CN109841278A - A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method - Google Patents
A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method Download PDFInfo
- Publication number
- CN109841278A CN109841278A CN201711217203.6A CN201711217203A CN109841278A CN 109841278 A CN109841278 A CN 109841278A CN 201711217203 A CN201711217203 A CN 201711217203A CN 109841278 A CN109841278 A CN 109841278A
- Authority
- CN
- China
- Prior art keywords
- disease
- rote learning
- coronary disease
- coronary
- screening
- 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
Abstract
A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method, the specimen of multiple subjects is tested with the set group with multinomial cardiovascular mark, and the result of inspection and its corresponding coronary disease disease state are input in rote learning machine, then by selecting the variable value and corresponding coronary disease disease state that the optimal several cardiovascular marks of classification efficiency are examined in the rote learning machine, coronary disease disease prediction model is established via supervised rote learning operation method, new subject's specimen is examined into obtained data through multinomial cardiovascular mark case group again, it is input to the risk assessment for suffer from coronary disease disease in above-mentioned coronary disease disease prediction model, general group's coronary disease disease screening convenience can be not only improved by the above method, economy and correctness can more be improved.
Description
Technical field
The present invention is a kind of method of coronary disease disease screening, and espespecially one kind combines cardiovascular mark and rote learning operation method
Obtained from coronary disease disease screening the method that the screening of coronary disease disease is carried out with cardiovascular mark and rote learning operation method.
Background technique
Cardiovascular disease is the underlying cause of death of many developing countries, advanced country, especially coronary disease disease and its at any time may be used
Myocardial infarction caused by energy, and the treatment of coronary disease disease and look after, even more heavy burden is brought for society.If therefore in early stage
Coronary disease disease is diagnosed, then can reduce myocardial infarction and other complication, however previously in the related technology, and there is no real
The method for applying simple screening coronary disease disease, the tool of existing coronary disease disease screening is time-consuming, spends height, is radiant exposure, relatively hazardous, and must face upward
Rely artificial judgment.
For example, usually asymptomatic, relative healths group coronary disease disease screening to be carried out, can use following method: 1. hearts
Dirty nucleon radiography；2. cardiac catheter；3. coronary artery computerized tomography, to allow sufferer to be screened out in the case where non-evident sympton
Potential coronary disease disease.Although above-mentioned screening methods effect is good, all there is certain limitations.Firstly, heart nucleon radiography,
Cardiac catheterization, coronary artery computerized tomography checking process in all along with the radioactive exposure of a large amount.Cardiac catheterization
Though accuracy rate highest also has coronary artery rupture risk simultaneously.Coronary artery computerized tomography is invasive at present
Lower and high-accuracy simultaneously coronary disease disease screening methods, however this method must be dependent on the use of computerized tomography, remove
Other than the problem of exit dose exposure, the instrument of computerized tomography, inspection fee or relatively expensive are unable to satisfy group big
The demand of scale screening.
The another way system commonly used is penetrated and is tested using cardiovascular mark case group, however cardiovascular mark case group
Contain the inspection numerical value of quite multinomial cardiovascular mark, the conventional interpretation for still relying upon personnel mostly of medical treatment at present.In interpretation
Method part is mostly foundation using the threshold value of each cardiovascular mark, implies that: if the inspection numerical value of any cardiovascular mark exceeds
Threshold value, that is, predict that the testee has the risk of potential coronary disease disease higher.However, such way usually can not be to whole number
According to synthesis profile shape judged, thus influence its accurate efficiency in clinical use.
In general, these ways are not only not convenient, price is high, and it is potential iatrogenic that subject may be made to be exposed to
It is not ideal way under injury and amount of radiation.In summary, lack feasible technology at present, it is convenient to asymptomatic
Coronary disease disease screening is done in general group.
Summary of the invention
Therefore, the purpose of the present invention is to provide one kind carries out coronary disease disease sieve with angiocarpy mark and rote learning operation method
The method of inspection can be directed to the general group of non-evident sympton, provide one kind and carried out with angiocarpy mark and rote learning operation method
The method of coronary disease disease screening, convenience, economy and the correctness of coronary disease disease screening can be improved simultaneously.
To achieve the above object, the invention discloses one kind carries out coronary disease disease with angiocarpy mark and rote learning operation method
The method of screening, it is characterised in that comprise the steps of
A. multiple subject's specimen are tested with the set group with multinomial cardiovascular mark, and by the result of inspection and
Its corresponding coronary disease disease state is input in rote learning machine；
B. the variable value of the optimal several cardiovascular marks of classification efficiency is selected through the rote learning machine；
C. using selecting after variable value and corresponding coronary disease disease state, by supervised rote learning operation
Method establishes coronary disease disease prediction model；
D. new subject's specimen is examined into obtained result data through multinomial cardiovascular mark case group, be input to above-mentioned
Operation and analysis is compared in coronary disease disease prediction model, and makes the risk assessment for suffering from coronary disease disease.
Wherein, when the interpretation result of coronary disease disease prediction model have potential risk, will to new subject carry out pre-alert notification.
Wherein, the coronary disease disease state to be to there is coronary disease disease/state classification without coronary disease disease, or tight with coronary disease disease
Weight degree classification.
Wherein, the date of inspection for determining date and multinomial cardiovascular mark case group of coronary disease disease state, the two are separated by
Time is between 1 day to 3 years.
Wherein, the rote learning machine system of step B is selected using variable selection method, then pick out classification efficiency it is best
Several cardiovascular indexed variables.
Wherein, the multinomial cardiovascular mark that rote learning machine is selected is high-density lipoprotein, low-density lipoprotein, three acid
Glycerolipid, total cholesterol, blood glucose, micro- albumin, glycated hemoglobin, c reactive protein, homocysteine, lipoprotein, uric acid, the heart
Flesh troponin, creatine phosphatization ferment, Type B benefit natriuresis victory, primary Type B Li Na victory, precalcitonin, erythrosedimentation rate, cream
Acidohydrogenase, sodium ion, potassium ion, calcium ion, chloride ion, magnesium ion, ferrous ion, iron ion, urea nitrogen, creatinine, Guang
PROTEIN C, bilirubin, ketoboidies, pH-value or above-mentioned any combination.
Wherein, the specimen of subject is blood, urine, saliva, sweat, excrement, hydrothorax, ascites or the myelencephalon of human body
Liquid.
Wherein, the supervised rote learning operation method that rote learning machine uses be Multiple logistic regression, k adjacent to method, support to
Amount machine, the study of class neural network, decision tree, random forest, bayesian Decision Method or above-mentioned any combination.
Effect of the invention is: culvert can be obtained in unitary sampling in addition to possessing multinomial cardiovascular mark case group
It covers outside the screening result of a variety of potential coronary disease diseases, through supervised rote learning operation method is combined, is able to utmostly from a variety of
Difference in the integrated data of the screening result of potential coronary disease disease, analysis coronary disease disease and non-coronary disease disease case and its vessel landmarks distribution
It is different.Therefore, in addition to being greatly reduced subject's Check-Out Time, increase the outer of convenience and timeliness, be more the reduction of many possibility
Iatrogenic injury and radioactive exposure；In addition, the present invention can find out classification foundation from whole integrated data distribution complexion,
Timeliness, correctness and the reproducibility when general group's coronary disease disease screening are provided, through combining clinical medicine and information engineering
The advantage of calculation can be applied widely in the screening of coronary disease disease, and then increase the progress of medical diagnosis.
Detailed description of the invention
Fig. 1 is block flow diagram, illustrates process and its risk assessment that coronary disease disease prediction model is established.
Fig. 2 is column diagram, and the prediction of cardiovascular mark and rote learning operation method is illustrated using area under ROC curve as index
Efficiency.
Specific embodiment
Before the present invention is described in detail, it shall be noted that in the numerical value of the following description content, when cannot be limited with this
The range that the present invention is implemented.
Refering to fig. 1, the process and its risk assessment of the foundation of coronary disease disease prediction model are illustrated for the present invention, includes following step
It is rapid: firstly, by specimen such as the blood of multiple subjects, urine, saliva, sweat, excrement, hydrothorax, ascites or celiolymphs, with tool
There is the set group of multinomial cardiovascular mark to test, and the result of inspection and its corresponding coronary disease disease state are input to
In rote learning machine, the coronary disease disease state can be according to there is coronary disease disease/state classification without coronary disease disease, or with coronary disease
Disease severity classification；Then the method that can be selected in rote learning machine using variable, it is optimal to pick out classification efficiency
The variable value of several cardiovascular marks, then coronary disease disease is established by the supervised rote learning operation method inside rote learning machine
Prediction model；Finally, new subject's specimen is examined obtained result data through multinomial cardiovascular mark case group, it is input to
Operation and analysis are carried out in above-mentioned coronary disease disease prediction model, can assess the risk whether new subject has potential coronary disease disease,
When the interpretation result of coronary disease disease prediction model is will to carry out pre-alert notification with potential risk to new subject, and remind subject
Follow-up action can be taken, such as the treatment means etc. taken are discussed with doctor.
It is noted that the date of inspection for determining date and multinomial cardiovascular mark case group of coronary disease disease state, two
Person is separated by a period of time, and interval time may be between 1 day to 3 years due to the difference of application.
Among the above, multinomial cardiovascular mark be high-density lipoprotein, low-density lipoprotein, triglyceride, total cholesterol,
Blood glucose, micro- albumin, glycated hemoglobin, c reactive protein, homocysteine, lipoprotein, uric acid, cardiac troponin, creatine phosphorus
Change ferment, Type B benefit natriuresis victory, primary Type B Li Na victory, precalcitonin, erythrosedimentation rate, lactic acid dehydrogenase, sodium ion, potassium
Ion, calcium ion, chloride ion, magnesium ion, ferrous ion, iron ion, urea nitrogen, creatinine, Cystatin C, bilirubin, ketoboidies,
PH-value or above-mentioned any combination.
Among the above, the supervised rote learning operation method that rote learning machine uses is Multiple logistic regression (Logistic
Regression), k is adjacent to method (k Nearest Neighbor), support vector machines (Support Vector Machine), class
Neural network learns (Artificial Neural Network), decision tree operation method (Decision Tree Induction
Algorithm), random forest operation method (Random Forest Algorithm), bayesian Decision Method (Bayesian
Classification Algorithms) or above-mentioned any combination.
Specific embodiment is as follows:
One, condition (being included in, exclusion condition), the number of subject:
Subject is the cardiovascular marker screening set group inspection of receiving greater than 20 years old adult.The present embodiment is returned using case history
It traces back, is not required to other subjects recruitment.
Two, design and method:
Major Clinical information examines measured value for gender (sex), age (age), body quality index (Body Mass
Index, BMI), hypertension history (Hypertension), diabetes medical history (Diabetes mellitus), high density lipoprotein level
White (High Density Lipoprotein, HDL), low-density lipoprotein (Low Density Lipoprotein, LDL), three
The measurement of acid glyceride (Triglycerol, TG), glycated hemoglobin (glycosylated hemoglobin, HbA1C).This
543 adults, blood drawing all has simultaneously receives cardiac catheterization to confirm disease condition coronarius.
Feature Selection: after carrying out preliminary data cleaning, the present embodiment uses univariate analysis, and the selection of Dependent variable, characteristic is suitable
When univariate statistics method (chi-square test and t examine and determine), triglyceride, low density cholesterol, total cholesterol, saccharification can be selected
The preferable feature of the classifying qualities such as hemochrome, high density cholesterol, the feature as following model training.
After Feature Selection completion, the present embodiment establishes several supervised learning models according to this, includes: k is adjacent to method, branch
Hold vector machine, class neural network.
Three, during data backtracking, during the present embodiment execution:
Data is recalled therebetween from the March 31st, 1 day 1 of September in 2010.
Four, the assessment and statistical method of result:
The present embodiment calculates the potassium ion distribution of various decentraction vessel landmarks object data, and variable value and its numerical value according to this
Training prediction model.The present embodiment verifies the predictive ability of each model for the group of internal verification.Model training use with
The method of machine grouping and cross-validation takes out model to carry out the training and study of each supervised learning model first
Training group data gradually selects optimum model parameter using cross-validation method.The effect of prediction model will be grasped with recipient
Make indicatrix (ROC curve) to be assessed, and calculates its area under the curve simultaneously.
Fig. 2 is that area, using single check mark, there is three acid as index evaluation coronary disease disease prediction efficiency using under ROC curve
Glycerolipid, low density cholesterol, total cholesterol, glycated hemoglobin, high density cholesterol) and rote learning machine (support vector machines,
K is neighbouring, class neural network), analyzing cardiovascular examines the coronary disease disease prediction efficiency of set group, various decentraction vessel landmarks whereby,
And different supervised rote learning operation method analyzing cardiovasculars examine the coronary disease disease screening efficiency after set group.Its operating characteristic curve
Lower area demonstrates, the area under the curve at most 0.7 or so of single angiocarpy check mark；If but being transported with supervised rote learning
After algorithm analyzing cardiovascular examines set group (containing multiple cardiovascular marks), the efficiency of coronary disease disease prediction can be substantially improved to
0.9 or so.This example demonstrates analyze suitable cardiovascular mark case group (containing more using supervised rote learning operation method
A angiocarpy check mark), accurate coronary disease disease prediction can be made.The heart is carried out through different supervised rote learning operation methods
The study of vessel landmarks set group data, can significantly improve the screening efficiency of potential coronary disease disease.
In conclusion the present invention can be obtained and cover in unitary sampling in addition to possessing multinomial cardiovascular mark case group
Outside the testing result of a variety of cardiovascular marks, through supervised rote learning operation method is combined, it is able to utmostly from a variety of hearts
Difference in the detection data of vessel landmarks, analysis coronary disease disease and non-coronary disease disease case and its vessel landmarks distribution, from entirety
Classification foundation is found out in integrated data distribution complexion, and train the coronary disease disease prediction model completed that can also be copied in many aspects to make
The terminating machine of user is carried out using therefore can widely the screening of coronary disease disease being applied, and then increase the progress of medical diagnosis, correct
Property, in timeliness, economic benefit and reproducibility, with traditional artificial interpretation compared to all acquisition major improvements.
Claims (8)
1. a kind of method for carrying out the screening of coronary disease disease with cardiovascular mark and rote learning operation method, it is characterised in that comprising following
Step:
A. multiple subject's specimen are tested with the set group with multinomial cardiovascular mark, and by the result and its phase of inspection
Corresponding coronary disease disease state is input in rote learning machine；
B. the variable value of the optimal several cardiovascular marks of classification efficiency is selected through the rote learning machine；
C. using selecting after variable value and corresponding coronary disease disease state, built by supervised rote learning operation method
Vertical coronary disease disease prediction model；
D. new subject's specimen is examined into obtained result data through multinomial cardiovascular mark case group, is input to above-mentioned coronary disease
Operation and analysis are compared in disease prediction model, and makes the risk assessment for suffering from coronary disease disease.
2. the method for carrying out the screening of coronary disease disease as described in claim 1 with cardiovascular mark and rote learning operation method, special
Sign is, when the interpretation result of coronary disease disease prediction model has potential risk, will carry out pre-alert notification to new subject.
3. the method for carrying out the screening of coronary disease disease as described in claim 1 with cardiovascular mark and rote learning operation method, special
Sign is, the coronary disease disease state is to there is coronary disease disease/state classification without coronary disease disease, or with coronary disease disease severity
Classification.
4. the method for carrying out the screening of coronary disease disease as claimed in claim 3 with cardiovascular mark and rote learning operation method, special
Sign is that the date of inspection for determining date and multinomial cardiovascular mark case group of coronary disease disease state, the two interval time is 1
It is between 3 years.
5. the method for carrying out the screening of coronary disease disease as described in claim 1 with cardiovascular mark and rote learning operation method, special
Sign is that the rote learning machine system of step B is selected using variable selection method, then pick out classification efficiency it is optimal several
Cardiovascular indexed variable.
6. the method for carrying out the screening of coronary disease disease as described in claim 1 with cardiovascular mark and rote learning operation method, special
Sign is that the multinomial cardiovascular mark that rote learning machine is selected is high-density lipoprotein, low-density lipoprotein, three acid glycerols
Rouge, total cholesterol, blood glucose, micro- albumin, glycated hemoglobin, c reactive protein, homocysteine, lipoprotein, uric acid, myocardium myo
Calcium albumen, creatine phosphatization ferment, Type B benefit natriuresis victory, primary Type B Li Na victory, precalcitonin, erythrosedimentation rate, lactic acid are gone
Hydrogen enzyme, sodium ion, potassium ion, calcium ion, chloride ion, magnesium ion, ferrous ion, iron ion, urea nitrogen, creatinine, Guang albumen
C, bilirubin, ketoboidies, pH-value or above-mentioned any combination.
7. the method for carrying out the screening of coronary disease disease as described in claim 1 with cardiovascular mark and rote learning operation method, special
Sign is that the specimen of subject is blood, urine, saliva, sweat, excrement, hydrothorax, ascites or the celiolymph of human body.
8. the method for carrying out the screening of coronary disease disease as described in claim 1 with cardiovascular mark and rote learning operation method, special
Sign is that the supervised rote learning operation method that rote learning machine uses is Multiple logistic regression, k adjacent to method, support vector machines, class
Neural network study, decision tree, random forest, bayesian Decision Method or above-mentioned any combination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711217203.6A CN109841278A (en) | 2017-11-28 | 2017-11-28 | A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711217203.6A CN109841278A (en) | 2017-11-28 | 2017-11-28 | A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109841278A true CN109841278A (en) | 2019-06-04 |
Family
ID=66881072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711217203.6A Pending CN109841278A (en) | 2017-11-28 | 2017-11-28 | A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109841278A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110349675A (en) * | 2019-07-16 | 2019-10-18 | 广东工业大学 | A kind of pre- measurement equipment of cardiovascular disease and device |
| TWI755731B (en) * | 2020-05-13 | 2022-02-21 | 國家中山科學研究院 | Biomedical data risk prediction system and method |
| WO2022229444A3 (en) * | 2021-04-30 | 2022-12-01 | Roche Diagnostics Gmbh | Pct marker panels for early detection of sepsis |
| CN115862843A (en) * | 2022-12-12 | 2023-03-28 | 哈尔滨医科大学 | Auxiliary identification system and equipment for myocardial muscle calcium egg rising type and cardiovascular diseases |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050020903A1 (en) * | 2003-06-25 | 2005-01-27 | Sriram Krishnan | Systems and methods for automated diagnosis and decision support for heart related diseases and conditions |
| CN105512477A (en) * | 2015-12-03 | 2016-04-20 | 万达信息股份有限公司 | Unplanned readmission risk assessment prediction model based on dimension reduction combination classification algorithm |
| CN106874663A (en) * | 2017-01-26 | 2017-06-20 | 中电科软件信息服务有限公司 | Cardiovascular and cerebrovascular disease Risk Forecast Method and system |
| CN106909769A (en) * | 2016-09-30 | 2017-06-30 | 马立明 | Big data risk of cardiovascular diseases monitoring system |
-
2017
- 2017-11-28 CN CN201711217203.6A patent/CN109841278A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050020903A1 (en) * | 2003-06-25 | 2005-01-27 | Sriram Krishnan | Systems and methods for automated diagnosis and decision support for heart related diseases and conditions |
| CN105512477A (en) * | 2015-12-03 | 2016-04-20 | 万达信息股份有限公司 | Unplanned readmission risk assessment prediction model based on dimension reduction combination classification algorithm |
| CN106909769A (en) * | 2016-09-30 | 2017-06-30 | 马立明 | Big data risk of cardiovascular diseases monitoring system |
| CN106874663A (en) * | 2017-01-26 | 2017-06-20 | 中电科软件信息服务有限公司 | Cardiovascular and cerebrovascular disease Risk Forecast Method and system |
Non-Patent Citations (1)
| Title |
|---|
| 杨进等: "冠心病的风险评估研究", 《西北大学学报（自然科学版）》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110349675A (en) * | 2019-07-16 | 2019-10-18 | 广东工业大学 | A kind of pre- measurement equipment of cardiovascular disease and device |
| TWI755731B (en) * | 2020-05-13 | 2022-02-21 | 國家中山科學研究院 | Biomedical data risk prediction system and method |
| WO2022229444A3 (en) * | 2021-04-30 | 2022-12-01 | Roche Diagnostics Gmbh | Pct marker panels for early detection of sepsis |
| CN115862843A (en) * | 2022-12-12 | 2023-03-28 | 哈尔滨医科大学 | Auxiliary identification system and equipment for myocardial muscle calcium egg rising type and cardiovascular diseases |
| CN115862843B (en) * | 2022-12-12 | 2024-02-02 | 哈尔滨医科大学 | Auxiliary identification system and equipment for myocardial troponin elevation type and cardiovascular diseases |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Coronary artery calcium score quantification using a deep-learning algorithm | |
| CN109841278A (en) | A method of the screening of coronary disease disease is carried out with angiocarpy mark and rote learning operation method | |
| CN107153774A (en) | The disease forecasting system of the structure and application of chronic disease risk assessment the hyperbolic model model | |
| Khanna et al. | Performance evaluation of 10-year ultrasound image-based stroke/cardiovascular (CV) risk calculator by comparing against ten conventional CV risk calculators: A diabetic study | |
| Bozorgmanesh et al. | CVD-predictive performances of “a body shape index” versus simple anthropometric measures: Tehran lipid and glucose study | |
| CN106339593A (en) | Kawasaki disease classification and prediction method based on medical data modeling | |
| Garcia et al. | Diagnostic performance of an artificial intelligence-driven cardiac-structured reporting system for myocardial perfusion SPECT imaging | |
| Raynor et al. | Associations of retrospective and concurrent lipid levels with subclinical atherosclerosis prediction after 20 years of follow-up: the Coronary Artery Risk Development in Young Adults (CARDIA) study | |
| CN113490987A (en) | Methods for assessing muscle-related disorders | |
| Hari | Muscular abnormalities in liver cirrhosis | |
| Scheel et al. | The inter-rater reliability and individual reviewer performance of the 2012 world heart federation guidelines for the echocardiographic diagnosis of latent rheumatic heart disease | |
| Athilingam et al. | Predictive validity of NYHA and ACC/AHA classifications of physical and cognitive functioning in heart failure | |
| Bakogiannis et al. | Hypertrophic cardiomyopathy or athlete’s heart? A systematic review of novel cardiovascular magnetic resonance imaging parameters | |
| Liu et al. | A joint multi-modal learning method for early-stage knee osteoarthritis disease classification | |
| Avram et al. | Automated assessment of cardiac systolic function from coronary angiograms with video-based artificial intelligence algorithms | |
| TWI641963B (en) | Method for screening coronary heart disease by cardiovascular marker and mechanical learning algorithm | |
| CN113707272A (en) | Evaluation model for suitable population for interventional therapy of radial artery access liver cancer and construction method | |
| CN113128654B (en) | Improved random forest model for coronary heart disease pre-diagnosis and pre-diagnosis system thereof | |
| CN113257416B (en) | COPD patient personalized management and optimization device and equipment based on deep learning | |
| Sancar et al. | Body mass index estimation by using an adaptive neuro fuzzy inference system | |
| CN108346471A (en) | A kind of analysis method and device of pathological data | |
| Horbal et al. | Morphomic calcification score from clinical CT scans: A proxy for coronary artery calcium | |
| Aditya et al. | Machine Learning approach for cardiovascular risk and coronary artery calcification score | |
| Jegelevicius et al. | Web-based health services and clinical decision support | |
| Philipp et al. | Annotation-Efficient Strategy for Segmentation of 3D Body Composition |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190604 |