CN112992370A - Unsupervised electronic medical record-based medical behavior compliance assessment method - Google Patents
Unsupervised electronic medical record-based medical behavior compliance assessment method Download PDFInfo
- Publication number
- CN112992370A CN112992370A CN202110489454.XA CN202110489454A CN112992370A CN 112992370 A CN112992370 A CN 112992370A CN 202110489454 A CN202110489454 A CN 202110489454A CN 112992370 A CN112992370 A CN 112992370A
- Authority
- CN
- China
- Prior art keywords
- medical
- data
- patient
- diagnosis
- treatment
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT 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)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Medical Informatics (AREA)
- Mathematical Physics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Epidemiology (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Probability & Statistics with Applications (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- Algebra (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Molecular Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Fuzzy Systems (AREA)
- Automation & Control Theory (AREA)
- Pathology (AREA)
- Operations Research (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses an unsupervised medical behavior compliance assessment method based on an electronic medical record, and the method comprises the following steps of S1, collecting, cleaning and preprocessing case data; s2, classifying the case data of the patient; s3, clustering the order data of the patients; s4, fusing the medical advice data after patient clustering and the operation data with time series, and mining a diagnosis and treatment process model according to the patient category and the effect of the patients after diagnosis and treatment; and S5, aligning the actual diagnosis and treatment sequence with the excavated diagnosis and treatment process model based on the cost function, positioning the position of the abnormality and calculating the deviation degree of the abnormality. The invention reduces the dependence of prior knowledge, can deeply utilize data, has strong clinical interpretability of an evaluation result, and has high preset logic complexity of the state of illness, physical condition and the like of a patient.
Description
Technical Field
The invention relates to the field of medical data processing and analysis, in particular to an unsupervised electronic medical record-based medical behavior compliance assessment method.
Background
In recent years, along with the increasing living standard of people, the development of the medical health industry also meets a plurality of problems. On one hand, medical expenses are increasing at a relatively fast speed, and in a clinical diagnosis and treatment process, due to interference of benefits of medical institutions, a phenomenon that clinical medical behaviors are unreasonable exists in some patients, so that medical resources are wasted, economic burden of the patients is increased, and even physical health of the patients can be possibly damaged. On the other hand, in the clinical diagnosis and treatment process, the problems of insufficient intervention flow and standard mastering of medical staff to the guideline requirements, insufficient compliance to the guideline requirements and the like exist, so that the phenomenon of non-compliance of medical behaviors is caused, the number of hospitalization days of a patient is increased, and the infection rate and the death rate of the patient are correspondingly increased.
With the advent of the big data era, a lot of valuable medical data are recorded in electronic medical records, but how to utilize artificial intelligence and informatization technology to enable the electronic medical record data to be better mined and utilized is a difficult problem which needs to be solved urgently. By establishing a set of medical behavior compliance assessment system based on the electronic medical records, the conventional electronic medical record data is mined and analyzed, technical assistance can be provided for medical workers, and the quality and efficiency of clinical diagnosis and treatment are greatly improved.
Most of traditional evaluation systems based on machine learning methods rely on prior knowledge of experts in related fields to label data, and machine learning methods based on behavior analysis learning have long learning time, but actually have relatively less labeled data, and data mining of electronic medical records is more suitable for semi-supervised or unsupervised data driving methods;
in the prior art, most of the prior art only utilizes the information of the patient charge item, and the information of the patient admission examination result, the allergy condition, the medical advice and the like is not considered comprehensively, so that the information utilization condition is not deep enough;
the prior art does not consider the clinical value of each index, and the evaluation result has insufficient clinical interpretability;
the existing model considers the uniformity of the model too much and depends on the difference of the illness state and the physical condition of different patients, so that the precision is low, the adaptability is poor, and meanwhile, the preset logic rule of the early warning system is more specific to a single disease type and a simple clinical scene.
Disclosure of Invention
The invention aims to provide an unsupervised medical behavior compliance assessment method based on an electronic medical record, which reduces the dependence of prior knowledge, can deeply utilize data, has strong clinical interpretability of assessment results and high preset logic complexity of patient conditions, physical conditions and the like.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
the invention discloses an unsupervised electronic medical record-based medical behavior compliance assessment method, which comprises the following steps of:
s1, case data are collected, cleaned and preprocessed, the case data comprise personal information, admission data, medical history data, examination data, diagnosis and treatment results, medical operation data and hospitalization data of a patient, the medical operation data comprise medical order data and operation data, and the medical order data and the operation data are in a time series form;
s2, classifying the case data of the patient according to the personal information, the medical history data, the examination data, the diagnosis data and the diagnosis and treatment results of the patient, and constructing a fuzzy concept with similar index values;
s3, clustering the order data of the patients;
s4, integrating the medical advice data and the operation data after patient clustering, and mining a diagnosis and treatment process model according to the category of the fuzzy concept to which the patient belongs and the effect of the patient after diagnosis and treatment;
s5, customizing a cost function of the diagnosis and treatment process model, aligning the actual diagnosis and treatment sequence with the excavated diagnosis and treatment process model based on the cost function, positioning the position of the abnormality and calculating the deviation degree of the abnormality.
Preferably, in step S2, using fuzzy form concept analysis theory, each historical patient with a complete clinical path is regarded as an object of the fuzzy form background, each type of index is regarded as an attribute of the fuzzy form background, the value of the form background is normalized, and a threshold is set for each attribute and similar disease patients are merged, so as to simplify the fuzzy form background and construct fuzzy concepts, each fuzzy concept represents a specific patient group with similar index values.
Preferably, in step S3, the order data is first clustered by using a multi-granularity topic model, then the order data after topic clustering is clustered by using a K-means + + algorithm to cluster the order data after topic clustering by day, so as to reduce the difficulty of medical behavior compliance assessment,
if the number of the subjects of the medical advice data istThen the patient is treatediAnd the patientjIn the first placemDay and daynThe day similarity is described below:
wherein
The data representing the order of the medical order,
representing the patientiIn the first placemSkytThe probability distribution of the topics of the dimensional topic vector,pthe probability of the representative subject matter,krepresents the weight of the corresponding subject matter,
representing the patientiAnd the patientjIn the first placemDay and daynThe degree of similarity of the days is,
representing the patientiFirst, themThe first daytThe topic probability distribution of the dimensional topic vector.
Preferably, in step S4, the subdivided "cured" or "improved" patient data is mined using the Imf process discovery algorithm in the ProM process mining software.
Preferably, in step S5, on the premise of the frequency of the specific medical action, the cost function of the medical procedure model based on the TF-IDF weighting technology is used to quantify the cost of the inserted or skipped medical action, and the alignment between the actual medical procedure sequence and the standard medical procedure model is realized through the pnetpply plug-in the ProM procedure mining software, so as to determine the position and deviation degree of the abnormality.
Preferably, in step S5, the medical sequenceSeqInserting event ofxTime, insertion cost
The specific description is as follows:
wherein the content of the first and second substances,Nis the total number of the medical sequences in the sample,
for containing and not containing insertion eventsxMedical treatment sequence ofSeqThe total number of occurrences is,
to contain an insertion eventxMedical treatment sequence ofSeqThe number of times of occurrence of the event,Nfor the number of samples of all the medical sequences,
to include an insertion eventxThe number of the medical treatment sequences of (a),
TF-IDF (Term Frequency-Inverse Document Frequency) is a commonly used weighting technique for information retrieval and data mining, where TF is Term Frequency (Term Frequency) in equation (3) and IDF is Inverse text Frequency index (Inverse Document Frequency) in equation (4).
The invention has the beneficial effects that:
1. the invention provides a mining scheme integrating more comprehensive data, and more comprehensively considers information of various aspects of patients.
2. The invention introduces the fuzzy form concept analysis theory to subdivide the patient population, so that the range of the reference data of the process model is finer, and the adaptation degree of the standard process model and the actual diagnosis and treatment sequence is improved.
3. The Multi-granularity Topic Model clustering method (M-GTM, Multi-gain Topic Model) adopted by the invention is obviously superior to the common LDA Topic Model clustering in use effect.
4. The method of the invention introduces TF-IDF weighting technology when calculating the frequency of medical behaviors, thereby improving the accuracy of cost function calculation.
5. The medical behavior compliance evaluation flow based on the electronic medical record does not need to label abnormal medical data, and is an unsupervised data driving method.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the present invention comprises the steps of:
s1: collecting, cleaning and preprocessing case data, wherein the case data comprises admission data, hospitalization data, medical history data, various examination data and medical advice data of a patient;
s2: classifying the patients according to the different conditions of the personal information, various examination data, the patient and family medical history and diagnosis data of the patients and the effect after diagnosis and treatment of the case data;
s3: clustering medical advice data of a patient, wherein the medical advice data is in a time series form;
s4: medical advice data after patient clustering and medical operation data with time series such as operations are fused, the effect of the patients after diagnosis and treatment is comprehensively considered according to the subdivided patient categories, and a relatively more effective diagnosis and treatment process model is excavated;
s5: and customizing a cost function of the diagnosis and treatment process model, aligning the actual diagnosis and treatment sequence with the excavated diagnosis and treatment process model based on the cost function, positioning the position of the abnormality and calculating the deviation degree of the abnormality.
In step S2, a fuzzy form concept analysis theory is introduced, each patient with a history of complete clinical paths is regarded as an object of the fuzzy form background, and each type of index is regarded as an attribute of the fuzzy form background. And then, carrying out normalization processing on the values of the form background, setting a threshold value for each attribute and combining similar disease patients so as to simplify the fuzzy form background. Construction of a grid of fuzzy concepts may then be performed, each fuzzy concept representing a particular patient population with similar metric values.
In step S3, the medical order data is first clustered by using a Multi-granularity Topic Model (Multi-gain Topic Model), and then the medical order data after Topic clustering is clustered by day by using a K-means + + algorithm, so as to reduce the difficulty of medical behavior compliance evaluation. If the number of subjects in the order data is t, the similarity between the patient i and the patient j on the m-th day and the n-th day is described as follows:
wherein the content of the first and second substances,
the topic probability distribution of the t-dimensional topic vector of the patient i on the m day is represented, p represents the topic probability, and k represents the weight of the corresponding topic.
In step S4, a Imf (Inductive Miner-frequency) process discovery algorithm in the ProM process mining software is used to mine the diagnosis and treatment process model for the subdivided "cured" or "improved" patient data.
In step S5, on the premise of how frequently the specific medical action is performed, the cost of the medical action in the form of insertion or skipping is quantified by using the cost function of the diagnosis and treatment process model based on the TF-IDF weighting technique. Taking the insertion event x of a certain medical sequence Seq as an example, the insertion cost (x) is specifically described as follows:
wherein, N is the total number of medical sequences in a sample, N (Seq) is the total number of occurrences of the medical sequence Seq including and not including the insertion event x, N (Seq) x is the number of occurrences of the medical sequence Seq including the insertion event x, N is the number of samples of all medical sequences, and N (x) is the number of medical sequences including the insertion event x.
And then, aligning the actual diagnosis and treatment sequence with the standard diagnosis and treatment process model through a PNetRelyer plug-in the ProM process mining software, and finally judging the abnormal position and the deviation degree.
In practical use, taking a large vessel disease as an example, the implementation process is as follows:
1. collecting, cleaning and preprocessing an electronic medical record of a patient with a large vascular disease:
collecting and sorting the electronic medical record data of all patients with the major vascular disease, and selecting admission data, hospitalization data, medical history data, various examination data and medical advice data of the patients in the electronic medical records.
Then data cleaning and preprocessing are carried out, including unifying the naming of similar items or diagnosis and treatment operations, eliminating the clinical path of midway insertion or exit and cases of invalid treatment, merging the same diagnosis and treatment operations or medical orders at the same time and the like.
2. Patient type classification:
(1) according to the fuzzy form concept analysis theory, firstly, a fuzzy form background is constructed: the method comprises the steps of sorting different conditions of personal information, various examination data, patient and family medical history and diagnosis data of a patient and information of several dimensionalities of the effect after diagnosis and treatment, and setting the attribute of a fuzzy form background according to the information.
(2) The specific information of each patient is converted into the membership degree corresponding to each attribute, and the membership degrees of all the attributes need to be normalized.
