CN107341345A - A kind of clinical rational drug use methods of risk assessment based on electronic health record big data - Google Patents
A kind of clinical rational drug use methods of risk assessment based on electronic health record big data Download PDFInfo
- Publication number
- CN107341345A CN107341345A CN201710483547.5A CN201710483547A CN107341345A CN 107341345 A CN107341345 A CN 107341345A CN 201710483547 A CN201710483547 A CN 201710483547A CN 107341345 A CN107341345 A CN 107341345A
- Authority
- CN
- China
- Prior art keywords
- drug
- risk
- medicine
- mrow
- medication
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000003814 drug Substances 0.000 title claims abstract description 230
- 229940079593 drug Drugs 0.000 title claims abstract description 158
- 230000036541 health Effects 0.000 title claims abstract description 43
- 238000012502 risk assessment Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 title claims abstract description 17
- 201000010099 disease Diseases 0.000 claims abstract description 49
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 49
- 206010013710 Drug interaction Diseases 0.000 claims abstract description 21
- 238000003745 diagnosis Methods 0.000 claims abstract description 20
- 239000000890 drug combination Substances 0.000 claims abstract description 20
- 238000013210 evaluation model Methods 0.000 claims abstract description 20
- 239000000284 extract Substances 0.000 claims abstract description 4
- 239000004615 ingredient Substances 0.000 claims description 29
- 239000011159 matrix material Substances 0.000 claims description 12
- 230000003993 interaction Effects 0.000 claims description 9
- 238000012986 modification Methods 0.000 claims description 5
- 230000004048 modification Effects 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 238000013136 deep learning model Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000010845 search algorithm Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 206010061623 Adverse drug reaction Diseases 0.000 description 1
- 208000030453 Drug-Related Side Effects and Adverse reaction Diseases 0.000 description 1
- 206010020751 Hypersensitivity Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000007815 allergy Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000000642 iatrogenic effect Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/22—Social work or social welfare, e.g. community support activities or counselling services
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Child & Adolescent Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a kind of clinical rational drug use methods of risk assessment based on electronic health record big data, comprise the following steps:S1 is from electronic health record extracting data rational use of medicines record and wrong medicine event, according to factors such as the physiological characteristic of patient, medical diagnosis on disease result, drug combination situations, the risk evaluation model on rational uses of medicines such as dosage, drug interaction, medication contraindications is established automatically;S2 extracts the factors such as its corresponding patient physiological characteristic, medical diagnosis on disease result, drug combination situation, using the rational use of medicines risk evaluation model, show that the drug risk of individuation is assessed, decision-making foundation is provided for clinical pharmaceutical administration to new electronic health record.Clinical rational drug use methods of risk assessment of the present invention, clinical application risk can be assessed automatically, and result accurate and effective, decision-making foundation can be provided for Practice of Clinical Rational Use of Drugs.
Description
Technical field
The invention belongs to clinical pharmaceutical administration field, and in particular to a kind of clinical rational based on electronic health record big data is used
Medicine methods of risk assessment.
Background technology
With the fast development of China's medical information, hospital information management system constantly improve, have accumulated on patient
Magnanimity electronic health record data.Electronic health record big data contains detailed record of the patient in hospital diagnosis therapeutic process, to prison
Diagnosis and the drug level for surveying and evaluating medical institutions are significant.
The rational use of medicines refer to safely, effectively, economically use medicine.Clinically, it is necessary to be known with the system of medicine and disease
Based on knowledge, optimal medicine and its preparation is selected according to kinds of Diseases, status of patient and pharmacology theory, formulate or adjust to
Prescription case.
Existing rational use of medicines methods of risk assessment is primarily present following both sides problem.On the one hand, its conjunction used
Manage the foundation of medication normative database and renewal relies on manual type and enters edlin arrangement to medication pertinent literature, it is necessary to disappear mostly
Substantial amounts of manpower and materials are consumed, especially, for the medicine of new listing, the problems such as data missing, renewal hysteresis often be present, cause
It is difficult to set up accurately and effectively medication and examines rule;On the other hand, rule-based medication examines that module intelligence degree is not high,
Examine that rule lacks flexibility, it is impossible to which the physiological status, condition-inference and drug combination situation for considering patient carry out individual
Change the rational use of medicines to assess, cause invalid warning ratio higher.Two above problem significantly limit traditional rational use of medicines decision-making
Effect of the support system in clinical pharmaceutical administration.
The content of the invention
It is an object of the invention to overcome above-mentioned the deficiencies in the prior art, there is provided a kind of facing based on electronic health record big data
Bed rational use of medicines methods of risk assessment.
To achieve the above object, the present invention uses following technical scheme:
A kind of clinical rational drug use methods of risk assessment based on electronic health record big data, it is characterised in that including following
Step:
S1, rational use of medicines record is extracted from electronic health record database, according to the physiological characteristic of patient, medical diagnosis on disease result
And its corresponding rational use of medicines dosage, using deep learning model, modeling medicine dosage is on patient physiological characteristic and disease
The relation of sick diagnostic result, dosage risk probability of each similar drug in the case of Different Individual is assessed, establishes dosage
Risk evaluation model;
S2, to new electronic health record, its corresponding patient physiological characteristic, medical diagnosis on disease result are extracted, using the medication
Dosage risk evaluation model predicts rational use of medicines dosage, and the medication that will be outputed in the rational use of medicines dosage of prediction and electronic health record
Dosage is compared, and is calculated dosage risk factor, if dosage risk factor is more than given threshold, is carried out using medicament
Amount warning.
Further, dosage risk factor η1Calculated by following formula,
In formula,The rational use of medicines dosage predicted by dosage risk evaluation model, y are what is outputed in electronic health record
Dosage, the threshold value are 15%.
Further, step S1 also includes establishing drug interaction risk evaluation model, is specially:By classifying drugs,
From drug ingredient corresponding to package insert acquisition, rational use of medicines record and drug interaction are extracted not from the database of case history
Good reaction event is as model training sample, and using drug ingredient as node, drug interaction event is side, establishes probability artwork
Type:If drug interaction Adverse Event occurred between drug ingredient, the weight on the side between its corresponding node adds
One;Otherwise the weight on the side between node is set to zero;Iteration renewal drug ingredient interaction probability graph, establishes medicine phase interaction
Use risk evaluation model;
Step S2 also includes the risk assessment of drug interaction, is specially:Carried first from electronic health record to be assessed
Drug combination situation and its corresponding drug ingredient are taken, then the node using drug ingredient used as figure, structure node set Vs,
Subgraph search algorithm is run in the drug ingredient interaction probability graph G constructed in S12, acquisition includes node set Vs's
Subgraph Gs;If subgraph GsThe middle side that weight be present and be not zero, then medicine list and its weight corresponding to node where exporting the side
Drug interaction risk factor η representated by w2,
WhereinFor the average weight on drug ingredient interaction probability graph G non-zero side.
Further, step S1 also includes the foundation of medication contraindication risk evaluation model, is specially:By disease and medicine
Classification, the disease medication event of rational use of medicines record and mistake is extracted from electronic health record database, using disease-medicine as two
Dimension, establish government image:If medication contraindication event occurred between a kind of disease and a kind of medicine, its corresponding square
The value of array element element adds one;If medication contraindication event does not occur for corresponding disease-drug combination, using drug ingredient similitude as according to
According to the medication contraindication based on existing similar medicine is distributed, and infers medication contraindication risk probability corresponding to the medicine;Iteration
Medication contraindication probability matrix model is updated, establishes medication contraindication risk evaluation model;
Step S2 also includes the risk assessment of medication contraindication:First medical diagnosis on disease is extracted from electronic health record to be assessed
List I ' and medicine list J ' used, for each single item disease i ∈ I ' in the case history and each single item medicine j ∈ J ', utilizes step
The medication contraindication probability matrix model M constructed in rapid S13 is calculated with the presence or absence of medication taboo risk:If corresponding disease-medicine
The risk probability of product combination is not zero, then exports risk warning and its corresponding risk factor η3,
WhereinFor the average value of the nonzero element of medication contraindication probability matrix model M.
Further, medicine is carried out to disease-drug combination that medication contraindication event does not occur using collaborative filtering
Medication contraindication risk probability corresponding to product is inferred;
First, the similitude of two kinds of medicines is calculated according to drug ingredient:Remember two kinds of medicine j1And j2Contained composition is set
C1With set C2, then set of computations C1With set C2Jie Kade similarity factors,
Secondly, the similarity measurement based on medicine infers unknown medication contraindication risk probability:For medicine j, choose
Maximum and more than given threshold the medicine j ' with its similarity factor, and medicine is assigned to the corresponding medication contraindication risks of j '
Matrix element corresponding to j.
Further, the rational use of medicines record and wrong medicine event are given birth to by the medication dirty bit of electronic health record
Into, if medication dirty bit is not found in case history, one rational use of medicines record of generation;If medication modification mark is found in case history
Will position, then to one wrong medicine event of modification generation every time.
After adopting the above technical scheme, the present invention has the following advantages that compared with background technology:Clinic proposed by the present invention
Rational use of medicines methods of risk assessment, it can be not required to automatically from magnanimity history medical record data learning and modeling rational use of medicines pattern
Want manual maintenance medication rule database;Drug risk intelligently can be predicted according to patient's actual conditions, realize individuation
Rational use of medicines risk assessment.The present invention will provide accurately and effectively decision-making foundation for medical institutions' Practice of Clinical Rational Use of Drugs, drop
Low iatrogenic drug risk, effective guarantee is provided for the drug safety of the people.
Brief description of the drawings
Fig. 1 is the specific implementation theory diagram of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Embodiment
As shown in figure 1, electronic health record database 1 can pass through the hospital information system (Hospital of medical institutions
Information System, HIS) etc. approach obtain.In the present embodiment, using certain Grade A hospital outpatient service of 1 year in 2015
Electronic health record database, after data de-noising, cleaning, obtain about 3,000,000 parts of effective electron case histories.
Clinical rational drug use methods of risk assessment based on electronic health record big data, comprises the following steps:
S1, rational use of medicines record and wrong medicine event, and corresponding patient life are extracted from electronic health record database
The information such as feature, medical diagnosis on disease result, drug combination situation are managed, are established on dosage, drug interaction, medicine contraindication
The risk evaluation model of disease etc..
S2, the case history to newly outputing, patient physiological characteristic, medical diagnosis on disease result, drug combination situation etc. corresponding to extraction
Factor, using rational use of medicines risk evaluation model, draw personalized medicine risk assessment, for Hospital pharmacy management provide decision-making according to
According to.
The step of S1 is specially:
S11, rational use of medicines database of record 8 and wrong medicine event database 9 are extracted from electronic health record database 1.
Electronic health record has medication dirty bit, is audited, at adverse drug reaction event before doctor goes together comment, pharmacists's dispensing
In the links such as reason, medication problem and its reason can be marked in the flag bit, and case history modification is carried out by flow.If do not sent out in case history
Current medicine dirty bit, then generate a rational use of medicines record;If medication dirty bit is found in case history, to repairing every time
Change one wrong medicine event of generation.
S12, patient physiological characteristic is extracted from electronic health record database 1, specifically include weight in patients, age, blood group, both
Toward medical history, whether special population (such as children, pregnant woman), whether have allergies critical field, and do sliding-model control, establish and suffer from
Person's physiological characteristic data storehouse 2;Medical diagnosis on disease result is extracted from electronic health record database 1, according to the injury of international disease and the cause of the death
Criteria for classification the tenth edition (ICD10) carries out structuring filing, establishes medical diagnosis on disease result database 3;Extracted from electronic health record
Drug combination situation, structuring filing is carried out by medicament categories, establishes drug combination situation database 4.
S13, establish dosage risk evaluation model.Specifically, a kind of medicine is given, from patient physiological characteristic's database
Patient physiological characteristic corresponding to the medicine is extracted in 2 to record, and comprising above-mentioned field, is expressed as X1;From medical diagnosis on disease result data
Medical diagnosis on disease result corresponding to the medicine is extracted in storehouse 3 to record, and is taken the ICD10 of first, second, and third diagnosis to number and is used as three
Individual field, the zero setting if without corresponding diagnose, is expressed as X2.Extract from rational use of medicines database of record 8 and used corresponding to the medicine
Pharmaceutical quantities, it is expressed as Y.With X=[X1, X2] input layer is used as, for Y as output layer, training one has the depth god of 8 hidden layers
Through network model f so that:Y~f (X).In the present embodiment, the record that 70% is randomly selected from the data set of said extracted enters
Row model training, remaining 30% is used for model cross validation.Each single item medicine in drug combination database 4 is repeated above-mentioned
Operation, dosage risk evaluation model corresponding to foundation.
S14, establish drug interaction risk evaluation model.First, drug combination situation database 4 is traveled through, according to medicine
Product specification obtains its corresponding drug ingredient set V.Then, using set V as node, probability graph model G=(V, E) is built,
The set on sides of the wherein E between node.A case history is given, if wherein having drug combination and drug interaction thing not occurring
Part, then side is generated between its corresponding drug ingredient node, and the weight on side is set to zero;If wherein there are drug combination and hair
Raw drug interaction event, then generate side, and the weight on side is added into one between its corresponding drug ingredient.Traversal is rationally used
Medicine database of record 8 and wrong medicine event database 9, iteration update drug ingredient interaction probability graph model.
S15, establish medication contraindication risk evaluation model.First, medical diagnosis on disease result database 3 is traveled through, obtains disease
The vectorial I of diagnostic result classification;Drug combination situation database 4 is traveled through, obtains medicine categorization vector J.Then, with disease-medicine
Product are two dimensions, establish government image MI×J.If medication contraindication thing occurred between a kind of disease and a kind of medicine
Part, then the value of its homography element add one;If medication taboo event did not occurred for corresponding disease-drug combination, may be somebody's turn to do
Data missing or the disease medication event not made a mistake be present, it is necessary to do further risk assessment according to drug ingredient in medicine.
In the present embodiment, unknown disease-medicine contraindication risk is assessed using collaborative filtering.Specifically, first
The similitude of two kinds of medicines is calculated according to drug ingredient.Remember two kinds of medicine j1And j2Contained composition is set C1With set C2.So
Set of computations C afterwards1With set C2Jie Kade similarity factors (Jaccard Similarity Coefficient):
On this basis, the similarity measurement based on medicine infers unknown medication contraindication risk probability.Specifically,
For medicine j, maximum and more than first threshold the medicine j ' with its similarity factor is chosen, and with medication contraindication corresponding j '
Risk is assigned to matrix element corresponding to medicine j.Preferably, first threshold 90%.
The step of S2 is specially:
S21, assess dosage risk.From case history to be assessed, medicine list J ' used in case history is extracted.For it
Middle each single item medicine j ∈ J ', call its rational use of medicines dosage of deep learning model prediction corresponding to the medicine.Specifically, it is defeated
Enter the patient physiological characteristic x in the case history1With medical diagnosis on disease result x2, measure in advanceAs the reasonable of recommendation
Dosage value.Then, by the recommended drug dosageCompared with the dosage y outputed in electronic health record, medication is calculated
Dosage risk factor η1:
If risk factor η1More than Second Threshold, then dosage warning is carried out.In the present embodiment, Second Threshold is
15%.
S22, assess drug interaction risk.First from case history to be assessed, extraction combination medicine and its correspondingly into
The list C ' divided.Use wherein drug ingredient c ∈ C ', structure node set Vs.Then in drug ingredient interaction probability graph
Operation subgraph search algorithm in model G=(V, E), acquisition include node set VsSubgraph Gs=(Vs, Es) so that:
Es=Vs×Vs∩E
Next to subgraph GsAnalyzed.If GsThe middle side that weight be present and be not zero, then export the node pair where the side
The drug ingredient answered warns as drug interaction risk, and the drug interaction risk factor representated by its weight w
η2:
WhereinFor the average weight on the non-zero side in probability graph model G.
S23, assess medication contraindication risk.First from electronic health record to be assessed, extraction medical diagnosis on disease list I ' and
It is combined medicine list J '.For each single item disease i ∈ I ' in the case history and each single item medicine j ∈ J ', institute in step (1) is utilized
The medication contraindication probability matrix model M of construction calculates to be avoided with the presence or absence of medication, that is, judges whether M (i, j) is zero.If M (i,
J) it is not zero, then it is assumed that corresponding disease-drug combination has medication taboo risk, the risk warning of output medication contraindication, and counts
Risk factor η corresponding to calculation3:
WhereinFor the average value of nonzero element in matrix model M.
In the present embodiment, on electronic health record database used carry out rational use of medicines risk assessment, and with manual review knot
Fruit is contrasted, and its accurate rate (Precision) and recall rate (Recall) respectively reach 93.17% and 96.25%.Thus may be used
See, the inventive method can be accurately and effectively assessed reasonable drug risk, had in clinical pharmaceutical administration good
Practical value.
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto,
Any one skilled in the art the invention discloses technical scope in, the change or replacement that can readily occur in,
It should all be included within the scope of the present invention.Therefore, protection scope of the present invention should be with scope of the claims
It is defined.
Claims (6)
1. a kind of clinical rational drug use methods of risk assessment based on electronic health record big data, it is characterised in that including following step
Suddenly:
S1, from electronic health record database extract the rational use of medicines record, according to the physiological characteristic of patient, medical diagnosis on disease result and its
Corresponding rational use of medicines dosage, using deep learning model, modeling medicine dosage is examined on patient physiological characteristic and disease
The relation of disconnected result, assesses dosage risk probability of each similar drug in the case of Different Individual, establishes dosage risk
Assessment models;
S2, to new electronic health record, its corresponding patient physiological characteristic, medical diagnosis on disease result are extracted, using the dosage
Risk evaluation model predicts rational use of medicines dosage, and the dosage that will be outputed in the rational use of medicines dosage of prediction and electronic health record
It is compared, calculates dosage risk factor, if dosage risk factor is more than given threshold, carries out dosage police
Show.
2. a kind of clinical rational drug use methods of risk assessment based on electronic health record big data according to claim 1, its
It is characterised by:Dosage risk factor η1Calculated by following formula,
<mrow>
<msub>
<mi>&eta;</mi>
<mn>1</mn>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mover>
<mi>y</mi>
<mo>^</mo>
</mover>
<mo>-</mo>
<mi>y</mi>
<mo>|</mo>
</mrow>
<mover>
<mi>y</mi>
<mo>^</mo>
</mover>
</mfrac>
<mo>&times;</mo>
<mn>100</mn>
<mi>%</mi>
<mo>;</mo>
</mrow>
In formula,The rational use of medicines dosage predicted by dosage risk evaluation model, y are the medication outputed in electronic health record
Dosage, the threshold value are 15%.
3. a kind of clinical rational drug use methods of risk assessment based on electronic health record big data according to claim 1, its
It is characterised by:
Step S1 also includes establishing drug interaction risk evaluation model, is specially:By classifying drugs, obtained from package insert
Corresponding drug ingredient is taken, rational use of medicines record and drug interaction Adverse Event conduct are extracted from the database of case history
Model training sample, using drug ingredient as node, drug interaction event is side, establishes probability graph model, if drug ingredient
Between drug interaction Adverse Event occurred, then the weight on the side between its corresponding node adds one;Otherwise node it
Between the weight on side be set to zero;Iteration renewal drug ingredient interaction probability graph, establishes drug interaction risk assessment mould
Type;
Step S2 also includes the risk assessment of drug interaction, is specially:First medicine is extracted from electronic health record to be assessed
Internet of Things situation and its corresponding drug ingredient, the then node using drug ingredient used as figure, structure node set Vs,
Subgraph search algorithm is run in the drug ingredient interaction probability graph G constructed in S12, acquisition includes node set VsSon
Scheme Gs:If subgraph GsThe middle side that weight be present and be not zero, then medicine list and its weight w corresponding to node where exporting the side
Representative drug interaction risk factor η2,
<mrow>
<msub>
<mi>&eta;</mi>
<mn>2</mn>
</msub>
<mo>=</mo>
<mfrac>
<mi>w</mi>
<mover>
<mi>w</mi>
<mo>&OverBar;</mo>
</mover>
</mfrac>
<mo>&times;</mo>
<mn>100</mn>
<mi>%</mi>
<mo>,</mo>
</mrow>
WhereinFor the average weight on drug ingredient interaction probability graph G non-zero side.
4. a kind of clinical rational drug use methods of risk assessment based on electronic health record big data according to claim 1, its
It is characterised by:
Step S1 also includes the foundation of medication contraindication risk evaluation model, is specially:By disease and classifying drugs, from electronics disease
Go through in database and extract the disease medication event of rational use of medicines record and mistake, using disease-medicine as two dimensions, establish two dimension
Matrix model:If medication contraindication event occurred between a kind of disease and a kind of medicine, the value of its homography element adds
One;If medication contraindication event does not occur for corresponding disease-drug combination, using drug ingredient similitude as foundation, based on existing
Similar medicine medication contraindication distribution, infer medication contraindication risk probability corresponding to the medicine;Iteration more novel drugs are prohibited
Avoid disease probability matrix model, establish medication contraindication risk evaluation model;
Step S2 also includes the risk assessment of medication contraindication:Medical diagnosis on disease list is extracted first from electronic health record to be assessed
I ' and medicine list J ' used, for each single item disease i ∈ I ' in the case history and each single item medicine j ∈ J ', utilizes step S13
Middle constructed medication contraindication probability matrix model M is calculated with the presence or absence of medication taboo risk:If corresponding disease-drug combination
Risk probability be not zero, then export risk warning and its corresponding risk factor η3,
<mrow>
<msub>
<mi>&eta;</mi>
<mn>3</mn>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mi>M</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
<mover>
<mi>M</mi>
<mo>&OverBar;</mo>
</mover>
</mfrac>
<mo>&times;</mo>
<mn>100</mn>
<mi>%</mi>
<mo>;</mo>
</mrow>
WhereinFor the average value of the nonzero element of medication contraindication probability matrix model M.
5. a kind of clinical rational drug use methods of risk assessment based on electronic health record big data according to claim 4, its
It is characterised by:Disease-drug combination that medication contraindication event does not occur is carried out corresponding to medicine using collaborative filtering
Medication contraindication risk probability is inferred;
First, the similitude of two kinds of medicines is calculated according to drug ingredient:Remember two kinds of medicine j1And j2Contained composition is set C1With
Set C2, then set of computations C1With set C2Jie Kade similarity factors,
<mrow>
<msub>
<mi>S</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mo>|</mo>
<msub>
<mi>C</mi>
<mn>1</mn>
</msub>
<mo>&cap;</mo>
<msub>
<mi>C</mi>
<mn>2</mn>
</msub>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mi>C</mi>
<mn>1</mn>
</msub>
<mo>&cup;</mo>
<msub>
<mi>C</mi>
<mn>2</mn>
</msub>
<mo>|</mo>
</mrow>
</mfrac>
<mo>;</mo>
</mrow>
Secondly, the similarity measurement based on medicine infers unknown medication contraindication risk probability:For medicine j, choose and its
Similarity factor is maximum and is more than the medicine j ' of given threshold, and is assigned to j pairs of medicine with the corresponding medication contraindication risks of j '
The matrix element answered.
A kind of 6. clinical rational drug use risk assessment based on electronic health record big data according to claim any one of 1-5
Method, it is characterised in that:The rational use of medicines record and wrong medicine event are given birth to by the medication dirty bit of electronic health record
Into, if medication dirty bit is not found in case history, one rational use of medicines record of generation;If medication modification mark is found in case history
Will position, then to one wrong medicine event of modification generation every time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710483547.5A CN107341345A (en) | 2017-06-22 | 2017-06-22 | A kind of clinical rational drug use methods of risk assessment based on electronic health record big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710483547.5A CN107341345A (en) | 2017-06-22 | 2017-06-22 | A kind of clinical rational drug use methods of risk assessment based on electronic health record big data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107341345A true CN107341345A (en) | 2017-11-10 |
Family
ID=60220754
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710483547.5A Pending CN107341345A (en) | 2017-06-22 | 2017-06-22 | A kind of clinical rational drug use methods of risk assessment based on electronic health record big data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107341345A (en) |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108447568A (en) * | 2018-03-19 | 2018-08-24 | 杭州逸曜信息技术有限公司 | Medication information processing method based on drug ingredient |
CN108648794A (en) * | 2018-05-14 | 2018-10-12 | 深圳医翼健康科技有限公司 | A kind of electronic prescription drug safety control system |
CN108717862A (en) * | 2018-04-10 | 2018-10-30 | 四川骏逸富顿科技有限公司 | A kind of careful square evolution model of the intelligence based on machine learning |
CN109035094A (en) * | 2018-08-01 | 2018-12-18 | 杭州认知网络科技有限公司 | Teaching method, device and terminal device based on artificial intelligence |
CN109243567A (en) * | 2018-08-14 | 2019-01-18 | 山东科技大学 | A kind of drug recommended method excavated based on prescription data |
CN109273096A (en) * | 2018-09-05 | 2019-01-25 | 南京邮电大学 | A kind of risk management grading evaluation method based on machine learning |
CN109508869A (en) * | 2018-10-23 | 2019-03-22 | 平安医疗健康管理股份有限公司 | A kind of risk checking method and device based on data processing |
CN109545320A (en) * | 2018-10-27 | 2019-03-29 | 平安医疗健康管理股份有限公司 | Medicine method of specifying error and terminal device are taken based on data processing |
CN109616176A (en) * | 2018-12-04 | 2019-04-12 | 平安医疗健康管理股份有限公司 | Method, apparatus, equipment and the storage medium that auxiliary doctor prescribes |
CN109637620A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Prescription based on data analysis generates monitoring method, device, equipment and medium |
CN109859815A (en) * | 2019-01-24 | 2019-06-07 | 西安交通大学 | A kind of adjuvant drug decision-making technique and intelligent adjuvant drug system |
CN110223751A (en) * | 2019-05-16 | 2019-09-10 | 平安科技(深圳)有限公司 | Prescription evaluation method, system and computer equipment based on medical knowledge map |
CN110289098A (en) * | 2019-05-17 | 2019-09-27 | 天津科技大学 | A kind of Risk Forecast Method for intervening data based on clinical examination and medication |
CN110534168A (en) * | 2019-08-30 | 2019-12-03 | 北京百度网讯科技有限公司 | Medicine advises indicating risk method, apparatus, electronic equipment and storage medium |
CN110782965A (en) * | 2019-10-12 | 2020-02-11 | 平安医疗健康管理股份有限公司 | Data processing method, data processing device, computer equipment and storage medium |
CN110827951A (en) * | 2019-12-24 | 2020-02-21 | 陕西省中医医院 | Clinical intelligent decision platform |
CN111009322A (en) * | 2019-10-21 | 2020-04-14 | 四川大学华西医院 | Perioperative risk assessment and clinical decision intelligent auxiliary system |
CN111223536A (en) * | 2020-02-18 | 2020-06-02 | 中国医学科学院北京协和医院 | Method and device for predicting heparin dosage |
CN111383761A (en) * | 2018-12-28 | 2020-07-07 | 医渡云(北京)技术有限公司 | Medical data analysis method and device, electronic equipment and computer readable medium |
CN111383726A (en) * | 2018-12-28 | 2020-07-07 | 国家食品药品监督管理总局药品评价中心 | Electronic medical record data processing method and device, electronic equipment and readable medium |
CN108630322B (en) * | 2018-04-27 | 2020-08-14 | 厦门大学 | Drug interaction modeling and risk assessment method, terminal device and storage medium |
CN111968715A (en) * | 2020-06-30 | 2020-11-20 | 厦门大学 | Drug recommendation modeling method based on medical record data and drug interaction risk |
CN111986754A (en) * | 2020-08-21 | 2020-11-24 | 南通大学 | Electronic medical record management model construction method based on diabetes |
CN112151140A (en) * | 2020-11-25 | 2020-12-29 | 西藏自治区人民政府驻成都办事处医院 | Antibacterial drug clinical application system, electronic device and computer readable storage medium |
CN112562869A (en) * | 2021-02-24 | 2021-03-26 | 北京中医药大学东直门医院 | Drug combination safety evaluation system, method and device |
CN112786213A (en) * | 2020-12-26 | 2021-05-11 | 四川大学华西医院 | System and method for comprehensively evaluating multiple medication risks |
CN113517044A (en) * | 2021-08-13 | 2021-10-19 | 南京焰帆生物技术有限公司 | Clinical data processing method and system for evaluating citicoline based on pharmacokinetics |
CN113724830A (en) * | 2021-08-31 | 2021-11-30 | 平安国际智慧城市科技股份有限公司 | Medicine taking risk detection method based on artificial intelligence and related equipment |
CN113870974A (en) * | 2021-09-28 | 2021-12-31 | 平安国际智慧城市科技股份有限公司 | Risk prediction method and device based on artificial intelligence, electronic equipment and medium |
CN114049970A (en) * | 2021-11-10 | 2022-02-15 | 重庆大学 | Medication management method and system based on multi-terminal data sharing |
CN114283949A (en) * | 2021-11-08 | 2022-04-05 | 湖南云视数据科技有限责任公司 | ADR (adaptive Doppler ratio) aided decision making system based on deep learning |
CN117116504A (en) * | 2023-07-24 | 2023-11-24 | 北京同仁堂互联网医院管理有限公司 | Management method, device, equipment and readable medium for traditional Chinese medicine medication safety |
CN117352178A (en) * | 2023-11-10 | 2024-01-05 | 西安艾派信息技术有限公司 | Big data-based drug risk assessment system and method |
-
2017
- 2017-06-22 CN CN201710483547.5A patent/CN107341345A/en active Pending
Cited By (46)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108447568A (en) * | 2018-03-19 | 2018-08-24 | 杭州逸曜信息技术有限公司 | Medication information processing method based on drug ingredient |
CN108717862A (en) * | 2018-04-10 | 2018-10-30 | 四川骏逸富顿科技有限公司 | A kind of careful square evolution model of the intelligence based on machine learning |
CN108630322B (en) * | 2018-04-27 | 2020-08-14 | 厦门大学 | Drug interaction modeling and risk assessment method, terminal device and storage medium |
CN108648794A (en) * | 2018-05-14 | 2018-10-12 | 深圳医翼健康科技有限公司 | A kind of electronic prescription drug safety control system |
CN109035094A (en) * | 2018-08-01 | 2018-12-18 | 杭州认知网络科技有限公司 | Teaching method, device and terminal device based on artificial intelligence |
CN109243567B (en) * | 2018-08-14 | 2021-11-02 | 山东科技大学 | Medicine recommendation method based on prescription data mining |
CN109243567A (en) * | 2018-08-14 | 2019-01-18 | 山东科技大学 | A kind of drug recommended method excavated based on prescription data |
CN109273096A (en) * | 2018-09-05 | 2019-01-25 | 南京邮电大学 | A kind of risk management grading evaluation method based on machine learning |
CN109508869B (en) * | 2018-10-23 | 2023-09-22 | 平安医疗健康管理股份有限公司 | Risk detection method and device based on data processing |
CN109508869A (en) * | 2018-10-23 | 2019-03-22 | 平安医疗健康管理股份有限公司 | A kind of risk checking method and device based on data processing |
CN109545320A (en) * | 2018-10-27 | 2019-03-29 | 平安医疗健康管理股份有限公司 | Medicine method of specifying error and terminal device are taken based on data processing |
CN109616176A (en) * | 2018-12-04 | 2019-04-12 | 平安医疗健康管理股份有限公司 | Method, apparatus, equipment and the storage medium that auxiliary doctor prescribes |
CN109637620A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Prescription based on data analysis generates monitoring method, device, equipment and medium |
CN111383761A (en) * | 2018-12-28 | 2020-07-07 | 医渡云(北京)技术有限公司 | Medical data analysis method and device, electronic equipment and computer readable medium |
CN111383761B (en) * | 2018-12-28 | 2023-05-12 | 医渡云(北京)技术有限公司 | Medical data analysis method, medical data analysis device, electronic equipment and computer readable medium |
CN111383726A (en) * | 2018-12-28 | 2020-07-07 | 国家食品药品监督管理总局药品评价中心 | Electronic medical record data processing method and device, electronic equipment and readable medium |
CN109859815A (en) * | 2019-01-24 | 2019-06-07 | 西安交通大学 | A kind of adjuvant drug decision-making technique and intelligent adjuvant drug system |
CN110223751A (en) * | 2019-05-16 | 2019-09-10 | 平安科技(深圳)有限公司 | Prescription evaluation method, system and computer equipment based on medical knowledge map |
CN110289098B (en) * | 2019-05-17 | 2022-11-25 | 天津科技大学 | Risk prediction method based on clinical examination and medication intervention data |
CN110289098A (en) * | 2019-05-17 | 2019-09-27 | 天津科技大学 | A kind of Risk Forecast Method for intervening data based on clinical examination and medication |
CN110534168A (en) * | 2019-08-30 | 2019-12-03 | 北京百度网讯科技有限公司 | Medicine advises indicating risk method, apparatus, electronic equipment and storage medium |
CN110782965A (en) * | 2019-10-12 | 2020-02-11 | 平安医疗健康管理股份有限公司 | Data processing method, data processing device, computer equipment and storage medium |
CN110782965B (en) * | 2019-10-12 | 2022-10-11 | 深圳平安医疗健康科技服务有限公司 | Data processing method, data processing device, computer equipment and storage medium |
CN111009322B (en) * | 2019-10-21 | 2022-08-02 | 四川大学华西医院 | Perioperative risk assessment and clinical decision intelligent auxiliary system |
CN111009322A (en) * | 2019-10-21 | 2020-04-14 | 四川大学华西医院 | Perioperative risk assessment and clinical decision intelligent auxiliary system |
CN110827951A (en) * | 2019-12-24 | 2020-02-21 | 陕西省中医医院 | Clinical intelligent decision platform |
CN111223536A (en) * | 2020-02-18 | 2020-06-02 | 中国医学科学院北京协和医院 | Method and device for predicting heparin dosage |
CN111223536B (en) * | 2020-02-18 | 2024-05-10 | 中国医学科学院北京协和医院 | Method and device for predicting heparin dosage |
CN111968715A (en) * | 2020-06-30 | 2020-11-20 | 厦门大学 | Drug recommendation modeling method based on medical record data and drug interaction risk |
CN111968715B (en) * | 2020-06-30 | 2022-11-01 | 厦门大学 | Drug recommendation modeling method based on medical record data and drug interaction risk |
CN111986754A (en) * | 2020-08-21 | 2020-11-24 | 南通大学 | Electronic medical record management model construction method based on diabetes |
CN112151140A (en) * | 2020-11-25 | 2020-12-29 | 西藏自治区人民政府驻成都办事处医院 | Antibacterial drug clinical application system, electronic device and computer readable storage medium |
CN112151140B (en) * | 2020-11-25 | 2021-03-12 | 西藏自治区人民政府驻成都办事处医院 | Antibacterial drug clinical application system, electronic device and computer readable storage medium |
CN112786213B (en) * | 2020-12-26 | 2023-03-10 | 四川大学华西医院 | System and method for comprehensively evaluating multiple medication risks |
CN112786213A (en) * | 2020-12-26 | 2021-05-11 | 四川大学华西医院 | System and method for comprehensively evaluating multiple medication risks |
CN112562869A (en) * | 2021-02-24 | 2021-03-26 | 北京中医药大学东直门医院 | Drug combination safety evaluation system, method and device |
CN113517044A (en) * | 2021-08-13 | 2021-10-19 | 南京焰帆生物技术有限公司 | Clinical data processing method and system for evaluating citicoline based on pharmacokinetics |
CN113724830A (en) * | 2021-08-31 | 2021-11-30 | 平安国际智慧城市科技股份有限公司 | Medicine taking risk detection method based on artificial intelligence and related equipment |
CN113724830B (en) * | 2021-08-31 | 2024-04-30 | 深圳平安智慧医健科技有限公司 | Medication risk detection method based on artificial intelligence and related equipment |
CN113870974A (en) * | 2021-09-28 | 2021-12-31 | 平安国际智慧城市科技股份有限公司 | Risk prediction method and device based on artificial intelligence, electronic equipment and medium |
CN113870974B (en) * | 2021-09-28 | 2024-09-10 | 深圳平安智慧医健科技有限公司 | Risk prediction method and device based on artificial intelligence, electronic equipment and medium |
CN114283949A (en) * | 2021-11-08 | 2022-04-05 | 湖南云视数据科技有限责任公司 | ADR (adaptive Doppler ratio) aided decision making system based on deep learning |
CN114049970A (en) * | 2021-11-10 | 2022-02-15 | 重庆大学 | Medication management method and system based on multi-terminal data sharing |
CN117116504A (en) * | 2023-07-24 | 2023-11-24 | 北京同仁堂互联网医院管理有限公司 | Management method, device, equipment and readable medium for traditional Chinese medicine medication safety |
CN117116504B (en) * | 2023-07-24 | 2024-04-19 | 北京同仁堂互联网医院管理有限公司 | Management method, device, equipment and readable medium for traditional Chinese medicine medication safety |
CN117352178A (en) * | 2023-11-10 | 2024-01-05 | 西安艾派信息技术有限公司 | Big data-based drug risk assessment system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107341345A (en) | A kind of clinical rational drug use methods of risk assessment based on electronic health record big data | |
CN108595683B (en) | Medical advice rationality monitoring method based on medical knowledge map | |
CA2702408C (en) | Self-improving method of using online communities to predict health-related outcomes | |
Monane et al. | Improving prescribing patterns for the elderly through an online drug utilization review intervention: a system linking the physician, pharmacist, and computer | |
US10311975B2 (en) | Rules-based system for care management | |
CN106529177A (en) | Patient portrait drawing method and device based on medical big data | |
Archenaa et al. | Health recommender system using big data analytics | |
CN108182973A (en) | A kind of Intelligent Diagnosis Technology of knowledge based collection of illustrative plates reasoning | |
CN110827951A (en) | Clinical intelligent decision platform | |
Hasanpour et al. | A critical appraisal of heterogeneity in obsessive-compulsive disorder using symptom-based clustering analysis | |
Hu et al. | Network analysis and visualisation of opioid prescribing data | |
Hu et al. | Detecting anomalies in controlled drug prescription data using probabilistic models | |
Katiyara et al. | MedEstimate: Patient Treatment Recommendation Model | |
Al-Majmar et al. | Development a Model for Drug Interaction Prediction Based on Patient State | |
Raghu et al. | Fall-Related injuries and their prevention strategies of in-patient population in tertiary health care setup | |
TWI704509B (en) | Patient accident risk evaluation system and method | |
Wu et al. | Relationship Between Pain and Delirium in Critically Ill Adults | |
US20230307141A1 (en) | Models for accurate patient acuity | |
Goh et al. | Exploring the use of a network model in drug prescription support for dental clinics | |
CN106845052B (en) | Medical information processing method and medical information processing apparatus | |
Song et al. | Study on the numerical value and the distance of symptoms on the Medical Decision Support System | |
Hussain et al. | Detecting Adverse Drug Reaction (ADR) with data mining and predicting its intensity with machine learning | |
Zhou et al. | Machine learning methods for anticipating the psychological distress in patients with alzheimer’s disease | |
CN113241172A (en) | ICU discrimination system for postoperative infection of neurosurgical patient | |
Vethamani et al. | A Bayesian-Based Machine Learning Analysis |
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: 20171110 |
|
RJ01 | Rejection of invention patent application after publication |