CN106951917A - The intelligent classification system and method for a kind of lymthoma histological type - Google Patents
The intelligent classification system and method for a kind of lymthoma histological type Download PDFInfo
- Publication number
- CN106951917A CN106951917A CN201710112852.3A CN201710112852A CN106951917A CN 106951917 A CN106951917 A CN 106951917A CN 201710112852 A CN201710112852 A CN 201710112852A CN 106951917 A CN106951917 A CN 106951917A
- Authority
- CN
- China
- Prior art keywords
- lymthoma
- information
- feature
- pathology
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The present invention relates to a kind of intelligent classification system of lymthoma histological type and method, doctor workstation is used to receive input information and display classification results;Server includes data obtaining module, pretreatment module, model training module, information receiving module, lymthoma pathology grader.Wherein, data obtaining module obtains electronic data, therefrom obtains training set;Pretreatment module is used to extract feature pathological information, and information is pre-processed, and is characterized word generation character pair numerical value;Model training module utilizes character numerical value, trains classification analysis model, obtains housebroken lymthoma pathology grader;Information receiving module receives the information that user is inputted by doctor workstation, and transfers information to pretreatment module;Housebroken lymthoma pathology grader draws lymthoma pathology classification of type result.The present invention realizes mechanized classification, saves manpower and independent of doctor's subjective judgement.
Description
Technical field
The present invention relates to the technical field of machine learning, more particularly to a kind of intelligent classification system of lymthoma histological type
And method.
Background technology
At present, the big data epoch are in, the data for the large-scale data amount that grows on trees, simple rule of the prior art
Processing is difficult to the value for playing these data.The high speed development of hardware provides condition to the application of big data.High-performance calculation
So that data learning time and data processing cost based on large-scale data greatly reduce;Mass data storage so that
Can faster, cost smaller handle large-scale data.Due to the development of hardware and algorithm so that solved using machine learning
After the problem of data analysis, abundanter income can be obtained.
Existing machine learning techniques mainly apply to the internet arenas such as news, ecommerce, in traditional field, especially
Be medical domain application it is very deficient.Therefore, the present invention provides one kind and applies to machine learning method in medical domain, enters
The intelligent method for classifying of row lymthoma histological type, to make up in the prior art, the classification of lymthoma relies primarily on the master of doctor
See and judge, the accuracy of judgement depends on the working experience of doctor, different doctors may return to the judged result of conditions of patients
Inconsistent, this is easily caused mistaken diagnosis, in some instances it may even be possible to the problem of being delayed conditions of patients.
The content of the invention
In view of above-mentioned analysis, the present invention is intended to provide the intelligent classification system and method for a kind of lymthoma histological type,
The problem of to solve labor intensive, rely on doctor's subjective judgement.
The purpose of the present invention is mainly achieved through the following technical solutions:
A kind of intelligent classification system of lymthoma histological type, it is characterised in that including doctor workstation and server,
Doctor workstation is used to receive input information and display classification results;
Server includes data obtaining module, pretreatment module, model training module, information receiving module, lymphomatosis
Manage grader;
Data obtaining module obtains the electronic data for the case for being diagnosed as lymthoma, therefrom obtains training set;
Pretreatment module extracts the feature pathological information of each case in training set, by the feature pathological information of extraction
It is corresponding with the Classification of Lymphoma result judged in advance one by one;The pathological information of extraction is pre-processed again, generation training is special
Set of words is levied, and corresponding character numerical value is generated for the Feature Words in set, character numerical value is input to model training module;In advance
Processing module is additionally operable to carry out user input information the extraction of feature pathological information, pretreatment and generation character numerical value, and should
Character numerical value is input to trained lymthoma pathology grader;
Model training module utilizes character numerical value, trains classification analysis model, obtains trained lymthoma pathology point
Class device;
Information receiving module receives the information that user is inputted by doctor workstation, and the information transfer of reception is located to pre-
Manage module;
Trained lymthoma pathology grader handles the characteristic that user input information is obtained according to pretreatment module
Value, draws lymthoma pathology classification of type result, and be output to doctor workstation.
The feature pathological information that the pretreatment module is extracted includes:The pathology title of lympha tumour, pathology description, pathology
Ownership classification.
Pretreatment in the pretreatment module includes:Participle is carried out to every information in the pathological information of extraction, obtained
Obtain text word set;And noise word rejecting processing is carried out to word segmentation result.
After noise word rejecting processing, also including being replaced to the synonym in feature set of words.
It is preferred that, noise word rejecting processing is matched and rejected by way of setting up noise vocabulary.
The pretreatment module also includes carrying out feature selecting with the Feature Words in feature set of words to training;And use
Hash algorithm is that the Feature Words in training feature set of words generate corresponding character numerical value.
The classification analysis model is based on algorithm of support vector machine.
Further, server also includes test module, for being surveyed to trained lymthoma pathology grader
Examination;When test result is undesirable, change Method of The Classification Analysis or the quantity for changing feature selecting, with mould of classifying to lymphomatosis
Type is adjusted;Lymthoma pathology grader is tested based on the lymphomatosis disaggregated model regained, then by test module;
By constantly adjusting lymphomatosis disaggregated model and being tested, until test result is up to standard.
The present invention also provides a kind of intelligent method for classifying of lymthoma histological type, comprises the following steps:
Step S1. obtains the electronic data for the case for being diagnosed as lymthoma, and training set is obtained from electronic data;
Step S2. extract training set in each case feature pathological information, by the feature pathological information of extraction by
It is individual corresponding with Classification of Lymphoma result that is judging in advance;
Step S3. is pre-processed to the pathological information of extraction, generates training feature set of words;
Step S4. is that the Feature Words in training feature set of words generate corresponding character numerical value;
Step S5. utilizes the character numerical value that previous step is generated, and trains classification analysis model, obtains lymthoma pathological classification
Device;
Step S6. receives user input, description information to pathology;
Step S7. carries out the extraction of feature pathological information, pretreatment and the processing of generation character numerical value to input information;
Step S8. lymthoma pathology graders draw the pouring of user input information according to the character numerical value of information to be sorted
Bar knurl disease classification results.
The present invention has the beneficial effect that:
By building lymthoma pathology grader, the artificial cost that data analyze with classification can be saved, is directly passed through
Computer program carries out the intelligent classification of lymthoma pathology, obtains corresponding classification results data, it is no longer necessary to largely manually go
Macro or mass analysis;And independent of the subjective judgement of doctor, be conducive to helping doctor to be diagnosed.
Other features and advantages of the present invention will be illustrated in the following description, also, the partial change from specification
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by the explanations write
Specifically noted structure is realized and obtained in book, claims and accompanying drawing.
Brief description of the drawings
Accompanying drawing is only used for showing the purpose of specific embodiment, and is not considered as limitation of the present invention, in whole accompanying drawing
In, identical reference symbol represents identical part.
Fig. 1 is the schematic diagram of lymthoma histological type intelligent classification system;
Fig. 2 is the flow chart of the intelligent method for classifying of lymthoma histological type.
Embodiment
The preferred embodiments of the present invention are specifically described below in conjunction with the accompanying drawings, wherein, accompanying drawing constitutes the application part, and
It is used for the principle for explaining the present invention together with embodiments of the present invention.
The specific embodiment of the present invention, discloses a kind of intelligent classification system of lymthoma histological type, such as Fig. 1 institutes
Show, including:Including doctor workstation and server,
Doctor workstation is used to receive input information and display classification results;
Server further comprises data obtaining module, pretreatment module, model training module, information receiving module, pouring
Bar knurl pathological classification device.
Data obtaining module, the electronic data for obtaining the case for being diagnosed as lymthoma, is trained from electronic data
Set, and store to memory cell.The electronic data derives from the electronic text of passing papery case history, existing electronic health record
With the Research statistics text of hospital internal.
It is preferred that, it is specifically that electronic data is divided into training set and test set to obtain training set from electronic data
Close.Test, which is integrated into, to be obtained after lymthoma pathology grader, the test for carrying out grader accuracy.
Pretreatment module, the feature pathological information for extracting each case in training set, by the feature disease of extraction
Manage information corresponding with the Classification of Lymphoma result judged in advance one by one;The pathological information of extraction is pre-processed again, generation instruction
Practice and use feature set of words, and corresponding character numerical value is generated for the Feature Words in set, and be input to model training module.
Pretreatment module is additionally operable to carry out user input information the extraction of feature pathological information, pretreatment and generation characteristic
Value, and this feature numerical value is input to trained lymthoma pathology grader.
Wherein, the feature pathological information specifically extracted includes:The pathology title of lympha tumour, pathology description, disease in case
Reason ownership classification etc..
Further, after the corresponding lymthoma pathological information of each case in extracting electronic data, removal is not inconsistent
The problem of conjunction is required data, such as null value, substantially problematic data, be not inconsistent logical data.
It is that the pathological information of extraction is parsed into entry, each of which disease to the pretreatment that the pathological information of extraction is carried out
The multiple entries of example correspondence.
Further, pretreatment includes the operations such as participle, noise word are rejected, synonym is handled, and obtains training Feature Words
Set, specifically includes following step:
1. carrying out participle to every information in the pathological information of extraction using participle instrument, text word set is obtained.If
Text is Chinese, then using Chinese word segmentation machine;If English, then participle is carried out using space, and after the completion of English string segmentation
The mode extracted using stem normalizes tense and single plural number.
Specifically, ICTCLAS (Institute of Computing Technology, Chinese can be used
Lexical Analysis System, Chinese lexical analysis system) and the participle instrument such as IK Analyzer (IK segmenter) make
For Chinese word segmentation machine.
2. pair word segmentation result carries out noise word rejecting processing, word or word of the removal without practical significance are specifically included, such as
Although ", not only but also, still " etc., and some rarely used words and additional character also remove the information with lympha tumour
Unrelated vocabulary.
Further, noise word rejecting processing can be matched and rejected by way of setting up noise vocabulary.
3. the synonym in feature set of words is replaced using the synonym table pre-established so that all synonyms
Represented with a word, obtain training feature set of words.
In the step for after pretreatment, further comprising carrying out training with Feature Words in feature set of words feature selecting
Suddenly.If Feature Words extract excessively, characteristic dimension can be caused too high, the problem of being unfavorable for classifier training.
Corresponding character numerical value is generated for the Feature Words in training feature set of words, the character numerical value obtained using calculating
To carry out character representation to entry.It is preferred that, specifically hash algorithm is used for each pathology in training feature set of words
Feature Words generate corresponding character numerical value.
Model training module is trained classification analysis model, is obtained trained lymthoma using the character numerical value of generation
Pathological classification device.
The classification analysis refers to being grouped into the set of physics or abstract object into the multiple classes being made up of similar object
Analysis process, the purpose is to collect data on the basis of similar to classify.The classification analysis model used in the present invention, can
With using following several:
Naive Bayesian 1. (Naive Bayes, NB) disaggregated model
If conditional independence assumption is set up, NB will be more convergent faster than discriminating model (such as Logistic recurrence), institute
To only need to a small amount of training data.Even if conditional independence assumption is invalid, NB still can obtain preferable result in practice.
2.Logistic returns (Logistic Regression, LR) model
LR has many methods to come to model regularization.Conditional independence assumption compared with NB, LR is without the concern for sample
No is related.Different with SVMs (SVM) from decision tree, NB has good probability interpretation, and is readily available new
Training data carrys out more new model (using online gradient descent method).If necessary to probabilistic information (e.g., for easier adjustment point
Class threshold value, the uncertainty classified, obtains confidential interval), it is desirable alternatively to have can easily be updated during more data in the future
Improved model, then can use LR models.
Decision tree 3. (Decision Tree, DT) model
DT models be it is non-parametric, it is no need to worry wild point (or outlier) and data whether linear separability the problem of (example
As DT models can easily handle such case:The feature x values for belonging to the sample of A classes are often very small or very big,
And belong to the feature x values of the sample of B classes in intermediate range).But, the major defect of DT models is easy over-fitting.
4. SVMs (Support Vector Machine, SVM)
SVMs has high classification accuracy rate, has theoretical guarantee well to over-fitting, chooses suitable core letter
Number, can also put up a good show in face of the inseparable problem of characteristic line.SVM is adapted to the higher text classification of dimension.
Lymphomatosis is carried out 3 grades of classification by the present embodiment, and its one-level is than a fraction in further detail.According to test result,
It is preferred that SVMs (SVM) algorithm carries out the intelligent classification of lymphomatosis.
Further, the intelligent classification system of lymthoma histological type includes test module, for being drenched to trained
Bar knurl pathological classification device is tested.It is specific to utilize the method used in pretreatment module (method of processing training set) processing
Test set, by according to test set symphysis into test lymphomatosis disaggregated model, and and standard knots are input to character numerical value
Fruit is contrasted, and obtains test result.
Further, in the case where test module is undesirable to the test result of lymthoma pathology grader, change point
Alanysis algorithm or the quantity for changing feature selecting, to be adjusted to lymphomatosis disaggregated model.New pouring will be regained
Bar knurl disease disaggregated model, then carry out the test of lymthoma pathology grader.By constantly adjusting lymphomatosis disaggregated model, Zhi Daoshi
Test result up to standard.
Information receiving module, for receiving user input, description information to pathology, and the information transfer of reception is arrived
Pretreatment module.The description information of the input can be multiple sentences, phrase, entry, numerical value, number range or sentence, short
Language, entry, the combination of numerical value;Wherein sentence, phrase and entry can be separated by branch.
Pretreatment module carries out the extraction of feature pathological information, pretreatment and generation character numerical value to the input information of user
Processing, processing method obtains the feature set of words of information to be sorted and corresponding with the process step in above-mentioned pretreatment module
Character numerical value.
Trained lymthoma pathology grader handles the characteristic that user input information is obtained according to pretreatment module
Value, draws lymthoma pathology classification of type result, and be output to doctor workstation.
By data obtaining module, pretreatment module, model training module, information receiving module, lymthoma pathology grader,
Test module is packaged, and storage is called by doctor workstation, used in the server, for user.
The present invention also provides another specific embodiment, a kind of intelligent method for classifying of lymthoma histological type, such as Fig. 2 institutes
Show, comprise the following steps:
Step S1. obtains the electronic data for the case for being diagnosed as lymthoma, and training set is obtained from electronic data.
Wherein, from electronic data obtain training set method be:Electronic data is divided into training set and test set
Close.Test, which is integrated into, to be obtained after lymthoma pathology grader, the test for carrying out grader accuracy.
Step S2. extract training set in each case feature pathological information, by the feature pathological information of extraction by
It is individual corresponding with Classification of Lymphoma result that is judging in advance.
Wherein, the feature pathological information of extraction includes:The pathology title of lympha tumour, pathology description, pathology are returned in case
Category classification etc..
Further, after the corresponding lymthoma pathological information of each case in extracting electronic data, removal is not inconsistent
The problem of conjunction is required data, such as null value, substantially problematic data, be not inconsistent logical data.
Step S3. is pre-processed to the pathological information of extraction, generates training feature set of words.
Wherein, pretreatment is that the pathological information of extraction is parsed into entry, the multiple entries of each of which case correspondence.
Pretreatment further comprises the operations such as participle, noise word are rejected, synonym is handled, and obtains training feature word set
Close, specifically include following step:
Participle is carried out to every information in the pathological information of extraction using participle instrument, text word set is obtained;
The rejecting of noise word is carried out to word segmentation result to handle, specifically include word or word of the removal without practical significance, such as again
Although ", not only but also, still " etc., and some rarely used words and additional character also remove the information with lympha tumour
Unrelated vocabulary.It is preferred that, noise word rejecting processing can be matched and rejected by way of setting up noise vocabulary;
The synonym in feature set of words is replaced using the synonym table pre-established so that all synonyms are equal
Represented with a word, obtain training feature set of words.
Further, after pretreatment, also including carrying out feature selecting with Feature Words in feature set of words to training
The step of.If Feature Words extract excessively, characteristic dimension can be caused too high, the problem of being unfavorable for classifier training.
Step S4. is that the Feature Words in training feature set of words generate corresponding character numerical value.It is preferred that, using Hash
Algorithm generates corresponding character numerical value for the Feature Words of each pathology in training feature set of words.
Step S5. utilizes the character numerical value that previous step is generated, and trains classification analysis model, obtains lymthoma pathological classification
Device.
The classification analysis model used in the present invention, can be classified mould using naive Bayesian (Naive Bayes, NB)
Type, Logistic return (Logistic Regression, LR) model, decision tree (Decision Tree, DT) model, support
Vector machine (Support Vector Machine, SVM) etc..According to test result, SVMs (SVM) algorithm is preferably based on
Carry out the intelligent classification of lymphomatosis.
The above-mentioned test to lymthoma pathology grader, refers to using the method process test collection described in step S2~S4
Close, by according to test set symphysis into test be input to lymphomatosis disaggregated model with character numerical value, and carried out with standard results
Contrast, obtains test result.
Further, in the case where the test result of lymthoma pathology grader is undesirable, Method of The Classification Analysis is changed
Or the quantity of change feature selecting, to be adjusted to lymphomatosis disaggregated model.New lymphomatosis classification will be regained
Model, then carry out the test of lymthoma pathology grader.By constantly adjusting lymphomatosis disaggregated model, until experimental result reaches
Mark.
Step S6. receives user input, description information to pathology.
Step S7. carries out the extraction of feature pathological information, pretreatment and the processing of generation character numerical value, processing to input information
Method obtains the feature set of words and corresponding character numerical value of information to be sorted with step S2~S4.
Step S8. lymthoma pathology graders draw the pouring of user input information according to the character numerical value of information to be sorted
Bar knurl disease classification results.
In the present embodiment, the electronic data obtained in step S1 is divided into training set and test set.In training classification point
Model is analysed, obtains after lymphomatosis disaggregated model, test set is carried out after the operation such as feature extraction, Feature Selection, is input to
In model after training, the accuracy rate of three-level classification is obtained.In theory, the thinner data for needing to support are classified more.This reality
The test result for applying example meets our expection.If test result is too big with anticipated deviation, need to consider to be data
Source is out of joint, the problem of Feature Selection or algorithmic issue, now needs correspondence to change.
In summary, the embodiments of the invention provide a kind of intelligent classification system of lymthoma histological type and method, lead to
Structure lymthoma pathology grader is crossed, the artificial cost that data analyze with classification can be saved, directly pass through computer program
The intelligent classification of lymthoma pathology is carried out, corresponding classification results data are obtained, classification automation, standardization is realized, no longer needs
Largely manually to remove Macro or mass analysis.In the case where tackling mass data, cost of labor can be greatly reduced using the present invention.
In addition, the characteristics of intelligent method for classifying for the lymthoma histological type that the present invention is provided also has unitized, defeated
Enter information it is consistent in the case of, classification results will not vary with each individual, will not be different because of the personal experience of doctor and result is different.Pin
The classification of illness can be provided to the patient of relative symptom, the diagnosis to doctor provides reference, is so advantageously examined in disease
Disconnected accuracy.
It will be understood by those skilled in the art that realizing all or part of flow of above-described embodiment method, meter can be passed through
Calculation machine program instructs the hardware of correlation to complete, and described program can be stored in computer-readable recording medium.Wherein, institute
It is disk, CD, read-only memory or random access memory etc. to state computer-readable recording medium.
The foregoing is intended to be a preferred embodiment of the present invention, but protection scope of the present invention is not limited thereto,
Any one skilled in the art the invention discloses technical scope in, the change or replacement that can be readily occurred in,
It should all be included within the scope of the present invention.
Claims (10)
1. a kind of intelligent classification system of lymthoma histological type, it is characterised in that including doctor workstation and server,
Doctor workstation is used to receive input information and display classification results;
Server includes data obtaining module, pretreatment module, model training module, information receiving module, lymthoma pathology point
Class device;
Data obtaining module obtains the electronic data for the case for being diagnosed as lymthoma, therefrom obtains training set;
Pretreatment module extracts the feature pathological information of each case in training set, by the feature pathological information of extraction one by one
It is corresponding with the Classification of Lymphoma result judged in advance;The pathological information of extraction is pre-processed again, training Feature Words are generated
Set, and corresponding character numerical value is generated for the Feature Words in set, character numerical value is input to model training module;Pretreatment
Module is additionally operable to carry out user input information feature pathological information extraction, pretreatment and generation character numerical value, and by this feature
Numerical value is input to trained lymthoma pathology grader;
Model training module utilizes character numerical value, trains classification analysis model, obtains trained lymthoma pathology grader;
Information receiving module receives the information that is inputted by doctor workstation of user, and by the information transfer of reception to pre-processing mould
Block;
Trained lymthoma pathology grader handles the character numerical value that user input information is obtained according to pretreatment module, obtains
Go out lymthoma pathology classification of type result, and be output to doctor workstation.
2. the intelligent classification system of lymthoma histological type according to claim 1, it is characterised in that the pretreatment mould
The feature pathological information that block is extracted includes:The pathology title of lympha tumour, pathology description, pathology ownership classification.
3. the intelligent classification system of lymthoma histological type according to claim 1, it is characterised in that the pretreatment mould
Pretreatment in block includes:Participle is carried out to every information in the pathological information of extraction, text word set is obtained;And to participle
As a result noise word rejecting processing is carried out.
4. the intelligent classification system of lymthoma histological type according to claim 3, it is characterised in that rejected in noise word
After processing, also including being replaced to the synonym in feature set of words.
5. the intelligent classification system of lymthoma histological type according to claim 3, it is characterised in that the noise word is picked
Except processing is matched and rejected by way of setting up noise vocabulary.
6. the intelligent classification system of the lymthoma histological type according to claim 1 or 3, it is characterised in that the pre- place
Reason module also includes carrying out feature selecting with the Feature Words in feature set of words to training.
7. the intelligent classification system of lymthoma histological type according to claim 1, it is characterised in that the pretreatment mould
Block uses hash algorithm to generate corresponding character numerical value for the Feature Words in training feature set of words.
8. the intelligent classification system of lymthoma histological type according to claim 1, it is characterised in that the classification analysis
Model is based on algorithm of support vector machine.
9. the intelligent classification system of lymthoma histological type according to claim 1, it is characterised in that server also includes
Test module, for testing trained lymthoma pathology grader;When test result is undesirable, change classification
Parser or the quantity for changing feature selecting, to be adjusted to lymphomatosis disaggregated model;Based on the lymph regained
Knurl disease disaggregated model, then lymthoma pathology grader is tested by test module;By constantly adjusting lymphomatosis disaggregated model
And tested, until test result is up to standard.
10. a kind of intelligent method for classifying of lymthoma histological type, it is characterised in that comprise the following steps:
Step S1. obtains the electronic data for the case for being diagnosed as lymthoma, and training set is obtained from electronic data;
Step S2. extract training set in each case feature pathological information, by the feature pathological information of extraction one by one with
The Classification of Lymphoma result correspondence judged in advance;
Step S3. is pre-processed to the pathological information of extraction, generates training feature set of words;
Step S4. is that the Feature Words in training feature set of words generate corresponding character numerical value;
Step S5. utilizes the character numerical value that previous step is generated, and trains classification analysis model, obtains lymthoma pathology grader;
Step S6. receives user input, description information to pathology;
Step S7. carries out the extraction of feature pathological information, pretreatment and the processing of generation character numerical value to input information;
Step S8. lymthoma pathology graders draw the lymthoma of user input information according to the character numerical value of information to be sorted
Sick classification results.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710112852.3A CN106951917A (en) | 2017-02-28 | 2017-02-28 | The intelligent classification system and method for a kind of lymthoma histological type |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710112852.3A CN106951917A (en) | 2017-02-28 | 2017-02-28 | The intelligent classification system and method for a kind of lymthoma histological type |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106951917A true CN106951917A (en) | 2017-07-14 |
Family
ID=59468114
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710112852.3A Pending CN106951917A (en) | 2017-02-28 | 2017-02-28 | The intelligent classification system and method for a kind of lymthoma histological type |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106951917A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107833603A (en) * | 2017-11-13 | 2018-03-23 | 医渡云(北京)技术有限公司 | Electronic medical record document sorting technique, device, electronic equipment and storage medium |
CN108648822A (en) * | 2018-05-03 | 2018-10-12 | 中国医学科学院肿瘤医院 | A kind of esophageal cancer lymph is carried down the quantitative analysis results output method and device of shifting |
CN109585015A (en) * | 2019-01-21 | 2019-04-05 | 兰州大学 | A kind of lymthoma histological type intelligent classification system and method |
CN111681756A (en) * | 2020-05-29 | 2020-09-18 | 吾征智能技术(北京)有限公司 | Disease symptom prediction system based on sputum character cognition |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1477581A (en) * | 2003-07-01 | 2004-02-25 | �Ϻ���ͨ��ѧ | Predictive modelling method application to computer-aided medical diagnosis |
CN104517038A (en) * | 2014-12-30 | 2015-04-15 | 杨关乐 | Disease diagnosis system server principle |
CN104915561A (en) * | 2015-06-11 | 2015-09-16 | 万达信息股份有限公司 | Intelligent disease attribute matching method |
CN106095996A (en) * | 2016-06-22 | 2016-11-09 | 量子云未来(北京)信息科技有限公司 | Method for text classification |
-
2017
- 2017-02-28 CN CN201710112852.3A patent/CN106951917A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1477581A (en) * | 2003-07-01 | 2004-02-25 | �Ϻ���ͨ��ѧ | Predictive modelling method application to computer-aided medical diagnosis |
CN104517038A (en) * | 2014-12-30 | 2015-04-15 | 杨关乐 | Disease diagnosis system server principle |
CN104915561A (en) * | 2015-06-11 | 2015-09-16 | 万达信息股份有限公司 | Intelligent disease attribute matching method |
CN106095996A (en) * | 2016-06-22 | 2016-11-09 | 量子云未来(北京)信息科技有限公司 | Method for text classification |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107833603A (en) * | 2017-11-13 | 2018-03-23 | 医渡云(北京)技术有限公司 | Electronic medical record document sorting technique, device, electronic equipment and storage medium |
CN108648822A (en) * | 2018-05-03 | 2018-10-12 | 中国医学科学院肿瘤医院 | A kind of esophageal cancer lymph is carried down the quantitative analysis results output method and device of shifting |
CN109585015A (en) * | 2019-01-21 | 2019-04-05 | 兰州大学 | A kind of lymthoma histological type intelligent classification system and method |
CN111681756A (en) * | 2020-05-29 | 2020-09-18 | 吾征智能技术(北京)有限公司 | Disease symptom prediction system based on sputum character cognition |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108831559B (en) | Chinese electronic medical record text analysis method and system | |
WO2022227294A1 (en) | Disease risk prediction method and system based on multi-modal fusion | |
US9779085B2 (en) | Multilingual embeddings for natural language processing | |
CN112001177A (en) | Electronic medical record named entity identification method and system integrating deep learning and rules | |
CN110705293A (en) | Electronic medical record text named entity recognition method based on pre-training language model | |
CN106934223A (en) | A kind of blood disease intelligent classification system and method based on big data | |
CN107301171A (en) | A kind of text emotion analysis method and system learnt based on sentiment dictionary | |
CN117744654A (en) | Semantic classification method and system for numerical data in natural language context based on machine learning | |
Carchiolo et al. | Medical prescription classification: a NLP-based approach | |
CN106951917A (en) | The intelligent classification system and method for a kind of lymthoma histological type | |
CN109492105B (en) | Text emotion classification method based on multi-feature ensemble learning | |
CN108009135A (en) | The method and apparatus for generating documentation summary | |
CN106778046A (en) | The system and method that a kind of feature based value carries out prostate cancer recurrence prediction | |
CN112035846A (en) | Unknown vulnerability risk assessment method based on text analysis | |
Sheshikala et al. | Natural language processing and machine learning classifier used for detecting the author of the sentence | |
Wadud et al. | Text coherence analysis based on misspelling oblivious word embeddings and deep neural network | |
CN111339777A (en) | Medical related intention identification method and system based on neural network | |
Rakhsha et al. | Detecting adverse drug reactions from social media based on multichannel convolutional neural networks modified by support vector machine | |
CN113312907A (en) | Remote supervision relation extraction method and device based on hybrid neural network | |
CN112784601A (en) | Key information extraction method and device, electronic equipment and storage medium | |
KR102126911B1 (en) | Key player detection method in social media using KeyplayerRank | |
Lin et al. | Automated malaria cells detection from blood smears under severe class imbalance via importance-aware balanced group softmax | |
CN111627566A (en) | Indication information processing method and device, storage medium and electronic equipment | |
CN116976321A (en) | Text processing method, apparatus, computer device, storage medium, and program product | |
Marerngsit et al. | A two-stage text-to-emotion depressive disorder screening assistance based on contents from online community |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170714 |
|
RJ01 | Rejection of invention patent application after publication |