KR20200143153A - Apparatus and systems for verifying the efficacy of a drug and food acting on a target mechanism using artificial intelligence - Google Patents

Abstract

One disclosure provides a method of verifying efficacy of drugs acting on a target gene by using artificial intelligence, capable of effectively analyzing the efficacy of the drugs. The method includes: periodically acquiring medical treatment data of patients to which a first drug prepared to cure a disease is administered; classifying medical treatment data of patients of which diseases are cured after the administration of the first drug among the medical treatment data into a first category, and classifying medical treatment data of patients of which diseases are not cured after the administration of the first drug among the medical treatment data into a second category; deriving a target gene from genes of the patients included in the first category by using a first artificial intelligence learning model in which correlation between a gene expression distribution due to the disease and a gene expression distribution due to a component of the first drug is trained; acquiring a biomarker profile from gene expression information of the patients included in the second category by analyzing an expression state of the target gene and an expression state of other genes by the first drug; and determining whether to administer a second drug, which is usable to treat the diseases of the patients included in the second category, and a content of the second drug by using a second artificial intelligence learning model in which correlation of the target gene with the component and a content of the first drug is trained based on the biomarker profile.

Description

Apparatus and systems for verifying the efficacy of a drug and food acting on a target mechanism using artificial intelligence.

According to one disclosure, the present invention is an apparatus and method for verifying the effectiveness of a specific drug and derive a new drug by analyzing the genes of a patient group with symptom relief and a patient group whose symptoms are not alleviated by using artificial intelligence. It is about.

Artificial Intelligence (AI) system is a computer system that realizes human-level intelligence, and unlike existing rule-based smart systems, machines learn, judge, and become smarter. As the artificial intelligence system is used, the recognition rate improves and the user's taste can be understood more accurately, and the existing rule-based smart system is gradually being replaced by a deep learning-based artificial intelligence system.

Artificial intelligence technology consists of machine learning (deep learning) and component technologies using machine learning. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and element technology is a technology that utilizes machine learning algorithms such as deep learning, such as linguistic understanding, visual understanding, reasoning/prediction, knowledge expression, motion control, etc. Consists of the technical fields of.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology that recognizes and applies/processes human language/text, and includes natural language processing, machine translation, dialogue systems, question and answer, and speech recognition/synthesis. Visual understanding is a technology that recognizes and processes objects like human vision, and includes object recognition, object tracking, image search, human recognition, scene understanding, spatial understanding, and image improvement. Inference prediction is a technique that logically infers and predicts information by judging information, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification), knowledge management (data utilization), and the like. Motion control is a technology that controls autonomous driving of a vehicle and movement of a robot, and includes movement control (navigation, collision, driving), operation control (behavior control), and the like.

With the advancement of artificial intelligence and digital technology, healthcare technology is undergoing major changes, and it has become an important task to effectively manage new data by creating, collecting and tracking new data. In particular, in the field of medical technology, as the importance of not only personal health but also various industrial scalability and health welfare management is increasing, interest in digital medical technology that can manage medical data and derive appropriate medical diagnosis is increasing. Are doing.

Korean Patent Publication No. 2019-0043639 (published on April 19, 2019) Korean Patent Registration Publication No. 2019-0036620 (published on April 5, 2019)

An object of the present invention is to provide an apparatus and method capable of effectively analyzing the effectiveness of a drug by monitoring whether a drug specialized for a specific disease and a disease cure for a patient receiving the drug.

According to the first embodiment, a method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence is provided, and the method periodically provides treatment data of patients receiving the first drug prepared to cure a disease. The step of obtaining, among the treatment data, classifying treatment data of patients whose disease has been cured after the first drug administration into a first category, and classifying treatment data of patients whose disease has not been cured after the first drug administration into a second category, Deriving a target gene from the genes of patients included in the first category using a first artificial intelligence learning model that learned the association between the gene expression distribution due to the disease and the gene expression distribution by the component of the first drug, 2 Obtaining a biomarker profile by analyzing the expression state of the target gene and the expression state of other genes by the first drug from the gene expression information of the patients included in the category, and the target gene and the first Using a second artificial intelligence learning model that has learned the association between the components and contents of drugs, determine whether to administer the second drug and the amount of the second drug that can be used to treat diseases of patients included in the second category. It may include the step of.