(3) The fuzzy form background is reduced by setting appropriate thresholds for the degree of membership of each attribute and merging similar patients. The membership degree of the attribute can be reasonably divided by means of expert experience, or the change condition of the membership degree of the attribute can be fitted into a normal distribution according to historical data, a proper confidence interval (for example, a confidence interval of 80% is set, and a single-side or double-side confidence interval is set) is selected, the membership degree outside the confidence interval is set to be 0, and the membership degree in the confidence interval is set to be 1.
(4) And constructing a concept lattice. The granularity of patient classification may be determined by selecting a hierarchy of concept lattices.
3. Doctor advice data clustering module:
for similar patients, the diagnosis and treatment schemes of each day are similar, so that firstly, medical order data are clustered by adopting a Multi-granularity Topic Model M-GTM (Multi-gain Topic Model), if the Topic clustering precision is improved, a little prior knowledge of medical order data classification can be added, then, the medical order data after Topic clustering are clustered by adopting a K-means + + algorithm, and the medical order data after Topic clustering are clustered by day according to the similarity, so that the difficulty of medical behavior compliance evaluation is reduced. If the number of subjects in the order data is t, the similarity between the patient i and the patient j on the m-th day and the n-th day is described as follows:
wherein the content of the first and second substances,
the topic probability distribution of the t-dimensional topic vector of the patient i on the m day is represented, p represents the topic probability, and k represents the weight of the corresponding topic.
4. Excavating a diagnosis and treatment process model:
medical order data after patient clustering and medical operation data with time series such as operations are fused, the effect of the patients after diagnosis and treatment is comprehensively considered according to subdivided patient categories, Imf (Inductive Miner-frequency) process discovery algorithm is selected through Prom process mining software, the patient category data of 'cured' and 'improved' are selected for mining of diagnosis and treatment process models, and relatively more effective diagnosis and treatment process models are mined.
5. And (3) discovering the abnormal medical behavior:
(1) on the premise of the frequency of specific medical behaviors related to the large vessel diseases, the cost of the medical behaviors in the forms of insertion or skipping and the like is quantified by adopting a diagnosis and treatment process model cost function based on the TF-IDF weighting technology. Taking the insertion event x of a certain medical sequence Seq as an example, the insertion cost (x) is specifically described as follows:
wherein, N is the total number of medical sequences in a sample, N (Seq) is the total number of occurrences of the medical sequence Seq including and not including the insertion event x, N (Seq) x is the number of occurrences of the medical sequence Seq including the insertion event x, N is the number of samples of all medical sequences, and N (x) is the number of medical sequences including the insertion event x.
(2) And aligning the actual diagnosis and treatment sequence with the excavated diagnosis and treatment process standard model based on the cost function through a PNetRelyer plug-in the ProM software, and finally judging the abnormal position and the deviation degree.
There are, of course, many other embodiments of the invention and modifications and variations which will be apparent to those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (6)
1. An unsupervised electronic medical record-based medical behavior compliance assessment method is characterized by comprising the following steps of:
s1, case data are collected, cleaned and preprocessed, the case data comprise personal information, admission data, medical history data, examination data, diagnosis and treatment results, medical operation data and hospitalization data of a patient, the medical operation data comprise medical order data and operation data, and the medical order data and the operation data are in a time series form;
s2, classifying the case data of the patient according to the personal information, the medical history data, the examination data, the diagnosis data and the diagnosis and treatment results of the patient, and constructing a fuzzy concept with similar index values;
s3, clustering the order data of the patients;
s4, integrating the medical advice data and the operation data after patient clustering, and mining a diagnosis and treatment process model according to the category of the fuzzy concept to which the patient belongs and the effect of the patient after diagnosis and treatment;
s5, customizing a cost function of the diagnosis and treatment process model, aligning the actual diagnosis and treatment sequence with the excavated diagnosis and treatment process model based on the cost function, positioning the position of the abnormality and calculating the deviation degree of the abnormality.
2. The unsupervised electronic medical record-based medical behavior compliance assessment method according to claim 1, wherein: in step S2, using the fuzzy form concept analysis theory, each patient with a history of complete clinical paths is regarded as an object of the fuzzy form background, each type of index is regarded as an attribute of the fuzzy form background, the value of the form background is normalized, a threshold is set for each attribute and similar disease patients are merged, so as to simplify the fuzzy form background and construct fuzzy concepts, each fuzzy concept represents a specific patient group with similar index values.
3. The unsupervised electronic medical record-based medical behavior compliance assessment method according to claim 1, wherein: in step S3, firstly, the medical advice data is clustered by adopting a multi-granularity topic model, then the medical advice data after topic clustering is clustered by adopting a K-means + + algorithm according to the day to reduce the difficulty of the medical behavior compliance evaluation,
if the number of the subjects of the medical advice data istThen the patient is treatediAnd suffer fromAjIn the first placemDay and daynThe day similarity is described below:
wherein
The data representing the order of the medical order,
representing the patientiIn the first placemSkytThe probability distribution of the topics of the dimensional topic vector,pthe probability of the representative subject matter,krepresents the weight of the corresponding subject matter,
representing the patientiAnd the patientjIn the first placemDay and daynThe degree of similarity of the days is,
representing the patientiFirst, themThe first daytThe topic probability distribution of the dimensional topic vector.
4. The unsupervised electronic medical record-based medical behavior compliance assessment method according to claim 1, wherein: in step S4, a Imf process discovery algorithm in the ProM process mining software is used to mine the diagnosis and treatment process model for the subdivided "cured" or "improved" patient data.
5. The unsupervised electronic medical record-based medical behavior compliance assessment method according to claim 1, wherein: in step S5, on the premise of the frequency of the specific medical action, the cost function of the diagnosis and treatment process model based on the TF-IDF weighting technology is used to quantify the cost of the inserted or skipped medical action, and the alignment between the actual diagnosis and treatment sequence and the standard diagnosis and treatment process model is realized by the pnetpply plug-in the ProM process mining software, so as to determine the position and deviation degree of the abnormality.
6. The unsupervised electronic medical record-based medical behavior compliance assessment method according to claim 5, wherein: in step S5, in the medical treatment sequenceSeqInserting event ofxTime, insertion cost
The specific description is as follows:
wherein the content of the first and second substances,Nis the total number of the medical sequences in the sample,
for containing and not containing insertion eventsxMedical treatment sequence ofSeqThe total number of occurrences is,
to contain an insertion eventxMedical treatment sequence ofSeqThe number of times of occurrence of the event,Nfor the number of samples of all the medical sequences,
to include an insertion eventxThe number of medical treatment sequences.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110489454.XA CN112992370B (en) | 2021-05-06 | 2021-05-06 | Unsupervised electronic medical record-based medical behavior compliance assessment method |
| PCT/CN2021/132173 WO2022233121A1 (en) | 2021-05-06 | 2021-11-22 | Unsupervised medical behavior compliance assessment method based on electronic medical record |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110489454.XA CN112992370B (en) | 2021-05-06 | 2021-05-06 | Unsupervised electronic medical record-based medical behavior compliance assessment method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112992370A true CN112992370A (en) | 2021-06-18 |
| CN112992370B CN112992370B (en) | 2021-07-30 |
Family
ID=76336976
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110489454.XA Active CN112992370B (en) | 2021-05-06 | 2021-05-06 | Unsupervised electronic medical record-based medical behavior compliance assessment method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112992370B (en) |
| WO (1) | WO2022233121A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113553399A (en) * | 2021-07-16 | 2021-10-26 | 山东建筑大学 | Text search method and system based on fuzzy language approximate concept lattice |
| CN115083616A (en) * | 2022-08-16 | 2022-09-20 | 之江实验室 | Chronic nephropathy subtype mining system based on self-supervision graph clustering |
| WO2022233121A1 (en) * | 2021-05-06 | 2022-11-10 | 四川大学华西医院 | Unsupervised medical behavior compliance assessment method based on electronic medical record |
| CN115910387A (en) * | 2022-11-08 | 2023-04-04 | 北京健康在线技术开发有限公司 | Data processing method, device and equipment based on time sequence and storage medium |
| CN116453637A (en) * | 2023-03-20 | 2023-07-18 | 杭州市卫生健康事业发展中心 | Health data management method and system based on regional big data |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN106202477A (en) * | 2016-07-18 | 2016-12-07 | 北京千安哲信息技术有限公司 | Medical expense method for digging and device |
| CN106951719A (en) * | 2017-04-10 | 2017-07-14 | 荣科科技股份有限公司 | The construction method and constructing system of clinical diagnosis model, clinical diagnosing system |
| US20190007208A1 (en) * | 2017-06-30 | 2019-01-03 | Microsoft Technology Licensing, Llc | Encrypting existing live unencrypted data using age-based garbage collection |
| CN110335684A (en) * | 2019-06-14 | 2019-10-15 | 电子科技大学 | The intelligent dialectical aid decision-making method of Chinese medicine based on topic model technology |
| US20200202988A1 (en) * | 2014-02-21 | 2020-06-25 | Intelligent Medical Objects, Inc. | System and Method for Generating and Updating a User Interface to Evaluate an Electronic Medical Record |
| CN111696640A (en) * | 2020-06-12 | 2020-09-22 | 上海联影医疗科技有限公司 | Method, device and storage medium for automatically acquiring medical record template |
| CN111832298A (en) * | 2020-07-14 | 2020-10-27 | 北京百度网讯科技有限公司 | Quality inspection method, device and equipment for medical records and storage medium |
| CN112735597A (en) * | 2020-12-31 | 2021-04-30 | 荆门汇易佳信息科技有限公司 | Medical text disorder identification method driven by semi-supervised self-learning |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150046181A1 (en) * | 2014-02-14 | 2015-02-12 | Brighterion, Inc. | Healthcare fraud protection and management |
| US9779407B2 (en) * | 2014-08-08 | 2017-10-03 | Brighterion, Inc. | Healthcare fraud preemption |
| US10886013B1 (en) * | 2017-11-15 | 2021-01-05 | Iodine Software, LLC | Systems and methods for detecting documentation drop-offs in clinical documentation |
| CN109243567B (en) * | 2018-08-14 | 2021-11-02 | 山东科技大学 | Medicine recommendation method based on prescription data mining |
| CN111667927A (en) * | 2020-06-05 | 2020-09-15 | 山东凯鑫宏业生物科技有限公司 | ZigBee network intelligent medical system and acquisition node networking method thereof |
| CN111916191A (en) * | 2020-07-22 | 2020-11-10 | 复旦大学 | Medical behavior operation compliance evaluation system based on medical behavior data |
| CN112992370B (en) * | 2021-05-06 | 2021-07-30 | 四川大学华西医院 | Unsupervised electronic medical record-based medical behavior compliance assessment method |
-
2021
- 2021-05-06 CN CN202110489454.XA patent/CN112992370B/en active Active
- 2021-11-22 WO PCT/CN2021/132173 patent/WO2022233121A1/en active Application Filing
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200202988A1 (en) * | 2014-02-21 | 2020-06-25 | Intelligent Medical Objects, Inc. | System and Method for Generating and Updating a User Interface to Evaluate an Electronic Medical Record |
| 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 |
| CN106202477A (en) * | 2016-07-18 | 2016-12-07 | 北京千安哲信息技术有限公司 | Medical expense method for digging and device |
| CN106951719A (en) * | 2017-04-10 | 2017-07-14 | 荣科科技股份有限公司 | The construction method and constructing system of clinical diagnosis model, clinical diagnosing system |
| US20190007208A1 (en) * | 2017-06-30 | 2019-01-03 | Microsoft Technology Licensing, Llc | Encrypting existing live unencrypted data using age-based garbage collection |
| CN110335684A (en) * | 2019-06-14 | 2019-10-15 | 电子科技大学 | The intelligent dialectical aid decision-making method of Chinese medicine based on topic model technology |
| CN111696640A (en) * | 2020-06-12 | 2020-09-22 | 上海联影医疗科技有限公司 | Method, device and storage medium for automatically acquiring medical record template |
| CN111832298A (en) * | 2020-07-14 | 2020-10-27 | 北京百度网讯科技有限公司 | Quality inspection method, device and equipment for medical records and storage medium |
| CN112735597A (en) * | 2020-12-31 | 2021-04-30 | 荆门汇易佳信息科技有限公司 | Medical text disorder identification method driven by semi-supervised self-learning |
Non-Patent Citations (2)
| Title |
|---|
| JIAO ZHANG等: "Semi-Supervised Patient Similarity Clustering Algorithm Based on Electronic Medical Records", 《IEEE ACCESS》 * |
| 杨锦锋 等: "电子病历命名实体识别和实体关系抽取研究综述", 《自动化学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022233121A1 (en) * | 2021-05-06 | 2022-11-10 | 四川大学华西医院 | Unsupervised medical behavior compliance assessment method based on electronic medical record |
| CN113553399A (en) * | 2021-07-16 | 2021-10-26 | 山东建筑大学 | Text search method and system based on fuzzy language approximate concept lattice |
| CN115083616A (en) * | 2022-08-16 | 2022-09-20 | 之江实验室 | Chronic nephropathy subtype mining system based on self-supervision graph clustering |
| CN115083616B (en) * | 2022-08-16 | 2022-11-08 | 之江实验室 | Chronic nephropathy subtype mining system based on self-supervision graph clustering |
| JP7404581B1 (en) | 2022-08-16 | 2023-12-25 | 之江実験室 | Chronic nephropathy subtype mining system based on self-supervised graph clustering |
| CN115910387A (en) * | 2022-11-08 | 2023-04-04 | 北京健康在线技术开发有限公司 | Data processing method, device and equipment based on time sequence and storage medium |
| CN116453637A (en) * | 2023-03-20 | 2023-07-18 | 杭州市卫生健康事业发展中心 | Health data management method and system based on regional big data |
| CN116453637B (en) * | 2023-03-20 | 2023-11-07 | 杭州市卫生健康事业发展中心 | Health data management method and system based on regional big data |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022233121A1 (en) | 2022-11-10 |
| CN112992370B (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112992370B (en) | Unsupervised electronic medical record-based medical behavior compliance assessment method | |
| Yang et al. | A sequential three-way approach to multi-class decision | |
| Chen et al. | Semi-supervised learning via regularized boosting working on multiple semi-supervised assumptions | |
| CN109993100B (en) | Method for realizing facial expression recognition based on deep feature clustering | |
| CN113688248B (en) | Medical event identification method and system under condition of small sample weak labeling | |
| Jader et al. | Fast and Accurate Artificial Neural Network Model for Diabetes Recogni-tion | |
| CN110890146A (en) | Bedside intelligent interaction system for intelligent ward | |
| Swain et al. | Appositeness of optimized and reliable machine learning for healthcare: a survey | |
| Zhang et al. | Classification of canker on small datasets using improved deep convolutional generative adversarial networks | |
| Abdul-Jabbar | Data analytics and techniques | |
| Qian et al. | Incomplete label distribution feature selection based on neighborhood-tolerance discrimination index | |
| Sande et al. | Statistical Learning in Medical Research with Decision Threshold and Accuracy Evaluation. | |
| Alves et al. | Variational autoencoders for medical image retrieval | |
| Angayarkanni | Predictive analytics of chronic kidney disease using machine learning algorithm | |
| Appari et al. | An Improved CHI 2 Feature Selection Based a Two-Stage Prediction of Comorbid Cancer Patient Survivability | |
| Chen | Applicability Analysis of Data-driven Methods for Adolescents Mental Health Surveillance | |
| Bhardwaj et al. | Protein Enzyme Sequence Class Prediction using Computational Model | |
| Chakraborty et al. | Data Mining-Based Variant Subset Features | |
| Vinod et al. | Classification of Incomplete Data Handling Techniques- An Overview | |
| Kaur et al. | A review on Naive Baye’s (NB), J48 and K-means based mining algorithms for medical data mining | |
| Awari | Diseases prediction model using machine learning technique | |
| Demigha | Mining Knowledge of the Patient Record | |
| CN116110594B (en) | Knowledge evaluation method and system of medical knowledge graph based on associated literature | |
| AU2020103756A4 (en) | Noisy text prediction in large volume clinical data using reinforcement learning | |
| Jiyun et al. | Patient similarity measuring with graph embedded learning and triplet network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |