CN105574351B - Medical data processing method - Google Patents
Medical data processing method Download PDFInfo
- Publication number
- CN105574351B CN105574351B CN201511029760.6A CN201511029760A CN105574351B CN 105574351 B CN105574351 B CN 105574351B CN 201511029760 A CN201511029760 A CN 201511029760A CN 105574351 B CN105574351 B CN 105574351B
- Authority
- CN
- China
- Prior art keywords
- patient
- class
- treatment
- data
- matrix
- 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
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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a medical data processing method. The medical data processing method comprises the following steps: S1: dividing medical data in regard to a plurality of patients into patient feature data and treatment data, and respectively converting the feature data and the treatment data into a matrix through normalization; S2: finding out the patients with the similar treatment manner from the treatment data by utilizing hierarchical clustering; S3: obtaining a common treatment manner aiming at each type of patients; S4: obtaining common features aiming at each type of patients, and relating the treatment manner and the features corresponding to each type of patients aiming at each type of patients, so as to obtain the corresponding relation of the treatment manner and the features of the patients.
Description
Technical field
The application is related to medical data mining field, and the relation relating more specifically to disease treatment mode with patient characteristicses is dug
Pick method.
Background technology
During disease finds, treats, doctor is diagnosed accordingly according to the different characteristic of patient.Therefore, send out
Existing relation between patient characteristicses and therapeutic modality can select suitable medicine and therapeutic modality to have directive function doctor.Existing
Patient information and therapeutic modality are generally carried out simple hypothesis testing by some correlation analysis methods successively, and lack certainly
The implementation of dynamicization.Therefore, can automatically obtain medical treatment in batch patient mode and the relation of patient characteristicses are expected
Method.
Content of the invention
For solving the above-mentioned problems in the prior art, presently filed embodiment proposes a kind of medical data and processes
Method, including step S1：Medical data with regard to multiple patients is divided into patient characteristicses data and treatment data, and by this feature
Data and treatment data are converted into matrix by normalization respectively；Step S2：Find out tool using hierarchical clustering from treatment data
There is the patient of similar therapeutic mode；Step S3：Obtain common therapeutic modality for each class patient；And step S4：For
Each class patient obtains common feature, and to each class patient, associates the corresponding therapeutic modality of such patient and feature, obtain
Therapeutic modality and the corresponding relation of patient characteristicses.
Brief description
Fig. 1 shows the schematic flow diagram of the method that medical data according to the embodiment of the present invention is processed.
Specific embodiment
Below in conjunction with the accompanying drawings embodiments of the present invention are described in detail.
Fig. 1 shows the schematic flow diagram of medical data processing method according to the embodiment of the present invention.With reference to Fig. 1,
In embodiments of the present invention, there is provided medical data processing method, the method can include：
Step S1：The medical data with regard to multiple patients collected is divided into patient characteristicses data and treatment data, and will
This feature data and treatment data are converted into matrix by normalization respectively.In this application, described medical data mainly may be used
To be divided into two classes.The first kind is patient characteristicses data, such as clinical test information, the front urine for the treatment of before patient basis, treatment
Vital sign etc. before Biochemical Information before routine information, treatment, treatment.Equations of The Second Kind is treatment data, such as medication information, treatment side
Formula etc..For example, feature " sex " can switch to -1,1 according to man, female to data normalization；Feature " positive " can switch to 0 according to value,
1；Medication information can be according to whether switch to 1,0 etc. with this medicine.
The input of this step can be the patient data collecting, including patient characteristicses data and treatment data.Output can be
Patient characteristicses matrix and treatment matrix, patient characteristicses matrix behavioural characteristic, are classified as patient, are worth for original record after conversion
Value.Treatment matrix behaviour therapy information, is classified as patient, is worth the value after conversion for original record.
In embodiments of the present invention, step S1 can include：
Step S1-1：Medical data is divided into patient characteristicses data and treatment data.Patient characteristicses data can be treatment
Biochemical letter before routine urinalysis information before clinical test information before front patient information, such as patient basis, treatment, treatment, treatment
Vital sign etc. before breath, treatment.Treatment data can be, for example, medication information, therapeutic modality etc..
Step S1-2：By patient characteristicses Data Discretization, quantize.Patient characteristicses data can be divided into discrete type, connect
Ideotype.For the factor of discrete type value, for example numerical value can be converted into centrifugal pump 1,2 ....And for continuous type value
Factor, retains numerical value.
Step S1-3：Patient characteristicses data normalization by discretization and after quantizing.By patient characteristicses data according under
Formula (1) normalizes, thus obtaining patient characteristicses matrix X.In formula (1), x is patient characteristicses data, xiRepresent the spy of i-th patient
Levy data, xmaxRepresent the maximum of patient characteristicses data, xminRepresent the minimum of a value of patient characteristicses data.
Formula (1)
This patient characteristics matrix X is for example shown below, and behavial factor is classified as patient, and each value is a numerical value.Should
In formula, f represents feature, total n feature, p patient.xijRepresent the value of the ith feature of j-th patient.
Step S1-4：By treatment data discretization, quantize.Treatment data can be divided into discrete type, continuous type.For
Numerical value for example can be converted into centrifugal pump 1,2 ... by the factor of discrete type value.For the factor of continuous type value, encumbrance
Value.
Step S1-5：Treatment data normalization by discretization and after quantizing.By treatment data (3) normalizing according to the following formula
Change, thus obtaining Case treatment matrix Y.In formula (3), y is treatment data, yiRepresent the treatment data of i-th patient, ymaxTable
Show the maximum of Case treatment data, yminRepresent the minimum of a value of Case treatment data.
Formula (3)
Case treatment matrix Y is for example shown below, and behaviour therapy information is classified as patient, and each value is a numerical value.
In formula, t represents treatment information, total m feature, p patient.yijRepresent the value of the i-th treatment information of j-th patient.
Step S2：Therapeutic modality clusters：The disease with similar therapeutic mode is found out using hierarchical clustering from treatment data
People.The input of this step can be Case treatment matrix Y, and output can be patient classification.Here therapeutic modality cluster may refer to
The patient with similar therapeutic mode is found out using hierarchical clustering from treatment data.The form for the treatment of data can be for suffering from the disease
People and its corresponding treatment information.For each patient, these treatment information can form vector.Treatment information dimension is entered
Row hierarchical clustering, selects classification N, patient is divided into N number of classification.Each class patient has similar therapeutic modality.
Specifically, step S2 can include：
Step S2-1：Case treatment mode vector Euclidean distance two-by-two is calculated according to following formula (4).If Pi=(y1i,y2i…
ymi) vectorial for the corresponding therapeutic modality of patient i, in formula (4), DijRepresent the Euclidean distance of vector, yaiRepresent i-th patient's
The value of a-th treatment information, yajRepresent the value of the a-th treatment information of j-th patient, total m feature, p patient.
Formula (4)
Step S2-2：P patient is divided into p class, that is, each class comprises only a patient, (5) calculate class according to the following formula
Between distance.In formula (5), PiRepresent i-th patient corresponding therapeutic modality vector, PjRepresent the corresponding therapeutic modality of j-th patient to
Amount, DC represents between class distance, and C represents classification, DCrsRepresent classification CrWith classification CsBetween class distance.
Formula (5)
Step S2-3：Merge two minimum classifications of distance.
Step S2-4：Repeat step S2-3, until last patient is merged, forms hierarchical clustering result.
Step S2-5：Select classification number N, patient is divided into N class (N is positive integer).
Step S3：Obtain way of concurrent therapy：Obtain common therapeutic modality for each class patient.The input of this step can
Think patient classification, treatment matrix Y, output can be the corresponding therapeutic modality of patient classification.This acquisition way of concurrent therapy can
To refer to find common therapeutic modality to each class patient.Each class to multiclass (such as N class) patient, by current research class
As positive sample, from remaining N-1 class, randomly select the negative sample of equal samples amount.To each therapeutic modality, false using t
If inspection obtains significant therapeutic modality.Through this step, obtain the way of concurrent therapy that each class patient has.
Specifically, this step S3 can include：
Step S3-1：Using the current research class of N class patient as positive sample, randomly select from remaining N-1 class
The negative sample of sample size.
Step S3-2：Retain the row being associated with the patient that positive sample comprises in treatment matrix Y, delete in treatment matrix Y
Remaining row, to form matrix A；Retain the row being associated with the patient that negative sample comprises in treatment matrix Y, delete treatment matrix
Remaining row in Y, to form matrix B.
Step S3-3：For each therapeutic modality, t hypothesis testing is carried out to matrix A and B.Select statistical significance<
0.01 therapeutic modality is as the way of concurrent therapy of current research class patient.
Step S3-4：Each class to N class patient, repeat step S3-1, to step S3-3, obtains every class patient corresponding
Therapeutic modality.
Step S4：Obtain common feature for each class patient.The input of this step can be patient classification, eigenmatrix
X, output can be the corresponding feature of patient classification.Obtain common trait to refer to find common feature to each class patient.Example
As each class to N class patient, using current research class as positive sample, random selects equal samples amount from remaining N-1 class
Negative sample.To each patient characteristics, obtain significant feature using t hypothesis testing.Through this step, obtain each class disease
The common trait that people has, the way of concurrent therapy having in conjunction with each class patient, obtain significant therapeutic modality special with patient
Levy corresponding relation.
Specifically, this step S4 can include：
Step S4-1：Using the current research class of N class patient as positive sample, randomly select from remaining N-1 class
The negative sample of sample size.
Step S4-2：The row being associated with the patient that positive sample comprises in keeping characteristics matrix X, are deleted in eigenmatrix X
Remaining row, with formed matrix A '；The row being associated with the patient that negative sample comprises in keeping characteristics matrix Y, delete feature square
Battle array Y in remaining row, with formed matrix B '.
Step S4-3：To each feature, to matrix A ' and B' carry out t hypothesis testing.Select statistical significance<0.01
Feature is as the common trait of current research class patient.
Step S4-4：Each class to N class patient, repeat step S4-1, to step S4-3, obtains every class patient corresponding
Feature.
Step S4-5：Each class to N class patient, the association corresponding therapeutic modality of such patient and feature, obtain medical treatment
Mode and the corresponding relation of patient characteristicses.
Alternatively, step S4 can include step S4-6：Using the related therapeutic modality of Computer display and patient characteristicses.
Step S5：Adjusting parameter is to obtain therapeutic modality and the patient characteristicses of correlation.Through step S1 to S4, without
Meet therapeutic modality and the patient characteristicses of condition, can be obtained using this step.Here parameter adjustment refers to without satisfaction
The therapeutic modality of condition and patient characteristicses, compute repeatedly, by adjusting classification number, therapeutic modality and the patient spy to obtain correlation
Levy.
Specifically, step S5 can include：
Step S5-1：To certain class in N class patient, if step S3-3 does not obtain the corresponding co-therapies of such patient
Mode, then can improve statistical significance threshold value, for example, bring up to 0.05.
Step S5-2：To certain class in N class patient, if step S4-3 does not obtain the corresponding common spy of such patient
Levy, then can improve statistical significance threshold value, for example, bring up to 0.05.
Step S5-3：Reduce or raise the value of classification number in step S2-5, the class number of patient is adjusted,
Repeat step S3 and/or S4, obtain therapeutic modality and the patient characteristicses of correlation in different rent fine granularities.
The t hypothesis testing mentioned in above-mentioned steps S3 and step S4 can be the known method of inspection in art,
Hereafter simply introduce the basic thought of t hypothesis testing.
If overall X and Y is independent, σ1Represent the standard deviation of X, μ1Represent the average of X, S1Represent the sample variance of X,Represent X
Sample average, σ2Represent the standard deviation of Y, μ2Represent the average of X, S2Represent the sample variance of Y,Represent the sample average of Y, p
Represent statistical significance.If the variance of two normal populations is equal, that is,Consider two-sided hypothesis test problem：
H0:μ1=μ2 H1:μ1≠μ2
At this moment in μ1=μ2Under can obtain：
Wherein
The region of rejection thus obtaining above-mentioned Hypothesis Testing Problem is
{ | T | ＞ t1-p/2(n1+n2-2)}
It will be appreciated by those skilled in the art that above-mentioned method can pass through the combination of software, hardware and software and hardware
Mode implementing.
During disease finds, treats, doctor can be diagnosed accordingly according to the different characteristic of patient.Find
Relation between patient characteristicses and therapeutic modality can instruct doctor to select suitable medicine and therapeutic modality.The enforcement of the present invention
Mode compared with prior art has at least one of advantages below.
1. existing correlation analysis method generally lacks the implementation of automation.Embodiments of the present invention propose one
Before planting disease treatment mode and treating, the relation excavation method of factor, can automatically find out with phase in large-scale patient
Like the patient of therapeutic modality, and detect the common trait that similar therapeutic mode patient has, thus the mode that obtains medical treatment
Relation with patient characteristicses.
2. patient information and therapeutic modality are generally carried out hypothesis testing by existing correlation analysis method successively, and in practice
Similitude is there is between patient.Step S2 of embodiments of the present invention from overall angle, using the thought of cluster
Find out the patient with similar therapeutic mode from treatment data, patient is divided into different classifications and is studied.
3. it is generally deficient of finer analysis after existing correlation analysis method classification.The step of embodiments of the present invention
S3 is directed to each class patient and obtains common therapeutic modality, and step S4 is directed to each class patient and obtains common feature.Step S3
With S4 on the basis of overall angle, finer analysis is carried out to each class.
4. the usual execution step of existing correlation analysis method more single it is assumed that analysis result is not inconsistent with expection, lack
Feedback procedure.Step S5 in embodiments of the present invention proposes the feedback procedure of regulation parameter, to obtain meeting actual feelings
The result of condition.
It will be understood by those skilled in the art that the medical data processing method of the embodiment of present invention as described above
Step is not required to execute according to the order of shown step.As long as embodiments of the present invention can be implemented, can be by
Other sequentially carry out execution step, or at least some step can synchronously be carried out.For example, any one step of step S3 and step S4
Suddenly can first be performed, or be performed simultaneously.For example, in the combination of step S1-2, S1-3 and the combination of step S1-4, S1-5
Any one combination can first be performed, or be performed simultaneously.
Although the application is to describe the present invention by way of describing specific embodiment, those skilled in the art can
To understand that these specific embodiments are schematic and nonrestrictive.Those skilled in the art pass through reading the application's
Embodiment can carry out various modifications, deformation and replacement to embodiment in the case of understanding the design of the application.
Claims (7)
1. a kind of medical data processing method, the method includes：
Step S1：Medical data with regard to multiple patients is divided into patient characteristicses data and treatment data, and by this feature data
Respectively matrix is converted into by normalization with treatment data；
Step S2：The patient with similar therapeutic mode is found out using hierarchical clustering from treatment data；
Step S3：Obtain common therapeutic modality for each class patient, specifically include following steps：
Step S3-1：Using the current research class of N class patient as positive sample, randomly from remaining N-1 class, select equal samples
The negative sample of amount；
Step S3-2：Retain the row being associated with the patient that positive sample comprises in treatment matrix Y, delete its in treatment matrix Y
Yu Lie, to form matrix A；Retain the row being associated with the patient that negative sample comprises in treatment matrix Y, delete in treatment matrix Y
Remaining row, to form matrix B；
Step S3-3：For each therapeutic modality, matrix A and B are carried out with t hypothesis testing, select statistical significance<0.01
Therapeutic modality is as the way of concurrent therapy of current research class patient；
Step S3-4：Each class to N class patient, repeating said steps S3-1, to described step S3-3, obtain every class patient couple
The therapeutic modality answered；And
Step S4：Obtain common feature for each class patient, and to each class patient, associate the corresponding treatment of such patient
Mode and feature, the corresponding relation of the mode that obtains medical treatment and patient characteristicses, specifically include following steps：
Step S4-1：Using the current research class of N class patient as positive sample, randomly from remaining N-1 class, select equal samples
The negative sample of amount；
Step S4-2：The row being associated with the patient that positive sample comprises in keeping characteristics matrix X, delete eigenmatrix X in its
Yu Lie, with formed matrix A '；The row being associated with the patient that negative sample comprises in keeping characteristics matrix Y, are deleted in eigenmatrix Y
Remaining row, with formed matrix B '；
Step S4-3：To each feature, to matrix A ' and B' carry out t hypothesis testing.Select statistical significance<0.01 feature
Common trait as current research class patient；
Step S4-4：Each class to N class patient, repeating said steps S4-1, to step S4-3, obtain every class patient corresponding
Feature；
Step S4-5：Each class to N class patient, the association corresponding therapeutic modality of such patient and feature, obtain medical treatment mode
Corresponding relation with patient characteristicses.
2. method according to claim 1, wherein, described step S1 includes：
Step S1-1：Medical data is divided into patient characteristicses data and treatment data；
Step S1-2：By patient characteristicses Data Discretization, quantize；
Step S1-3：Patient characteristicses data by discretization and after quantizing normalizes according to formula (1), thus it is special to obtain patient
Levy matrix X,
Wherein, xiRepresent the characteristic of i-th patient, xmaxRepresent the maximum of patient characteristicses data, xminRepresent patient characteristicses
The minimum of a value of data；
Step S1-4：By treatment data discretization, quantize；
Step S1-5：Treatment data by discretization and after quantizing normalizes according to formula (3), thus obtaining Case treatment square
Battle array Y,
Wherein, yiRepresent the treatment data of i-th patient, ymaxRepresent the maximum of Case treatment data, yminRepresent Case treatment
The minimum of a value of data.
3. method according to claim 1, wherein, described step S2 includes：
Step S2-1：Case treatment mode vector Euclidean distance two-by-two is calculated according to formula (4)；
If Pi=(y1i,y2i…ymi) vectorial for the corresponding therapeutic modality of patient i, in formula (4), DijRepresent the Euclidean distance of vector,
yaiRepresent the value of the a-th treatment information of i-th patient, yajRepresent the value of the a-th treatment information of j-th patient,
Step S2-2：P patient is divided into p class, that is, each class comprises only a patient, (5) calculate class spacing according to the following formula
From in formula (5), DC represents between class distance, and C represents classification, DCrsRepresent classification CrWith classification CsBetween class distance,
Step S2-3：Merge two minimum classifications of distance；
Step S2-4：Repeat step S2-3, until last patient is merged, to form hierarchical clustering result；
Step S2-5：Select classification number N, patient is divided into N class.
4. method according to claim 1, the method also includes：
Step S5：Adjusting parameter is to obtain therapeutic modality and the patient characteristicses of correlation.
5. method according to claim 4, wherein, described step S5 includes：
To the class in N class patient, if described step S3-3 does not obtain the corresponding way of concurrent therapy of such patient, carry
High statistical significance threshold value, repeats described step S3.
6. method according to claim 4, wherein, described step S5 includes：
To the class in N class patient, if described step S4-3 does not obtain the corresponding common trait of such patient, can improve
Statistical significance threshold value, repeats described step S4.
7. method according to claim 4, wherein, described step S5 includes：
Described step S2-5 reduces or raises class number, repeating said steps S3 and/or described step S4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511029760.6A CN105574351B (en) | 2015-12-31 | 2015-12-31 | Medical data processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511029760.6A CN105574351B (en) | 2015-12-31 | 2015-12-31 | Medical data processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105574351A CN105574351A (en) | 2016-05-11 |
| CN105574351B true CN105574351B (en) | 2017-02-15 |
Family
ID=55884476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511029760.6A Active CN105574351B (en) | 2015-12-31 | 2015-12-31 | Medical data processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105574351B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650299B (en) * | 2017-01-18 | 2019-01-25 | 浙江大学 | A kind of quick calculation method of patient's similarity analysis |
| CN106951710B (en) * | 2017-03-22 | 2020-11-03 | 华东师范大学 | CAP data system and method based on privilege information learning support vector machine |
| CN108109700B (en) * | 2017-12-19 | 2021-05-25 | 中国科学院深圳先进技术研究院 | Method and device for evaluating curative effect of chronic disease |
| CN108346474B (en) * | 2018-03-14 | 2021-09-28 | 湖南省蓝蜻蜓网络科技有限公司 | Electronic medical record feature selection method based on word intra-class distribution and inter-class distribution |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704961A (en) * | 2004-05-26 | 2005-12-07 | 冲博子 | Analysis method for diagnosis and treatment behavior and administration |
| CN101334843A (en) * | 2007-06-29 | 2008-12-31 | 中国科学院自动化研究所 | Pattern recognition characteristic extraction method and apparatus |
| CN103745227A (en) * | 2013-12-31 | 2014-04-23 | 沈阳航空航天大学 | Method for identifying benign and malignant lung nodules based on multi-dimensional information |
| CN104915560A (en) * | 2015-06-11 | 2015-09-16 | 万达信息股份有限公司 | Method for disease diagnosis and treatment scheme based on generalized neural network clustering |
| CN104951649A (en) * | 2015-05-27 | 2015-09-30 | 华南农业大学 | HBV classifying method based on Gaussian blur integrals |
-
2015
- 2015-12-31 CN CN201511029760.6A patent/CN105574351B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704961A (en) * | 2004-05-26 | 2005-12-07 | 冲博子 | Analysis method for diagnosis and treatment behavior and administration |
| CN101334843A (en) * | 2007-06-29 | 2008-12-31 | 中国科学院自动化研究所 | Pattern recognition characteristic extraction method and apparatus |
| CN103745227A (en) * | 2013-12-31 | 2014-04-23 | 沈阳航空航天大学 | Method for identifying benign and malignant lung nodules based on multi-dimensional information |
| CN104951649A (en) * | 2015-05-27 | 2015-09-30 | 华南农业大学 | HBV classifying method based on Gaussian blur integrals |
| CN104915560A (en) * | 2015-06-11 | 2015-09-16 | 万达信息股份有限公司 | Method for disease diagnosis and treatment scheme based on generalized neural network clustering |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105574351A (en) | 2016-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220375611A1 (en) | Determination of health sciences recommendations | |
| Müller-Putz et al. | Better than random: a closer look on BCI results | |
| CN105574351B (en) | Medical data processing method | |
| US7809660B2 (en) | System and method to optimize control cohorts using clustering algorithms | |
| Çalişir et al. | A new intelligent hepatitis diagnosis system: PCA–LSSVM | |
| Prichep et al. | Classification of traumatic brain injury severity using informed data reduction in a series of binary classifier algorithms | |
| Alpan et al. | Classification of diabetes dataset with data mining techniques by using WEKA approach | |
| Liao et al. | Appropriate medical data categorization for data mining classification techniques | |
| Poon et al. | A novel approach in discovering significant interactions from TCM patient prescription data | |
| CN106228000A (en) | Over-treatment detecting system and method | |
| Lee et al. | High-throughput analysis of clinical flow cytometry data by automated gating | |
| Gerber et al. | Automated discovery of functional generality of human gene expression programs | |
| Prabhakar et al. | Factor analysis, Hessian Local Linear Embedding and Isomap for epilepsy classification from EEG | |
| Kumari et al. | A deep learning-based approach for accurate diagnosis of alcohol usage severity using EEG signals | |
| Seiler et al. | Uncertainty quantification in multivariate mixed models for mass cytometry data | |
| Liu et al. | Cross-generation and cross-laboratory predictions of Affymetrix microarrays by rank-based methods | |
| Liu et al. | A hybrid double-density dual-tree discrete wavelet transformation and marginal Fisher analysis for scoring sleep stages from unprocessed single-channel electroencephalogram | |
| Liang et al. | Abnormal discharge detection using adaptive neuro-fuzzy inference method with probability density-based feature and modified subtractive clustering | |
| Balli | Use of XGBoost Algorithm in Classification of EEG Signals | |
| CN110957010A (en) | Immune age model learning method | |
| Ying et al. | Nursing scheme based on back propagation neural network and probabilistic neural network in chronic kidney disease | |
| CN105653866B (en) | Disease factor data processing method and system | |
| Venkatesh et al. | Classification of cancer gene expressions from micro-array analysis | |
| Noertjahjani et al. | Classification of epileptic and non-epileptic EEG events by feature selection f-score | |
| Fowler et al. | Dynamic Bayesian clustering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |