CN108597614A - A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record - Google Patents

A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record Download PDF

Info

Publication number
CN108597614A
CN108597614A CN201810324593.5A CN201810324593A CN108597614A CN 108597614 A CN108597614 A CN 108597614A CN 201810324593 A CN201810324593 A CN 201810324593A CN 108597614 A CN108597614 A CN 108597614A
Authority
CN
China
Prior art keywords
disease
electronic health
health record
symptom
data
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
Application number
CN201810324593.5A
Other languages
Chinese (zh)
Inventor
闫凤麒
曲艺
陆明名
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Hee Hee Mdt Infotech Ltd
Original Assignee
Shanghai Hee Hee Mdt Infotech Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Hee Hee Mdt Infotech Ltd filed Critical Shanghai Hee Hee Mdt Infotech Ltd
Priority to CN201810324593.5A priority Critical patent/CN108597614A/en
Publication of CN108597614A publication Critical patent/CN108597614A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70ICT 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
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Epidemiology (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

Auxiliary diagnosis decision-making technique is that one in Clinical Decision Support Systems (Clinical decision support system, CDSS) actively applies branch very much.The exact classification that it electronic health record data set using Chinese that current auxiliary diagnosis decision-making technique fails to well, can not effectively excavate the correlation rule of symptom and disease in Chinese electronic health record, can not reduce symptom characteristic dimension well, can not carry out a variety of diseases in higher-dimension symptom characteristic space very accurately.For these problems, The present invention gives a kind of auxiliary diagnosis decision-making techniques based on Chinese electronic health record.The disease that Chinese electronic health record is concentrated is cleaned with symptom information first, then the correlation rule of disease and symptom affairs is excavated, the present invention is ranked up according to disease and the confidence level of symptom affairs correlation rule, and therefrom carry out feature selecting on the basis of base grader classification results, classification of diseases is carried out with decision tree classifier according to the feature vector that feature selecting goes out, realizes the prediction to disease.

Description

A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record
Technical field
The present invention relates to Data Minings, and in particular to feature selecting algorithm, sorting algorithm.
Background technology
Earliest auxiliary diagnosis decision-making technique mainly all concentrates on Knowledge based engineering CDSS researchs and realizes, system structure Substantially it is made of man-machine interface, knowledge base and inference machine three parts.Knowledge base is by the rule from medical literature, clinical guidelines Etc. building, and according to these knowledge transformations at rule and determined by human-computer interaction interface offer with certain inference mechanism Plan supports service.But medicine and medical knowledge are developing always, are changing, and knowledge in knowledge base to be enable to keep up with development Step need huge input and consumption.So the CDSS based on non-knowledge, i.e. the auxiliary diagnosis decision based on electronic health record The research of method has gradually become the hot spot and main flow direction of Current Diagnostic Study on DSS.
Different from Knowledge based engineering CDSS, the auxiliary diagnosis decision-making technique based on electronic health record passes through to extensive electronics disease Go through being associated the operations such as rule digging, classification, recurrence and capable of constantly finding new knowledge to help doctor in disease for data set Better decision is made during disease diagnosis.The Ministry of Public Health formulated the fundamental norms of electronic health record with 2011, and in the whole nation The pilot work of 110 hospital electronic health records in range.Through development in a few years, electronic medical record system has become measurement doctor One major criterion of institute's IT application level, electronic medical record system are also accumulated from considerable data volume, and these Data are all the truthful datas of patient, this all has diagnosis and treatment and medical research of disease etc. huge value and wide development Foreground, applying especially based on electronic health record have played important function during the medical diagnosis on disease of CDSS, and into One step advances the research and development of domestic disease diagnosing system.
Auxiliary diagnosis Research of Decision based on Chinese electronic health record is still at an early stage, at home, Zhou Zhihua, ginger It is remote et al. that machine learning model is analyzed for disease forecasting but is Single diseases disease forecasting model, it is difficult to be directly applied for In the prediction of the multi-class medical data collection of multiple features.Although a large amount of artificial neural network has been used to medical diagnosis, refreshing Need to extract feature from a large amount of sample with Training diagnosis prediction model, it is difficult to be directly used in structure minority class through network model Medical conditions diagnostic model.2007, Jiang Lin etc. used SVM technologies, and 2 type glycosurias are established to 14 features of 436 cases Disease forecasting model, to improve estimated performance, it is proposed that a kind of optimal feature subset selection method finally selects 4 kinds of features, this four A feature corresponds to index highest (susceptibility 86.66%, specificity 64.22%, accuracy rate 70.14%).Meanwhile also using decision Tree, multi-layer perception (MLP) method are tested, the results showed that support the effect of SVM best.
It is existing to study mainly for Knowledge based engineering auxiliary diagnosis decision-making technique, while in Chinese electronic health record data set Because there are a large amount of characteristic attribute, causing characteristic attribute dimension height to cause, matrix operation amount is big, training sample is sparse and excessively quasi- The problems such as conjunction, limits always the classification quality of traditional classifier, so feature selecting is can not ignore in classification of diseases problem One important engineering.
Invention content
The present invention provides a kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record, utilizes Apriori correlation rules Algorithm (being already belonging to existing algorithm) carries out the excavation of illness rule, and feature row is carried out using the confidence level of illness rule as standard Then sequence is selected to divide forward, be selected most according to the sequence that the classifying quality of base grader is standard progress character subset again It is trained with decision tree classifier centering text electronic health record after excellent character subset, finally completes trained grader again Auxiliary diagnosis decision task.
It is achieved through the following technical solutions:
A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record, which is characterized in that include the following steps, it is a kind of right Disease and symptom information carry out the data preprocessing method of data cleansing in Chinese electronic health record data set, one kind based on disease with The method that symptom is associated rule digging, a method of feature selecting being carried out based on illness correlation rule, one kind is based on pass The feature selection approach of connection rule carries out the grader of disease forecasting after selecting feature vector;
The sub- medical record data of a kind of centering message concentrates the data prediction of disease and symptom information progress data cleansing Method, this method first have to Chinese electronic health record data of the removal containing value of having vacant position and (are already belonging to routine techniques in the prior art Means), it is secondly standardized the attribute value in i.e. uniform data source to the data information of disease and symptom, then carries out feature two Value is to convert character type data to the two-value data of Boolean type, finally carries out label coding to disease category.
A kind of method being associated rule digging based on disease and symptom, core operation are to be calculated with Apriori Method is associated rule digging to disease and symptom.
A kind of method carrying out feature selecting based on illness correlation rule, core operation are two frequent episodes to illness Collection rule carries out confidence level sequence, and the classification performance of base grader is then recycled to carry out sequence as character subset evaluation criterion Selection method SFS selects character subset forward.
A kind of point carrying out disease forecasting after selecting feature vector based on the feature selection approach of correlation rule Class device, core operation are the character subset selected first according to above method, then recycle decision tree classifier to Chinese Electronic health record is trained and predicts.
Advantageous effect
Disease and the data of symptom information progress data cleansing is concentrated to locate in advance the sub- medical record data of middle message in the present invention In reason method:Make symptom and disease data standardized format by building near synonym phrase, then by the disease in Chinese electronic health record Shape characteristic information is converted into the higher-dimension sparse matrix of Boolean type.
It is associated in the method for rule digging based on disease and symptom in the present invention:Using Apriori algorithm to disease During disease is associated rule digging with symptom, the illness that the method for closing Frequent Set item eliminates partial redundance is used to close Connection rule.
In the method for the illness correlation rule progress feature selecting of the present invention:It is calculated by using Apriori correlation rules Method carries out rule digging to the data set of disease and symptom, then carries out feature ordering by the confidence level of two frequent item sets rule After being divided as the character subset of standard using base grader classifying quality, feature more preferable than traditional feature selection approach effect Subset smaller.
Disease forecasting is carried out after selecting feature vector based on the feature selection approach of correlation rule in the present invention In the method for grader:After obtaining character subset by the above-mentioned feature selection approach based on correlation rule, in conjunction with decision tree point Disease is trained with symptom data collection, adjusts ginseng in class device centering text electronic health record, final to obtain classification of diseases model, with tradition Grader classifying quality is compared, accuracy higher of classifying.
Compared to traditional auxiliary diagnosis decision-making technique, the present invention fundamentally improves complicated more in Chinese electronic health record The disease forecasting accuracy rate of symptom characteristic.This research method, which is suitable for providing a large amount of medicine for clinician, to be supported, to help Clinician is helped to make most rational diagnosis, selection optimal treatment measure.
Description of the drawings
Attached drawing is for carrying a further understanding of the present invention, and a part for constitution instruction, with following tool Body embodiment is used to explain the disclosure together, but is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the Technology Roadmap of the auxiliary diagnosis decision-making technique based on Chinese electronic health record;
Fig. 2 is the flow chart of Chinese electronic health record data prediction.
Fig. 3 is the flow chart of 2 Frequent Sets of excavation illness based on Apriori algorithm.
The flow chart of feature selecting algorithm of the positions Fig. 4 based on correlation rule.
Specific implementation mode
Auxiliary diagnosis decision-making technique be Clinical Decision Support Systems (Clinical decision support system, CDSS one in) is actively applied branch very much.Current auxiliary diagnosis decision-making technique fails to utilize Chinese electronics well Medical record data collection can not effectively excavate the correlation rule of symptom and disease in Chinese electronic health record, can not reduce disease well Shape characteristic dimension, the exact classification that a variety of diseases can not be carried out in higher-dimension symptom characteristic space very accurately.It is asked for these Topic, The present invention gives a kind of auxiliary diagnosis decision-making techniques based on Chinese electronic health record.Chinese electronic health record is concentrated first Disease cleaned with symptom information, then excavate disease and symptom affairs correlation rule, the present invention is according to disease and disease The confidence level of shape affairs correlation rule is ranked up, and therefrom carries out feature selecting on the basis of base grader classification results, Classification of diseases is carried out with decision tree classifier according to the feature vector that feature selecting goes out, realizes the prediction to disease.
In order to deepen the understanding of the present invention, below in conjunction with existing method and attached drawing 2, the invention will be further described, Existing method is only used for explaining the present invention, is not intended to limit the scope of the present invention..
Content included by the present invention:A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record includes four portions Point:A kind of sub- medical record data of centering message concentrates the data preprocessing method of disease and symptom information progress data cleansing;It is a kind of The method for being associated rule digging based on disease and symptom;A method of feature selecting is carried out based on illness correlation rule; A kind of grader carrying out disease forecasting after selecting feature vector based on the feature selection approach of correlation rule.
The first step:The sub- medical record data of centering message is concentrated carries out pretreatment work to disease and symptom information, such as Fig. 2 institutes Show.
1.1:Chinese electronic health record data of the removal containing value of having vacant position;
1.2:The attribute value in i.e. uniform data source is standardized to the data information of disease and symptom;
1.3:It is to convert character type data to the two-value data of Boolean type to carry out feature binaryzation;
1.4:Label coding is carried out to disease category.
1.5:Finally obtain illness data set.
Second step:Based on the method that disease and symptom are associated rule digging, that is, it is based on Apriori algorithm principle and excavates Illness binomial Frequent Set, as shown in Figure 3.
2.1:Minimum support threshold value min_sup is defined first, then scans Chinese electronic health record illness data set;
2.2:Judge whether to also need to generate Candidate Set Ci (maximum value of acquiescence i is 2), enter step 2.3 if necessary, If you do not need to being generated into step 2.4;
2.3:Generate the Candidate Set Ci of Chinese electronic health record illness data set;
2.4:It calculates in Candidate Set Ci so the support sup of element, then is compared with minimum support threshold value min_sup Compared with, if Ci set in also have element j be put into step 2.5 more than minimum support threshold value min_sup, be otherwise just put into step Rapid 2.6;
2.5:Generate frequent item set Li;
2.6:The frequent 2 item collection L2 that illness is filtered out from frequent 2 item collection, 2 item collections of remaining non-illness is filtered out, most Throughout one's life at the frequent 2 item collection L2 of symptom.
Third walks:Feature selecting is carried out based on illness correlation rule, as shown in Figure 4.
3.1:Calculate the confidence level of each characteristic attribute in the frequent 2 item collection L2 of illness;
3.2:Two frequent item set L2 of illness are ranked up according to confidence level parameter first;
3.3:Then sequence is carried out to the binomial Frequent Set L2 after sequence again and divides character subset forward, if do not divided It finishes, then enters step 3.4, step 3.5 is put into if dividing to finish;
3.4:Setting divides step-length, and it is 1 that acquiescence, which divides step-length,;
3.5:Finally recycle base grader FtTrained classification accuracy ftAs character subset evaluation criterion 3.6:Iteration Filter out the optimal feature vector V of classification accuracy best grader F and final output.
4th step:The feature vector V selected is walked based on third, then in decision tree classifier centering text electronic health record Disease is trained with symptom data collection, adjusts ginseng, finally obtains optimal classification of diseases device.
Innovative point
Propose a kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record and existing auxiliary diagnosis decision-making technique It compares, it is the training that data source carries out classification of diseases model that the present invention, which has used a large amount of Chinese electronic health record,.To middle message The feature of disease and symptom and classification information have carried out the work of the data predictions such as data cleansing in sub- medical record data, and by disease The matrix of a Boolean is converted into symptom data information completely, in order to reduce the sparse of disease and symptom Boolean matrix Property, the dimension of symptom characteristic value is reduced, present invention firstly provides a kind of feature selection approach based on correlation rule, effectively The dimension for reducing symptom characteristic value improves the accuracy of classification of diseases.
Method proposed by the present invention is concentrated with good performance in the electronic health record data of the more disease categories of more symptom characteristics, It improves the accuracy rate of a variety of classification of diseases predictions and reduces the dimension of feature vector.

Claims (5)

1. a kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record, which is characterized in that based on the auxiliary of Chinese electronic health record The diagnosis decision method is helped to include:
Step 1: a kind of sub- medical record data of centering message concentrates the data prediction side of disease and symptom information progress data cleansing Method;
Step 2: a kind of method being associated rule digging based on disease and symptom;
Step 3: a kind of method carrying out feature selecting based on illness correlation rule;(crucial characteristic step)
Step 4: a kind of classification carrying out disease forecasting after selecting feature vector based on the feature selection approach of correlation rule Device.
2. the auxiliary diagnosis decision-making technique according to claim 1 based on Chinese electronic health record, which is characterized in that described one The sub- medical record data of kind centering message concentrates the data preprocessing method of disease and symptom information progress data cleansing:
This method first has to removal containing the Chinese electronic health record data of value of having vacant position, secondly to the data information of disease and symptom into Row standardization is the attribute value in uniform data source, then carries out feature binaryzation and converts character type data to the two of Boolean type Value Data finally carries out label coding to disease category.
3. the auxiliary diagnosis decision-making technique according to claim 1 based on Chinese electronic health record, which is characterized in that described one The method that kind is associated rule digging based on disease and symptom, core operation are with Apriori algorithm to disease and symptom It is associated rule digging.
4. the auxiliary diagnosis decision-making technique according to claim 1 based on Chinese electronic health record, which is characterized in that Yi Zhongji In the method that illness correlation rule carries out feature selecting:Core operation is to carry out confidence level row to two frequent item set rules of illness Then sequence recycles the classification performance of base grader to carry out sequence selection method SFS choosings forward as character subset evaluation criterion Select character subset.
5. the auxiliary diagnosis decision-making technique according to claim 1 based on Chinese electronic health record, which is characterized in that Yi Zhongji The grader of disease forecasting is carried out after selecting feature vector in the feature selection approach of correlation rule, core operation is first Then the character subset selected according to above method recycles decision tree classifier centering text electronic health record to be trained and in advance It surveys.
CN201810324593.5A 2018-04-12 2018-04-12 A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record Pending CN108597614A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810324593.5A CN108597614A (en) 2018-04-12 2018-04-12 A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810324593.5A CN108597614A (en) 2018-04-12 2018-04-12 A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record

Publications (1)

Publication Number Publication Date
CN108597614A true CN108597614A (en) 2018-09-28

Family

ID=63622072

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810324593.5A Pending CN108597614A (en) 2018-04-12 2018-04-12 A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record

Country Status (1)

Country Link
CN (1) CN108597614A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109659033A (en) * 2018-12-18 2019-04-19 浙江大学 A kind of chronic disease change of illness state event prediction device based on Recognition with Recurrent Neural Network
CN110781216A (en) * 2019-11-05 2020-02-11 广东工业大学 Traditional Chinese medicine symptom association rule mining method and device and storage medium
CN111192680A (en) * 2019-12-25 2020-05-22 山东众阳健康科技集团有限公司 Intelligent auxiliary diagnosis method based on deep learning and collective classification
CN111292818A (en) * 2020-01-17 2020-06-16 同济大学 Query reconstruction method for electronic medical record description
CN111341454A (en) * 2018-12-19 2020-06-26 中国电信股份有限公司 Data mining method and device
CN111785372A (en) * 2020-05-14 2020-10-16 浙江知盛科技集团有限公司 Collaborative filtering disease prediction system based on association rule and electronic equipment thereof
CN111968747A (en) * 2020-08-20 2020-11-20 卫宁健康科技集团股份有限公司 VTE intelligent prevention and control management system
CN112560900A (en) * 2020-09-08 2021-03-26 同济大学 Multi-disease classifier design method for sample imbalance
CN112635070A (en) * 2020-12-14 2021-04-09 创业慧康科技股份有限公司 Patient clinical information acquisition and display method and device
CN113421653A (en) * 2021-06-23 2021-09-21 平安科技(深圳)有限公司 Medical information pushing method and device, storage medium and computer equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110201529A1 (en) * 2010-02-12 2011-08-18 Liang-Tsung Huang System for analyzing and screening disease related genes using microarray database
CN107358014A (en) * 2016-11-02 2017-11-17 华南师范大学 The clinical pre-treating method and system of a kind of physiological data
CN107451416A (en) * 2017-08-28 2017-12-08 昆明理工大学 A kind of sle auxiliary diagnostic equipment and method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110201529A1 (en) * 2010-02-12 2011-08-18 Liang-Tsung Huang System for analyzing and screening disease related genes using microarray database
CN107358014A (en) * 2016-11-02 2017-11-17 华南师范大学 The clinical pre-treating method and system of a kind of physiological data
CN107451416A (en) * 2017-08-28 2017-12-08 昆明理工大学 A kind of sle auxiliary diagnostic equipment and method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
王美华 等: ""一种基于关联规则与支持向量机的基因表达数据分类模型"", 《计算机应用与软件》 *
胡可云、田凤占、黄厚宽: "《数据挖掘理论与应用》", 30 April 2008, 北京交通大学出版社 *
钟颖、胡雪蕾、陆建峰: ""基于关联规则和决策树的中医胃炎诊断分析"", 《中国中医药信息杂志》 *
陈希蔓: ""维度缩减的原因"", 《HTTPS://SAMANTHACHEN.GITHUB.IO/2016/08/08/机器学习_周志华_笔记11/》 *

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109659033A (en) * 2018-12-18 2019-04-19 浙江大学 A kind of chronic disease change of illness state event prediction device based on Recognition with Recurrent Neural Network
CN111341454B (en) * 2018-12-19 2023-07-25 中国电信股份有限公司 Data mining method and device
CN111341454A (en) * 2018-12-19 2020-06-26 中国电信股份有限公司 Data mining method and device
CN110781216A (en) * 2019-11-05 2020-02-11 广东工业大学 Traditional Chinese medicine symptom association rule mining method and device and storage medium
CN111192680A (en) * 2019-12-25 2020-05-22 山东众阳健康科技集团有限公司 Intelligent auxiliary diagnosis method based on deep learning and collective classification
CN111192680B (en) * 2019-12-25 2021-06-01 山东众阳健康科技集团有限公司 Intelligent auxiliary diagnosis method based on deep learning and collective classification
CN111292818A (en) * 2020-01-17 2020-06-16 同济大学 Query reconstruction method for electronic medical record description
CN111292818B (en) * 2020-01-17 2022-04-19 同济大学 Query reconstruction method for electronic medical record description
CN111785372A (en) * 2020-05-14 2020-10-16 浙江知盛科技集团有限公司 Collaborative filtering disease prediction system based on association rule and electronic equipment thereof
CN111968747A (en) * 2020-08-20 2020-11-20 卫宁健康科技集团股份有限公司 VTE intelligent prevention and control management system
CN111968747B (en) * 2020-08-20 2023-12-12 卫宁健康科技集团股份有限公司 VTE intelligent control management system
CN112560900A (en) * 2020-09-08 2021-03-26 同济大学 Multi-disease classifier design method for sample imbalance
CN112560900B (en) * 2020-09-08 2023-01-20 同济大学 Multi-disease classifier design method for sample imbalance
CN112635070A (en) * 2020-12-14 2021-04-09 创业慧康科技股份有限公司 Patient clinical information acquisition and display method and device
CN113421653B (en) * 2021-06-23 2022-09-09 平安科技(深圳)有限公司 Medical information pushing method and device, storage medium and computer equipment
CN113421653A (en) * 2021-06-23 2021-09-21 平安科技(深圳)有限公司 Medical information pushing method and device, storage medium and computer equipment

Similar Documents

Publication Publication Date Title
CN108597614A (en) A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record
Ye et al. Explainable AI for COVID-19 CT classifiers: an initial comparison study
CN107016665B (en) CT pulmonary nodule detection method based on deep convolutional neural network
CN108198621B (en) Database data comprehensive diagnosis and treatment decision method based on neural network
CN113436698B (en) Automatic medical term standardization system and method integrating self-supervision and active learning
CN113892148A (en) Interpretable AI (xAI) platform for computational pathology
CN102799579B (en) Statistical machine translation method with error self-diagnosis and self-correction functions
CN111223553A (en) Two-stage deep migration learning traditional Chinese medicine tongue diagnosis model
CN105808712A (en) Intelligent system and method for converting text type medical reports into structured data
CN112381178B (en) Medical image classification method based on multi-loss feature learning
CN108875809A (en) The biomedical entity relationship classification method of joint attention mechanism and neural network
Doan et al. SONNET: A self-guided ordinal regression neural network for segmentation and classification of nuclei in large-scale multi-tissue histology images
CN111767952B (en) Interpretable lung nodule benign and malignant classification method
US20230230241A1 (en) System and method for detecting lung abnormalities
CN113255728A (en) Depression classification method based on map embedding and multi-modal brain network
CN108335756B (en) Nasopharyngeal carcinoma database and comprehensive diagnosis and treatment decision method based on database
CN109492106A (en) Text code combined automatic classification method for defect reasons
CN109102498A (en) A kind of method of cluster type nucleus segmentation in cervical smear image
CN114021584A (en) Knowledge representation learning method based on graph convolution network and translation model
CN110288577A (en) A kind of 3D realizes the control method and control device of multiple organ lesion integration
CN113920379A (en) Zero sample image classification method based on knowledge assistance
CN116350231A (en) Semi-supervised electrocardiogram myocardial infarction positioning method based on hierarchical different granularity class constraint
Minh et al. An improvement in integrating clustering method and neural network to extract rules and application in diagnosis support
CN116759076A (en) Unsupervised disease diagnosis method and system based on medical image
CN116228759A (en) Computer-aided diagnosis system and apparatus for renal cell carcinoma type

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20180928

RJ01 Rejection of invention patent application after publication