CN109859815A - A kind of auxiliary medicine decision-making method and intelligent auxiliary medicine system - Google Patents
A kind of auxiliary medicine decision-making method and intelligent auxiliary medicine system Download PDFInfo
- Publication number
- CN109859815A CN109859815A CN201910067659.1A CN201910067659A CN109859815A CN 109859815 A CN109859815 A CN 109859815A CN 201910067659 A CN201910067659 A CN 201910067659A CN 109859815 A CN109859815 A CN 109859815A
- Authority
- CN
- China
- Prior art keywords
- data
- drug
- patient
- medication
- decision
- 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
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
本发明公开了一种辅助用药决策方法及智能辅助用药系统，包括以下步骤：步骤1，采集获取患者医疗数据，将其进行预处理，获得结构统一的患者标准临床数据；步骤2，根据步骤1获得的患者标准临床数据，通过分类回归树算法对患者适用药物进行预测，结合药品数据库内药品信息，自动生成用药提议，完成辅助用药决策；用药提议用于提供给医师审核使用。本发明可智能化的提供具有科学性、专业性与精确性的用药依据，能够辅助临床医生及临床药师制定药物最佳使用决策，从而进一步规范药物使用。
The invention discloses an auxiliary medication decision-making method and an intelligent auxiliary medication system, comprising the following steps: step 1, collecting and obtaining patient medical data, preprocessing the same, and obtaining patient standard clinical data with a unified structure; step 2, according to step 1 The obtained standard clinical data of patients is used to predict the suitable drugs for patients through the classification and regression tree algorithm, and combined with the drug information in the drug database, the drug recommendations are automatically generated to complete the auxiliary drug decision-making; the drug recommendations are provided to the physicians for review and use. The present invention can intelligently provide scientific, professional and accurate medication basis, and can assist clinicians and clinical pharmacists to make optimal drug use decisions, so as to further standardize drug use.
Description
Technical field
The invention belongs to medical domain and technical field of data processing, in particular to a kind of adjuvant drug decision-making technique and intelligence
It can adjuvant drug system.
Background technique
In current therapeutic regimen formulation process, it is the medication suggestion provided according to package insert first, is manually looked into
Patient medical data is ask, then formulates initial therapeutic regimen according to experience and drug operation instructions by doctor or pharmacist, and
It is adjusted optimization over the course for the treatment of.Current method has the skill of apparent hysteresis quality, adjustment effect and doctor or pharmacist
Art level is related, the influence vulnerable to human factor.
In addition, infectious diseases disease incidence height, the course of disease are fast in patient with severe symptoms, antimicrobial DP finish pharmacokinetic parameter is because of weight
Disease patient sign difference is to generate greatly variation.For how effectively to formulate optimal antibacterials dosage regimen, mention
The problem of high antibacterial Medication safety and validity, at present more effective patient with severe symptoms's infectious diseases dosage regimen system
The method of determining include: be that the medication suggestion provided according to package insert is treated first, pass through multiple blood over the course for the treatment of
Concentration sample detecting adjusts dosage.According to the dosage regimen that above-mentioned existing method is formulated, artificial enquiry patient is needed
Medical data, then initial therapeutic regimen is formulated according to experience and drug operation instructions by doctor or pharmacist, and treating
It is adjusted optimization in the process.The method has apparent hysteresis quality, and adjustment effect is related with the technical level of doctor or pharmacist,
It is affected by human factors seriously, causes anti-infective medication to be unable to reach desired effect, and then patient with severe symptoms's treatment of infection can be made
Failure rate is high, patient survival is low.
In conclusion needing a kind of novel adjuvant drug decision-making technique.
Summary of the invention
It is above-mentioned to solve the purpose of the present invention is to provide a kind of adjuvant drug decision-making technique and intelligent adjuvant drug system
?.The present invention, which intelligentized can provide, has scientific, the professional medication foundation with accuracy, Neng Goufu
Clinician and clinical pharmacist is helped to formulate drug optimal use decision, so that further specification drug uses.
In order to achieve the above objectives, the invention adopts the following technical scheme:
A kind of adjuvant drug decision-making technique, comprising the following steps:
Step 1, acquisition obtains patient medical data, is pre-processed, and it is clinical to obtain the unified patient criteria of structure
Data；
Step 2, the patient criteria's clinical data obtained according to step 1, is applicable in medicine to patient by post-class processing algorithm
Object is predicted, in conjunction with medicine information in drug data bank, is automatically generated medication proposal, is realized adjuvant drug decision；Medication mentions
View audits use for being supplied to doctor.
It further, further include step 3 when drug used is antimicrobial；
Step 3, antimicrobial population pharmacokinetics used are calculated, population pharmacokinetics model is established；According to
Patient criteria's clinical data that step 1 obtains is modified population pharmacokinetics model；It calculates and obtains individual medicine for power
Parameter is learned, individual pharmacokinetics model is constructed；By individual pharmacokinetics model, the blood concentration for predicting patient becomes
Change, the blood concentration variation obtained according to prediction generates medication and proposes.
Further, further includes: step 4, sample detecting is carried out to patient's blood concentration in the given time；If detected value
Meet predicted value, then former dosage is maintained to carry out medication, if detected value and predicted value difference are more than threshold value, more new individual medicine generation is dynamic
Mechanical model parameter, recalculates dosage, and the medication after obtaining optimization is proposed.
Further, in step 1, by data warehouse technology directly from the different business systems number of hospital or medical institutions
Patient medical data is obtained according to extracting in library, data warehouse is arrived in storage after converting by data cleansing；Patient medical data includes:
The diagnosis information of patient, individual sign information, clinical examination and result of laboratory test, diagnosis report and treatment and medication history.
Further, step 1 specifically includes: step 1.1, operation system data interconnection: to including all kinds of medical numbers of patient
According to e commerce transactions system establish Open Database Connection, realize the data sharing between heterogeneous database；
Step 1.2, data pick-up: data extraction tool is used, the data of needs is extracted in the database, passes through TCP/IP
Data delivery to target side is carried out parsing reduction by agreement；
Step 1.3, data cleansing: the data being drawn into are pre-processed, so that data format meets preset use
It is required that；
Step 1.4, data normalization: by step 1.3, treated that data format, and obtains the unified trouble of structure
Person's standard clinical data.
Further, step 1.3 specifically includes:
Step 1.3.1 the characteristics of according to data are extracted, carries out data analysis, determines data type, data scale and each
The distribution situation of data under attributive character；
Step 1.3.2 finds the missing values in simultaneously completion data；
Step 1.3.3 judges the exceptional value in garbled data.
Further, step 2 specifically includes:
Step 2.1, drug data bank is constructed according to drug operation instructions；
Step 2.2, the drug data in drug data bank is grouped repeatedly using post-class processing algorithm, recursive generation
Several child nodes stop decision tree growth when data set is inseparable, obtain drug decision-tree model；Wherein each leaf section
Point represents a kind of conclusion, and each non-leaf nodes represents a kind of attributive character；
Step 2.3, patient criteria's clinical data that step 1 obtains is mapped to fixed length N-dimensional patient characteristic vector, dimension N
Parameter or attribute type included in representation vector；
Step 2.4, in the drug decision-tree model patient characteristic vector input step 2.2 that step 2.3 obtains constructed,
Is generated by medication and is proposed by patient classification by judging to complete item by item.
Further, step 3 specifically includes:
Step 3.1, using nonlinear mixed-effect model, antimicrobial population pharmacokinetics used are calculated, and
Establish population pharmacokinetics model；
Step 3.2, population pharmacokinetics step 3.1 established using patient criteria's clinical data that step 1 obtains
Model is modified；By Bayesian feedback method, individual pharmacokinetics parameter is obtained, building obtains individual pharmacokinetics mould
Type；By individual pharmacokinetics model, predicts blood concentration in patient's predetermined time and change, become according to blood concentration
Metaplasia is proposed at medication.
A kind of intelligence adjuvant drug system, comprising:
Reading data and preprocessing module obtain patient medical data for acquiring, are pre-processed, obtain structure
Unified patient criteria's clinical data；
Drug intelligent Matching module, patient criteria's clinical data for being obtained according to reading data and preprocessing module,
Drug is applicable in patient by post-class processing algorithm to predict, in conjunction with medicine information in drug data bank, automatically generates use
Medicine is proposed, adjuvant drug decision is completed；Medication is proposed to audit use for being supplied to doctor.
Further, when drug used is antimicrobial, further includes: personalized medicine module, for calculating antibacterial used
Drug population pharmacokinetics establish population pharmacokinetics model；It is obtained according to reading data and preprocessing module
Patient criteria's clinical data is modified population pharmacokinetics model；It calculates and obtains individual pharmacokinetics parameter, building
Individual pharmacokinetics model；By individual pharmacokinetics model, the blood concentration variation of patient is predicted, is obtained according to prediction
The blood concentration variation obtained generates medication and proposes.
Compared with prior art, the invention has the following advantages:
In the present invention, according to drug using standard, patient characteristic parameter, population pharmacokinetics, bass leaf feedback transmitter and
Post-class processing etc. establishes drug medication and medication Optimized model, intelligentized can provide have it is scientific, professional with it is smart
The medication foundation of true property formulates antibacterials optimal treatment decision with adjuvant clinical doctor and clinical pharmacist, to improve drug
Safety, validity and economy in use.
Further, the present invention provides the antimicrobial DP finish intelligence of a kind of combination computer field and pharmaceutical field the relevant technologies
Energy adjuvant drug system, by the extraction of patients clinical data, drug matching and personalized medicine solution formulation can be quick, quasi-
It really formulates and meets the therapeutic regimen of patient clinical parameter, to improve safety of the antimicrobial DP finish in anti-infective therapy, effectively
Property and economy, patient with severe symptoms's infectious age and cause of disease Bacteria drug tolorance rate can be reduced, specification antimicrobial DP finish uses.
Detailed description of the invention
Fig. 1 is a kind of schematic process flow diagram of antimicrobial DP finish adjuvant drug decision-making technique of the embodiment of the present invention；
Fig. 2 is the drug decision-tree model schematic diagram constructed in the embodiment of the present invention；
Fig. 3 is a kind of structural schematic block diagram of intelligent adjuvant drug system of the embodiment of the present invention.
Specific embodiment
Invention is further described in detail in the following with reference to the drawings and specific embodiments.
Referring to Fig. 1, a kind of antimicrobial DP finish adjuvant drug decision-making technique of the invention, specifically includes the following steps:
Step 1, reading data and pretreatment: by ETL (Extract-Transform-Load, data warehouse technology),
Patient medical data directly is extracted from the different business systems database of hospital or medical institutions, after converting by data cleansing
Store data warehouse.Patient medical data includes that diagnosis information, individual sign information, clinical examination and the chemical examination of patient is tied
Fruit, diagnosis report and treatment and medication history.
In step 1, reading data includes: with pretreatment concrete methods of realizing
(1) operation system data interconnect: under general feelings, patient medical data can be stored in different electronic systems, and each
A system cannot achieve effective data sharing between system by specific storage format and reading manner, therefore be unable to complete pair
The unified access and extraction of the complete medical data of patient.ODBC is established to the electronic system comprising all kinds of medical datas of patient (to open
Put Database Connectivity), it is only necessary to partial software upgrading is carried out on the hardware foundation of original electronic system, is provided for database access
Unified interface, thus the data sharing between reaching heterogeneous database.It is low in cost in this way, can directly with existing electricity
Subsystem docking, and operational effect is stablized.
(2) data pick-up: data extraction tool is used, the data of needs is extracted in the database, passes through ICP/IP protocol
By data delivery to target side, parsing reduction is carried out.
(3) data cleansing: in order to make the data format being drawn into meet the basic demand that further uses, need to its into
Row pretreatment.
Data cleansing specific steps include:
Step1: according to data characteristics are extracted, data analysis is carried out, determines data type, data scale and each attributive character
The distribution situation of lower data.
Step2: missing values in data are found.Diagnosis information and individual sign information for patient, can be patrolled by establishing
Regression model is collected, missing data is predicted, obtains missing values in data.For patient clinical inspection and result of laboratory test, examine
Disconnected report, treatment and the shortage of data in medication history, need to be by artificial completion.
Step3: judge exceptional value in data.For parameters and medication history in patient clinical inspection and result of laboratory test
In dosage data establish model, filter out data value that model cannot be fitted as exceptional value by manual examination and verification.
(4) standardization of data: formatting the data extracted from different business systems, and such as unified metering is single
Position, uniform data form of expression etc., to obtain the unified normal data of structure.
Through the above steps, it ultimately generates and meets requirement and meet the structuring of Uniform Access, standardized patient doctor
Treat data.
Step 2, the normal data unified according to the structure of acquisition, by post-class processing algorithm to patient be suitble to drug into
Row prediction automatically generates several medications and proposes in conjunction with medicine information in drug data bank, including antimicrobial DP finish type, medication
Metering and drug side effect prompt, then audited by clinician and pharmacist.Wherein, had using post-class processing algorithm
Advantage includes: that construction method is simple, and the speed of service is very fast, and prediction result is more accurate.
Referring to Fig. 2, step 2 specifically includes the following steps:
(1) it constructs anti-microbial type drug data bank: constructing structured database, major parameter packet according to drug operation instructions
It includes indication, adapt to patients, different sexes and the reference of age dosage, medication taboo and adverse drug reflection.
(2) drug decision-tree model generates: using the drug data in drug data bank, passing through CART (post-class processing)
Algorithm establishes antibacterial similar drug decision tree, as shown in Figure 2.The concrete mode that drug decision-tree model generates are as follows: calculated using CART
Method is grouped drug data repeatedly, recursive to generate several child nodes, stops decision tree growth when data set is inseparable.Its
In each leaf node represent a kind of conclusion, each non-leaf nodes represents a kind of attributive character.
(3) it constructs patient parameter feature vector: the normal data obtained by module one being mapped to using word2vec and is determined
Long N-dimensional vector, parameter or attribute type included in dimension N representation vector, value represent the occurrence of certain attribute.
(4) medication matches: patient characteristic vector being inputted in drug decision-tree model, by judging to complete item by item to patient
Classification, obtains suitable drug and exports relevant information, generates medication and proposes.
Step 3, using patient clinical physical sign parameters and antimicrobial population pharmacokinetics model has been chosen, has used Bayes
Feedback algorithm obtains accurate individual pharmacokinetics model, makes personalized therapeutic regimen.Meanwhile passing through blood concentration
Monitor value adjusts therapeutic regimen in real time, realizes the accurate medication of the individuation of different patients.
Step 3 specifically includes the following steps:
(1) population pharmacokinetics are calculated: using nonlinear mixed-effect model, calculate antimicrobial group used
Pharmacokinetic parameter, and establish population pharmacokinetics model.
(2) individual pharmacokinetics parameter is calculated: using the patient criteria's clinical data obtained to population pharmacokinetics
Model is modified, and by Bayesian feedback method, is obtained individual pharmacokinetics parameter, is constructed individual pharmacokinetics model；
By individual pharmacokinetics model, predicts blood concentration in patient's certain time and change, change to obtain according to blood concentration
Accurately initial medication is proposed.
(3) therapeutic drug monitoring and dosage are corrected: sample detecting is carried out to patient's blood concentration within a certain period of time,
If meeting predicted value, former dosage is maintained to be treated, if differing greatly with predicted value, by updating individual pharmacokinetics
Model parameter recalculates dosage, and the medication after obtaining optimization is proposed, to reach safely and effectively blood concentration.
To sum up, a kind of artificial intelligence antimicrobial DP finish adjuvant drug system advantage of the invention is: it is directed to general patient,
By the present invention in that being matched with standard clinical data of the decision Tree algorithms to patient with drug storage, can fast and accurately mention
Propose to refer to for doctor or pharmacist for medication, optimizes medical experience while guaranteeing patient's correctly taking drugs.For needing antibacterial
The patient with severe symptoms for the treatment of when personalized medicine module used in the present invention can be according to patient criteria's clinical data, establishes quasi-
True individual pharmacokinetics model, adjuvant clinical doctor and pharmacist fast and accurately formulate initial therapeutic regimen, and realization is directed to
The personalized administration services of different patients.The present invention can be detected by blood concentration simultaneously, feedback drug effect promptly and accurately
Fruit, and quickly complete the optimization of dosage.The accurate calculating of antimicrobial DP finish dosage is helped to reduce pathogen
Resistant rate improves medication validity, reduces mortality, so that the medication effect and medication peace of antimicrobial DP finish be effectively ensured
Entirely.For clinician and clinical pharmacist, intelligence medication auxiliary system proposed by the present invention combines patient clinical data and uses
Medicine knowledge can be eliminated substantially and bring medication inaccuracy, dosage inaccurate by clinical experience, mistake, used being promoted
It ensure that therapeutic effect while medicine quality.
The present invention is based on web systems to build, and code and data are hosted in SQL Server remote server, data storage
In distributed system.User is in use, only need conventional browser that relevant operation can be completed, and without installation, other are subsidiary soft
Part.Meanwhile the platform is not limited by operating system, can be used for a variety of operating platforms such as Windows, Mac and Linux.In addition,
It is calculated when system-computed dosage regimen by server GPU, platform property is avoided to be limited by user terminal calculated performance.
Referring to Fig. 3, with reference to the accompanying drawing with clinical application specific steps to a kind of antimicrobial DP finish of the present invention
Artificial intelligence adjuvant drug system.A kind of antimicrobial DP finish adjuvant drug system of the invention.This system is mainly by three modules
It constitutes, is described in detail as follows:
Module one, reading data and preprocessing module: the module mainly realizes following functions: the data of calibration-based hearing loss evaluation
It extracts, pretreatment.By reading data and preprocessing module read patient in hospital or medical institutions it is medical, check, chemical examination,
The medical informations such as diagnosis and sign.By data mining ETL technology, misspelling because manual operation factor generates, no is removed
Legitimate value, null value and repetition indicate, generate structuring, standardized medical data.
Module two, antimicrobial DP finish intelligent Matching module: the module mainly realizes following functions: patient information and anti-microbial type
Drug matching.By the age of patient, gender, weight, infection site, symptom, result of laboratory test (hepatic and renal function, blood etc.), allergy
History, bacteria cultivation results etc., by inquiry antibacterial similar drug knowledge base carry out similarity mode, search out match it is several
Antimicrobial DP finish simultaneously provides suggestion dosage, and therapeutic regimen is suggested in output.It is recommended that therapeutic regimen by clinician or pharmacist into
Row audit passes through, or generates initial administration scheme after carrying out manual change, can be used for patient anti-infective therapy.
Module three, personalized medicine module: the module mainly realizes following functions: passing through patient clinical parameter and antimicrobial
Object population pharmacokinetics model foundation model, calculates patient's blood concentration in certain medication cycle, and predict and obtain
The target value for obtaining optimum therapeuticing effect, as doctor and pharmacist's audit, the auxiliary information for correcting initial therapeutic regimen.According to patient
Blood concentration measured value corrects individual pharmacokinetics parameter, recalculates dosage, it is ensured that medication effect.
Specific step is as follows in clinical application for system of the invention:
The first step, patient information extract.Clinician or pharmacist use personal workstation's proposition antimicrobial DP finish prescription
Case formulates request, by reading data and preprocessing module, obtains treated clinical characteristic data and is audited,
Manual amended record missing information.
Second step, medication matching.The first step obtains antibacterial in complete patient data and antimicrobial DP finish intelligent Matching module
Class drug medication library is matched by post-class processing algorithm, is obtained and clinical characteristic drug the most matched, standard
Dosage, and drug side-effect is prompted.For general patient, by generating initial use after clinician or pharmacist's audit
Prescription case.For patient with severe symptoms, goes to third step and carry out personalized medicine solution formulation.
Third step calls drug pharmacokinetic model by the determined types of medicines of antimicrobial DP finish intelligent Matching module,
Accurate individual pharmacokinetics parameter is calculated, to the blood concentration desired value of optimum therapeuticing effect according to clinical characteristic parameter
With initial administration scheme, manual examination and verification are carried out by clinician or pharmacist, if there is particular/special requirement conscientious can modify parameter simultaneously by hand
It recalculates.
4th step is compared with prediction result by blood concentration testing result, is counted automatically by personalized medicine module
Dosage is calculated and corrected, expected blood concentration is finally reached.Meanwhile by calculating, patient's blood concentration trend is carried out quasi-
Really prediction.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The above embodiments are merely illustrative of the technical scheme of the present invention and are not intended to be limiting thereof, although referring to above-described embodiment pair
The present invention is described in detail, those of ordinary skill in the art still can to a specific embodiment of the invention into
Row modification perhaps equivalent replacement these without departing from any modification of spirit and scope of the invention or equivalent replacement, applying
Within pending claims of the invention.
Claims (10)
1. a kind of adjuvant drug decision-making technique, which comprises the following steps:
Step 1, acquisition obtains patient medical data, is pre-processed, and the unified patient criteria's clinical data of structure is obtained；
Step 2, according to step 1 obtain patient criteria's clinical data, by post-class processing algorithm to patient be applicable in drug into
Row prediction automatically generates medication proposal, realizes adjuvant drug decision in conjunction with medicine information in drug data bank；Medication is proposed to use
Use is audited in being supplied to doctor.
2. a kind of adjuvant drug decision-making technique according to claim 1, which is characterized in that drug used is antimicrobial
When, it further include step 3；
Step 3, antimicrobial population pharmacokinetics used are calculated, population pharmacokinetics model is established；According to step 1
Patient criteria's clinical data of acquisition is modified population pharmacokinetics model；It calculates and obtains individual pharmacokinetics ginseng
Number constructs individual pharmacokinetics model；By individual pharmacokinetics model, the blood concentration variation of patient, root are predicted
It is predicted that the blood concentration variation obtained generates medication and proposes.
3. a kind of adjuvant drug decision-making technique according to claim 2, which is characterized in that further include:
Step 4, sample detecting is carried out to patient's blood concentration in the given time；If detected value meets predicted value, remain former
Dosage carries out medication, if detected value and predicted value difference are more than threshold value, update individual pharmacokinetics model parameter, counts again
Dosage is calculated, the medication after obtaining optimization is proposed.
4. a kind of adjuvant drug decision-making technique according to claim 1, which is characterized in that in step 1, pass through data warehouse
Technology extracts directly from the different business systems database of hospital or medical institutions and obtains patient medical data, clear by data
Data warehouse is arrived in storage after washing conversion；Patient medical data includes: the diagnosis information of patient, individual sign information, clinical examination
With result of laboratory test, diagnosis report and treatment and medication history.
5. a kind of adjuvant drug decision-making technique according to claim 1, which is characterized in that step 1 specifically includes:
Step 1.1, operation system data interconnect: establishing open data to the e commerce transactions system comprising all kinds of medical datas of patient
Library interconnection, realizes the data sharing between heterogeneous database；
Step 1.2, data pick-up: data extraction tool is used, the data of needs is extracted in the database, passes through ICP/IP protocol
By data delivery to target side, parsing reduction is carried out；
Step 1.3, data cleansing: the data being drawn into are pre-processed, so that data format meets preset requirement；
Step 1.4, data normalization: by step 1.3, treated that data format, and obtains the unified patient's mark of structure
Quasi- clinical data.
6. a kind of adjuvant drug decision-making technique according to claim 5, which is characterized in that step 1.3 specifically includes:
Step 1.3.1, according to extract data the characteristics of, carry out data analysis, determine data type, data scale and each attribute
The distribution situation of data under feature；
Step 1.3.2 finds the missing values in simultaneously completion data；
Step 1.3.3 judges the exceptional value in garbled data.
7. a kind of adjuvant drug decision-making technique according to claim 1, which is characterized in that step 2 specifically includes:
Step 2.1, drug data bank is constructed according to drug operation instructions；
Step 2.2, the drug data in drug data bank is grouped repeatedly using post-class processing algorithm, recursive generation is several
Child node stops decision tree growth when data set is inseparable, obtains drug decision-tree model；Wherein each leaf node generation
A kind of conclusion of table, each non-leaf nodes represent a kind of attributive character；
Step 2.3, patient criteria's clinical data that step 1 obtains is mapped to fixed length N-dimensional patient characteristic vector, dimension N is represented
Parameter or attribute type included in vector；
Step 2.4, in the drug decision-tree model patient characteristic vector input step 2.2 that step 2.3 obtains constructed, pass through
Judgement completes that patient classification is generated medication and proposed item by item.
8. a kind of adjuvant drug decision-making technique according to claim 2, which is characterized in that step 3 specifically includes:
Step 3.1, using nonlinear mixed-effect model, antimicrobial population pharmacokinetics used are calculated, and are established
Population pharmacokinetics model；
Step 3.2, the population pharmacokinetics model that the patient criteria's clinical data obtained using step 1 establishes step 3.1
It is modified；By Bayesian feedback method, individual pharmacokinetics parameter is obtained, building obtains individual pharmacokinetics model；
By individual pharmacokinetics model, predicts blood concentration in patient's predetermined time and change, changed according to blood concentration
Medication is generated to propose.
9. a kind of intelligence adjuvant drug system characterized by comprising
Reading data and preprocessing module obtain patient medical data for acquiring, are pre-processed, it is unified to obtain structure
Patient criteria's clinical data；
Drug intelligent Matching module passes through for patient criteria's clinical data according to reading data and preprocessing module acquisition
Post-class processing algorithm is applicable in drug to patient and predicts, in conjunction with medicine information in drug data bank, automatically generates medication and mentions
View completes adjuvant drug decision；Medication is proposed to audit use for being supplied to doctor.
10. a kind of intelligent adjuvant drug system according to claim 9, which is characterized in that drug used is antimicrobial
When, further includes:
Personalized medicine module establishes population pharmacokinetics for calculating antimicrobial population pharmacokinetics used
Model；Population pharmacokinetics model is repaired according to patient criteria's clinical data that reading data and preprocessing module obtain
Just；It calculates and obtains individual pharmacokinetics parameter, construct individual pharmacokinetics model；By individual pharmacokinetics model,
The blood concentration variation for predicting patient, the blood concentration variation obtained according to prediction generate medication and propose.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910067659.1A CN109859815A (en) | 2019-01-24 | 2019-01-24 | A kind of auxiliary medicine decision-making method and intelligent auxiliary medicine system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910067659.1A CN109859815A (en) | 2019-01-24 | 2019-01-24 | A kind of auxiliary medicine decision-making method and intelligent auxiliary medicine system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109859815A true CN109859815A (en) | 2019-06-07 |
Family
ID=66895884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910067659.1A Pending CN109859815A (en) | 2019-01-24 | 2019-01-24 | A kind of auxiliary medicine decision-making method and intelligent auxiliary medicine system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859815A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322946A (en) * | 2019-07-11 | 2019-10-11 | 河南大学 | Optimal medication granularity calculation method based on multi-granularity decision model |
| CN110444288A (en) * | 2019-07-24 | 2019-11-12 | 卓尔智联(武汉)研究院有限公司 | Auxiliary diagnostic equipment, method and readable storage medium storing program for executing based on complex network |
| CN110827951A (en) * | 2019-12-24 | 2020-02-21 | 陕西省中医医院 | Clinical intelligent decision platform |
| CN111768830A (en) * | 2020-05-27 | 2020-10-13 | 医利捷(上海)信息科技有限公司 | Cloud computing-based personalized medicine administration management system and method |
| CN112102962A (en) * | 2020-09-10 | 2020-12-18 | 兰州市第二人民医院 | Method and system for clinical intervention and effect prognosis data analysis of adjuvant medication |
| CN112259251A (en) * | 2020-10-20 | 2021-01-22 | 昭衍(北京)医药科技有限公司 | Method and system for simplifying and analyzing pharmacokinetic parameters |
| CN112988850A (en) * | 2021-04-27 | 2021-06-18 | 明品云(北京)数据科技有限公司 | Article information analysis management method, system, equipment and medium |
| CN113469203A (en) * | 2020-03-31 | 2021-10-01 | 日本电气株式会社 | Method, electronic device and computer program product for evaluating operation results |
| CN113889221A (en) * | 2021-12-08 | 2022-01-04 | 首都医科大学宣武医院 | System for accurately selecting treatment dosage of tacrolimus of myasthenia gravis patient and application of system |
| CN113936810A (en) * | 2021-08-26 | 2022-01-14 | 上海交通大学医学院附属瑞金医院 | Method and system for monitoring blood concentration of antibiotics |
| CN113971997A (en) * | 2021-10-12 | 2022-01-25 | 浙江太美医疗科技股份有限公司 | Method and system for determining dispensing doses in clinical trials |
| 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 |
| CN115083622A (en) * | 2022-06-14 | 2022-09-20 | 西藏自治区人民医院 | Regional related medication guidance system |
| CN115547511A (en) * | 2021-06-29 | 2022-12-30 | 广州静远医药研究有限公司 | Individual auxiliary administration strategy making method, system and storage medium |
| CN115602284A (en) * | 2022-10-18 | 2023-01-13 | 浙江海心智惠科技有限公司（Cn） | Medication reminder method, system, device, equipment and storage medium |
| CN116779096A (en) * | 2023-06-28 | 2023-09-19 | 南栖仙策(南京)高新技术有限公司 | Medication policy determination method, device, equipment and storage medium |
| CN116936025A (en) * | 2023-07-19 | 2023-10-24 | 上海市第七人民医院(上海中医药大学附属第七人民医院) | A medication monitoring system for clinical treatment of cardiovascular diseases |
| CN118039061A (en) * | 2024-03-11 | 2024-05-14 | 暨南大学附属第一医院（广州华侨医院） | A hypoglycemic drug treatment auxiliary guidance system for mild to moderate diabetes |
| CN118262858A (en) * | 2022-12-28 | 2024-06-28 | 首都医科大学附属北京安定医院 | Method, device, storage medium and electronic device for recommending drug dosage |
| CN118841128A (en) * | 2024-06-21 | 2024-10-25 | 首都医科大学宣武医院 | Intelligent interactive perioperative safe medication auxiliary decision-making system and method |
| CN119207704A (en) * | 2024-11-27 | 2024-12-27 | 山东高呈网络科技有限公司 | A drug management method and system for the elderly |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102004863A (en) * | 2010-12-15 | 2011-04-06 | 中国人民解放军第四军医大学 | Clinical safe and reasonable administration decision supporting method |
| US20140288956A1 (en) * | 2004-09-10 | 2014-09-25 | Ldm Group, Llc | Systems and Methods for Providing an Inducement of a Purchase in Conjunction with a Prescription |
| CN104584021A (en) * | 2014-11-07 | 2015-04-29 | 曹庆恒 | A method, system and device for intelligent drug selection |
| CN105468900A (en) * | 2015-11-20 | 2016-04-06 | 邹远强 | Intelligent medical record input platform based on knowledge base |
| CN105868531A (en) * | 2016-03-23 | 2016-08-17 | 遵义医学院 | Medical big data based automatic auxiliary prescribing system and method |
| CN105956382A (en) * | 2016-04-26 | 2016-09-21 | 北京工商大学 | Traditional Chinese medicine constitution optimized classification method based on improved CART decision-making tree and fuzzy naive Bayes combined model |
| CN107341345A (en) * | 2017-06-22 | 2017-11-10 | 厦门大学 | A kind of clinical rational drug use methods of risk assessment based on electronic health record big data |
| CN107845411A (en) * | 2017-12-04 | 2018-03-27 | 青岛大学附属医院 | Clinical medication decision support system |
| CN109215756A (en) * | 2018-09-25 | 2019-01-15 | 福建中医药大学 | A kind of medicine assisting in diagnosis and treatment system |
-
2019
- 2019-01-24 CN CN201910067659.1A patent/CN109859815A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140288956A1 (en) * | 2004-09-10 | 2014-09-25 | Ldm Group, Llc | Systems and Methods for Providing an Inducement of a Purchase in Conjunction with a Prescription |
| CN102004863A (en) * | 2010-12-15 | 2011-04-06 | 中国人民解放军第四军医大学 | Clinical safe and reasonable administration decision supporting method |
| CN104584021A (en) * | 2014-11-07 | 2015-04-29 | 曹庆恒 | A method, system and device for intelligent drug selection |
| CN105468900A (en) * | 2015-11-20 | 2016-04-06 | 邹远强 | Intelligent medical record input platform based on knowledge base |
| CN105868531A (en) * | 2016-03-23 | 2016-08-17 | 遵义医学院 | Medical big data based automatic auxiliary prescribing system and method |
| CN105956382A (en) * | 2016-04-26 | 2016-09-21 | 北京工商大学 | Traditional Chinese medicine constitution optimized classification method based on improved CART decision-making tree and fuzzy naive Bayes combined model |
| CN107341345A (en) * | 2017-06-22 | 2017-11-10 | 厦门大学 | A kind of clinical rational drug use methods of risk assessment based on electronic health record big data |
| CN107845411A (en) * | 2017-12-04 | 2018-03-27 | 青岛大学附属医院 | Clinical medication decision support system |
| CN109215756A (en) * | 2018-09-25 | 2019-01-15 | 福建中医药大学 | A kind of medicine assisting in diagnosis and treatment system |
Non-Patent Citations (4)
| Title |
|---|
| 张志军: "《大数据技术在高校中的应用研究》", 30 September 2017, 北京：北京邮电大学出版社 * |
| 郭涛: "《新编药物动力学》", 31 January 2005, 北京：中国科学技术出版社 * |
| 陈翔等: "数据仓库技术在糖尿病健康教育中的应用研究", 《医学信息》 * |
| 陈鹏展: "《个体行为的机器识别与决策协同》", 31 July 2018, 北京：知识产权出版社 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322946B (en) * | 2019-07-11 | 2022-09-16 | 河南大学 | A computing device for optimal medication granularity based on multi-granularity decision-making model |
| CN110322946A (en) * | 2019-07-11 | 2019-10-11 | 河南大学 | Optimal medication granularity calculation method based on multi-granularity decision model |
| CN110444288A (en) * | 2019-07-24 | 2019-11-12 | 卓尔智联(武汉)研究院有限公司 | Auxiliary diagnostic equipment, method and readable storage medium storing program for executing based on complex network |
| CN110827951A (en) * | 2019-12-24 | 2020-02-21 | 陕西省中医医院 | Clinical intelligent decision platform |
| CN113469203A (en) * | 2020-03-31 | 2021-10-01 | 日本电气株式会社 | Method, electronic device and computer program product for evaluating operation results |
| CN113469203B (en) * | 2020-03-31 | 2026-02-03 | 日本电气株式会社 | Method, electronic device and computer program product for evaluating operation result |
| CN111768830A (en) * | 2020-05-27 | 2020-10-13 | 医利捷(上海)信息科技有限公司 | Cloud computing-based personalized medicine administration management system and method |
| CN112102962A (en) * | 2020-09-10 | 2020-12-18 | 兰州市第二人民医院 | Method and system for clinical intervention and effect prognosis data analysis of adjuvant medication |
| CN112259251B (en) * | 2020-10-20 | 2021-07-13 | 昭衍(北京)医药科技有限公司 | Method and system for simplifying and analyzing pharmacokinetic parameters |
| CN112259251A (en) * | 2020-10-20 | 2021-01-22 | 昭衍(北京)医药科技有限公司 | Method and system for simplifying and analyzing pharmacokinetic parameters |
| CN112988850A (en) * | 2021-04-27 | 2021-06-18 | 明品云(北京)数据科技有限公司 | Article information analysis management method, system, equipment and medium |
| CN115547511A (en) * | 2021-06-29 | 2022-12-30 | 广州静远医药研究有限公司 | Individual auxiliary administration strategy making method, system and storage medium |
| CN113936810A (en) * | 2021-08-26 | 2022-01-14 | 上海交通大学医学院附属瑞金医院 | Method and system for monitoring blood concentration of antibiotics |
| CN113971997A (en) * | 2021-10-12 | 2022-01-25 | 浙江太美医疗科技股份有限公司 | Method and system for determining dispensing doses in clinical trials |
| 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 |
| CN113889221A (en) * | 2021-12-08 | 2022-01-04 | 首都医科大学宣武医院 | System for accurately selecting treatment dosage of tacrolimus of myasthenia gravis patient and application of system |
| CN115083622A (en) * | 2022-06-14 | 2022-09-20 | 西藏自治区人民医院 | Regional related medication guidance system |
| CN115602284A (en) * | 2022-10-18 | 2023-01-13 | 浙江海心智惠科技有限公司（Cn） | Medication reminder method, system, device, equipment and storage medium |
| CN118262858A (en) * | 2022-12-28 | 2024-06-28 | 首都医科大学附属北京安定医院 | Method, device, storage medium and electronic device for recommending drug dosage |
| CN116779096A (en) * | 2023-06-28 | 2023-09-19 | 南栖仙策(南京)高新技术有限公司 | Medication policy determination method, device, equipment and storage medium |
| CN116779096B (en) * | 2023-06-28 | 2024-04-16 | 南栖仙策(南京)高新技术有限公司 | Medication policy determination method, device, equipment and storage medium |
| CN116936025A (en) * | 2023-07-19 | 2023-10-24 | 上海市第七人民医院(上海中医药大学附属第七人民医院) | A medication monitoring system for clinical treatment of cardiovascular diseases |
| CN118039061A (en) * | 2024-03-11 | 2024-05-14 | 暨南大学附属第一医院（广州华侨医院） | A hypoglycemic drug treatment auxiliary guidance system for mild to moderate diabetes |
| CN118841128A (en) * | 2024-06-21 | 2024-10-25 | 首都医科大学宣武医院 | Intelligent interactive perioperative safe medication auxiliary decision-making system and method |
| CN118841128B (en) * | 2024-06-21 | 2025-04-18 | 首都医科大学宣武医院 | Intelligent interactive perioperative safe medication decision-making support system and method |
| CN119207704A (en) * | 2024-11-27 | 2024-12-27 | 山东高呈网络科技有限公司 | A drug management method and system for the elderly |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109859815A (en) | A kind of auxiliary medicine decision-making method and intelligent auxiliary medicine system | |
| CN111986770B (en) | Prescription medication auditing method, device, equipment and storage medium | |
| CN111613289B (en) | Individualized drug dose prediction method, device, electronic equipment and storage medium | |
| CN111009322A (en) | Perioperative risk assessment and clinical decision-making intelligent assistance system | |
| CN104766259A (en) | Medical clinical quality monitoring and evaluation system based on single-disease model | |
| CN110136838A (en) | Data Matching decision-making technique and system based on multiple knowledge base reasoning | |
| WO2021151327A1 (en) | Triage data processing method and apparatus, and device and medium | |
| CN109805898A (en) | Prediction method of critically ill death based on attention mechanism time series convolutional network algorithm | |
| CN112530602B (en) | Drug side effect analysis method, device, electronic device and computer storage medium | |
| CN109243620A (en) | Drug effect optimization method and device based on therapeutic drug monitoring | |
| CN109243546B (en) | Method and device for establishing anticoagulant drug effect optimization model | |
| WO2017017554A1 (en) | Reliability measurement in data analysis of altered data sets | |
| CN113140281B (en) | A management method and management system for chemotherapy drugs for patients with hematological tumors | |
| CN109524083A (en) | Based on the rational method of medical big data monitoring adjuvant drug and Related product | |
| CN111612278A (en) | Life state prediction method, device, electronic device and storage medium | |
| CN112967803A (en) | Early mortality prediction method and system for emergency patients based on integrated model | |
| CN119479982A (en) | A method and system for intelligent processing of hospital pharmacy dispensing information | |
| CN118053540A (en) | Intelligent prescriptions prepositive auditing system | |
| CN112071431B (en) | Clinical path automatic generation method and system based on deep learning and knowledge graph | |
| CN118629633B (en) | A method and system for assisting decision making based on medical big data | |
| CN114220541A (en) | Disease prediction method, device, electronic device and storage medium | |
| CN113517044A (en) | Clinical data processing method and system for evaluating citicoline based on pharmacokinetics | |
| CN120089272A (en) | A cue-based large language model drug recommendation method | |
| CN113744870B (en) | System and method for main diagnosis and prediction of medical records front page | |
| CN117688226B (en) | Intelligent pre-diagnosis self-service bill making method and system based on similar child patient matching |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |