Drug use information recommendation method and related products

Abstract

The invention provides a drug use information recommendation method and related products. According to a specific embodiment of the drug use information recommendation method, a candidate drug identifier set is preliminarily selected from specified selectable drugs (namely, a preset drug identifier set) according to the effective rate of each drug in historical drugs of different genotypes and a gene detection result of a patient, and the screening range is narrowed; and further determining the drug effective rate of the corresponding candidate drug identifier for the target patient by using the drug effective rate prediction model of each candidate drug identifier, and finally determining the to-be-recommended drug identifier. As for each preset drug identifier, the drug effective rate prediction model is pre-trained by using the gene detection result of the clinical patient using the drug indicated by the preset drug identifier and the corresponding drug use effect label, and compared with a statistical analysis conclusion based on a small number of cases, the drug effective rate prediction model has the advantages that the prediction accuracy is improved; the reliability and robustness are high, and the reliability of drug use information recommendation can be finally improved.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of artificial intelligence technology, and specifically to a medication information recommendation method and related products. Background Art

[0002] Drug therapy is currently the main treatment for many diseases. Faced with a variety of drugs, it is crucial to quickly select the most effective drug for patients. Clinical treatment currently still remains at the stage of selecting therapeutic drugs for patients based on the subjective experience of doctors, and the trial and error cost is high, that is, multiple attempts are required to determine the most effective drug for individual patients.

[0003] Pharmacogenomics (PGx) provides a certain reference for individualized medication selection from a genetic perspective, that is, based on the patient's genetic test results and combined with relevant research, appropriate medication is recommended for the patient. If the SLC6A4 gene of a patient with depression is S / S type, then according to the research of Kim H et.al and Hicks JK et.al, the patient may have a better effect with sertraline. However, the level of evidence for this method is low. Some studies have pointed out that the clinical reference of this method is low. One reason is that the relevant research used as evidence is generally based on dozens of control studies and statistical analyses to draw conclusions, and the reliability and robustness are poor; another reason is that the standards for efficacy judgment are not unified, resulting in differences between different PGx tools, resulting in the inability to form reliable medication information recommendation results. Summary of the invention

[0004] The embodiments of the present disclosure provide a method, device, electronic device, storage medium and computer program product for recommending medication information.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for recommending medication information, the method comprising: Obtaining a genetic test result of a target patient, wherein the genetic test result includes a genotype of each gene locus in a preset gene locus set; For each of the preset gene loci, determining a set of candidate drug identifiers for the target patient according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus; For each of the candidate drug identifiers, based on the drug effectiveness prediction model corresponding to the candidate drug identifier, the drug effectiveness of the target patient is predicted using the gene test results of the target patient to obtain the drug effectiveness of the candidate drug identifier for the target patient, wherein each preset drug identifier corresponds to a drug effectiveness prediction model for characterizing the corresponding relationship between the gene test results and the drug effectiveness; According to the drug effectiveness of each candidate drug identification for the target patient, a drug identification to be recommended is determined from each candidate drug identification.

[0006] In some optional embodiments, for each of the preset gene loci, determining the candidate drug marker set for the target patient according to the historical drug efficacy of each preset drug marker in the preset drug marker set for the genotype of the target patient at the preset gene locus comprises: For each of the preset gene loci, according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus, determine a set of highly effective drug identifiers for the target patient at the preset gene locus; The highly effective drug identification sets of the target patient at each of the preset gene sites are merged to obtain a candidate drug identification set of the target patient.

[0007] In some optional embodiments, determining a set of highly effective drug markers for the target patient at the preset gene locus according to the historical drug effectiveness of each preset drug marker in the preset drug marker set for the genotype of the target patient at the preset gene locus comprises: The L preset drug identifiers in the preset drug identifier set with the highest historical drug efficacy for the genotype of the target patient at the preset gene locus, and / or the preset drug identifiers in the preset drug identifier set with a historical drug efficacy for the genotype of the target patient at the preset gene locus greater than a preset historical drug efficacy threshold, are determined as the high-efficiency drug identifier set for the target patient at the preset gene locus, where L is a natural number.

[0008] In some optional embodiments, the historical drug effectiveness rate of each preset drug identification in the preset drug identification set for each genotype of each preset gene locus is obtained through the following historical drug effectiveness determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each genotype of each of the preset gene loci, the clinical drug effect data whose gene test result at the preset gene locus is the genotype in the clinical drug effect data set is determined as the clinical drug effect data subset of the genotype, and based on the clinical drug effect data subset of the genotype, the following drug effectiveness determination operation is performed: for each preset drug identifier in the preset drug identifier set, the ratio of the number of drug-effective patients of the preset drug identifier divided by the number of patients using the drug of the preset drug identifier is determined as the historical drug effectiveness of the preset drug identifier for the genotype of the preset gene locus, wherein the number of drug-effective patients of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier and whose drug effect is effective, and the number of patients using the drug of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier.

[0009] In some optional embodiments, the determining of a clinical medication effect data set includes: Acquire a real clinical data set, wherein the real clinical data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and M drug effect determination labels, wherein M is a natural number greater than or equal to 2, and the M drug effect determination labels are used to represent whether the drug treatment for the clinical patients is effective or ineffective according to each of the M preset drug efficacy determination criteria; For each of the preset drug efficacy determination criteria, a weight coefficient of the preset drug efficacy determination criteria is determined according to the number of valid data of the preset drug efficacy determination criteria, wherein the weight coefficient of the preset drug efficacy determination criteria is negatively correlated with the number of valid data of the preset drug efficacy determination criteria, and the number of valid data of the preset drug efficacy determination criteria is the number of real clinical data in the real clinical data set that are determined to be valid according to the preset drug efficacy determination criteria; For each real clinical data in the real clinical data set, performing weighted summation on each medication effect determination label in the real clinical data according to the weight coefficient of the corresponding preset medication efficacy determination standard to obtain a medication effect evaluation value of the real clinical data, and determining the medication effect label of the real clinical data according to a comparison result between the medication effect evaluation value of the real clinical data and a preset medication effect evaluation value threshold; The corresponding clinical medication effect data in the clinical medication effect data set are generated using the gene test results, drug identification and medication effect label of each of the real clinical data.

[0010] In some optional embodiments, the real clinical data set is clinical data of clinical patients diagnosed with depression, and the M preset drug efficacy judgment criteria include at least two of the following judgment criteria: If the final Hamilton Depression Rating Scale HAMD score is less than or equal to 7 points, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If the reduction rate of the last HAMD score compared to the first HAMD score is greater than or equal to 50%, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If, during the treatment of a clinical patient, a certain antidepressant is used for more than or equal to 14 days without changing the type and dosage of the drug, the medication effect label corresponding to the antidepressant is set to valid; If a clinical patient is treated with an antidepressant for less than 14 days without changing the type and dosage of the drug, the medication effect label of the antidepressant will be set to invalid.

[0011] In some optional embodiments, the step of determining the drug identification to be recommended from each of the candidate drug identifications according to the drug effectiveness of each of the candidate drug identifications for the target patient comprises: For each of the candidate drug identifications, in response to determining that the drug effectiveness of the candidate drug identification for the target patient is greater than the average historical drug effectiveness of the candidate drug identification, the candidate drug identification is determined as a drug identification to be recommended.

[0012] In some optional implementations, the average historical drug effectiveness rate corresponding to each preset drug identification in the preset drug identification set is obtained by the following average historical drug effectiveness rate determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each preset drug identifier in the preset drug identifier set, the ratio of the number of patients with effective drugs for the preset drug identifier divided by the number of patients using the drugs for the preset drug identifier is determined as the average historical drug effectiveness rate of the preset drug identifier, wherein the number of patients with effective drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier and whose drug effects are effective, and the number of patients using the drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier.

[0013] In some optional embodiments, the drug effectiveness prediction model corresponding to each preset drug identification in the preset drug identification set is pre-trained through the following model training steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each of the preset drug identifiers, based on the data subset including the preset drug identifier in the clinical medication effect data set, the initial drug effectiveness prediction model corresponding to the preset drug identifier is trained to obtain the drug effectiveness prediction model corresponding to the preset drug identifier.

[0014] In a second aspect, an embodiment of the present disclosure provides a medication information recommendation device, the device comprising: A test result acquisition module is configured to acquire a gene test result of a target patient, wherein the gene test result includes a genotype of each gene locus in a preset gene locus set; The primary screening module is configured to determine, for each of the preset gene loci, a set of candidate drug identifiers for the target patient according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus; The efficacy prediction module is configured to predict the drug efficacy of the target patient for each candidate drug identifier based on the drug efficacy prediction model corresponding to the candidate drug identifier and using the gene test result of the target patient to obtain the drug efficacy of the candidate drug identifier for the target patient, wherein each preset drug identifier corresponds to a drug efficacy prediction model for characterizing the corresponding relationship between the gene test result and the drug efficacy; The recommended drug determination module is configured to determine a drug identification to be recommended from each candidate drug identification according to the drug effectiveness of each candidate drug identification for the target patient.

[0015] In some optional embodiments, the primary screening module includes: The single gene site primary screening unit is configured to determine, for each of the preset gene sites, a set of highly effective drug markers for the target patient at the preset gene site according to the historical drug effectiveness of each preset drug marker in the preset drug marker set for the genotype of the target patient at the preset gene site; The drug identification merging unit is configured to merge the high-efficiency drug identification sets of the target patient at each of the preset gene sites to obtain a candidate drug identification set of the target patient.

[0016] In some optional embodiments, the single gene site primary screening unit is further configured as follows: The L preset drug identifiers in the preset drug identifier set with the highest historical drug efficacy for the genotype of the target patient at the preset gene locus, and / or the preset drug identifiers in the preset drug identifier set with a historical drug efficacy for the genotype of the target patient at the preset gene locus greater than a preset historical drug efficacy threshold, are determined as the high-efficiency drug identifier set for the target patient at the preset gene locus, where L is a natural number.

[0017] In some optional embodiments, the historical drug effectiveness rate of each preset drug identification in the preset drug identification set for each genotype of each preset gene locus is obtained through the following historical drug effectiveness determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each genotype of each of the preset gene loci, the clinical drug effect data whose gene test result at the preset gene locus is the genotype in the clinical drug effect data set is determined as the clinical drug effect data subset of the genotype, and based on the clinical drug effect data subset of the genotype, the following drug effectiveness determination operation is performed: for each preset drug identifier in the preset drug identifier set, the ratio of the number of drug-effective patients of the preset drug identifier divided by the number of patients using the drug of the preset drug identifier is determined as the historical drug effectiveness of the preset drug identifier for the genotype of the preset gene locus, wherein the number of drug-effective patients of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier and whose drug effect is effective, and the number of patients using the drug of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier.

[0018] In some optional embodiments, the determining of a clinical medication effect data set includes: Acquire a real clinical data set, wherein the real clinical data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and M drug effect determination labels, wherein M is a natural number greater than or equal to 2, and the M drug effect determination labels are used to represent whether the drug treatment for the clinical patients is effective or ineffective according to each of the M preset drug efficacy determination criteria; For each of the preset drug efficacy determination criteria, a weight coefficient of the preset drug efficacy determination criteria is determined according to the number of valid data of the preset drug efficacy determination criteria, wherein the weight coefficient of the preset drug efficacy determination criteria is negatively correlated with the number of valid data of the preset drug efficacy determination criteria, and the number of valid data of the preset drug efficacy determination criteria is the number of real clinical data in the real clinical data set that are determined to be valid according to the preset drug efficacy determination criteria; For each real clinical data in the real clinical data set, performing weighted summation on each medication effect determination label in the real clinical data according to the weight coefficient of the corresponding preset medication efficacy determination standard to obtain a medication effect evaluation value of the real clinical data, and determining the medication effect label of the real clinical data according to a comparison result between the medication effect evaluation value of the real clinical data and a preset medication effect evaluation value threshold; The corresponding clinical medication effect data in the clinical medication effect data set are generated using the gene test results, drug identification and medication effect label of each of the real clinical data.

[0019] In some optional embodiments, the real clinical data set is clinical data of clinical patients diagnosed with depression, and the M preset drug efficacy judgment criteria include at least two of the following judgment criteria: If the final Hamilton Depression Rating Scale HAMD score is less than or equal to 7 points, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If the reduction rate of the last HAMD score compared to the first HAMD score is greater than or equal to 50%, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If, during the treatment of a clinical patient, a certain antidepressant is used for more than or equal to 14 days without changing the type and dosage of the drug, the medication effect label corresponding to the antidepressant is set to valid; If a clinical patient is treated with an antidepressant for less than 14 days without changing the type and dosage of the drug, the medication effect label of the antidepressant will be set to invalid.

[0020] In some optional embodiments, the recommended drug determination module is further configured to: For each of the candidate drug identifications, in response to determining that the drug effectiveness of the candidate drug identification for the target patient is greater than the average historical drug effectiveness of the candidate drug identification, the candidate drug identification is determined as a drug identification to be recommended.

[0021] In some optional implementations, the average historical drug effectiveness rate corresponding to each preset drug identification in the preset drug identification set is obtained by the following average historical drug effectiveness rate determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each preset drug identifier in the preset drug identifier set, the ratio of the number of patients with effective drugs for the preset drug identifier divided by the number of patients using the drugs for the preset drug identifier is determined as the average historical drug effectiveness rate of the preset drug identifier, wherein the number of patients with effective drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier and whose drug effects are effective, and the number of patients using the drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier.

[0022] In some optional embodiments, the drug effectiveness prediction model corresponding to each preset drug identification in the preset drug identification set is pre-trained through the following model training steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each of the preset drug identifiers, based on the data subset including the preset drug identifier in the clinical medication effect data set, the initial drug effectiveness prediction model corresponding to the preset drug identifier is trained to obtain the drug effectiveness prediction model corresponding to the preset drug identifier.

[0023] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0024] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by one or more processors, implements the method described in any implementation manner in the first aspect.

[0025] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program / instruction, which implements the method described in any implementation manner in the first aspect when the computer program / instruction is executed by a processor.

[0026] In order to solve the problem that reliable medication recommendation information cannot be formed due to inconsistent standards such as efficacy judgment in the current drug selection process based on pharmacogenomics, the present disclosure provides a medication information recommendation method, device, electronic device, storage medium and computer program product, which first obtains the genetic test results of the target patient, the genetic test results include the genotype of each gene locus in the preset gene locus set, and then for each preset gene locus, according to the historical drug effectiveness of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus, a set of candidate drug identifiers for the target patient is determined; then for each candidate drug identifier, based on the drug effectiveness prediction model corresponding to the candidate drug identifier, the target patient is predicted to have a drug effectiveness prediction model for the target patient using the genetic test results of the target patient, and the drug effectiveness of the candidate drug identifier for the target patient is obtained, wherein each preset drug identifier corresponds to a drug effectiveness prediction model for characterizing the corresponding relationship between the genetic test results and the drug effectiveness; finally, according to the drug effectiveness of each candidate drug identifier for the target patient, a drug identifier to be recommended is determined in each candidate drug identifier. That is, based on the statistical analysis results between drugs and patient genes, that is, the historical drug efficacy of each drug in different genotypes, a candidate drug identification set is preliminarily selected from the specified optional drugs (i.e., the preset drug identification set) according to the patient's genetic test results to narrow the screening range; then the drug efficacy prediction model of each candidate drug identification is used to further determine the drug efficacy of the corresponding candidate drug identification for the target patient, and then finally determine the drug identification to be recommended. Since the drug efficacy prediction model is pre-trained for each preset drug identification using the genetic test results of clinical patients who have used the drug indicated by the preset drug identification and the corresponding drug effect labels, the drug efficacy prediction model has higher reliability and robustness than the statistical analysis conclusions based on a small number of cases, and can ultimately improve the reliability of drug information recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other features, objects and advantages of the present disclosure will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the drawings: Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied; Figure 2A is a flow chart of an embodiment of a method for recommending medication information according to the present disclosure; Figure 2B is a decomposed flow chart of one embodiment of step 202 according to the present disclosure; Figure 3Ais a flow chart of one embodiment of the steps of determining historical drug effectiveness according to the present disclosure; Figure 3B is a decomposed flow chart of one embodiment of step 301 according to the present disclosure; Figure 4 is a flow chart of an embodiment of the steps of determining the average historical drug effectiveness rate according to the present disclosure; Figure 5 is a flowchart of an embodiment of a model training step according to the present disclosure; Figure 6 is a structural schematic diagram of an embodiment of a medication information recommendation device according to the present disclosure; Figure 7 It is a structural diagram of a computer system of an electronic device suitable for implementing an embodiment of the present disclosure. DETAILED DESCRIPTION

[0028] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0029] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0030] Figure 1 An exemplary system architecture 100 is shown to which embodiments of the medication information recommendation method, apparatus, electronic device, and storage medium of the present disclosure can be applied.

[0031] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0032] The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 101, 102, 103, such as medication information recommendation applications, etc.

[0033] Terminal devices 101, 102, 103 can be hardware or software. When terminal devices 101, 102, 103 are hardware, they can be various electronic devices with information input devices (e.g., keyboard, mouse, touch screen, microphone, camera, etc.) and information output devices (e.g., display screen, speaker, etc.), including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Compression Standard Audio Layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Compression Standard Audio Layer 4) players, laptop computers and desktop computers, etc. When terminal devices 101, 102, 103 are software, they can be installed in the terminal devices listed above. It can be implemented as multiple software or software modules (for example, used to provide medication information recommendation services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0034] In some cases, the medication information recommendation method provided by the present disclosure may be executed by the terminal devices 101, 102, 103, and accordingly, the medication information recommendation apparatus may be disposed in the terminal devices 101, 102, 103. In this case, the system architecture 100 may not include the server 105.

[0035] In some cases, the medication information recommendation method provided by the present disclosure can be jointly executed by the terminal devices 101, 102, 103 and the server 105. For example, the step of "obtaining the genetic test results of the target patient" can be executed by the terminal devices 101, 102, 103, and the steps of "for each preset gene locus, according to the historical drug effectiveness of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus, determine the candidate drug identifier set for the target patient" can be executed by the server 105. The present disclosure does not limit this. Accordingly, the medication information recommendation device can also be respectively set in the terminal devices 101, 102, 103 and the server 105.

[0036] In some cases, the medication information recommendation method provided in the present disclosure may be executed by the server 105 , and accordingly, the medication information recommendation device may also be set in the server 105 . In this case, the system architecture 100 may not include the terminal devices 101 , 102 , and 103 .

[0037] It should be noted that the server 105 can be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server 105 is software, it can be implemented as multiple software or software modules (for example, for providing distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0038] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0039] Continue to refer Figure 2A , which shows a process 200 of an embodiment of a medication information recommendation method according to the present disclosure, the medication information recommendation method comprises the following steps: Step 201, obtaining the genetic test results of the target patient.

[0040] Here, the genetic test result of the target patient is the result obtained by performing genetic testing on each gene locus of the target patient in the preset gene locus set, and specifically may include the genotype of each gene locus of the target patient in the preset gene locus set.

[0041] For example, the genotype for genetic testing of the SLC6A4 gene may be: LL, LS, SL, SS, etc.

[0042] In practice, the preset gene locus set can be set according to actual conditions.

[0043] Optionally, the preset gene locus set is a gene locus set for a preset disease. As an example, the preset disease may be depression. In order to find a drug suitable for patients with depression, the preset gene locus set may include, for example, CYP2D6, CYP2C19, CYP1A2, HTR1A, SLC6A4, COMT and BDNF. Accordingly, the target patient may be subjected to genetic testing of the above-mentioned gene loci, and the corresponding test results may be obtained as the genetic test results of the target patient.

[0044] Step 202 , for each preset gene locus, a candidate drug identification set for the target patient is determined according to the historical drug efficacy of each preset drug identification in the preset drug identification set for the genotype of the target patient at the preset gene locus.

[0045] Here, for each preset drug identifier in the preset drug identifier set, each genotype of each locus in the preset gene locus set is associated with a corresponding historical drug effectiveness rate. Specifically, for each genotype of each preset gene locus, patients who historically had the genotype at the preset gene locus can be grouped according to the different drugs indicated in the preset drug identifier set that they have taken. For each preset drug identifier, the ratio of the number of patients who took the drug indicated by the preset drug identifier and were effective divided by the total number of patients who took the drug indicated by the preset drug identifier is determined as the historical drug effectiveness rate of the preset drug identifier for the genotype of the preset gene locus.

[0046] Optionally, the preset drug identification set may be a drug identification set of drugs for treating a preset disease. For example, drug identifications for depression may include, but are not limited to: fluoxetine, paroxetine, sertraline, citalopram, venlafaxine, duloxetine, bupropion, amitriptyline, clomipramine, phenelzine, agomelatine, ketamine and its derivatives, dextromethorphan, etc.

[0047] When choosing antidepressants, individualized treatment is required based on the patient's specific symptoms, side effects of the drugs, and individual differences of the patients, and the drugs should be used under the guidance of a doctor.

[0048] In some optional implementations, step 202 may include: Figure 2B The following steps 2021 and 2022 are shown: Step 2021, for each preset gene locus, according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus, determine a set of highly effective drug identifiers for the target patient at the preset gene locus.

[0049] That is, assuming that there are I gene loci in the preset gene locus set, specifically {G1, ..., G i , …, G I}, targeting gene locus G i Yes K i There are J preset drug identifiers in the preset drug identifier set, specifically {D1, ..., D j , …, D J For each gene locus in I gene loci (for example, gene locus G i ), each of the J preset drug identifiers (e.g., drug identifier D j ) for this gene locus G i Each genotype (e.g., genotype T k (k ranges from 1 to K iThe natural number)) corresponds to the corresponding historical drug efficacy E i,j,k .

[0050] Assume that the target patient is at gene locus G i The genotype is T p , p ranges from 1 to K i A natural number.

[0051] Here, step 2021 may be for each preset gene locus (e.g., gene locus G i ), according to J preset drug identifiers {D1, ..., D j , …, D J Each drug marker in} has a genotype of T p The historical drug efficacy i,1,p , …, E i,j,p , …, E i,J,p}, determine the target patient at gene locus G i The set of highly effective drug identifiers. That is, in {E i,1,p , …, E i,j,p , …, E i,J,p Find the relatively high E i,j,p , use the corresponding preset drug identifier D j Generate target patient at gene locus G i The high-efficiency drug identification set F i .

[0052] After step 221, a set F of highly effective drug identifiers for each gene locus of the target patient in I preset gene loci can be obtained. i , that is, we can get a set of I high-efficiency drug identifiers F i .

[0053] In some optional embodiments, the L preset drug identifications with the highest historical drug efficacy for the genotype of the target patient at the preset gene locus in the preset drug identification set, and / or the preset drug identifications with the historical drug efficacy for the genotype of the target patient at the preset gene locus greater than the preset historical drug efficacy threshold in the preset drug identification set, can be determined as the high-efficiency drug identification set for the target patient at the preset gene locus. Here, L is a natural number, and optionally, L is a natural number greater than or equal to 2.

[0054] Step 2022: merge the high-efficiency drug identification sets of the target patient at each preset gene site to obtain a candidate drug identification set for the target patient.

[0055] Continuing with the above example, after step 221, a set of highly effective drug identifiers F for each gene locus of the target patient in I preset gene loci can be obtained.i , that is, we can get a set of I high-efficiency drug identifiers F i . It is understandable that there may be duplicate drug identifiers between the high-efficiency drug identifier sets of two different gene loci, so the high-efficiency drug identifier sets of the target patient at each preset gene locus can be merged to obtain a candidate drug identifier set for the target patient. Furthermore, there is no duplication between the candidate drug identifiers in the candidate drug identifier set of the target patient obtained after the merger, and each candidate drug identifier has a relatively high historical drug efficacy for the genotype of the target patient at at least one preset gene locus, that is, the candidate drug identifier set obtained through the preliminary screening of step 202 can narrow the screening range of subsequent steps.

[0056] Step 203, for each candidate drug identifier, based on the drug effectiveness prediction model corresponding to the candidate drug identifier, the drug effectiveness of the target patient is predicted using the gene test results of the target patient to obtain the drug effectiveness of the candidate drug identifier for the target patient.

[0057] Here, each preset drug marker corresponds to a drug effectiveness prediction model for characterizing the corresponding relationship between the gene test result and the drug effectiveness.

[0058] Since each candidate drug identifier also comes from the preset drug identifier set, for each candidate drug identifier, based on the drug efficacy prediction model corresponding to the candidate drug identifier, the target patient's gene test results can be used to predict the drug efficacy of the target patient, and the drug efficacy of the candidate drug identifier for the target patient can be obtained. Here, the drug efficacy of the candidate drug identifier for the target patient is used to characterize the probability or degree of effectiveness of the drug indicated by the candidate drug identifier for the target patient.

[0059] Optionally, for each candidate drug marker, the gene test result of the target patient may be input into a drug effectiveness prediction model corresponding to the candidate drug marker to obtain the drug effectiveness of the candidate drug marker for the target patient.

[0060] Optionally, for each candidate drug marker, the genetic test results of the target patient can be used as a sample to perform SHAP (SHapley Additive ex Planations) analysis on the drug effectiveness prediction model corresponding to the candidate drug marker to obtain the SHAP value of the target patient at each preset gene site, and then the sum of the SHAP values ​​of each preset gene site is used as the drug effectiveness of the candidate drug marker for the target patient.

[0061] After step 203 , the drug effectiveness of each candidate drug identification for the target patient can be obtained.

[0062] Here, the drug effectiveness prediction model corresponding to the preset drug identification can be pre-trained through the following medication information recommendation steps: Step 204 , determining a drug identifier to be recommended from among the candidate drug identifiers according to the drug efficacy of each candidate drug identifier for the target patient.

[0063] After step 203, the drug effectiveness of each candidate drug identification for the target patient can be obtained. Various implementation methods can be adopted here to determine the drug identification to be recommended from each candidate drug identification according to the drug effectiveness of each candidate drug identification for the target patient.

[0064] In some optional embodiments, step 204 may be performed as follows: for each candidate drug identification, in response to determining that the drug effectiveness of the candidate drug identification for the target patient is greater than the average historical drug effectiveness of the candidate drug identification, the candidate drug identification is determined as a drug identification to be recommended.

[0065] Here, each preset drug identification corresponds to a corresponding average historical drug effectiveness rate. The average historical drug effectiveness rate corresponding to each preset drug identification is used to represent the proportion of patients who have been effectively treated with the drug indicated by the preset drug identification in standard history. Since the candidate drug identification comes from the preset drug identification set, each candidate drug identification also corresponds to an average historical drug effectiveness rate. If the drug effectiveness rate of a target patient of a candidate drug identification is greater than the average historical drug effectiveness rate of the candidate drug identification, indicating that the probability that the candidate drug identification is effective for the target patient has exceeded the average historical effectiveness rate, then the candidate drug identification can be determined as a drug identification to be recommended.

[0066] Optionally, the execution subject of the above-mentioned medication recommendation information method may further perform the following step 205 after step 204: Step 205: Generate medication recommendation information based on the identification of each drug to be recommended.

[0067] Here, the execution subject may generate medication recommendation information based on each to-be-recommended drug identifier in various ways. For example, medication recommendation information may be generated directly using each to-be-recommended drug identifier. For another example, medication recommendation information may be generated using each to-be-recommended drug identifier and the drug effectiveness of the corresponding to-be-recommended drug identifier for the target patient. For another example, each to-be-recommended drug identifier may be sorted in descending order of drug effectiveness for the target patient, and medication recommendation information may be generated using the sorted to-be-recommended drug identifiers.

[0068] Optionally, the above-mentioned execution entity may also output the generated medication recommendation information for doctors' reference in medication.

[0069] In some optional embodiments, the historical drug effectiveness of each preset drug identifier in the preset drug identifier set used in step 202 for each genotype of each gene locus in the preset gene locus set can be obtained by Figure 3A The historical drug effectiveness determination step 300 shown in FIG. 3 is obtained, and the historical drug effectiveness determination step 300 includes the following steps 301 and 302: Specifically, for each genotype of each preset gene locus, patients who historically had the genotype at the preset gene locus can be grouped according to the different drugs indicated in the preset drug identification set that they have taken. For each preset drug identification, the ratio of the number of patients who took the drug indicated by the preset drug identification and were effective divided by the total number of patients who took the drug indicated by the preset drug identification is used to determine the historical drug effectiveness rate of the preset drug identification for the genotype at the preset gene locus.

[0070] Step 301, determine a clinical medication effect data set.

[0071] Here, the clinical drug effect data may include genetic test results of clinical patients who have received drug treatment, drug identifiers of drugs used to treat clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective.

[0072] Optionally, the clinical patient here may be a patient diagnosed with a preset disease, and the clinical drug effect data of the clinical patient may also be the corresponding gene test results, drug identifiers of drug treatment, and corresponding drug effect labels of the clinical patient during the treatment of the preset disease. Accordingly, the drug identifiers in the clinical drug effect data may belong to a preset drug identifier set, the preset drug identifier set may also be a drug identifier set of drugs for treating the preset disease, and the preset gene locus set involved in the gene test results may also be a gene locus set for the preset disease.

[0073] Here, the medication effect label can be expressed in various specific ways. For example, 0 can be used to indicate ineffectiveness, and 1 can be used to indicate effectiveness.

[0074] The medication effect labels in the clinical medication effect data may be determined in various ways based on real clinical data.

[0075] For example, the medication effect label may be manually determined by a technical expert with professional medical knowledge.

[0076] In order to solve the problem that the results of determining whether a drug is effective in treating clinical patients are not uniform according to different efficacy determination standards, thereby resulting in the problem that reliable drug information recommendation results cannot be formed, optionally, step 301 may include the following: Figure 3BThe following steps 3011 to 3014 are shown: Step 3011, obtaining a real clinical data set.

[0077] Here, the real clinical data may include the genetic test results of clinical patients who have received drug treatment, the drug identification of the drug used for drug treatment of clinical patients, and M drug effect determination labels. Among them, M is a natural number greater than or equal to 2. The M drug effect determination labels are used to characterize whether the drug treatment indicated by the drug identification is effective or ineffective for clinical patients according to each of the M preset drug efficacy determination standards. That is, the M drug effect determination labels correspond one-to-one to the M preset drug efficacy determination standards. Each real clinical data includes N drug effect determination labels for M different preset drug efficacy determination standards.

[0078] As an example, when the preset drug identification set and the preset gene locus set are both for depression, the real clinical data set is clinical data for patients diagnosed with depression, and the M preset drug efficacy judgment criteria may include at least two of the following judgment criteria: Standard 1: If the final Hamilton Depression Rating Scale HAMD score is less than or equal to 7 points, the medication effect judgment label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of other antidepressants are set to invalid.

[0079] Standard 2: If the reduction rate of the last HAMD score compared to the first HAMD score is greater than or equal to 50%, the medication effect judgment label corresponding to the antidepressant drug finally used by the corresponding clinical patient is set to valid based on the clinical data, and if the clinical patient has taken other antidepressants, the medication effect labels of other antidepressants are set to invalid.

[0080] Standard 3: If, during the treatment of a clinical patient, an antidepressant is used for greater than or equal to 14 days without changing the type and dosage of the drug, the medication effect label corresponding to the antidepressant is set to valid.

[0081] Standard 4: If a clinical patient is treated with an antidepressant for less than 14 days without changing the type and dosage of the drug, the medication effect label of the antidepressant should be set as invalid.

[0082] Step 3012: for each preset drug efficacy determination standard, determine the weight coefficient of the preset drug efficacy determination standard according to the number of valid data of the preset drug efficacy determination standard.

[0083] Here, the weight coefficient of each drug efficacy judgment standard is negatively correlated with the number of valid data of the corresponding preset drug efficacy judgment standard. The number of valid data of each preset drug efficacy judgment standard is the number of real clinical data in the real clinical data set that are judged to be valid according to the preset drug efficacy judgment standard.

[0084] Since each real clinical data includes M drug effect judgment labels for M different drug efficacy judgment standards. Assume that there are U real clinical data in the real clinical data set. According to the 1st, ..., mth, ..., Mth drug efficacy judgment standards in the M preset drug efficacy judgment standards, the U real clinical data have N1, ..., N m ,…,N M The number of real clinical data that are judged to be valid is N1,…,N m ,…,N M are all less than or equal to U. Accordingly, the weight coefficient W of the mth drug efficacy judgment criterion m The number of real clinical data N that are determined to be valid according to the drug efficacy determination criteria in Article m m Negative correlation.

[0085] As an example, the following is a method for calculating the weight coefficient W of the mth drug efficacy judgment criterion: m The specific calculation formula is: In the above formula, m is a natural number between 1 and M.

[0086] Step 3013, for each real clinical data in the real clinical data set, weighted sum is performed on each medication effect judgment label in the real clinical data according to the weight coefficient of the corresponding preset medication efficacy judgment standard to obtain the medication effect evaluation value of the real clinical data, and based on the comparison result between the medication effect evaluation value of the real clinical data and the preset medication effect evaluation value threshold, the medication effect label of the real clinical data is determined.

[0087] As an example, 0 represents the ineffective medication effect determination label, and 1 represents the ineffective medication effect determination label. According to the above formula, step 3013 can be performed as follows: Then, for a real clinical data, the M drug effect judgment labels are O1, ..., O m , …, O M , where O m 0 means that the drug is ineffective according to the mth drug efficacy judgment standard, m 0 means that the drug is effective according to the mth drug efficacy judgment standard. The drug efficacy evaluation value of the real clinical data can be calculated as follows : From the above formula, it can be seen that for the mth drug efficacy judgment standard, the larger Wm is, the higher the drug efficacy evaluation value of the real clinical data is. Also bigger.

[0088] After obtaining the medication effect evaluation value of the real clinical data, the medication effect evaluation value of the real clinical data can be compared with the preset medication effect evaluation value threshold. If the medication effect evaluation value of the real clinical data is greater than the preset medication effect evaluation value threshold, the medication effect label of the real clinical data can be determined to be valid (for example, the medication effect label of the real clinical data is set to 1); conversely, if the medication effect evaluation value of the real clinical data is not greater than the preset medication effect evaluation value threshold, the medication effect label of the real clinical data can be determined to be invalid (for example, the medication effect label of the real clinical data is set to 0).

[0089] Step 3014, using the gene test results, drug identification and drug effect label of each real clinical data to generate corresponding clinical drug effect data in the clinical drug effect data set.

[0090] Here, the principle of the above optional implementation is explained as follows: If the mth drug efficacy judgment standard is judged to be valid, the number of real clinical data N m The larger the m-th drug efficacy judgment standard is, the looser it is, that is, the easier it is to judge that the drug is effective. Then the influence of the m-th drug efficacy judgment standard in the final process of generating the drug effect label for the real clinical data should be smaller, so as not to cause the m-th drug efficacy judgment standard to have an adverse effect on the final drug effect label generated for the real clinical data. That is, the weight coefficient W of the m-th drug efficacy judgment standard is m The smaller it should be.

[0091] On the contrary, if N m The smaller it is, the stricter the m-th drug efficacy judgment standard is, that is, it is more difficult to judge that the drug is effective; then the influence of the m-th drug efficacy judgment standard in the final generation of the drug effect label for the real clinical data should be greater, that is, the influence of the m-th drug efficacy judgment standard on the final generated drug effect label for the real clinical data should be increased, that is, the weight coefficient W of the m-th drug efficacy judgment standard m The larger it should be.

[0092] The medication effect labels generated by the real clinical data finally determined by the above-mentioned optional implementation method (i.e., the medication effect labels in the clinical medication effect data) comprehensively consider the medication effect determination labels generated according to different drug efficacy determination standards, and the influence of each preset drug efficacy determination standard on the finally generated medication effect label is related to the strictness of the corresponding preset drug efficacy determination standard, which can improve the reliability and robustness of the finally generated medication effect label.

[0093] Step 302, for each genotype of each preset gene locus, the clinical drug effect data of the genotype whose gene test result at the preset gene locus in each clinical drug effect data set is determined as the clinical drug effect data subset of the genotype, and based on the clinical drug effect data subset of the genotype, a drug effectiveness determination operation is performed.

[0094] Here, the drug effectiveness determination operation can be performed as follows: For each preset drug marker in the preset drug marker set, the ratio of the number of patients with effective drugs for the preset drug marker divided by the number of patients using the drug for the preset drug marker is determined as the historical drug effectiveness rate of the preset drug marker for the genotype of the preset gene locus. The number of patients with effective drugs for the preset drug identification is the number of clinical medication effect data in the clinical medication effect data subset of the genotype whose drug identification is the preset drug identification and whose medication effects are effective, and the number of patients using drugs for the preset drug identification is the number of clinical medication effect data in the clinical medication effect data subset of the genotype whose drug identification is the preset drug identification.

[0095] Here, for ease of understanding, continue to use the assumptions in step 2021 above: Assume that there are I gene loci in the preset gene loci set, specifically {G1, ..., G i , …, G I}, where for gene locus G i YesN i There are J preset drug identifiers in the preset drug identifier set, specifically {D1, ..., D j , …, D J}.

[0096] Then step 302 specifically refers to: For each genotype (e.g., genotype Tn among Ni genotypes of gene locus Gi) of each preset gene locus (e.g., gene locus Gi), the clinical drug effect data with a gene test result of Tn at gene locus Gi in each clinical drug effect data set is determined as a clinical drug effect data subset Sn of genotype Tn, and a drug effectiveness determination operation is performed based on the clinical drug effect data subset Sn of genotype Tn. Specifically, the drug effectiveness determination operation based on the clinical drug effect data subset Sn of genotype Tn can be performed as follows: is the calculated drug identifier D j Targeting gene locus G i Genotype T n historical drug effectiveness; G i Genotype T n A subset of clinical drug efficacy data S n The Chinese medicine identification is drug identification D j And the number of clinical medication effect data that are effective; G i Genotype T n A subset of clinical drug efficacy data S n The Chinese medicine identification is drug identification D j The number of clinical drug effect data.

[0097] In some optional implementations, each preset drug identifier used in step 204 corresponds to a corresponding average historical drug effectiveness rate, which can be obtained by Figure 4 The average historical drug effectiveness determination step is obtained, and the average historical drug effectiveness determination step includes the following steps 401 and 402: Step 401, determine a clinical medication effect data set.

[0098] Here, the specific implementation of step 401 and the technical effects it brings can refer to the relevant records of step 301 and its optional implementation methods, which will not be repeated here.

[0099] It should be noted that the genotype of the clinical patients is not of concern here, but the average historical drug effectiveness of the drug in all genotypes is of concern. Therefore, the clinical drug effect data here may not include the genetic test results of the clinical patients, or may include the genetic test results. However, the gene factor will not be considered in the subsequent statistical analysis and calculation in step 402.

[0100] Of course, it is understandable that in order to solve the problem of inconsistent results in determining whether a drug is effective in treating clinical patients according to different efficacy determination criteria, step 401 may also be used. Figure 3B The same or similar implementation of steps 3011 to 3014 shown in the figure determines a clinical medication effect data set.

[0101] Step 402: for each preset drug identification in the preset drug identification set, the ratio of the number of patients with effective drugs for the preset drug identification divided by the number of patients using the drugs for the preset drug identification is determined as the average historical drug effectiveness rate of the preset drug identification.

[0102] Here, the number of patients with effective drugs for the preset drug identifier is the number of clinical medication effect data in the clinical medication effect data set whose drug identifier is the preset drug identifier and whose medication effects are effective, and the number of patients using the drug for the preset drug identifier is the number of clinical medication effect data in the clinical medication effect data set whose drug identifier is the preset drug identifier.

[0103] In some optional embodiments, the drug efficacy prediction model corresponding to each preset drug identifier can be obtained by Figure 5 The model training steps shown are pre-trained, and the model training steps include the following steps 501 and 502: Step 501, determine a clinical medication effect data set.

[0104] Here, the specific implementation method of step 501 and the technical effects it brings can refer to the relevant records of step 301 and its optional implementation methods, which will not be repeated here.

[0105] Of course, it is understandable that in order to solve the problem of inconsistent results in determining whether a drug is effective in treating clinical patients according to different efficacy determination criteria, step 501 may also be used. Figure 3B Steps 3011 to 3014 are similarly implemented to determine a clinical medication effect data set.

[0106] Step 502, for each preset drug identifier, based on the drug identifier in the clinical medication effect data set including the data subset of the preset drug identifier, an initial drug effectiveness prediction model corresponding to the preset drug identifier is trained to obtain a drug effectiveness prediction model corresponding to the preset drug identifier.

[0107] Specifically, the initial drug efficacy prediction model corresponding to each preset drug identifier can be various regression models used to calculate the drug efficacy based on the gene test results. As an example, it can be the following regression models: linear regression models such as multivariate linear regression model, ridge regression, lasso regression, elastic net regression, or polynomial regression, support vector regression (SVR), Gaussian process regression, or nonlinear regression models such as decision tree regression, gradient boosting tree regression, extreme gradient boosting tree regression (XGBoost Regression), light gradient boosting tree regression (LightGBM Regression), classification and regression tree (CART) regression, random forest regression, or Bayesian regression, generalized linear model (GLM), etc.

[0108] During the training process, the genetic test results in the data subset including the preset drug identification in the clinical drug effect data set can be used as input data, and the genetic test results can be input into the initial drug effectiveness prediction model of the preset drug identification to obtain the actual output drug effectiveness for the preset drug identification, and then based on the difference between the above-mentioned output drug effectiveness for the preset drug identification and the medication effect label corresponding to the input data, the model parameters of the initial drug effectiveness prediction model of the preset drug identification are adjusted, and finally the trained drug effectiveness prediction model of the preset drug identification is obtained.

[0109] The drug efficacy prediction model for each preset drug marker trained by the above optional method is based on real and effective clinical data, which can improve the reliability of the drug efficacy finally calculated based on the drug efficacy prediction model for each preset drug marker. Figure 3B Steps 3011 to 3014 shown are determined in a similar implementation manner. The medication effect label comprehensively considers the medication effect determination labels generated according to different drug efficacy determination standards, and the influence of each preset drug efficacy determination standard on the finally generated medication effect label is related to the strictness of the corresponding preset drug efficacy determination standard, which can improve the reliability and robustness of the medication effect label, and the reliability and robustness of the drug effectiveness prediction model trained based on the clinical medication effect data including the above-mentioned medication effect labels can also be improved accordingly.

[0110] The above-mentioned embodiments of the present disclosure provide a method for recommending medication information, which first obtains the genetic test results of the target patient, wherein the genetic test results include the genotype of each gene locus in a preset gene locus set, and then for each preset gene locus, according to the historical drug effectiveness of each preset drug identification in the preset drug identification set for the genotype of the target patient at the preset gene locus, a set of candidate drug identifications for the target patient is determined; then for each candidate drug identification, based on a drug effectiveness prediction model corresponding to the candidate drug identification, the genetic test results of the target patient are used to predict the drug effectiveness of the target patient, and the drug effectiveness of the candidate drug identification for the target patient is obtained, wherein each preset drug identification corresponds to a drug effectiveness prediction model for characterizing the corresponding relationship between the genetic test results and the drug effectiveness; finally, according to the drug effectiveness of each candidate drug identification for the target patient, a drug identification to be recommended is determined in each candidate drug identification. That is, based on the statistical analysis results between drugs and patient genes, that is, the historical drug efficacy of each drug in different genotypes, a candidate drug identification set is preliminarily selected from the specified optional drugs (i.e., the preset drug identification set) according to the patient's genetic test results to narrow the screening range; then the drug efficacy prediction model of each candidate drug identification is used to further determine the drug efficacy of the corresponding candidate drug identification for the target patient, and then finally determine the drug identification to be recommended. Since the drug efficacy prediction model is pre-trained for each preset drug identification using the genetic test results of clinical patients who have used the drug indicated by the preset drug identification and the corresponding drug effect labels, the drug efficacy prediction model has higher reliability and robustness than the statistical analysis conclusions based on a small number of cases, and can ultimately improve the reliability of drug information recommendations.

[0111] Further references Figure 6 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a medication information recommendation device. Figure 2A Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0112] like Figure 6As shown, the medication information recommendation device 600 of this embodiment includes: a test result acquisition module 601, a preliminary screening module 602, an effectiveness prediction module 603 and a recommended drug determination module 604. Among them: the test result acquisition module 601 is configured to obtain the gene test result of the target patient, and the gene test result includes the genotype of each gene locus in the preset gene locus set; the initial screening module 602 is configured to determine the candidate drug identification set of the target patient for each of the preset gene loci according to the historical drug effectiveness of each preset drug identification in the preset drug identification set for the genotype of the target patient at the preset gene locus; the effectiveness prediction module 603 is configured to predict the drug effectiveness of the target patient for each of the candidate drug identifications based on the drug effectiveness prediction model corresponding to the candidate drug identification and using the gene test result of the target patient to obtain the drug effectiveness of the candidate drug identification for the target patient, wherein each preset drug identification corresponds to a drug effectiveness prediction model for characterizing the corresponding relationship between the gene test result and the drug effectiveness; the recommended drug determination module 604 is configured to determine the drug identification to be recommended in each of the candidate drug identifications according to the drug effectiveness of each candidate drug identification for the target patient.

[0113] In this embodiment, the specific processing of the detection result acquisition module 601, the initial screening module 602, the efficiency prediction module 603 and the recommended drug determination module 604 of the medication information recommendation device 600 and the technical effects thereof can be referred to in Figure 2A The relevant descriptions of step 201, step 202, step 203 and step 204 in the corresponding embodiment are not repeated here.

[0114] In some optional embodiments, the primary screening module 602 may include: The single gene site primary screening unit 6021 is configured to determine, for each of the preset gene sites, a set of highly effective drug markers for the target patient at the preset gene site according to the historical drug effectiveness of each preset drug marker in the preset drug marker set for the genotype of the target patient at the preset gene site; The drug identification merging unit 6022 is configured to merge the high-efficiency drug identification sets of the target patient at each of the preset gene sites to obtain a candidate drug identification set of the target patient.

[0115] In some optional embodiments, the single gene site primary screening unit 6021 can be further configured as follows: The L preset drug identifiers in the preset drug identifier set with the highest historical drug efficacy for the genotype of the target patient at the preset gene locus, and / or the preset drug identifiers in the preset drug identifier set with a historical drug efficacy for the genotype of the target patient at the preset gene locus greater than a preset historical drug efficacy threshold, are determined as the high-efficiency drug identifier set for the target patient at the preset gene locus, where L is a natural number.

[0116] In some optional embodiments, the historical drug effectiveness rate of each preset drug marker in the preset drug marker set for each genotype of each preset gene locus may be obtained through the following historical drug effectiveness rate determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each genotype of each of the preset gene loci, the clinical drug effect data whose gene test result at the preset gene locus is the genotype in the clinical drug effect data set is determined as the clinical drug effect data subset of the genotype, and based on the clinical drug effect data subset of the genotype, the following drug effectiveness determination operation is performed: for each preset drug identifier in the preset drug identifier set, the ratio of the number of drug-effective patients of the preset drug identifier divided by the number of patients using the drug of the preset drug identifier is determined as the historical drug effectiveness of the preset drug identifier for the genotype of the preset gene locus, wherein the number of drug-effective patients of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier and whose drug effect is effective, and the number of patients using the drug of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier.

[0117] In some optional embodiments, the determining of the clinical medication effect data set may include: Acquire a real clinical data set, wherein the real clinical data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and M drug effect determination labels, wherein M is a natural number greater than or equal to 2, and the M drug effect determination labels are used to represent whether the drug treatment for the clinical patients is effective or ineffective according to each of the M preset drug efficacy determination criteria; For each of the preset drug efficacy determination criteria, a weight coefficient of the preset drug efficacy determination criteria is determined according to the number of valid data of the preset drug efficacy determination criteria, wherein the weight coefficient of the preset drug efficacy determination criteria is negatively correlated with the number of valid data of the preset drug efficacy determination criteria, and the number of valid data of the preset drug efficacy determination criteria is the number of real clinical data in the real clinical data set that are determined to be valid according to the preset drug efficacy determination criteria; For each real clinical data in the real clinical data set, performing weighted summation on each medication effect determination label in the real clinical data according to the weight coefficient of the corresponding preset medication efficacy determination standard to obtain a medication effect evaluation value of the real clinical data, and determining the medication effect label of the real clinical data according to a comparison result between the medication effect evaluation value of the real clinical data and a preset medication effect evaluation value threshold; The corresponding clinical medication effect data in the clinical medication effect data set are generated using the gene test results, drug identification and medication effect label of each of the real clinical data.

[0118] In some optional embodiments, the real clinical data set may be clinical data of clinical patients diagnosed with depression, and the M preset drug efficacy determination criteria may include at least two of the following determination criteria: If the final Hamilton Depression Rating Scale HAMD score is less than or equal to 7 points, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If the reduction rate of the last HAMD score compared to the first HAMD score is greater than or equal to 50%, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If, during the treatment of a clinical patient, a certain antidepressant is used for more than or equal to 14 days without changing the type and dosage of the drug, the medication effect label corresponding to the antidepressant is set to valid; If a clinical patient is treated with an antidepressant for less than 14 days without changing the type and dosage of the drug, the medication effect label of the antidepressant will be set to invalid.

[0119] In some optional implementations, the recommended drug determination module 604 may be further configured to: For each of the candidate drug identifications, in response to determining that the drug effectiveness of the candidate drug identification for the target patient is greater than the average historical drug effectiveness of the candidate drug identification, the candidate drug identification is determined as a drug identification to be recommended.

[0120] In some optional implementations, the average historical drug effectiveness rate corresponding to each preset drug identification in the preset drug identification set is obtained by the following average historical drug effectiveness rate determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each preset drug identifier in the preset drug identifier set, the ratio of the number of patients with effective drugs for the preset drug identifier divided by the number of patients using the drugs for the preset drug identifier is determined as the average historical drug effectiveness rate of the preset drug identifier, wherein the number of patients with effective drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier and whose drug effects are effective, and the number of patients using the drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier.

[0121] In some optional embodiments, the drug effectiveness prediction model corresponding to each preset drug identification in the preset drug identification set is pre-trained through the following model training steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each of the preset drug identifiers, based on the data subset including the preset drug identifier in the clinical medication effect data set, the initial drug effectiveness prediction model corresponding to the preset drug identifier is trained to obtain the drug effectiveness prediction model corresponding to the preset drug identifier.

[0122] It should be noted that the implementation details and technical effects of each module in the medication information recommendation device provided in the embodiments of the present disclosure can be referred to the description of other embodiments in the present disclosure, and will not be repeated here.

[0123] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 suitable for implementing the electronic device of the present disclosure. Figure 7 The computer system 700 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0124] like Figure 7 As shown, the computer system 700 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage device 708 to a random access memory (RAM) 703. Various programs and data required for the operation of the computer system 700 are also stored in the RAM 703. The processing device 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0125] Typically, the following devices may be connected to the I / O interface 705: input devices 706 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, etc.; output devices 707 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 708 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 709. The communication devices 709 may allow the computer system 700 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 The computer system 700 of the electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0126] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 709, or installed from a storage device 708, or installed from a ROM 702. When the computer program is executed by the processing device 701, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0127] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0128] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0129] The computer readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device implements the following Figure 2A The illustrated embodiment and its optional implementations illustrate a method for recommending medication information.

[0130] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Python, Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0131] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0132] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, a test result acquisition module may also be described as a "module for acquiring the genetic test results of a target patient."

[0133] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

Claims

1. A method for recommending medication information, comprising: Obtaining a genetic test result of a target patient, wherein the genetic test result includes a genotype of each gene locus in a preset gene locus set; For each of the preset gene loci, determining a set of candidate drug identifiers for the target patient according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus; For each of the candidate drug identifiers, based on the drug effectiveness prediction model corresponding to the candidate drug identifier, the drug effectiveness of the target patient is predicted using the gene test results of the target patient to obtain the drug effectiveness of the candidate drug identifier for the target patient, wherein each preset drug identifier corresponds to a drug effectiveness prediction model for characterizing the corresponding relationship between the gene test results and the drug effectiveness; According to the drug effectiveness of each candidate drug identification for the target patient, a drug identification to be recommended is determined from each candidate drug identification.

2. The method according to claim 1, wherein: For each of the preset gene loci, determining a set of candidate drug identifiers for the target patient according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus includes: For each of the preset gene loci, according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus, determine a set of highly effective drug identifiers for the target patient at the preset gene locus; The highly effective drug identification sets of the target patient at each of the preset gene sites are merged to obtain a candidate drug identification set of the target patient.

3. The method according to claim 2, wherein: The step of determining a set of highly effective drug markers for the target patient at the preset gene locus according to the historical drug effectiveness of each preset drug marker in the preset drug marker set for the genotype of the target patient at the preset gene locus comprises: The L preset drug identifiers in the preset drug identifier set with the highest historical drug efficacy for the genotype of the target patient at the preset gene locus, and / or the preset drug identifiers in the preset drug identifier set with a historical drug efficacy for the genotype of the target patient at the preset gene locus greater than a preset historical drug efficacy threshold, are determined as the high-efficiency drug identifier set for the target patient at the preset gene locus, where L is a natural number.

4. The method according to claim 2, wherein: The historical drug effectiveness rate of each preset drug identification in the preset drug identification set for each genotype of each preset gene locus is obtained through the following historical drug effectiveness determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each genotype of each of the preset gene loci, the clinical drug effect data whose gene test result at the preset gene locus is the genotype in the clinical drug effect data set is determined as the clinical drug effect data subset of the genotype, and based on the clinical drug effect data subset of the genotype, the following drug effectiveness determination operation is performed: for each preset drug identifier in the preset drug identifier set, the ratio of the number of drug-effective patients of the preset drug identifier divided by the number of patients using the drug of the preset drug identifier is determined as the historical drug effectiveness of the preset drug identifier for the genotype of the preset gene locus, wherein the number of drug-effective patients of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier and whose drug effect is effective, and the number of patients using the drug of the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data subset of the genotype whose drug identifier is the preset drug identifier.

5. The method according to claim 4, wherein: The clinical medication effect data set is determined, including: Acquire a real clinical data set, wherein the real clinical data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and M drug effect determination labels, wherein M is a natural number greater than or equal to 2, and the M drug effect determination labels are used to represent whether the drug treatment for the clinical patients is effective or ineffective according to each of the M preset drug efficacy determination criteria; For each of the preset drug efficacy determination criteria, a weight coefficient of the preset drug efficacy determination criteria is determined according to the number of valid data of the preset drug efficacy determination criteria, wherein the weight coefficient of the preset drug efficacy determination criteria is negatively correlated with the number of valid data of the preset drug efficacy determination criteria, and the number of valid data of the preset drug efficacy determination criteria is the number of real clinical data in the real clinical data set that are determined to be valid according to the preset drug efficacy determination criteria; For each real clinical data in the real clinical data set, performing weighted summation on each medication effect determination label in the real clinical data according to the weight coefficient of the corresponding preset medication efficacy determination standard to obtain a medication effect evaluation value of the real clinical data, and determining the medication effect label of the real clinical data according to a comparison result between the medication effect evaluation value of the real clinical data and a preset medication effect evaluation value threshold; The corresponding clinical medication effect data in the clinical medication effect data set are generated using the gene test results, drug identification and medication effect label of each of the real clinical data.

6. The method according to claim 5, wherein: The real clinical data set is clinical data of clinical patients diagnosed with depression, and the M preset drug efficacy judgment criteria include at least two of the following judgment criteria: If the final Hamilton Depression Rating Scale HAMD score is less than or equal to 7 points, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If the reduction rate of the last HAMD score compared to the first HAMD score is greater than or equal to 50%, the medication effect determination label corresponding to the antidepressant drug finally used by the clinical patient is set to valid, and if the clinical patient has taken other antidepressants, the medication effect labels of the other antidepressants are set to invalid; If, during the treatment of a clinical patient, a certain antidepressant is used for more than or equal to 14 days without changing the type and dosage of the drug, the medication effect label corresponding to the antidepressant is set to valid; If a clinical patient is treated with an antidepressant for less than 14 days without changing the type and dosage of the drug, the medication effect label of the antidepressant will be set to invalid.

7. The method according to claim 1, wherein: The step of determining the drug identification to be recommended from each of the candidate drug identifications according to the drug effectiveness of each of the candidate drug identifications for the target patient comprises: For each of the candidate drug identifications, in response to determining that the drug effectiveness of the candidate drug identification for the target patient is greater than the average historical drug effectiveness of the candidate drug identification, the candidate drug identification is determined as a drug identification to be recommended.

8. The method according to claim 1, wherein: The average historical drug effectiveness rate corresponding to each preset drug identification in the preset drug identification set is obtained through the following average historical drug effectiveness rate determination steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each preset drug identifier in the preset drug identifier set, the ratio of the number of patients with effective drugs for the preset drug identifier divided by the number of patients using the drugs for the preset drug identifier is determined as the average historical drug effectiveness rate of the preset drug identifier, wherein the number of patients with effective drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier and whose drug effects are effective, and the number of patients using the drugs for the preset drug identifier is the number of clinical drug effect data in the clinical drug effect data set whose drug identifier is the preset drug identifier.

9. The method according to claim 1, wherein: The drug effectiveness prediction model corresponding to each preset drug identification in the preset drug identification set is pre-trained through the following model training steps: Determine a set of clinical drug effect data, wherein the clinical drug effect data includes gene test results of clinical patients who have received drug treatment, drug identifiers of drugs used for drug treatment of the clinical patients, and drug effect labels indicating whether the drug treatment is effective or ineffective; For each of the preset drug identifiers, based on the data subset including the preset drug identifier in the clinical medication effect data set, the initial drug effectiveness prediction model corresponding to the preset drug identifier is trained to obtain the drug effectiveness prediction model corresponding to the preset drug identifier.

10. A medication information recommendation device, comprising: A test result acquisition module is configured to acquire a gene test result of a target patient, wherein the gene test result includes a genotype of each gene locus in a preset gene locus set; The primary screening module is configured to determine, for each of the preset gene loci, a set of candidate drug identifiers for the target patient according to the historical drug efficacy of each preset drug identifier in the preset drug identifier set for the genotype of the target patient at the preset gene locus; The efficacy prediction module is configured to predict the drug efficacy of the target patient for each candidate drug identifier based on the drug efficacy prediction model corresponding to the candidate drug identifier and using the gene test result of the target patient to obtain the drug efficacy of the candidate drug identifier for the target patient, wherein each preset drug identifier corresponds to a drug efficacy prediction model for characterizing the corresponding relationship between the gene test result and the drug efficacy; The recommended drug determination module is configured to determine a drug identification to be recommended from each candidate drug identification according to the drug effectiveness of each candidate drug identification for the target patient.

11. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, wherein: When the computer program is executed by one or more processors, the method according to any one of claims 1 to 9 is implemented.

13. A computer program product, comprising a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images