CN113270203A - Drug dose prediction method, device, electronic device and storage medium - Google Patents
Drug dose prediction method, device, electronic device and storage medium Download PDFInfo
- Publication number
- CN113270203A CN113270203A CN202110426774.0A CN202110426774A CN113270203A CN 113270203 A CN113270203 A CN 113270203A CN 202110426774 A CN202110426774 A CN 202110426774A CN 113270203 A CN113270203 A CN 113270203A
- Authority
- CN
- China
- Prior art keywords
- data
- drug
- dose prediction
- dose
- clinical
- 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 224
- 229940079593 drug Drugs 0.000 title claims abstract description 173
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000004140 cleaning Methods 0.000 claims abstract description 25
- 238000007781 pre-processing Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000012216 screening Methods 0.000 claims abstract description 19
- 229940126585 therapeutic drug Drugs 0.000 claims abstract description 14
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 13
- 230000001225 therapeutic effect Effects 0.000 claims abstract description 10
- 238000007689 inspection Methods 0.000 claims abstract description 9
- 238000010801 machine learning Methods 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 26
- 238000012360 testing method Methods 0.000 claims description 20
- 238000002790 cross-validation Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 6
- 238000011156 evaluation Methods 0.000 description 6
- QJJXYPPXXYFBGM-LFZNUXCKSA-N Tacrolimus Chemical compound C1C[C@@H](O)[C@H](OC)C[C@@H]1\C=C(/C)[C@@H]1[C@H](C)[C@@H](O)CC(=O)[C@H](CC=C)/C=C(C)/C[C@H](C)C[C@H](OC)[C@H]([C@H](C[C@H]2C)OC)O[C@@]2(O)C(=O)C(=O)N2CCCC[C@H]2C(=O)O1 QJJXYPPXXYFBGM-LFZNUXCKSA-N 0.000 description 5
- 229960001967 tacrolimus Drugs 0.000 description 5
- QJJXYPPXXYFBGM-SHYZHZOCSA-N tacrolimus Natural products CO[C@H]1C[C@H](CC[C@@H]1O)C=C(C)[C@H]2OC(=O)[C@H]3CCCCN3C(=O)C(=O)[C@@]4(O)O[C@@H]([C@H](C[C@H]4C)OC)[C@@H](C[C@H](C)CC(=C[C@@H](CC=C)C(=O)C[C@H](O)[C@H]2C)C)OC QJJXYPPXXYFBGM-SHYZHZOCSA-N 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000000890 drug combination Substances 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 108010081668 Cytochrome P-450 CYP3A Proteins 0.000 description 2
- 102000004328 Cytochrome P-450 CYP3A Human genes 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 230000036765 blood level Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000035622 drinking Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000391 smoking effect Effects 0.000 description 2
- 101150090724 3 gene Proteins 0.000 description 1
- 101150119038 ABCB1 gene Proteins 0.000 description 1
- 102100033350 ATP-dependent translocase ABCB1 Human genes 0.000 description 1
- 102000015081 Blood Coagulation Factors Human genes 0.000 description 1
- 108010039209 Blood Coagulation Factors Proteins 0.000 description 1
- 206010014357 Electric shock Diseases 0.000 description 1
- 206010071602 Genetic polymorphism Diseases 0.000 description 1
- 206010020751 Hypersensitivity Diseases 0.000 description 1
- 206010062016 Immunosuppression Diseases 0.000 description 1
- 238000000585 Mann–Whitney U test Methods 0.000 description 1
- 108010047230 Member 1 Subfamily B ATP Binding Cassette Transporter Proteins 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 238000011497 Univariate linear regression Methods 0.000 description 1
- 238000001467 acupuncture Methods 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 238000009098 adjuvant therapy Methods 0.000 description 1
- 208000026935 allergic disease Diseases 0.000 description 1
- 230000007815 allergy Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000003114 blood coagulation factor Substances 0.000 description 1
- 229940019700 blood coagulation factors Drugs 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000000747 cardiac effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 238000002635 electroconvulsive therapy Methods 0.000 description 1
- 239000003792 electrolyte Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001506 immunosuppresive effect Effects 0.000 description 1
- 239000003018 immunosuppressive agent Substances 0.000 description 1
- 229940124589 immunosuppressive drug Drugs 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000003907 kidney function Effects 0.000 description 1
- 230000003908 liver function Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005399 mechanical ventilation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 231100000417 nephrotoxicity Toxicity 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 231100001274 therapeutic index Toxicity 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 231100000331 toxic Toxicity 0.000 description 1
- 230000002588 toxic effect Effects 0.000 description 1
- 238000002054 transplantation Methods 0.000 description 1
- 210000002700 urine Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medicinal Chemistry (AREA)
- Pharmacology & Pharmacy (AREA)
- Toxicology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a medicine dose prediction method, a medicine dose prediction device, electronic equipment and a storage medium, which are applied to the technical field of data processing and can determine more accurate medicine use dose aiming at different patients. The method comprises the following steps: acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug use information, auxiliary therapeutic means, gene polymorphism and inspection information; performing data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening; and inputting the target characteristic data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for predicting a drug dose, an electronic device, and a storage medium.
Background
Tacrolimus is an immunosuppressive drug widely used in renal transplant recipients (KTRS). Due to the narrow therapeutic index, i.e. the close boundary between therapeutic and toxic blood levels, inappropriate blood levels of tacrolimus will cause acute rejection such as renal toxicity, renal infection, and new onset diabetes.
In the prior art, the dosage of tacrolimus used by clinicians is usually adjusted according to Therapeutic Drug Monitoring (TDM) in an early stage after transplantation until reaching a long-term maintenance level of immunosuppression.
However, the dosage required for achieving the targeted whole blood tacrolimus concentration varies from individual to individual, and factors influencing the dosage of tacrolimus are many, so how to determine the targeted dosage of the drug according to the individual characteristics of the patient is a problem to be solved urgently at present.
Disclosure of Invention
The invention provides a medicine dose prediction method, a medicine dose prediction device, electronic equipment and a storage medium, which are used for solving the problem that the targeted medicine use dose cannot be determined according to the personal characteristics of patients in the prior art and achieving the effect of determining more accurate medicine use dose aiming at different patients.
The invention provides a drug dose prediction method, which comprises the following steps: acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug use information, auxiliary therapeutic means, gene polymorphism and inspection information; performing data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening; and inputting the target characteristic data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.
According to the present invention, before inputting the target characteristic data into the drug dose prediction model, the method further comprises: acquiring clinical raw data of a plurality of patients; determining a modeled data set affecting patient medication dosage from the clinically raw data for the plurality of patients; and taking the dosage of the patient in unit time as a target variable, and constructing the drug dosage prediction model by adopting a preset machine learning algorithm based on the modeling data set.
According to the present invention, there is provided a method for predicting a medication dose, the method for determining a modeling data set influencing a medication dose for a patient based on clinical raw data of a plurality of patients, comprising: performing data cleaning on the clinical raw data of the plurality of patients, and performing normalized coding on categorical variables in the clinical raw data of the plurality of patients to obtain first feature data, wherein the data cleaning comprises at least one of the following items: deleting, deduplication and sorting; determining a correlation and significance of the first characteristic data with a patient medication dose, and determining second characteristic data from the first characteristic data according to the correlation and significance; determining an importance score of the second feature data, and determining third feature data from the second feature data according to the importance score; and screening the third characteristic data by a stepwise regression method to obtain the modeling data set.
According to the present invention, there is provided a method for predicting a dose of a drug, the method further comprising, after determining the modeling data set: dividing the modeling dataset into experimental group data and control group data; the constructing of the drug dose prediction model based on the modeling dataset by using a preset machine learning algorithm comprises: training the drug dose prediction model by adopting a preset machine learning algorithm based on the experimental group data; testing the drug dose prediction model based on the control group data.
According to the method for predicting the drug dosage provided by the invention, the construction of the drug dosage prediction model by adopting a preset machine learning algorithm comprises the following steps: and automatically learning and adjusting parameters of the medicine dosage prediction model by an automatic machine learning method.
According to the present invention, there is provided a method for predicting a dose of a drug, the method further comprising: and determining a model with the optimal correlation coefficient and the optimal root mean square error as the drug dose prediction model by a five-fold cross validation method.
The present invention also provides a drug dose prediction device comprising: the device comprises an acquisition module, a processing module and a prediction module; the acquisition module is used for acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug information, auxiliary therapeutic means, gene polymorphism and inspection information; the processing module is used for carrying out data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening; and the prediction module is used for inputting the target characteristic data into a medicine dose prediction model to obtain the medicine dose of the target patient in unit time.
According to the present invention, there is provided a medication dose prediction device, the device further comprising a training module: the acquisition module is further used for acquiring clinical raw data of a plurality of patients before inputting the target characteristic data into a drug dose prediction model; the processing module is further configured to determine a modeling dataset from the clinical raw data of the plurality of patients that affects patient medication dosage; and the training module is used for constructing the medicine dose prediction model by using the medicine dose of the patient in unit time as a target variable and adopting a preset machine learning algorithm based on the modeling data set.
According to the present invention, there is provided a medication dose prediction device, wherein the processing module is specifically configured to: performing data cleaning on the clinical raw data of the plurality of patients, and performing normalized coding on categorical variables in the clinical raw data of the plurality of patients to obtain first feature data, wherein the data cleaning comprises at least one of the following items: deleting, deduplication and sorting; determining a correlation and significance of the first characteristic data with a patient medication dose, and determining second characteristic data from the first characteristic data according to the correlation and significance; determining an importance score of the second feature data, and determining third feature data from the second feature data according to the importance score; and screening the third characteristic data by a stepwise regression method to obtain the modeling data set.
According to the present invention, there is provided a drug dose prediction device, wherein the processing module is further configured to, after determining the modeling data set, divide the modeling data set into experimental group data and control group data; the training module is specifically configured to: training the drug dose prediction model by adopting a preset machine learning algorithm based on the experimental group data; testing the drug dose prediction model based on the control group data.
According to the present invention, there is provided a medication dose prediction device, wherein the training module is specifically configured to: and automatically learning and adjusting parameters of the medicine dosage prediction model by an automatic machine learning method.
According to the medicine dose prediction device provided by the invention, the training module is also used for determining a model with the optimal correlation coefficient and the optimal root mean square error as the medicine dose prediction model through a five-fold cross validation method.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method for predicting a drug dose as described in any of the above when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of drug dose prediction as described in any of the above.
The medicine dose prediction method, the device, the electronic equipment and the storage medium provided by the invention can acquire clinical original data of a target patient, perform data preprocessing on the clinical original data to obtain target characteristic data, and input the target characteristic data into a medicine dose prediction model to obtain the medicine dose of the target patient in unit time. Since the drug dose prediction model takes the preprocessed clinical raw data of the target patient, i.e. the target characteristic data, as input, the finally output drug dose can achieve the effect of determining a more accurate drug use dose for the target patient.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for predicting a dosage of a drug according to the present invention;
FIG. 2 is a second schematic flow chart of the method for predicting the dosage of a drug according to the present invention;
FIG. 3 is a schematic structural diagram of a drug dosage prediction device provided in the present invention;
FIG. 4 is a second schematic structural diagram of a drug dosage prediction device provided in the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that, in the embodiments of the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "e.g.," an embodiment of the present invention is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present concepts related in a concrete fashion.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Further, it should be noted that the scope of the methods and apparatus of embodiments of the present invention is not limited to performing functions in the order illustrated or discussed, but may include performing functions in a substantially simultaneous manner or in a reverse order based on the functions involved, e.g., the methods described may be performed in an order different than that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.
For the convenience of clearly describing the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", and the like are used for distinguishing the same items or similar items with basically the same functions and actions, and those skilled in the art can understand that the words "first", "second", and the like are not limited in number or execution order.
While certain exemplary embodiments of the invention have been described for purposes of illustration, it is to be understood that the invention may be practiced otherwise than as specifically described.
The above-described implementations are described in detail below with reference to specific embodiments and the accompanying drawings.
As shown in fig. 1, an embodiment of the present invention provides a drug dose prediction method, which may be applied to a drug dose prediction apparatus. The drug dose prediction method may include S101-S103:
s101, acquiring clinical original data of a target patient by a medicine dose prediction device.
The clinical raw data may include demographic information, therapeutic drug use information, drug combination information, adjuvant therapy means, gene polymorphism, and test information.
Specifically, the demographic information may include information such as age, sex, height, weight, smoking history, drinking history, allergy history, past history, etc.; the information on the use of the treatment medicine can include information on single dosage of the treatment medicine, frequency of administration of the treatment medicine, daily dosage of the treatment medicine, total dosage of the treatment medicine during hospitalization and the like; the drug combination information can comprise the use conditions of drugs such as prazole, MPA and the like; the auxiliary treatment means may include: whether physical auxiliary treatment therapies such as mechanical ventilation of a breathing machine, electric shock treatment, infrared treatment, laser, acupuncture, massage and the like exist; the gene polymorphism may include information on gene sites such as CYP3A5, ABCB1, etc.; the test information may include blood routine, urine routine, liver function, kidney function, electrolytes, cancer markers, cardiac markers, blood coagulation factors, etiology test, and the like.
It should be noted that the medication dose prediction apparatus may obtain the clinical raw data from a hospital database, and the above is an exemplary illustration of the clinical raw data, and in fact, the clinical raw data may also include other categories not specifically shown.
The therapeutic drug in the therapeutic drug use information is a drug of which a dose is to be predicted in the present invention, and is hereinafter simply referred to as a target drug; the drug in the combination information is a drug to be taken concurrently with the target drug.
S102, the medicine dose prediction device carries out data preprocessing on the clinical original data to obtain target characteristic data.
The data preprocessing may include at least one of data cleaning, data normalization coding, and data screening.
First, data cleansing of clinical raw data is required because the clinical raw data of a target patient may contain many repetitive, ambiguous, and dose-independent information. For example, the method may include deleting order information with ambiguous medication frequency, deleting order information with medication dose of 0, de-duplicating and combining diagnostic information belonging to the target patient, and sorting the medication information of the target patient and the TDM detection information from early to late in time; the target patient is screened for compliance with regulatory protocols for testing, medication and diagnostic information.
Second, the clinical raw data of the target patient includes continuous independent variable information and classified independent variable information. The continuous independent variable information refers to variable information including various results, such as age, height, frequency of medication and the like; classification independent variable means variable information including a limited number of results, for example, gene test results (including 3 gene points), whether there are other disease types (including both yes and no cases). The drug dose prediction device may encode the classification independent variable information in the clinical raw data in a normalized manner. For example, the results of the 0, 1,2 canonical CYP3A5, ABCB1 gene tests can be used to uniquely encode drug combinations and disease types.
It should be noted that the medicine dosage prediction device may only perform normalized coding on the classification independent variable information that needs to be further analyzed and determined, and does not need to perform normalized coding on the classification independent variable information that can be directly determined whether smoking, drinking, and the like.
Finally, because of the large number of dimensions of the raw clinical data, further screening is required based on the correlation of each dimension with the dose of the target drug. Specifically, the drug dose prediction device can screen clinical raw data through steps 10 to 12:
step 10: the drug dose prediction device can check the correlation between the continuous independent variable information and the drug dose of the target drug by a Pearson correlation check method, check the significance between the classified independent variable information and the drug dose of the target drug by a Mann-Whitney U check method, delete the irrelevant continuous independent variable information and the classified independent variable information with the significance lower than a first threshold value according to the detection result, and keep the relevant and significant variable information.
Step 11: based on the variable information screened in step 10, the medicine dose prediction device may use the XGBoost model to establish a target medicine dose model, calculate importance scores of the variables and sort in descending order, the higher the importance score is, the greater the influence of the variable on the unit time dose of the target medicine is, and finally, screen out the variable whose importance is greater than the second threshold.
Step 12: based on the variable information screened in step 10, the drug dosage prediction device may first establish a univariate regression equation y ═ a for each independent variable and dependent variable (i.e., the target drug dosage used per unit time)iXi+bi1, 2.. times.m. Then, test statistic F of regression coefficients in m unitary regression equations is calculated respectively, and the maximum value is obtainedIf it isStopping screening, otherwiseSelect into the variable set, at which time it willConsider as x1. Then respectively setting independent variable groups (x)1,x2),(x1,x3),...,(x1,xm) And dependent on the factorsQuantity-building a binary regression equation, (in this case x1Is that the above-mentioned). Then, x in the equation is calculated2,x3,...,xmRegression coefficient test statistic F, takingIf it isStopping screening, otherwiseSelect into the variable set, at which timeConsider as x2..... And iterating until the maximum F value of the independent variable is smaller than a critical value, wherein the regression equation is the optimal regression equation. Introducing independent variables into the model one by one, checking whether the introduction of the independent variables causes the model to have a significant change (F test), introducing the independent variables into the model if the model has the significant change, and otherwise, ignoring the independent variables until all the independent variables are considered. The variables are arranged from large to small according to the contribution degree, and are added in sequence to screen out target characteristic data.
S103, inputting the target characteristic data into a drug dose prediction model by the drug dose prediction device to obtain the drug dose of the target patient in unit time.
After the target characteristic data is determined, the medicine dose prediction device can input the target characteristic data into the medicine dose prediction model, and finally the medicine dose of the target patient in unit time, namely the dose of the target medicine used by the target patient in unit time, is obtained.
Alternatively, the dosage of the target patient in the unit time may be a daily dosage of the target patient, and the daily dosage may be a product of a daily single dose and a frequency of administration.
In the embodiment of the invention, the clinical original data of the target patient can be obtained, the data preprocessing is carried out on the clinical original data to obtain the target characteristic data, and the target characteristic data is input into the drug dose prediction model to obtain the drug dose of the target patient in unit time. Since the drug dose prediction model takes the preprocessed clinical raw data of the target patient, i.e. the target characteristic data, as input, the finally output drug dose can achieve the effect of determining a more accurate drug use dose for the target patient.
Optionally, as shown in fig. 2, the method for predicting the dosage of the drug may further include: S104-S106:
s104, acquiring clinical raw data of a plurality of patients by the medicine dosage prediction device.
S105, determining a modeling data set influencing the medication dosage of the patient according to clinical raw data of a plurality of patients by the medication dosage prediction device.
Alternatively, the drug dose prediction means may determine the modeled data set by the following steps 20-23.
Step 20, performing data cleaning on clinical raw data of a plurality of patients, and performing normalized coding on classification variables in the clinical raw data of the plurality of patients to obtain first feature data, where the data cleaning may include at least one of: delete, deduplication, and sort.
The process of performing data cleaning and normalized coding on clinical raw data of a plurality of patients by the drug dose prediction device may refer to the description of performing data cleaning and normalized coding on clinical raw data of a target patient in S102, and will not be described herein again.
It should be noted that, during the data cleaning process of the clinical raw data of a plurality of patients, the drug dosage prediction device can also delete the patient information with incomplete medication information and no TDM detection information.
And step 21, determining the correlation and significance of the first characteristic data and the medication dosage of the patient, and determining second characteristic data from the first characteristic data according to the correlation and significance.
The drug dose prediction device can determine the correlation between continuous independent variable information in the first characteristic data of a plurality of patients and the drug dosage of the target drug by a Pearson correlation test method, determine the significance between the classified independent variable information and the drug dosage of the target drug by a Mann-Whitney U test method, delete irrelevant continuous independent variable information and classified independent variable information with the significance lower than a first threshold value according to the final result, and keep the variable information with strong correlation and significance to obtain second characteristic data.
And step 22, determining the importance score of the second characteristic data, and determining third characteristic data from the second characteristic data according to the importance score.
The medicine dose prediction device can establish a target medicine dose model by using the XGboost model, calculate importance scores of all variables, perform descending order, and finally screen out variables with importance greater than a second threshold value to obtain third feature data.
And 23, screening the third characteristic data through a stepwise regression method to obtain a modeling data set.
The process of modeling the data set by the drug dose prediction device through stepwise regression method can refer to the description related to determining the target characteristic data in step 12, and will not be described herein again.
Optionally, after determining the modeling data set, the drug dose prediction device may further divide the modeling data set into experimental group data and control group data; then training a drug dose prediction model by adopting a preset machine learning algorithm based on experimental group data; drug dose prediction models were tested based on control data.
Specifically, the drug dose prediction device may use the modeling data set as a basis for dividing experimental group data and control group data, and perform tendency score matching on the experimental group data and the control group data, so that baseline levels of the experimental group data and the control group data are substantially consistent, thereby eliminating confounding factors in the modeling data set, and further analyzing variables that have significant effects on the dose of the target drug used in unit time.
S106, the medicine dose prediction device takes the medicine dose of a patient in unit time as a target variable, and a preset machine learning algorithm is adopted to construct a medicine dose prediction model based on a modeling data set.
The medicine dose prediction device can use 'daily use dose of target medicine' as a target variable, screened experimental group data is used as an independent variable, and a medicine dose prediction model is constructed based on an XGboost algorithm. In order to solve the problem of a large amount of missing data in experimental group data, the medicine dosage prediction device can carry out automatic learning and parameter adjustment on a medicine dosage prediction model through an automatic machine learning method. Meanwhile, the drug dose prediction device can also introduce a five-fold cross validation method, evaluate the prediction capability of the drug dose prediction model by using the correlation coefficient and the root mean square error, and determine the model with the optimal correlation coefficient and the optimal root mean square error as the drug dose prediction model.
It should be noted that the goal of automated machine learning is to use automated data-driven approaches to make decisions. The automatic machine learning system can automatically determine the optimal scheme as long as a user provides data, and the automatic machine learning system not only comprises algorithm selection, hyper-parameter optimization and neural network architecture search, but also covers each step of a machine learning workflow, such as automatic data preparation, automatic feature selection, automatic algorithm selection and the like.
It should be noted that the implementation process of the five-fold cross validation is as follows: dividing the data set into five equal parts, sequentially taking one part as a test set and the remaining four parts as a training set to construct a prediction model, adopting the training effect of the test set prediction model, outputting the prediction evaluation index result of each compromise test set, and finally taking the average value of each evaluation index in the five folds as the standard for the model prediction capability evaluation index.
Specifically, the modeling process of the drug dose prediction device is as follows:
(1) canonical learning objective
To learn the set of functions used in the model, the following regularization targets are first minimized:where i is a slightly convex loss function for measuring the difference between the predicted value and the target value. Second itemIs the complexity of the penalty model. The term includes two parts, one is the total number of leaf nodes and one is the L2 regularization term derived from the leaf nodes. The added regularization term can smooth the learning weight of each leaf node to avoid overfitting.
(2)Shinkage and Column Subsampling
To prevent overfitting, Shinkage and Column Subsampling techniques may be introduced on the basis of (1) to further prevent overfitting of the established model. shinkage can reduce the influence of each independent tree and reserve space for future trees to optimize a model; column Subsampling is more resistant to overfitting than traditional line sampling, and also speeds up the computation of parallel algorithms when constructing features.
(3) Sparse value processing (sparse-aware Split filling)
There are many possible reasons for sparsity: missing values, frequent zero entries in statistics, or feature engineering exist in the data. It is very important for the XGBoost algorithm to know the sparse pattern in the data, and in order to better solve this problem, a default direction may be added to each tree node. When there is no value in the sparse matrix x, the sample is classified as the default direction. There are two default directions chosen in each branch, the best default direction being learned from the data. The key improvement is to access only non-missing entries. The proposed algorithm treats a non-existing value as a missing value and learns the best direction to handle the missing value.
(4) Parallel Block Learning (Column Block for Parallel Learning)
The most time-consuming part in the decision tree learning is to sequence data according to characteristic values, in order to reduce the sequencing cost, the XGboost stores the data in a memory unit, when the nodes are split, the gain of each characteristic is calculated, and the characteristic with the maximum gain is selected to be split, so that the multi-process is realized.
(5) Automatic machine learning
In order to make the training process of the XGboost model simpler, the XGboost parameters can be optimized through automatic machine learning (auto-ml), the configuration strategy selects a super-parameter search method to optimize the parameters such as n _ estimator, learning _ rate, gamma, subsample, collemplempere _ byte, max _ depth and the like, and the optimal parameters are selected to establish the model.
(6) Cross validation
And (5) introducing a five-fold cross validation method for training according to the model established by the optimal parameters in the step (5), constructing a prediction model by using a training set, predicting an evaluation index result, and taking the mean value of each evaluation index obtained in the five-fold cross validation as an optimal standard.
(7) Test model
And selecting the correlation coefficient, the root mean square error and the average absolute error of the test set and the prediction set as evaluation indexes of the model, and evaluating the accuracy, the stability and the robustness of the model.
In the embodiment of the invention, based on patient clinical original data such as demographic information, therapeutic drug use information, combined drug information, auxiliary treatment means, gene polymorphism, inspection information and the like, the data are preprocessed by using a statistical method in combination with a machine learning algorithm, a modeling data set is screened out, and a drug dose prediction model is constructed for predicting the personalized dose of a target drug for different patients. Because the loss of the left and right subtrees can be calculated through the decision tree, the optimal one is selected to fit the drug prediction model, and the variable with the maximum correlation can be quickly and effectively screened out from high-dimensional data, the accurate prediction result of the target drug dose can be obtained according to different clinical characteristics and treatment conditions of patients.
The scheme provided by the embodiment of the invention is mainly introduced from the perspective of a method. To implement the above functions, it includes hardware structures and/or software modules for performing the respective functions. Those of skill in the art will readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
According to the method for predicting the drug dose provided by the embodiment of the invention, the execution main body can be a drug dose prediction device or a control module for predicting the drug dose in the drug dose prediction device. In the embodiment of the present invention, a method for predicting a drug dose performed by a drug dose prediction apparatus is taken as an example, and the drug dose prediction apparatus provided in the embodiment of the present invention is described.
It should be noted that, the embodiment of the present invention may divide the function modules of the drug dosage prediction apparatus according to the above method, for example, each function module may be divided according to each function, or two or more functions may be integrated into one processing module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. Optionally, the division of the modules in the embodiment of the present invention is schematic, and is only a logic function division, and there may be another division manner in actual implementation.
As shown in fig. 3, an embodiment of the present invention provides a medication dose prediction apparatus 300. The drug dose prediction device 300 includes: an acquisition module 301, a processing module 302, and a prediction module 303. The obtaining module 301 may be configured to obtain clinical raw data of a target patient, where the clinical raw data includes demographic information, therapeutic drug use information, drug combination information, auxiliary therapeutic measures, genetic polymorphisms, and test information; the processing module 302 may be configured to perform data preprocessing on the clinical raw data to obtain target feature data, where the data preprocessing includes at least one of data cleaning, data normalization coding, and data screening; the prediction module 303 may be configured to input the target characteristic data into a drug dose prediction model to obtain a drug dose of the target patient per unit time.
Optionally, in conjunction with fig. 3, as shown in fig. 4, the apparatus 300 may further include a training module 304. The obtaining module 301 may be further configured to obtain clinical raw data of a plurality of patients before inputting the target characteristic data into a drug dose prediction model; the processing module 302 may be further configured to determine a modeling dataset affecting patient dosage based on the clinical raw data of the plurality of patients; the training module 304 may be configured to construct the drug dose prediction model by using a preset machine learning algorithm based on the modeling data set with the drug dose of the patient per unit time as a target variable.
Optionally, the processing module 302 may specifically be configured to: performing data cleaning on the clinical raw data of the plurality of patients, and performing normalized coding on categorical variables in the clinical raw data of the plurality of patients to obtain first feature data, wherein the data cleaning comprises at least one of the following items: deleting, deduplication and sorting; determining a correlation and significance of the first characteristic data with a patient medication dose, and determining second characteristic data from the first characteristic data according to the correlation and significance; determining an importance score of the second feature data, and determining third feature data from the second feature data according to the importance score; and screening the third characteristic data by a stepwise regression method to obtain the modeling data set.
Optionally, the processing module 302 may be further configured to, after determining the modeling data set, divide the modeling data set into experimental group data and control group data; the training module 304 may be specifically configured to: training the drug dose prediction model by adopting a preset machine learning algorithm based on the experimental group data; testing the drug dose prediction model based on the control group data.
Optionally, the training module 304 may be specifically configured to: and automatically learning and adjusting parameters of the medicine dosage prediction model by an automatic machine learning method.
Optionally, the training module 304 may be further configured to determine, through a five-fold cross-validation method, that the model with the optimal correlation coefficient and the optimal root mean square error is the drug dose prediction model.
In the embodiment of the invention, the clinical original data of the target patient can be obtained, the data preprocessing is carried out on the clinical original data to obtain the target characteristic data, and the target characteristic data is input into the drug dose prediction model to obtain the drug dose of the target patient in unit time. Since the drug dose prediction model takes the preprocessed clinical raw data of the target patient, i.e. the target characteristic data, as input, the finally output drug dose can achieve the effect of determining a more accurate drug use dose for the target patient.
Fig. 5 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 5: a processor (processor)810, a communication Interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication Interface 820 and the memory 830 communicate with each other via the communication bus 840. The processor 810 may invoke logic instructions in the memory 830 to perform a method of drug dose prediction comprising: acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug use information, auxiliary therapeutic means, gene polymorphism and inspection information; performing data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening; and inputting the target characteristic data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.
In addition, the logic instructions in the memory 830 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform a method of drug dose prediction provided by the above methods, the method comprising: acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug use information, auxiliary therapeutic means, gene polymorphism and inspection information; performing data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening; and inputting the target characteristic data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the method of predicting a dose of a drug provided above, the method comprising: acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug use information, auxiliary therapeutic means, gene polymorphism and inspection information; performing data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening; and inputting the target characteristic data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (14)
1. A method of predicting a dosage of a drug, comprising:
acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug use information, auxiliary therapeutic means, gene polymorphism and inspection information;
performing data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening;
and inputting the target characteristic data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.
2. The method of claim 1, wherein prior to entering the target characteristic data into a drug dose prediction model, the method further comprises:
acquiring clinical raw data of a plurality of patients;
determining a modeled data set affecting patient medication dosage from the clinically raw data for the plurality of patients;
and taking the dosage of the patient in unit time as a target variable, and constructing the drug dosage prediction model by adopting a preset machine learning algorithm based on the modeling data set.
3. A method of predicting medication dose as recited in claim 2, wherein said determining a modeled data set that affects patient medication dose from said plurality of patient clinical raw data comprises:
performing data cleaning on the clinical raw data of the plurality of patients, and performing normalized coding on categorical variables in the clinical raw data of the plurality of patients to obtain first feature data, wherein the data cleaning comprises at least one of the following items: deleting, deduplication and sorting;
determining a correlation and significance of the first characteristic data with a patient medication dose, and determining second characteristic data from the first characteristic data according to the correlation and significance;
determining an importance score of the second feature data, and determining third feature data from the second feature data according to the importance score;
and screening the third characteristic data by a stepwise regression method to obtain the modeling data set.
4. A method of drug dose prediction according to claim 2 or 3, wherein after determining the modeled data set, the method further comprises:
dividing the modeling dataset into experimental group data and control group data;
the constructing of the drug dose prediction model based on the modeling dataset by using a preset machine learning algorithm comprises:
training the drug dose prediction model by adopting a preset machine learning algorithm based on the experimental group data;
testing the drug dose prediction model based on the control group data.
5. The method of predicting a dosage of a pharmaceutical according to claim 2 or 3, wherein the constructing the model of the pharmaceutical dosage using a pre-determined machine learning algorithm comprises:
and automatically learning and adjusting parameters of the medicine dosage prediction model by an automatic machine learning method.
6. A method of predicting a dose of a drug as in claim 2 or 3, further comprising:
and determining a model with the optimal correlation coefficient and the optimal root mean square error as the drug dose prediction model by a five-fold cross validation method.
7. A medication dose prediction device, comprising: the device comprises an acquisition module, a processing module and a prediction module;
the acquisition module is used for acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, combined drug information, auxiliary therapeutic means, gene polymorphism and inspection information;
the processing module is used for carrying out data preprocessing on the clinical original data to obtain target characteristic data, wherein the data preprocessing comprises at least one of data cleaning, data standardized coding and data screening;
and the prediction module is used for inputting the target characteristic data into a medicine dose prediction model to obtain the medicine dose of the target patient in unit time.
8. The medication dose prediction device of claim 7, further comprising a training module:
the acquisition module is further used for acquiring clinical raw data of a plurality of patients before inputting the target characteristic data into a drug dose prediction model;
the processing module is further configured to determine a modeling dataset from the clinical raw data of the plurality of patients that affects patient medication dosage;
and the training module is used for constructing the medicine dose prediction model by using the medicine dose of the patient in unit time as a target variable and adopting a preset machine learning algorithm based on the modeling data set.
9. The medication dose prediction device of claim 8, wherein the processing module is specifically configured to:
performing data cleaning on the clinical raw data of the plurality of patients, and performing normalized coding on categorical variables in the clinical raw data of the plurality of patients to obtain first feature data, wherein the data cleaning comprises at least one of the following items: deleting, deduplication and sorting;
determining a correlation and significance of the first characteristic data with a patient medication dose, and determining second characteristic data from the first characteristic data according to the correlation and significance;
determining an importance score of the second feature data, and determining third feature data from the second feature data according to the importance score;
and screening the third characteristic data by a stepwise regression method to obtain the modeling data set.
10. A medication dose prediction device according to claim 8 or 9,
the processing module is further used for dividing the modeling data set into experimental group data and control group data after determining the modeling data set;
the training module is specifically configured to: training the drug dose prediction model by adopting a preset machine learning algorithm based on the experimental group data; testing the drug dose prediction model based on the control group data.
11. A drug dose prediction device according to claim 8 or 9, wherein the training module is specifically configured to: and automatically learning and adjusting parameters of the medicine dosage prediction model by an automatic machine learning method.
12. The medication dose prediction device of claim 8 or 9, wherein the training module is further configured to determine a model with optimal correlation coefficients and root mean square error as the medication dose prediction model through a five-fold cross-validation method.
13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps in the method of drug dose prediction according to any of claims 1 to 6.
14. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the steps in the method of drug dose prediction according to any of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110426774.0A CN113270203A (en) | 2021-04-20 | 2021-04-20 | Drug dose prediction method, device, electronic device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110426774.0A CN113270203A (en) | 2021-04-20 | 2021-04-20 | Drug dose prediction method, device, electronic device and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113270203A true CN113270203A (en) | 2021-08-17 |
Family
ID=77229058
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110426774.0A Pending CN113270203A (en) | 2021-04-20 | 2021-04-20 | Drug dose prediction method, device, electronic device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113270203A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113469151A (en) * | 2021-09-03 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Method, device, equipment and medium for screening sensor in industrial manufacturing |
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 |
CN113948170A (en) * | 2021-08-18 | 2022-01-18 | 天津开心生活科技有限公司 | Treatment duration obtaining method and device, electronic equipment and storage medium |
CN114722976A (en) * | 2022-06-09 | 2022-07-08 | 青岛美迪康数字工程有限公司 | Medicine recommendation system and construction method |
CN115376649A (en) * | 2022-09-29 | 2022-11-22 | 中日友好医院(中日友好临床医学研究所) | Method and device for predicting dose of intrathecal opioid analgesic |
CN115424695A (en) * | 2022-11-03 | 2022-12-02 | 首都医科大学附属北京安贞医院 | Method for determining dosing dose prediction model and dosing dose prediction method |
CN116072259A (en) * | 2023-02-02 | 2023-05-05 | 山东大学 | Neonatal beta lactam medicine optimal dose selection method and system |
CN116994700A (en) * | 2023-03-31 | 2023-11-03 | 北京诺道认知医学科技有限公司 | Quetiapine dose individuation recommendation method and device based on deep learning |
CN118039062A (en) * | 2024-04-12 | 2024-05-14 | 四川省肿瘤医院 | Individualized chemotherapy dose remote control method based on big data analysis |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6658396B1 (en) * | 1999-11-29 | 2003-12-02 | Tang Sharon S | Neural network drug dosage estimation |
CN110010252A (en) * | 2019-04-01 | 2019-07-12 | 上海交通大学医学院附属新华医院 | Warfarin dosage prediction technique and device |
US20200250554A1 (en) * | 2019-02-01 | 2020-08-06 | Jabil Circuit (Shanghai) Co., Ltd. | Method and storage medium for predicting the dosage based on human physiological parameters |
CN111613289A (en) * | 2020-05-07 | 2020-09-01 | 浙江大学医学院附属第一医院 | Individualized drug dose prediction method, individualized drug dose prediction device, electronic equipment and storage medium |
-
2021
- 2021-04-20 CN CN202110426774.0A patent/CN113270203A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6658396B1 (en) * | 1999-11-29 | 2003-12-02 | Tang Sharon S | Neural network drug dosage estimation |
US20200250554A1 (en) * | 2019-02-01 | 2020-08-06 | Jabil Circuit (Shanghai) Co., Ltd. | Method and storage medium for predicting the dosage based on human physiological parameters |
CN110010252A (en) * | 2019-04-01 | 2019-07-12 | 上海交通大学医学院附属新华医院 | Warfarin dosage prediction technique and device |
CN111613289A (en) * | 2020-05-07 | 2020-09-01 | 浙江大学医学院附属第一医院 | Individualized drug dose prediction method, individualized drug dose prediction device, electronic equipment and storage medium |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113948170A (en) * | 2021-08-18 | 2022-01-18 | 天津开心生活科技有限公司 | Treatment duration obtaining method and device, electronic equipment and storage medium |
CN113469151A (en) * | 2021-09-03 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Method, device, equipment and medium for screening sensor in industrial manufacturing |
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 |
CN114722976A (en) * | 2022-06-09 | 2022-07-08 | 青岛美迪康数字工程有限公司 | Medicine recommendation system and construction method |
CN115376649A (en) * | 2022-09-29 | 2022-11-22 | 中日友好医院(中日友好临床医学研究所) | Method and device for predicting dose of intrathecal opioid analgesic |
CN115376649B (en) * | 2022-09-29 | 2023-07-07 | 中日友好医院(中日友好临床医学研究所) | Dose prediction method and device for intrathecal opioid analgesic |
CN115424695A (en) * | 2022-11-03 | 2022-12-02 | 首都医科大学附属北京安贞医院 | Method for determining dosing dose prediction model and dosing dose prediction method |
CN116072259A (en) * | 2023-02-02 | 2023-05-05 | 山东大学 | Neonatal beta lactam medicine optimal dose selection method and system |
CN116072259B (en) * | 2023-02-02 | 2023-09-08 | 山东大学 | Neonatal beta lactam medicine optimal dose selection method and system |
CN116994700A (en) * | 2023-03-31 | 2023-11-03 | 北京诺道认知医学科技有限公司 | Quetiapine dose individuation recommendation method and device based on deep learning |
CN116994700B (en) * | 2023-03-31 | 2024-06-04 | 北京诺道认知医学科技有限公司 | Quetiapine dose individuation recommendation method and device based on deep learning |
CN118039062A (en) * | 2024-04-12 | 2024-05-14 | 四川省肿瘤医院 | Individualized chemotherapy dose remote control method based on big data analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113270203A (en) | Drug dose prediction method, device, electronic device and storage medium | |
Rady et al. | Prediction of kidney disease stages using data mining algorithms | |
CN111613289B (en) | Individuation medicine dosage prediction method, device, electronic equipment and storage medium | |
CN111986770B (en) | Prescription medication auditing method, device, equipment and storage medium | |
CN109243620B (en) | Drug effect optimization method and device based on blood concentration monitoring | |
CN109036553A (en) | A kind of disease forecasting method based on automatic extraction Medical Technologist's knowledge | |
CN107785057B (en) | Medical data processing method, device, storage medium and computer equipment | |
WO2021151327A1 (en) | Triage data processing method and apparatus, and device and medium | |
CN109243546B (en) | Method and device for establishing anticoagulant drug effect optimization model | |
US20220246297A1 (en) | Causal Recommender Engine for Chronic Disease Management | |
CN111291131A (en) | Data processing method, data processing device, storage medium and electronic equipment | |
CN105009130B (en) | Detection means, detection method and detection program that the omen detection migrated to organism state based on network entropy is aided in | |
CN114999656B (en) | Alzheimer disease risk assessment system and module | |
CN107145715A (en) | A kind of clinical medical intelligent discriminating gear based on election algorithm | |
CN117133471A (en) | Blood transfusion prediction method, system, equipment and medium for acute aortic dissection diseases | |
CN113990502B (en) | ICU heart failure prediction system based on heterogeneous graph neural network | |
CN112071431B (en) | Clinical path automatic generation method and system based on deep learning and knowledge graph | |
Samet et al. | Predicting and staging chronic kidney disease using optimized random forest algorithm | |
US20240303544A1 (en) | Graph database techniques for machine learning | |
RU2723674C1 (en) | Method for prediction of diagnosis based on data processing containing medical knowledge | |
CN115376658A (en) | Artificial intelligent evaluation method for traditional Chinese medicine prescription based on fusion phenotype and molecular information of deep neural network | |
DE102005028975B4 (en) | A method of determining a biomarker for identifying a specific biological condition of an organism from at least one dataset | |
Oliveira et al. | Towards an intelligent systems to predict nosocomial infections in intensive care | |
Saranya et al. | Bd-Mdl: bipolar disorder detection using machine leanring and deep learning | |
CN117095820B (en) | Risk early warning method and equipment for family gout |
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: 20210817 |
|
RJ01 | Rejection of invention patent application after publication |