CN111540475B - Method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology - Google Patents
Method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology Download PDFInfo
- Publication number
- CN111540475B CN111540475B CN202010336891.3A CN202010336891A CN111540475B CN 111540475 B CN111540475 B CN 111540475B CN 202010336891 A CN202010336891 A CN 202010336891A CN 111540475 B CN111540475 B CN 111540475B
- Authority
- CN
- China
- Prior art keywords
- medical
- classifiers
- chinese medicine
- traditional chinese
- unmarked
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a method for mining a law of traditional Chinese medicine treatment based on a semi-supervised learning technology, which is applied to the field of big data processing and aims at the problem of low accuracy of mining the law of the traditional Chinese medicine treatment, and the method firstly trains 3 logistic regression classifiers on a marked medical record set; then predicting a treatment label corresponding to the unmarked medical record according to the trained 3 classifiers; finally, selecting a final prediction result by using a majority voting mechanism; the invention effectively reduces the classification error rate by maximizing the difference of the three classifiers; and the influence on the result of the traditional Chinese medicine treatment method is considered from the three aspects of symptoms, syndromes and pathogenesis, and the method is more in line with the treatment theory under the background of traditional Chinese medicine, so that the obtained result has reliable theoretical basis.
Description
Technical Field
The invention belongs to the field of big data mining, and particularly relates to a traditional Chinese medicine therapeutic law mining technology.
Background
The traditional Chinese medicine is a traditional medicine which takes the theory and practical experience of the traditional Chinese medicine as the main body and researches the human life activity health and disease transformation rule and the prevention, diagnosis, treatment, rehabilitation and health care of the human life activity health and disease transformation rule. The traditional Chinese medicine diagnosis and treatment process comprises the following steps: 1) Syndrome differentiation: analyzing symptoms and signs collected in the four diagnostic methods according to the theory of traditional Chinese medicine, and determining the syndrome; 2) Therapeutic method: determining a corresponding treatment method according to the syndrome differentiation result; 3) The formula is as follows: based on the therapeutic results, the drugs interact with each other and are combined to form the recipe. The clinical medical record is an important carrier of the traditional Chinese medicine diagnosis and treatment experience, is a precious resource for the research and development of the traditional Chinese medicine, comprises symptoms, causes and pathogenesis, syndromes, treatment methods, prescriptions and the like, and has abundant traditional Chinese medicine diagnosis and treatment rules, so that the diagnosis and treatment efficiency of doctors can be improved by effectively utilizing the diagnosis and treatment rules, and how to effectively disclose the diagnosis and treatment rules through the clinical medical record becomes a key problem. With the application of computer technology in the field of traditional Chinese medicine, it becomes possible to mine the law of disease treatment based on a large number of clinical cases. At present, the research on the law of traditional Chinese medicine treatment mainly adopts frequency analysis and a subject model method to analyze the law of treatment. For example, a frequency statistical method is used for analyzing the traditional Chinese medicine treatment law of breast cancer, and as a result, 21 common treatment methods (Yangyufeng, excessive rainfall and Liyuan, the traditional Chinese medicine treatment law after breast cancer operation is mined from data processing [ J ] medical research and education, 2017,34 (5): 16-23.) are found; the clinical practice of TCM is used as a document set, the symptoms of patients are regarded as obvious variables, the adopted treatment is regarded as hidden variables, and the subject model method is applied to excavate the implicit treatment distribution to generate a symptom-based treatment prediction subject model (L.Yao, Y.Zhang, B.Wei, et al.discovery treating patient pattern in Traditional Chinese medical science by applying the experimental topical topic model and domain knowledge [ J ]. Journal of Biomedical information, 2015,58 (c): 260-267).
The existing research for mining the therapeutic law from clinical medical records by using methods such as frequency analysis, a subject model and the like only considers symptom factors generally, and lacks of analysis on the relationship between the factors such as the disease position, the pathogenesis, the syndrome and the like and the therapeutic law, so that the mining accuracy of the therapeutic law is influenced.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for mining the law of the traditional Chinese medicine treatment based on the semi-supervised learning technology, simultaneously considers the multi-characteristic attributes of symptoms, syndromes, etiology and pathogenesis and the like of a patient, and provides an auxiliary decision for the treatment of the diagnosis and treatment process of a doctor of traditional Chinese medicine.
The technical scheme adopted by the invention is as follows: a method for mining a law of a traditional Chinese medicine treatment based on a semi-supervised learning technology comprises the following steps:
s1, training 3 logistic regression classifiers on a marked medical case set; the marked medical record set comprises a plurality of marked medical records, and each marked medical record corresponds to one treatment label;
s2, predicting a treatment label corresponding to the unmarked medical case according to the 3 classifiers trained in the step S1;
and S3, selecting a final prediction result by using a majority voting mechanism for the 3 legal labels obtained in the step S2.
In step S1, each marked medical record comprises 3 attributes of symptoms, syndromes and pathogenesis.
In step S1, each marked medical record is a D-dimensional 0-1 vector, where D is the total number of attributes included in the marked medical record set, and D = the total number of symptoms a + the total number of syndromes B + the total number of etiological mechanisms C, and if a certain symptom, syndrome, or etiological mechanism is included in a certain marked medical record, it is represented by 1, otherwise it is 0.
It also includes maximizing the variance among 3 classifiers: the D attributes are divided using 3 classifiers, each attribute if and only if it belongs to 1 of them.
Further comprising: the process of optimizing the classifier:
acquiring a label-free medical case set; the unmarked medical case set comprises a plurality of unmarked medical cases; calculating the prediction reliability of each unmarked medical case in the unmarked medical case set on each classifier; and for each classifier, sorting the unmarked medical cases from large to small according to the prediction reliability, then selecting the first 5 unmarked medical cases in each classifier and adding the unmarked medical cases into the marked medical case set so as to generate a new marked medical case set, and retraining 3 logistic regression classifiers based on the new marked medical case set until the unmarked medical case set is empty.
The method for calculating the prediction reliability of each unmarked medical case in the unmarked medical case set on each classifier specifically comprises the following steps:
wherein, the first and the second end of the pipe are connected with each other,
representing the reliability of the prediction of the unmarked case x' on the classifiers ε, θ, v, ε, θ, v being used to represent said 3 classifiers H i (x ', t) represents the reliability of the unmarked medical case x' on the treatment label t, K represents the total number of treatment labels, and U represents the unmarked medical case set.
The invention has the beneficial effects that: the method simultaneously considers the multi-characteristic attributes of symptoms, syndromes, etiology and pathogenesis and the like of the patient, and provides an auxiliary decision for the treatment method of the diagnosis and treatment process of the doctor of traditional Chinese medicine; compared with the result obtained by analyzing the law of the traditional Chinese medicine treatment method by using a common statistical method, the method disclosed by the invention can achieve higher accuracy and provide an auxiliary decision for a doctor; the method of the invention has the following advantages:
1) The invention maximizes the difference of the three classifiers and effectively reduces the classification error rate: the method comprises the steps that predefined attribute division does not exist, 3 initialized attribute views are obtained through an attribute division method, an initialized classifier is trained on the basis of the 3 initialized views, and the difference between the classifiers is continuously improved in the attribute division process;
2) The influence on the result of the traditional Chinese medical treatment is considered from the three aspects of symptoms, syndromes and pathogenesis, and the method is more in line with the treatment theory under the background of traditional Chinese medicine, so that the obtained result has reliable theoretical basis.
Drawings
FIG. 1 is a flow chart of the scheme of the invention.
Detailed Description
In order to facilitate the understanding of the technical contents of the present invention by those skilled in the art, the following further explains the technical contents of the present invention with reference to fig. 1.
In the embodiment, the traditional Chinese medicine treatment method aided decision research is carried out by taking the chronic glomerulonephritis medical record as original data, symptoms, syndromes and pathogenesis are taken as characteristic attributes, the treatment method is taken as a classification label, and the treatment method prediction is carried out by taking the analysis process of the traditional Chinese medicine treatment method as a multi-classification problem. The method mainly comprises the following steps: 1) Learning 3 classifiers on the labeled dataset; 2) Using the 3 classifiers to predict the treatment without marking the medical record; 3) Optimizing 3 classifiers using a multi-view training algorithm; 4) The prediction results are selected using a "voting mechanism". The overall flow chart of the scheme is shown in figure 1.
Data set: the chronic glomerulonephritis medical record set M comprises 3 attributes of symptoms, syndromes and etiology and pathogenesis, each medical record is represented as a D-dimensional 0-1 vector, wherein D is the attribute total number contained in the medical record set M, D = the symptom total number A + the syndrome total number B + the etiology and pathogenesis total number C, if a certain symptom, syndrome or etiology and pathogenesis is contained in the medical record, the symptom, syndrome or etiology and pathogenesis is represented by 1, and otherwise, the symptom, syndrome or etiology and pathogenesis is 0.
Inputting:
1) A mark set L: m parts of medical records;
2) No marker set U: n parts of medical records;
3) And (4) label set Y: the medical record set M includes K treatment methods.
And (3) outputting: vector of K dimension therapy.
The method of the invention comprises the following steps:
1. 3 Logistic Regression (LR) classifiers were trained on the labeled dataset.
1) Therapeutic label space representation:
Y＝{1,2,…,K}
2) The 3 classifiers are represented using a matrix:
ε＝[ε 1 ,…,ε K ]∈R D×K
θ＝[θ 1 ,…,θ K ]∈R D×K
v＝[v 1 ,…,v K ]∈R D×K
wherein epsilon, theta and v respectively represent the matrix form of 3 classifiers; d is the number of attributes contained in L and U, which is mainly classified into 3 types of attributes, namely: symptoms, syndromes, etiologies and pathogenesis; epsilon K Representing D-dimensional attribute vectors corresponding to the classifiers epsilon under a therapeutic K, and representing the D-dimensional attribute vectors by using 0-1 vectors; theta.theta. K ,v K The same is true.
3) Constructing a loss function on each classifier based on logistic regression:
wherein L (epsilon; L) represents the loss value of the classifier epsilon on the label set L, e is an exponential function, and L (theta; L), L (v; L) are the same; (x, y) represents the therapeutic label y corresponding to the medical record x.
4) Minimize the loss function over 3 classifiers:
5) In order to maximize the difference among the 3 classifiers and ensure the performance superiority of the method, the following method is used to classify the attributes contained in the labeled medical record set L. Namely: dividing D attributes by using 3 classifiers epsilon, theta and v, wherein each attribute belongs to 1 classifier, and converting the attribute into a numerical form; wherein, each classifier is trained by 3 different attributes, namely: symptoms, syndromes, etiology and pathogenesis. For example: the values of the 3 classifiers are respectively epsilon on the attribute a a ＝0,θ a ＝1,v a =0, it means that the attribute a is divided on the classifier θ. The details are as follows:
wherein D is the number of attributes contained in L and U, epsilon a 、θ a 、v a The value of each classifier on the attribute a is shown, 0 indicates that the attribute is not included, and 1 indicates that the attribute is included.
6) In summary, the above is converted into an optimization problem, so as to learn 3 classifiers epsilon, theta, v, and the objective function is as follows:
2. the 3 classifiers epsilon, theta and v are used for predicting the therapeutic label of the unmarked medical case x' epsilon U respectively.
y ε (x′)＝(x′ Τ ε 1 ,…,x′ Τ ε K )
y θ (x′)＝(x′ Τ θ 1 ,…,x′ Τ θ K )
y v (x′)＝(x′ Τ v 1 ,…,x ′Τ v K )
Wherein, y ε (x ') represents the predicted result of the medical case x' on the classifier ε, y θ (x′),y v (x ') similarly, x' Τ Is the transposed vector, x 'of medical case x' Τ ε K Denotes the probability, x ', of predicting medical case x ' with a therapeutic label of K using classifier ε ' Τ θ K 、x′ Τ v K The same is true.
3. The classifier is optimized using a multi-view training algorithm.
1) The reliability of the prediction of the medical plan x' on the classifiers e, theta, v, respectively, is evaluated.
Wherein H i (x ', t) represents the reliability of the medical case x' on the therapeutic label t, as follows:
wherein, the first and the second end of the pipe are connected with each other,
the probability that the label representing the medical case x' is t is as follows:
2) The labeled unlabeled cases were passed between 3 classifiers: sorting the unlabeled medical records based on the prediction reliability, selecting the first 5 medical records obtained by prediction of each classifier, adding the medical records into the labeled medical record set so as to generate a new labeled medical record set, and retraining the three classifiers based on the new labeled medical record set until the unlabeled medical record set is empty, so as to optimize the classifiers.
4. Predictive outcome selection
The final legal prediction results were selected using the majority voting mechanism, as follows:
key points of the invention
1) The invention maximizes the difference of the three classifiers and effectively reduces the classification error rate: the method comprises the steps that predefined attribute division does not exist, 3 initialized attribute views are obtained through an attribute division method, an initialized classifier is trained on the basis of the 3 initialized views, and the difference between the classifiers is continuously improved in the attribute division process.
2) The influence on the prediction result of the traditional Chinese medicine treatment method is considered from the three aspects of symptoms, syndromes and pathogenesis, and the method is more in line with the treatment theory under the background of traditional Chinese medicine, so that the obtained result has reliable theoretical basis.
Compared with the result obtained by analyzing the law of the traditional Chinese medicine treatment by using a common statistical method, the method for mining the law of the treatment can achieve higher accuracy and provide an auxiliary decision for doctors.
TABLE 1 results recommended for chronic glomerulonephritis treatment
We invite Chinese physicians of Chengdu Chinese medicine university to analyze the treatment prediction results in Table 1, and the experiments found that the accuracy of the treatment prediction obtained by using the method reaches 86.4% based on 1959 cases of chronic kidney disease medical records, wherein each case of medical records is composed of 3 basic attributes of symptoms, syndromes, etiology and pathogenesis and classification label-treatment.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (4)
1. A method for mining a law of a traditional Chinese medicine treatment based on a semi-supervised learning technology is characterized by comprising the following steps:
s1, training 3 logistic regression classifiers on a marked medical case set; the marked medical record set comprises a plurality of marked medical records, and each marked medical record corresponds to one treatment label;
s2, predicting treatment labels corresponding to the unmarked medical records according to the 3 classifiers trained in the step S1; further comprising: the process of optimizing the classifier:
acquiring a label-free medical case set; the unmarked medical case set comprises a plurality of unmarked medical cases; calculating the prediction reliability of each unmarked medical case in the unmarked medical case set on each classifier; for each classifier, sorting the unmarked medical cases from large to small according to the prediction reliability, then selecting the first 5 unmarked medical cases in each classifier to be added to the marked medical case set so as to generate a new marked medical case set, and retraining 3 logistic regression classifiers based on the new marked medical case set until the unmarked medical case set is empty;
the method for calculating the prediction reliability of each unmarked medical case in the unmarked medical case set on each classifier specifically comprises the following steps:
wherein, the first and the second end of the pipe are connected with each other,
to representThe prediction reliability of the unmarked medical case x' on the classifiers ε, θ, v, ε, θ, v are used to represent the 3 classifiers, H i (x ', t) represents the reliability of the unmarked medical case x' on the therapeutic label t, K represents the total number of therapeutic labels, and U represents the unmarked medical case set;
and S3, selecting a final prediction result by using a majority voting mechanism for the 3 legal labels obtained in the step S2.
2. The method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology as claimed in claim 1, wherein each marked case in step S1 comprises 3 attributes of symptom, syndrome and pathogenesis.
3. The method for mining rules of traditional Chinese medicine treatment based on semi-supervised learning technology as recited in claim 2, wherein each labeled medical case in step S1 is a D-dimensional 0-1 vector, wherein D is the total number of attributes included in the labeled medical case set, D = the total number of symptoms a + the total number of syndromes B + the total number of etiological mechanisms C, if a certain symptom, syndrome or etiological mechanism is included in a certain labeled medical case, it is represented by 1, otherwise it is 0.
4. The method of claim 1, further comprising maximizing variability among 3 classifiers by mining a law of traditional Chinese medicine based on semi-supervised learning techniques: the D attributes are partitioned using 3 classifiers, each attribute if and only if it belongs to 1 classifier among them.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010336891.3A CN111540475B (en) | 2020-04-26 | 2020-04-26 | Method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010336891.3A CN111540475B (en) | 2020-04-26 | 2020-04-26 | Method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111540475A CN111540475A (en) | 2020-08-14 |
| CN111540475B true CN111540475B (en) | 2022-10-21 |
Family
ID=71970220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010336891.3A Active CN111540475B (en) | 2020-04-26 | 2020-04-26 | Method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111540475B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617623A (en) * | 2013-12-10 | 2014-03-05 | 江南大学 | Medical image segmentation method and system for fully-represented semi-supervised fast spectral clustering |
| CN104318242A (en) * | 2014-10-08 | 2015-01-28 | 中国人民解放军空军工程大学 | High-efficiency SVM active half-supervision learning algorithm |
| CN108549638A (en) * | 2018-04-19 | 2018-09-18 | 心承智慧（海口）健康科技有限公司 | A kind of Chinese medicine ancient literature semi-supervised learning method and system |
| CN110738052A (en) * | 2019-09-29 | 2020-01-31 | 中国中医科学院 | semi-supervised Chinese medicine named entity acquisition method based on characteristic words |
-
2020
- 2020-04-26 CN CN202010336891.3A patent/CN111540475B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617623A (en) * | 2013-12-10 | 2014-03-05 | 江南大学 | Medical image segmentation method and system for fully-represented semi-supervised fast spectral clustering |
| CN104318242A (en) * | 2014-10-08 | 2015-01-28 | 中国人民解放军空军工程大学 | High-efficiency SVM active half-supervision learning algorithm |
| CN108549638A (en) * | 2018-04-19 | 2018-09-18 | 心承智慧（海口）健康科技有限公司 | A kind of Chinese medicine ancient literature semi-supervised learning method and system |
| CN110738052A (en) * | 2019-09-29 | 2020-01-31 | 中国中医科学院 | semi-supervised Chinese medicine named entity acquisition method based on characteristic words |
Non-Patent Citations (2)
| Title |
|---|
| A Method of Collecting Four Character Medicine Effect Phrases in TCM Patents Based on Semi-supervised Learning;Deng Na;《CISIS 2019：Complex,Intelligent,and Software Intensive System》;20190621;462-473页 * |
| 面向疾病预测的半监督学习算法研究;潘媛;《中国优秀硕士学位论文全文数据库-信息科技辑》;20190215(第2期);第I140-226页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111540475A (en) | 2020-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pereira et al. | COVID-19 identification in chest X-ray images on flat and hierarchical classification scenarios | |
| Wagholikar et al. | Modeling paradigms for medical diagnostic decision support: a survey and future directions | |
| Chen et al. | Information fusion and artificial intelligence for smart healthcare: a bibliometric study | |
| CN113688248B (en) | Medical event identification method and system under condition of small sample weak labeling | |
| López-Úbeda et al. | Automatic medical protocol classification using machine learning approaches | |
| CN113035362A (en) | Medical prediction method and system based on semantic graph network | |
| TW202101477A (en) | Method for applying a label made after sampling to neural network training model | |
| CN115458108A (en) | Visual diabetes big data intelligent analysis system fusing multivariate technology | |
| Pendyala et al. | Automated medical diagnosis from clinical data | |
| Li et al. | Improving prediction for medical institution with limited patient data: Leveraging hospital-specific data based on multicenter collaborative research network | |
| CN111540475B (en) | Method for mining law of traditional Chinese medicine treatment based on semi-supervised learning technology | |
| Sarafraz et al. | Domain adaptation and generalization on functional medical images: A systematic survey | |
| Pan et al. | Multi-label classification for clinical text with feature-level attention | |
| CN114694841B (en) | Adverse event risk prediction method based on patient electronic health record | |
| Safa et al. | A survey on hybrid case-based reasoning and deep learning systems for medical data classification | |
| CN116682536A (en) | Traditional Chinese medicine prescription recommendation method based on improved Bert4Rec | |
| Perner | Intelligent data analysis in medicine—recent advances | |
| Chhabra et al. | Image pattern recognition for an intelligent healthcare system: An application area of machine learning and big data | |
| Chaithra et al. | A Review of Machine Learning Techniques Used in the Prediction of Heart Disease. | |
| Malakouti et al. | Hierarchical deep multi-task learning for classification of patient diagnoses | |
| Chantamit-O-Pas et al. | A case-based reasoning framework for prediction of stroke | |
| Madanan et al. | Applying an optimal feature ranking and selection algorithm and random forest classifier algorithm along with k-fold cross validation for classification of blood cancer cells | |
| Zhou et al. | An unsupervised deep clustering for Bone X-ray classification and anomaly detection | |
| Salem | Computational intelligence for brain tumors detection | |
| Demigha | Mining Knowledge of the Patient Record |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |