CN111968747A - VTE intelligent prevention and control management system - Google Patents
VTE intelligent prevention and control management system Download PDFInfo
- Publication number
- CN111968747A CN111968747A CN202010843967.1A CN202010843967A CN111968747A CN 111968747 A CN111968747 A CN 111968747A CN 202010843967 A CN202010843967 A CN 202010843967A CN 111968747 A CN111968747 A CN 111968747A
- Authority
- CN
- China
- Prior art keywords
- module
- vte
- text data
- historical text
- risk
- 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
- 230000002265 prevention Effects 0.000 title claims abstract description 15
- 238000012360 testing method Methods 0.000 claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims abstract description 20
- 230000009467 reduction Effects 0.000 claims abstract description 19
- 238000012502 risk assessment Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000007637 random forest analysis Methods 0.000 claims abstract description 5
- 238000012216 screening Methods 0.000 claims abstract description 4
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 8
- 208000024891 symptom Diseases 0.000 claims description 7
- 230000001186 cumulative effect Effects 0.000 claims description 6
- 238000012847 principal component analysis method Methods 0.000 claims description 5
- 238000012417 linear regression Methods 0.000 claims description 3
- 230000000474 nursing effect Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 206010014522 Embolism venous Diseases 0.000 description 44
- 208000004043 venous thromboembolism Diseases 0.000 description 44
- 238000011156 evaluation Methods 0.000 description 10
- 208000025747 Rheumatic disease Diseases 0.000 description 8
- 238000000034 method Methods 0.000 description 6
- 230000002411 adverse Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000003058 natural language processing Methods 0.000 description 3
- 206010008479 Chest Pain Diseases 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010002091 Anaesthesia Diseases 0.000 description 1
- 206010011224 Cough Diseases 0.000 description 1
- 239000003154 D dimer Substances 0.000 description 1
- 206010051055 Deep vein thrombosis Diseases 0.000 description 1
- 208000000059 Dyspnea Diseases 0.000 description 1
- 206010013975 Dyspnoeas Diseases 0.000 description 1
- 206010021143 Hypoxia Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 208000010378 Pulmonary Embolism Diseases 0.000 description 1
- 206010047249 Venous thrombosis Diseases 0.000 description 1
- 230000037005 anaesthesia Effects 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000539 dimer Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 230000024924 glomerular filtration Effects 0.000 description 1
- 238000001794 hormone therapy Methods 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000013077 scoring method Methods 0.000 description 1
- 208000013220 shortness of breath Diseases 0.000 description 1
- 230000008961 swelling Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000008733 trauma Effects 0.000 description 1
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
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Epidemiology (AREA)
- General Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
VTE intelligence prevention and cure management system: the word segmentation module is used for carrying out word segmentation on the preprocessed historical text data and screening out association influence factors; the calculation module calculates the support degree of the corresponding VTE when any correlation influence factor appears in all the historical text data and the confidence degree of the corresponding VTE in all the historical text data with the correlation influence factor; the judging module judges whether the support degree is greater than the corresponding set value or not and whether the confidence degree is greater than the corresponding set value or not, and the correlation influence factors are used as key influence factors; the dimension reduction module performs dimension reduction processing on the key influence factors to obtain optimal influence factors; the training module substitutes part of historical text data into a random forest model for training to obtain a trained risk assessment model; the testing module substitutes the optimal influence factors in the residual historical text data into a risk assessment model to test to obtain the probability value of VTE of the patient; and the grading module is used for grading to obtain the corresponding risk grade label of the tested patient by using a normal distribution 3 sigma principle.
Description
Technical Field
The invention relates to the technical field of venous thromboembolism, in particular to a VTE intelligent prevention and treatment management system.
Background
Venous Thromboembolism (VTE), including deep vein thrombosis and pulmonary thromboembolism, is a significant cause of unintended death in hospitalized patients. The disease can be widely existed in multiple departments, and high risk factors include long-term bed rest, operation, trauma, tumor and the like. The hidden onset of VTE, its high incidence, high disability rate, high mortality are the leading cause of unexpected death of patients in hospitals, and the prevention and control situation is severe. The scale evaluation is regarded as an important means for effectively preventing VTE, but the clinical real use rate is not high, different specialists have obvious cognitive difference on VTE, the evaluation lacks uniform specification, and the prevention rate is low. The traditional scale evaluation work greatly increases the burden of medical staff, clinical data required to be filled in the scale are scattered in each information system of a hospital and cannot be shared and utilized highly, a lot of time is required to be consumed for medical staff each time of manual evaluation, data is difficult to store and utilize, and a tool for automatic evaluation is also desired clinically to reduce the work burden. Furthermore, the existing manual scale assessment is difficult to dynamically control the key nodes of disease condition changes in real time, and in the actual clinical process, a target patient with potential risk needs to be accurately found through the dynamic changes of some subtle and important patients, so that a standardized and personalized diagnosis and treatment scheme is implemented as soon as possible. Therefore, the purpose of preventing VTE or avoiding serious consequences and adverse events for patients who have VTE is really achieved, the improvement of medical quality is achieved, the medical safety is guaranteed, and the benefit of the patients is maximized.
Disclosure of Invention
Aiming at the problems and the defects in the prior art, the invention provides a novel VTE intelligent prevention and control management system.
The invention solves the technical problems through the following technical scheme:
the invention provides a VTE intelligent prevention and control management system which is characterized by comprising a preprocessing module, a word segmentation module, a calculation module, a judgment module, a dimension reduction module, a training module, a test module and a grading module;
the preprocessing module is used for preprocessing the historical text data of each patient;
the word segmentation module is used for carrying out word segmentation on the preprocessed historical text data, screening different word segmentation symptoms related to VTE from the segmented word segmentation symptoms as associated influence factors, wherein the historical text data comprises patient basic data, examination and inspection result data, nursing and doctor course records, physical signs, various instrument record data and medical advice data;
the calculation module is used for calculating the support degree of the corresponding VTE when any one association influence factor appears in all historical text data and the confidence degree of the corresponding VTE in all the historical text data with the association influence factor;
the judging module is used for judging whether the support degree is greater than a corresponding support degree set value and the confidence degree is greater than a corresponding confidence degree set value, if yes, the associated influence factor is taken as a key influence factor, and if not, the associated influence factor is not taken as a key influence factor;
the dimension reduction module is used for carrying out dimension reduction processing on the key influence factors to obtain the key influence factors without collinear relation as optimal influence factors;
the training module is used for substituting the optimal influence factors of part of the historical text data in the preprocessed historical text data and whether the patient has VTE as training samples into a random forest model for model training so as to obtain a trained risk assessment model;
the testing module is used for substituting the optimal influence factors in the residual historical text data as testing samples into the trained risk assessment model for testing so as to obtain the probability value of VTE of the tested patient;
the grading module is used for converting each probability value into a rank, mapping the rank distribution to a normal distribution curve, and grading by using the 3 sigma principle of normal distribution so as to obtain a risk grade label corresponding to each test patient.
Preferably, the classification module is configured to convert each probability value into a rank, calculate corresponding cumulative frequencies based on the rank, convert the cumulative frequencies into corresponding probability units, substitute the probability units into a linear regression equation to calculate corresponding estimation values, map the distribution of the estimation values onto a normal distribution curve, and perform classification by using a 3 σ principle of normal distribution to obtain a risk class label corresponding to each test patient.
Preferably, the system further comprises an optimization module, and the optimization module is used for performing modification optimization on the trained risk assessment model.
Preferably, the optimization module is configured to modify the optimally trained risk assessment model based on a fixed rule formulated by delphire scoring.
Preferably, the dimension reduction module is configured to perform dimension reduction processing on the key influence factors by using a principal component analysis method.
Preferably, the system further comprises a prediction module for substituting the preprocessed text data of the new patient into the trained and tested risk assessment model for prediction to obtain the probability that the new patient has VTE and the corresponding risk level label.
Preferably, the risk profile labels include low risk, medium risk and high risk; the system further comprises a reminding module, wherein the reminding module is used for reminding corresponding workers based on different risk grade labels, and carrying out highest-grade risk reminding when any optimal influence factor in the optimal influence factors appears in text data of the new patient after the risk grade label that the new patient suffers from the VTE is obtained as the medium risk or the high risk.
Preferably, the calculation module is configured to calculate a support degree, which is the number of historical text data corresponding to the VTE with which the associated influence factor appears in all the historical text data/the total number of all the historical text data;
the calculation module is further used for calculating the confidence coefficient which is the number of the historical text data corresponding to the VTE in all the historical text data with the associated influence factor/the total number of the historical text data with the associated influence factor.
Preferably, the word segmentation module is configured to perform word segmentation on the preprocessed historical text data by using an NLP technique.
On the basis of the common knowledge in the field, the above preferred conditions can be combined randomly to obtain the preferred embodiments of the invention.
The positive progress effects of the invention are as follows:
the VTE intelligent prevention and treatment management system screens out the associated influence factors from historical text data, obtains the optimal influence factors after carrying out support degree judgment, confidence degree judgment and dimension reduction processing on the associated influence factors, trains a test model based on the optimal influence factors to obtain the probability value of VTE of a test patient, and carries out grading by applying the 3 sigma principle of normal distribution to obtain a risk grade label corresponding to each test patient; and then, substituting the preprocessed text data of the new patient into a risk assessment model for prediction to obtain the probability that the new patient has VTE and a corresponding risk grade label. According to the invention, through risk prediction and evaluation result medical care sharing, the effect of accurately identifying the target patient in an early stage is achieved, and the risk of VTE adverse consequences is avoided as much as possible. The system ensures the scientific and effective evaluation result, enables the evaluation process to be faster, saves more labor cost, can continuously update the evaluation rule and ensures the accuracy of evaluation.
Drawings
Fig. 1 is a block diagram of a VTE intelligent control management system according to a preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
As shown in fig. 1, the present embodiment provides a VTE intelligent prevention and treatment management system, which includes a preprocessing module 1, a word segmentation module 2, a calculation module 3, a judgment module 4, a dimension reduction module 5, a training module 6, a test module 7, a grading module 8, a prediction module 9, a reminding module 10, and an optimization module 11.
The preprocessing module 1 is used for preprocessing the historical text data of each patient, wherein the historical text data comprises patient basic data (age, sex, BMI and the like), examination and test result data (glomerular filtration rate, INR, platelet count and the like), nursing and doctor disease course record sheets, physical sign and various instrument record data and medical advice data.
The pretreatment comprises two steps: data cleansing and data type conversion. The first step is as follows: the acquired historical text data basically has the problems of data loss, data noise, data redundancy, outliers/outliers, data duplication and the like. The data are processed in different modes, for example, data are missing, for the missing different data, modes such as manual filling, automatic filling and elimination of the data are selected, for example, mode filling may be adopted when gender is missing, manual filling is performed through discussion when some clinical data are missing, and the data are eliminated when important characteristic factors are considered to be missing. The second step is that: data type conversion, such as age, converts age to 0 (0-40), 1 (41-60), 2 (61-70), 3 (71-85), 5(> 85), risk level 2 (high risk), 1 (medium risk), 0 (low risk), etc., which then lays a good foundation for model creation.
The word segmentation module 2 is used for performing word segmentation processing (rheumatic diseases, malignant tumors, hormone therapy, VTE medical history and the like) on the preprocessed historical text data by using an NLP (natural language processing) technology, and screening different word segmentation symptoms related to VTE from the segmented symptoms as associated influence factors (chest distress, chest pain, shortness of breath, hypoxia or low oxygen saturation, cough, swelling and the like).
The calculation module 3 is used for calculating the support degree corresponding to the suffering of VTE when any associated influence factor appears in all the historical text data and the confidence degree corresponding to the suffering of VTE in all the historical text data with the associated influence factor.
The calculating module 3 is configured to calculate a support degree, which is the number of historical text data/the total number of all historical text data corresponding to the VTE with the occurrence of the associated influence factor in all historical text data.
The calculating module 3 is further configured to calculate a confidence coefficient, which is the number of historical text data corresponding to the VTE in all historical text data having the associated influence factor/the total number of all historical text data having the associated influence factor.
For example: assuming that there are 100 historical text data, wherein there are 5 historical text data in which the patient has VTE when the rheumatic disease vocabulary appears, the support is 0.05; the ratio of the number of the historical text data with VTE in all the historical text data with rheumatic disease to the total number of the historical text data with rheumatic disease is the confidence coefficient, and the calculated confidence coefficient is assumed to be 30%. At this time, whether the rheumatic disease is taken as a VTE risk factor or not is judged, and the support degree and the confidence degree are both larger than corresponding set values, so that the rheumatic disease is considered to increase the VTE risk, and therefore, the rheumatic disease key words are taken as key influence factors.
The judging module 4 is configured to judge whether the support degree is greater than the corresponding support degree setting value and whether the confidence degree is greater than the corresponding confidence degree setting value, if yes, use the associated impact factor as a key impact factor, and if not, use the associated impact factor as no key impact factor.
And the dimension reduction module 5 is used for performing dimension reduction processing on the key influence factors by adopting a principal component analysis method to obtain the key influence factors without collinear relation as optimal influence factors.
After finding the key impact factors above, it was found that the key impact factors found were numerous, e.g., a total of 42 key impact factors, including 11 patient attributes, 8 examination findings, and 23 keywords. The number of the key influence factors is large, and each key influence factor has a correlation relationship, such as the age and the rheumatic disease have a high correlation relationship, and if it is obviously unreasonable to simultaneously incorporate the two influence factors into the evaluation model, a common factor needs to be found to replace the two key influence factors having the correlation relationship. Therefore, a commonly used data dimension reduction method is used for dimension reduction processing, namely a Principal Component Analysis (PCA) full name, and original data can be transformed into a group of linearly independent representations of each dimension through linear transformation, so that the main linear Component of the data can be extracted. After the principal component analysis method is used for dimensionality reduction, only 14 influence factors are left as optimal influence factors after the principal component analysis method is brought into model evaluation factors, and therefore accuracy and rapidness of the evaluation model are guaranteed.
And the training module 6 is used for substituting the optimal influence factor of part of the historical text data in the preprocessed historical text data and whether the patient has VTE as a training sample into the random forest model to carry out model training so as to obtain a trained risk assessment model. And whether the patient has VTE in the part of the historical text data is used as an output sample of the random forest model.
And the testing module 7 is used for substituting the optimal influence factors in the residual historical text data as testing samples into the trained risk assessment model for testing so as to obtain the probability value of the VTE of the test patient.
The grading module 8 is configured to convert each tested probability value into a rank, calculate corresponding cumulative frequencies based on each rank, convert each cumulative frequency into corresponding probability units, substitute each probability unit into a linear regression equation to calculate corresponding estimated values, map each estimated value distribution onto a normal distribution curve, and perform grading by using a 3 σ principle of normal distribution to obtain a risk grade label corresponding to each test patient.
The method has the advantages that the patients are graded, limited medical resources can be reasonably and scientifically distributed, high attention is paid to the high-risk patients, and the trend that the change trend of the patients is worsened is deeply discovered and known, so that adverse events in hospitals are avoided.
The prediction module 9 is configured to substitute the preprocessed text data of the new patient into the trained and tested risk assessment model for prediction to obtain the probability that the new patient has VTE and corresponding risk level labels, where the risk level labels include low risk, medium risk, and high risk.
The reminding module 10 is used for reminding corresponding staff based on different risk grade labels, and carrying out highest-grade risk reminding when any optimal influence factor appears in text data of a new patient after the risk grade label that the new patient has the VTE is the medium risk or the high risk.
For example: according to different degrees of risk levels, doctors (bed inpatients → operators on duty → attending physicians → chief physicians) → managers (medical group leader → department chief task → medical department → VTE control and management group leader in hospital) can be respectively reminded to ensure that the high-risk VTE patients are approved and paid attention by the clinicians.
The optimization module 11 is configured to modify the optimally trained risk assessment model based on the fixed rule formulated by the delphire scoring method.
In the process of clinical assistant decision making, not only the results of automatic classification need to be referred to, in order to ensure the safety of the risk assessment model, some fixed rules are made through the delphire grading method, and then the predicted results are updated through the rules, such as: when D dimer is changed to be greater than a certain threshold, for example: d dimer is greater than 2.5mg/L, and can be directly updated into high-risk patients and the like, so that the whole risk assessment model is optimized by continuously accumulating and continuously updating rules.
The system combines the medical care system with various clinical auxiliary systems (LIS \ HIS \ RIS \ hand anesthesia system \ electrocardiographic monitoring and other internet-of-things devices), comprehensively analyzes in real time, increases the capture of patient change keywords (such as symptom words and sign words) on the basis of general scale scoring, evaluates result medical care sharing through risk prediction and achieves the effect of accurately identifying target patients in early stage.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that these are by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (9)
1. A VTE intelligent prevention and control management system is characterized by comprising a preprocessing module, a word segmentation module, a calculation module, a judgment module, a dimension reduction module, a training module, a test module and a grading module;
the preprocessing module is used for preprocessing the historical text data of each patient;
the word segmentation module is used for carrying out word segmentation on the preprocessed historical text data, screening different word segmentation symptoms related to VTE from the segmented word segmentation symptoms as associated influence factors, wherein the historical text data comprises patient basic data, examination and inspection result data, nursing and doctor course records, physical signs, various instrument record data and medical advice data;
the calculation module is used for calculating the support degree of the corresponding VTE when any one association influence factor appears in all historical text data and the confidence degree of the corresponding VTE in all the historical text data with the association influence factor;
the judging module is used for judging whether the support degree is greater than a corresponding support degree set value and the confidence degree is greater than a corresponding confidence degree set value, if yes, the associated influence factor is taken as a key influence factor, and if not, the associated influence factor is not taken as a key influence factor;
the dimension reduction module is used for carrying out dimension reduction processing on the key influence factors to obtain the key influence factors without collinear relation as optimal influence factors;
the training module is used for substituting the optimal influence factors of part of the historical text data in the preprocessed historical text data and whether the patient has VTE as training samples into a random forest model for model training so as to obtain a trained risk assessment model;
the testing module is used for substituting the optimal influence factors in the residual historical text data as testing samples into the trained risk assessment model for testing so as to obtain the probability value of VTE of the tested patient;
the grading module is used for converting each probability value into a rank, mapping the rank distribution to a normal distribution curve, and grading by using the 3 sigma principle of normal distribution so as to obtain a risk grade label corresponding to each test patient.
2. The VTE intelligent control management system according to claim 1, wherein the grading module is configured to convert each probability value into a rank, calculate each corresponding cumulative frequency based on each rank, convert each cumulative frequency into a corresponding probability unit, substitute each probability unit into a linear regression equation to calculate a corresponding estimated value, map each estimated value distribution onto a normal distribution curve, and perform grading by using a 3 σ principle of normal distribution to obtain a risk grade label corresponding to each test patient.
3. The VTE intelligent control management system of claim 1, further comprising an optimization module to revise and optimize the trained risk assessment model.
4. The VTE intelligent control management system of claim 3, wherein the optimization module is configured to modify the optimally trained risk assessment model based on fixed rules formulated by Delphi scoring.
5. The VTE intelligent prevention and treatment management system according to claim 1, wherein the dimension reduction module is configured to perform dimension reduction processing on the key impact factors by using a principal component analysis method.
6. The VTE intelligent prevention management system of claim 1, further comprising a prediction module to substitute the preprocessed textual data for the new patient into a trained and tested risk assessment model to predict the probability that the new patient has a VTE and a corresponding risk profile label.
7. The VTE intelligent control management system of claim 6, wherein the risk profile tags comprise a low risk, a medium risk, and a high risk;
the system further comprises a reminding module, wherein the reminding module is used for reminding corresponding workers based on different risk grade labels, and carrying out highest-grade risk reminding when any optimal influence factor in the optimal influence factors appears in text data of the new patient after the risk grade label that the new patient suffers from the VTE is obtained as the medium risk or the high risk.
8. The VTE intelligent prevention and treatment management system according to claim 1, wherein the calculation module is configured to calculate a support degree, which is a number of historical text data/a total number of all historical text data corresponding to the VTE with the incidence factor appearing in all historical text data;
the calculation module is further used for calculating the confidence coefficient which is the number of the historical text data corresponding to the VTE in all the historical text data with the associated influence factor/the total number of the historical text data with the associated influence factor.
9. The VTE intelligent prevention and treatment management system of claim 1, wherein the word segmentation module is configured to perform word segmentation on the preprocessed historical text data by using NLP technology.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010843967.1A CN111968747B (en) | 2020-08-20 | 2020-08-20 | VTE intelligent control management system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010843967.1A CN111968747B (en) | 2020-08-20 | 2020-08-20 | VTE intelligent control management system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111968747A true CN111968747A (en) | 2020-11-20 |
CN111968747B CN111968747B (en) | 2023-12-12 |
Family
ID=73389499
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010843967.1A Active CN111968747B (en) | 2020-08-20 | 2020-08-20 | VTE intelligent control management system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111968747B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113723674A (en) * | 2021-08-18 | 2021-11-30 | 卫宁健康科技集团股份有限公司 | Medical risk prediction method based on big data correlation |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103020454A (en) * | 2012-12-15 | 2013-04-03 | 中国科学院深圳先进技术研究院 | Method and system for extracting morbidity key factor and early warning disease |
CN104766259A (en) * | 2015-03-31 | 2015-07-08 | 华据医疗评估信息技术(北京)有限公司 | Medical clinical quality monitoring and evaluation system based on single-disease model |
CN104866979A (en) * | 2015-06-08 | 2015-08-26 | 苏芮 | Traditional Chinese medicine case data processing method and system of emergent acute infectious disease |
CN108153734A (en) * | 2017-12-26 | 2018-06-12 | 北京嘉和美康信息技术有限公司 | A kind of text handling method and device |
CN108597614A (en) * | 2018-04-12 | 2018-09-28 | 上海熙业信息科技有限公司 | A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record |
WO2018215590A1 (en) * | 2017-05-24 | 2018-11-29 | Gendiag.Exe, S.L. | Cancer-associated venous thromboembolic events |
WO2018223005A1 (en) * | 2017-06-02 | 2018-12-06 | The Henry M. Jackson Foundation For The Advancement Of Military Medicine, Inc. | Predictive factors for venous thromboembolism |
CN110556186A (en) * | 2019-09-17 | 2019-12-10 | 田红燕 | Medical care interaction system for preventing and treating venous thromboembolism of inpatient |
WO2020019797A1 (en) * | 2018-07-23 | 2020-01-30 | 无锡慧方科技有限公司 | Method, device, computer, and readable storage medium for electronic medical record data analysis |
-
2020
- 2020-08-20 CN CN202010843967.1A patent/CN111968747B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103020454A (en) * | 2012-12-15 | 2013-04-03 | 中国科学院深圳先进技术研究院 | Method and system for extracting morbidity key factor and early warning disease |
CN104766259A (en) * | 2015-03-31 | 2015-07-08 | 华据医疗评估信息技术(北京)有限公司 | Medical clinical quality monitoring and evaluation system based on single-disease model |
CN104866979A (en) * | 2015-06-08 | 2015-08-26 | 苏芮 | Traditional Chinese medicine case data processing method and system of emergent acute infectious disease |
WO2018215590A1 (en) * | 2017-05-24 | 2018-11-29 | Gendiag.Exe, S.L. | Cancer-associated venous thromboembolic events |
WO2018223005A1 (en) * | 2017-06-02 | 2018-12-06 | The Henry M. Jackson Foundation For The Advancement Of Military Medicine, Inc. | Predictive factors for venous thromboembolism |
CN108153734A (en) * | 2017-12-26 | 2018-06-12 | 北京嘉和美康信息技术有限公司 | A kind of text handling method and device |
CN108597614A (en) * | 2018-04-12 | 2018-09-28 | 上海熙业信息科技有限公司 | A kind of auxiliary diagnosis decision-making technique based on Chinese electronic health record |
WO2020019797A1 (en) * | 2018-07-23 | 2020-01-30 | 无锡慧方科技有限公司 | Method, device, computer, and readable storage medium for electronic medical record data analysis |
CN110556186A (en) * | 2019-09-17 | 2019-12-10 | 田红燕 | Medical care interaction system for preventing and treating venous thromboembolism of inpatient |
Non-Patent Citations (2)
Title |
---|
WANG, X 等: "Comparing different venous thromboembolism risk assessment machine learning models in Chinese patients", JOURNAL OF EVALUATION IN CLINICAL PRACTICE, vol. 26, no. 1, pages 26 - 34 * |
李笠;魏蓉溪;姚晓东;: "医院VTE风险评估与预警监控系统的应用", 中国卫生信息管理杂志, no. 02, pages 99 - 102 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113723674A (en) * | 2021-08-18 | 2021-11-30 | 卫宁健康科技集团股份有限公司 | Medical risk prediction method based on big data correlation |
Also Published As
Publication number | Publication date |
---|---|
CN111968747B (en) | 2023-12-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Lecky et al. | Trauma scoring systems and databases | |
CN108198615B (en) | Online cognitive evaluation system | |
US8682693B2 (en) | Patient data mining for lung cancer screening | |
Bray et al. | Evaluation of data quality in the cancer registry: principles and methods. Part I: comparability, validity and timeliness | |
US7711404B2 (en) | Patient data mining for lung cancer screening | |
Ceniccola et al. | Relevance of AND-ASPEN criteria of malnutrition to predict hospital mortality in critically ill patients: A prospective study | |
Fetter et al. | Ambulatory visit groups: a framework for measuring productivity in ambulatory care. | |
JP2018067303A (en) | Diagnosis support method, program and apparatus | |
Ghosh et al. | Early Deterioration Indicator: Data-driven approach to detecting deterioration in general ward | |
EP0591439A1 (en) | A clinical information reporting system | |
Lee et al. | The minimal clinically important difference for PROMIS physical function in patients with thumb carpometacarpal arthritis | |
Grün et al. | Identifying heart failure in ECG data with artificial intelligence—a meta-analysis | |
Martín-Rodríguez et al. | Association of prehospital oxygen saturation to inspired oxygen ratio with 1-, 2-, and 7-day mortality | |
Humbertjean et al. | Predictive factors of brain death in severe stroke patients identified by organ procurement and transplant coordination in Lorrain, France | |
Fine et al. | Assessing population-level symptoms of anxiety, depression, and suicide risk in real time using NLP applied to social media data | |
Yonge et al. | The respiratory rate: a neglected triage tool for pre‐hospital identification of trauma patients | |
Truong et al. | Biceps tenodesis demonstrates lower reoperation rates compared to SLAP repair for treatment of SLAP tears in a large cross-sectional population | |
Raghu et al. | Deep learning to predict mortality after cardiothoracic surgery using preoperative chest radiographs | |
CN111968747A (en) | VTE intelligent prevention and control management system | |
Pirkle et al. | Validity and reliability of criterion based clinical audit to assess obstetrical quality of care in West Africa | |
Mulligan | Validation of a physiological track and trigger score to identify developing critical illness in haematology patients | |
EP4002382A1 (en) | Using unstructured temporal medical data for disease prediction | |
US8756234B1 (en) | Information theory entropy reduction program | |
McConchie et al. | The AusPSIs: the Australian version of the Agency of Healthcare Research and Quality patient safety indicators | |
Holt et al. | A nationwide adaptive prediction tool for coronary heart disease prevention. |
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 |