated characteristic information. Here, the correlation characteristic information is normalized through a TF-IDF (term frequency-inverse document frequency) technology, common terms in the correlation characteristic information can be filtered out, and important terms are retained, so that the formed key characteristic terms can accurately express the characteristics of the pain disease.

The model construction module 203 is configured to input semantic feature information corresponding to the disease knowledge base into a multivariate decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and train the multivariate decision tree model through the semantic feature information subset to obtain a pain disease discrimination classification model.

And the cognitive recognition module 204 is configured to receive the pain symptom information and the feature information input by the patient, perform similarity calculation based on the pain disease discrimination and classification model, obtain a pain disease discrimination and classification result of the patient, and output the result. The patient can input the symptom information and the characteristic information of the pain on a terminal user interface and then submit the information, the system responds based on the information submitted by the patient, similarity calculation is carried out based on the pain disease distinguishing and classifying model, and the pain disease distinguishing and classifying result of the patient is obtained and output. For example, the pain disease discrimination and classification model deduces the pain disease discrimination and classification result, and proposes a corresponding medical conditioning and treatment scheme to send to the patient terminal, where the patient terminal may also be a doctor terminal.

For example, heart problems, mainly manifested as precordial pain, and also manifested as pain on the medial side of the left arm, occur. But often accompanied by other symptoms such as chest distress, dyspnea, purple lips, cold limbs, etc.; problems with the liver, manifested by discomfort in the right upper abdomen and right upper shoulder, may accompany yellowing of the skin, yellow staining of the sclera of the eye, yellow color in urine, sweat, tears; problems of the gallbladder are manifested as pain in the right upper abdomen and right shoulder, but also manifested as pain in the scapular region of the back; problems occur in the lung, pain sensation in the chest can occur, and common symptoms are as follows: cough, expectoration, hemoptysis, bloody sputum and dyspnea, and lung diseases without position are different in expression, but chronic cough, dyspnea and incapability of activities are caused in the early stage; the major problem of appendicitis is right lower abdominal pain. Nausea, vomiting, diarrhea, etc. may also occur; problems of the intestines and stomach are manifested as pain in the abdomen, and besides, symptoms such as halitosis, flatulence in the lower abdomen, frequent diarrhea, inappetence, vomiting and the like are also manifested; the ureter has problems mainly manifested as pain in the groin area and on both sides of the lower abdomen. Urinary calculi not only feel pain, but also have the condition of hematuria; problems with other internal organs such as the pancreas, mainly manifested in left upper abdominal pain; the ovaries are problematic and mainly manifested as pain in the left, right, middle, lower abdomen, etc.

Therefore, the pain disease distinguishing and classifying device based on the multivariate decision tree model, provided by the embodiment of the application, solves the problem that the diagnosis of pain diseases in the related technology is difficult, and realizes the following beneficial effects: through the application of the multivariate decision tree model, the pain diseases are intelligently distinguished and classified, corresponding conditioning and treatment schemes are provided, a doctor is assisted to improve the diagnosis and treatment efficiency, and the misdiagnosis rate of the pain diseases is reduced.

An embodiment of the present application further provides an electronic device, including: one or more processors; a storage device for storing one or more programs which, when executed by one or more processors, cause the one or more processors to implement the multivariate decision tree model based pain disorder discrimination and classification methods of embodiments of the present application

Embodiments of the present application also provide a computer readable medium, on which a computer program is stored, which when executed by a processor, implements the multivariate decision tree model-based pain disease identification and classification method of embodiments of the present application.

Fig. 3 illustrates an exemplary system architecture 300 to which the multivariate decision tree model based pain disorder discrimination and classification method or apparatus of embodiments of the present application can be applied.

As shown in fig. 3, the system architecture 300 may include

terminal devices

301, 302, 303, a network 304, and a server 305. The network 304 serves as a medium for providing communication links between the

terminal devices

301, 302, 303 and the server 305. Network 304 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal device

301, 302, 303 to interact with the server 305 via the network 304 to receive or send messages or the like. The

terminal devices

301, 302, 303 may have various communication client applications installed thereon, such as a shopping application, a web browser application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.

The

terminal devices

301, 302, 303 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 305 may be a server providing various services, such as a background management server providing support for users using incoming and outgoing messages sent by the

terminal devices

301, 302, 303. The background management server can perform analysis and other processing after receiving the terminal device request, and feed back the processing result to the terminal device.

It should be noted that the pain disease determination and classification method based on the multivariate decision tree model provided in the embodiment of the present application is generally executed by the

terminal device

301, 302, 303 or the server 305, and accordingly, the pain disease determination and classification apparatus based on the multivariate decision tree model is generally disposed in the

terminal device

301, 302, 303 or the server 305.

It should be understood that the number of terminal devices, networks, and servers in fig. 3 are merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 4, shown is a block diagram of a computer system 400 suitable for use in implementing the electronic device of an embodiment of the present application. The computer system illustrated in FIG. 4 is only an example and should not impose any limitations on the functionality or scope of use of embodiments of the present application.

As shown in fig. 4, the computer system 400 includes a Central Processing Unit (CPU) 401 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 402 or a program loaded from a storage section 408 into a Random Access Memory (RAM) 403. In the RAM 403, various programs and data necessary for the operation of the system 400 are also stored. The CPU 401, ROM 402, and RAM 403 are connected to each other via a bus 404. An input/output (I/O) interface 405 is also connected to bus 404.

The following components are connected to the I/O interface 405: an input section 406 including a keyboard, a mouse, and the like; an output section 407 including a display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 408 including a hard disk and the like; and a communication section 409 including a network interface card such as a LAN card, a modem, or the like. The communication section 409 performs communication processing via a network such as the internet. A driver 410 is also connected to the I/O interface 405 as needed. A removable medium 411 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 410 as needed, so that a computer program read out therefrom is mounted in the storage section 408 as needed.

In particular, according to embodiments disclosed herein, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, embodiments disclosed herein include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 409 and/or installed from the removable medium 411. The above-described functions defined in the system of the present application are executed when the computer program is executed by a Central Processing Unit (CPU) 401.

It should be noted that the computer readable medium shown in the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present application may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a determination module, an extraction module, a training module, and a screening module. Where the names of these modules do not in some cases constitute a limitation of the module itself, for example, a determination module may also be described as a "module that determines a set of candidate users".

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.

The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A pain disease distinguishing and classifying method based on a multivariate decision tree model is characterized by comprising the following steps:

normalizing the associated characteristic information to obtain a characteristic information set of the associated characteristic information, and establishing a disease knowledge base associated with the characteristic information set;

and receiving the pain symptom information and the characteristic information input by the patient, and performing similarity calculation based on the pain disease distinguishing and classifying model to obtain and output a pain disease distinguishing and classifying result of the patient.

2. The multivariate decision tree model based pain disorder discrimination and classification method according to claim 1, wherein the associated feature information comprises at least one of: the name of the pain disorder, the cause of the disease, the site of occurrence, the typical symptoms, the accompanying symptoms, the concurrent symptoms, the degree of pain, the exacerbation or remission factor of the pain, and the time and duration of the occurrence.

3. The method of claim 1, wherein the normalizing the associated feature information to obtain the feature information set of the associated feature information comprises: and performing normalization processing on the associated characteristic information based on a word frequency-inverse document frequency algorithm, and extracting key characteristic words of the associated characteristic information to obtain a characteristic information set of the associated characteristic information.

4. The method of claim 3, wherein the normalization processing is performed on the associated feature information based on a word frequency-inverse document frequency algorithm, and the extracting key feature words of the associated feature information comprises: and calculating a TF-IDF value of each word in the text information and data samples through a word frequency-inverse document frequency algorithm, arranging each word in a descending order based on the TF-IDF value, and extracting N words arranged at the top as the key feature words, wherein N is a positive integer.

5. The method of claim 3, wherein before the semantic feature information corresponding to the disease knowledge base is input into the multivariate decision tree model for pre-training to obtain the semantic feature information subset of the associated feature information, the method further comprises: and converting the disease knowledge base into corresponding semantic feature information through a Bayesian network technology.

6. A pain disease distinguishing and classifying device based on a multivariate decision tree model is characterized by comprising:

the model construction module is used for inputting semantic feature information corresponding to the disease knowledge base into a multi-variable decision tree model for pre-training to obtain a semantic feature information subset of the associated feature information, and training the multi-variable decision tree model through the semantic feature information subset to obtain a pain disease distinguishing and classifying model;

7. The apparatus for pain diagnosis and classification based on multivariate decision tree model according to claim 6, wherein the normalization process of the associated feature information to obtain the feature information set of the associated feature information comprises: and performing normalization processing on the associated characteristic information based on a word frequency-inverse document frequency algorithm, extracting key characteristic words of the associated characteristic information, and obtaining a characteristic information set of the associated characteristic information.

8. The multivariate decision tree model-based pain disease discrimination and classification device according to claim 7, wherein the normalization processing of the associated feature information based on the word frequency-inverse document frequency algorithm is performed, and extracting key feature words of the associated feature information comprises: and calculating a TF-IDF value of each word in the text information and data samples through a word frequency-inverse document frequency algorithm, arranging each word in a descending order based on the TF-IDF value, and extracting N words arranged at the top as the key feature words, wherein N is a positive integer.

9. An electronic device, comprising a memory and a processor, the memory having stored therein a computer program which, when executed by the processor, causes the processor to carry out the steps of the multivariate decision tree model based pain disorder discrimination and classification method according to any one of claims 1-5.

10. A computer-readable storage medium, having stored thereon a computer program which, when executed by a processor, causes the processor to carry out the steps of the multivariate decision tree model based pain disorder discrimination and classification method according to any one of claims 1-5.