According to the second embodiment, an apparatus for verifying the effectiveness of a drug acting on a target gene using artificial intelligence is provided, the apparatus comprising a processor and a memory storing executable instructions, the processor executing the instructions, Periodically acquiring medical data of patients who have been administered the first drug manufactured to cure disease, and among the medical data, medical data of patients whose disease has been cured after the first drug is administered as the first category, after the first drug is administered. The first artificial intelligence learning model is used to classify the medical data of patients whose disease has not been cured into a second category, and learn the relationship between gene expression distribution, first drug component, first drug content, disease and cell function. Thus, by deriving the target gene from the genes of patients included in the first category, and analyzing the expression state of the target gene and the expression state of other genes by the first drug from the gene expression information of the patients included in the second category. Acquiring a biomarker profile, and using a second artificial intelligence learning model that learned the association between the target gene and the component and content of the first drug based on the biomarker profile, it is possible to treat diseases of patients included in the second category. At least one of the amount of the first drug that can be used, whether the second drug is administered, and the amount of the second drug may be determined.

According to the third embodiment, it is possible to provide a computer-readable non-transitory recording medium in which a program for implementing a method of verifying the effectiveness of a drug acting on a target gene using artificial intelligence is recorded.

By initiation, it is possible to verify the effectiveness of the drug by tracking whether patients with the same disease using the same drug have cured the disease, and in particular, analyze the same genome of the patient group whose disease has been cured and derive the target gene of the disease. It can lead to other prescriptions for patients who have not cured the disease.

In addition, by deriving a target gene of a disease according to the present invention and analyzing the same genome of a patient group that has not been cured of the disease, it is possible to provide an effect of deriving other drugs that can be effectively applied to the corresponding genome.

According to one disclosure, medical data may be analyzed based on existing medical information, and a digitization platform for various medical information may be constructed based on a result of analyzing a bio-signal obtained from a user.

1 is a view for explaining the flow of a method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence by one disclosure.
FIG. 2 is a diagram for describing a feature of generating a first artificial intelligence learning model according to one disclosure.
3 is a diagram for describing a feature of generating a second artificial intelligence learning model according to one disclosure.
Fig. 4 is a diagram for explaining a pathway for a drug to act on a gene according to one initiation.
5 is a view for explaining the flow of a method of verifying the effectiveness of a drug by one initiation and recommending another drug.
6 is a view for explaining the configuration of an apparatus for verifying the effectiveness of a drug acting on a target gene using artificial intelligence according to one disclosure.
FIG. 7 is a diagram illustrating a configuration of a processor that generates an artificial intelligence learning model through data learning according to one disclosure.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments disclosed below. In addition, parts irrelevant to the present invention are omitted in the drawings in order to clearly disclose the present invention, and the same or similar reference numerals denote the same or similar components in the drawings.

Objects and effects of the present invention may be naturally understood or more clearly understood by the following description, and the objects and effects of the present invention are not limited only by the following description.

Objects, features and advantages of the present invention will become more apparent through the following detailed description. In addition, in describing the present invention, when it is determined that a detailed description of a known technology related to the present invention may unnecessarily obscure the subject matter of the present invention, a detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Hereinafter, the device 100 for verifying the effectiveness of a drug acting on a target gene using artificial intelligence is abbreviated as the drug validity verification device 100 to be used.

1 is a view for explaining the flow of a method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence by one disclosure.

According to an initiation, in block 101, the drug validity verification apparatus 100 may periodically acquire treatment data of patients who have been administered the first drug manufactured to cure a disease.

By one initiation the disease is, for example, a cancer, a neoplasm, a tumor, and/or a form of metastasis therein, a metabolic disorder, an inflammatory disease, or an infectious disease. In some embodiments, the cancer, neoplasm, tumor, or metastatic form therein is, for example, leukemia, lymphoma, prostate cancer, lung cancer, breast cancer, liver cancer, colorectal cancer, or kidney cancer. All other diseases that occur in the human body may be included.

According to one initiation, the first drug is a drug formulated with a component suitable for curing a disease, and acts effectively on a target gene of the disease.

In block 102 according to the start of the day, the drug validity verification apparatus 100 includes the treatment data of patients whose disease has been cured after the first drug administration among the treatment data as a first category, and the patients whose disease has not been cured after the first drug administration. Treatment data can be classified into a second category.

Patients included in the first category by the start of the day refer to patients whose disease has been cured by administering a first drug prepared with a predetermined component. Herein, the patients included in the first category did not receive the first drug by the same dosage method, and may have received the first drug by different dosage methods according to each patient's condition. Patients included in the first category were cured of the disease due to the first drug, indicating that the first drug effectively acted on a specific gene of the patients included in the first category.

Patients included in the second category by the start of the day include patients whose disease has not been cured even though they have been administered the first drug prepared with a predetermined component. Here, the patients included in the second category did not receive the first drug by the same dosage method, and may have received the first drug by different dosage methods according to the condition of each patient. Patients included in the second category indicated that the disease was not cured by the first drug, and thus the first drug did not effectively act on specific genes of the patients included in the second category. In other words, patients included in the second category are patients who need other methods to cure the disease by improving the existing first drug or taking other drugs.

By the start of the day, in block 103, the drug validity verification apparatus 100 uses a first artificial intelligence learning model that learns the relationship between the gene expression distribution due to the disease and the gene expression distribution due to the component of the first drug. Target genes can be derived from genes of patients included in the category.

By the start of the day, the first artificial intelligence learning model is learned to derive a target gene of a disease, and the change in the expression state of the gene due to the disease, the change in the expression state of the gene due to the first drug, and the other due to the first drug. It was obtained based on the result of unsupervised learning of the relationship between changes in the expression state of genes.

Genes that are targets for analysis in DNA by one initiation may include MAM domain containing glycosylphosphatidylinositol anchor 1 (MDGA1). The MDGA1 may be composed of a polynucleotide including bases from 37617785 to 37618223 on chromosome 6 or a complementary polynucleotide thereof. In addition, the target gene may further include CYP4V2 (cytochrome P450 family 4 subfamily V member 2). The CYP4V2 may be composed of a polynucleotide including a base 187124479 to a base 187125005 on chromosome 4 or a complementary polynucleotide thereof. The target gene may further include one or more of ADAP1 (Arf-GAP with dual PH domain-containing protein 1), LHX6 (LIM homeobox 6), and RASA3 (Ras GTPase-activating protein 3). The ADAP1 may be composed of a polynucleotide including bases 948157 to 949189 bases on chromosome 7 or a complementary polynucleotide thereof. The LHX6 may be composed of a polynucleotide including bases from 124989781 th to 124989950 th base on chromosome 9 or a complementary polynucleotide thereof. The RASA3 may be composed of a polynucleotide including bases from 114813885 to 114814341 on chromosome 13 or a complementary polynucleotide thereof.

The target gene may further include one or more of a polymeric immunoglobulin receptor (PIGR), disintegrin and metalloproteinase domain-containing protein 12 (ADAM12), and chromosome 21 open reading frame 128 (C21orf128). In addition, the PIGR may be composed of a polynucleotide including bases from 207105468th to 207105762th base on chromosome 1 or a complementary polynucleotide thereof. The ADAM12 may be composed of a polynucleotide including bases from 127822964 th base to 127823174 th base on chromosome 10 or a complementary polynucleotide thereof. The C21orf128 may be composed of a polynucleotide including the 43528602th base to the 43529012th base or a complementary polynucleotide thereof on chromosome 21.

Additionally, the target gene may be a gene present on chromosome 2. Specifically, the target gene may further include at least one of G protein-coupled receptor 35 (GPR35) and stonin 1 (STON1).

GPR35 may be composed of a polynucleotide including bases from 241562679 to 241562995 on chromosome 2 or a complementary polynucleotide thereof. The STON1 may be composed of a polynucleotide including bases from 48983601 to 48983821 on chromosome 2 or a complementary polynucleotide thereof. The target gene may further include a Parkin coregulated gene protein (PACRG) present on chromosome 6.

By the start of the day, the drug validity verification device 100 in block 104 analyzes the expression state of the target gene and the expression state of other genes by the first drug from the gene expression information of the patients included in the second category, You can obtain the profile.

According to one disclosure, the biomarker profile refers to data for clarifying the reason why the same drug was not cured despite having the same disease. For example, let's assume a case where it is determined that the 6th gene is the target gene from people whose disease has been cured by the first drug. Those included in the second category did not cure the disease because the modified state (expression or inhibition) of gene 6 by the first drug was not sufficient to cure the disease. However, if people in the 2nd category commonly had a modification of the 12th gene, the data on the relationship between the 1st drug and the 12th gene could be the biomarker profile, and the 12th gene could be the biomarker. have.

In block 105 by the start of the day, the drug validity verification device 100 uses a second artificial intelligence learning model that learns the relationship between the target gene and the component and content of the first drug based on the biomarker profile, and uses the second category. Whether to administer the second drug and the content of the second drug that can be used for the treatment of diseases of patients included in may be determined.

That is, since the drug validation device 100 cannot cure patients included in the second category in the same manner as the patients included in the first category, other treatments are provided to cure the patients included in the second category. You can figure out a way. Since the second AI learning model was created using medical treatment data of patients whose disease has not been cured even though the first drug was administered, a new treatment method for each of the patients included in the second category can be created based on this. I can.

According to one disclosure, the second drug may include a drug having the same component as the first drug, but a different content of the component, and may be a drug including a component not included in the first drug. The second drug is used for the purpose of curing the same disease as acting on the first drug. By one initiation, the second drug may contain at least one new drug.

In some embodiments, the computer-based analysis program converts raw data generated by a detection assay (e.g., the presence, absence or amount of a given marker or marker) to a data value (e.g., drug target or drug (S) is used to translate as optional). Clinicians can access the data using any suitable means. Thus, in some preferred embodiments, the present invention provides the additional advantage that a clinician who is not educated in genetics or molecular biology does not need to understand the raw data. The data is provided directly to the clinician in its most useful form. The clinician can then immediately use the information to optimize the subject's healing.

The present invention contemplates any method by which information may be received, processed, and transmitted to or from laboratories, informants, healthcare workers and subjects performing assays. For example, in some embodiments of the present invention, a sample (e.g., a biopsy or other sample) is obtained from a subject, and anywhere in the world (e.g., the country in which the subject resides or the information is ultimately used). To a profiling service (e.g., a clinical laboratory in a medical facility, a genome profiling business, etc.), located in a country other than that of the country. If the sample contains a tissue or other biological sample, the subject may visit a medical center to provide the sample and send it to the profiling center, or the subject may himself collect a sample (e.g., a urine sample) and , You can send it directly to the profiling center. If the sample contains biological information that has already been measured, the information can be passed by the subject to the profiling center (e.g., an information card containing the information is scanned with a computer and the data is transferred to an electronic communication system. After delivery to the profiling service, the sample is processed, and a profile specific to the desired diagnostic, treatment or prognostic information for the subject is created (eg, protein Expression data).

By one disclosure, the profile is created in a format suitable for interpretation by the attending physician. For example, without providing raw expression data, the resulting format can represent a proposed therapeutic course of action (eg, specific drug for administration). Data can be presented to the clinician by any suitable method. For example, in some embodiments, the profiling service generates a report that can be printed for the clinician (eg, at a point of care) or presented to the clinician on a computer monitor.

In some embodiments, the information may first be analyzed at the treatment site or at a local facility. The raw data is then passed to a central processing facility for further analysis and/or to convert the raw data into useful information for the clinician or patient. Central processing facilities offer the advantages of privacy (all data is stored in a central facility with a uniform security protocol), speed, and uniformity in data analysis. The central processing facility can then control the fate of the data after the subject's treatment. For example, using an electronic communication system, a central facility can provide data to a clinician, subject, or researcher.

FIG. 2 is a diagram for describing a feature of generating a first artificial intelligence learning model according to one disclosure.

The first artificial intelligence learning model is learned to derive the target gene of a disease by the initiation of the day, and the change in the expression state of the gene due to the disease, the change in the expression state of the gene due to the first drug, and other genes due to the first drug. It was obtained based on the result of unsupervised learning of the relationship between the change in the expression state of.

The drug validity verification device 100 is based on a change in the expression state of a gene due to a disease, a change in the expression state of a gene due to the first drug, a change in the expression state of another gene due to the first drug, and the function of the cell and the type of disease. Iterative learning can be performed to derive a disease target gene. The drug validity verification apparatus 100 may generate a first artificial intelligence learning model for deriving a target gene of a disease after repeated learning.

The first artificial intelligence learning model may be generated as a result of unsupervised learning of the initial deep neural network by inputting training data into the input layer of the initial deep neural network, and an intermediate deep neural network with an updated initial connection strength. And the learning data may be information on expression change genes for each of a plurality of gene datasets.

Using a Deep Belief Network method of learning the current layer by performing propagation and backpropagation in the current layer by one start, and learning the next layer of the current layer when learning of the current layer is completed. The initial connection strength may be updated.

The drug validity verification apparatus 100 sets training data to an input layer of an intermediate deep neural network, sets an output value designated for a type of training data input to the input layer to an output layer of the intermediate deep neural network, and sets the input layer of the intermediate deep neural network A final deep neural network may be generated by supervised learning the intermediate deep neural network based on a value set in the and output layers.

The type of learning data by the drug validity verification device 100 includes any one of a first type representing a normal gene dataset extracted from a normal population and a second type representing a disease gene dataset extracted from a patient group, and the learning unit When the training data corresponding to the first type is input to the input layer of the intermediate deep neural network, the intermediate deep neural network is trained by setting the first output value to the output layer of the intermediate deep neural network, and the training data corresponding to the second type is stored in the intermediate deep layer. When inputting to the input layer of the neural network, the intermediate deep neural network may be trained by setting a second output value in the output layer of the intermediate deep neural network.

The drug validation device 100 compares a prediction value obtained by calculating propagation in the forward direction from the input layer of the intermediate deep neural network and an output value set in the output layer of the intermediate deep neural network to obtain an error, and then backpropagates in a direction to minimize the error. Thus, each connection strength of the intermediate deep neural network can be updated.

3 is a diagram for describing a feature of generating a second artificial intelligence learning model according to one disclosure.

According to one initiation, the drug validity verification device 100 includes a component of a first drug, a content of the first drug, a change in the expression state of a target gene due to the first drug, a change in the expression state of another gene due to the first drug, and a second drug. Based on the components and content of the drug, whether or not to alleviate the disease, and the function of cells, it is possible to perform repeated learning to find other treatment methods for the disease. The drug validity verification apparatus 100 may generate a second artificial intelligence learning model through repeated learning.

According to the start of the day, the second artificial intelligence learning model is learned to derive other treatment methods when the disease is not treated due to the first drug, and the expression state of the target gene according to the change in the composition and content of the first drug It was obtained based on the result of learning the relationship between the change and the change in the expression state of the target gene according to the change and the change in the component and content of the second drug.

By initiation, the second artificial intelligence learning model extracts the pathway in which the first drug acts on the target gene, and according to the change in the content of the first drug, the activation and inhibition reactions of other genes included in the pathway ( inhibition), and by analyzing the active response and inhibitory response of other genes included in the pathway according to the change in the content of the second drug, a new treatment method for patients included in the second category is determined. Here, the new treatment method includes a change in the content of the first drug, administration of another new drug, and the number of times the drug is administered.

In the second AI learning model, AGER, THBS2, CA3, MMP12, and AGER, THBS2, CA3, MMP12, and protein of the target gene of the second category are compared with the state of the protein of the target gene of the first category by testing a tissue sample from a patient diagnosed with a disease. MMP-1, MMP-7, MMP-9, MMP-13, MMP-8, MMP-10, MMP-2, PIGR, DCN, PGAM1, CD36, FABP, ACP5, CCDC80, PPBP, LYVE1, STC1, SPON1, At least one selected from IL17RC, MMP1, CA1, SERPINC1, TPSB2, CKB/CKBM, NAMPT/PBEF, PPBP/CTAPIII, F9, DCTPP1, F5, SPOCK2, CAT, PF4, MDK, BGN, CKM, POSTN, PGLYRP1 and CXCL12 The altered expression state of the protein of can be analyzed. According to one disclosure, the proteins listed above may be proteins that are targets of drugs.

In addition, according to the fold difference of at least one protein having an altered expression state of the analyzed protein, the content of the first drug may be increased or decreased, or the administration of at least one or more other drugs may be determined.

Fig. 4 is a diagram for explaining a pathway for a drug to act on a gene according to one initiation.

By initiation, the drug validity verification device 100 first extracts a pathway in which the drug acts on the target gene using the second artificial intelligence learning model, and the other genes included in the pathway according to the change in the content of the first drug By analyzing the activation and inhibition reactions of the second drug, and analyzing the activation and suppression reactions of other genes included in the pathway according to the change in the content of the second drug, new patients included in the second category You can decide how to treat it.

By initiation, the drug validity verification device 100 may recreate a new drug using a biological network including a plurality of gene information, and a biological network is an interaction between biological entities such as genes and proteins ( interaction).

Referring to (a) of FIG. 4, the right end of each pathway connecting between the acting gene, the intermediate gene and the disease gene is indicated by an arrow or a circle, where the arrow indicates activation, and the circle indicates Means inhibition.

The relationship between the drug 401 and the disease gene or the target disease can be derived based on the shortest path derived from the initiation and the activity/inhibition, and furthermore, the relationship between the drug 401 and the disease gene or the target disease A step of calculating a score by digitizing is performed.

This score can be calculated based on the number of intermediate genes arranged on the shortest path, whether the shortest path is active/inhibited, the relationship between the drug and the acting gene, and the state of the disease gene in the disease.

Referring to (B) of FIG. 4, the disease gene may be two of the first disease gene 403 and the second disease gene 404, which means that the drug 401 to be verified is two disease genes ( 404, 405).

On the other hand, the input drug 401 may act to treat the disease by reacting with the disease gene, and conversely, may act to make the disease worse. Looking at this in detail using (b) of FIG. 4, the drug 401 ) Acts to treat the disease by reacting with the first disease gene 404 that is suppressed in the disease state, and acts to cure the disease by inhibiting the second disease gene 405 that is activated in the disease state. do.

Extracting the shortest path between the effector genes 402 and 403 and the disease genes 404 and 405 includes extracting all the shortest paths that can link between the effector genes 402 and 403 and the disease genes 404 and 405, And extracting the shortest path between the working genes 402 and 403 and the disease genes 404 and 405 based on the number of at least one intermediate gene arranged on the path.

5 is a view for explaining the flow of a method of verifying the effectiveness of a drug by one initiation and recommending another drug.

The drug validity verification device 100 of the present invention may determine the effectiveness of the drug and determine the effectiveness of the dosage method.

First, the drug validity verification device 100 injects a first drug into patients suffering from the same disease, and refers to patients whose disease is cured by the first drug as a first category, and patients who are not cured as a second category. It can be classified as

The drug validity verification device 100 analyzes the expression state of the target gene of each of the patients included in the first category according to the component ratio of the first drug, the dose, the number of doses, the dose time, and the dose method. 1 The utility value of a drug can be calculated. In addition, the drug validity verification device 100 may determine a customized dosage method of the first drug for each of the patients included in the first category according to the utility value of the first drug.

As a result, although the disease has been cured using the same first drug, the drug validity verification device 100 may determine a dosage method that can effectively use the first drug in consideration of the patient's gene expression state more precisely.

In addition, the drug validity verification device 100 may check disease treatment states of patients included in the second category after administration of a new drug by the second artificial intelligence learning model. In addition, the drug validity verification device 100 is a component and content of the first drug, the component and content of the second drug, and the component and content of other drugs by the second artificial intelligence learning model re-learned by reflecting the disease treatment state. By re-determining, the treatment drugs of the patients of the second category may be determined. That is, the drug validity verification device 100 may provide a more accurate customized dosage method by reflecting the feedback from the patient and performing learning again based on this.

The drug validation device 100 analyzes the expression state of the target gene of each of the patients included in the second category according to the combination of the first drug and the second drug, and the synergy value of the combination of the first drug and the second drug Can be calculated.

In addition, the drug validity verification device 100 determines a customized dosage method of the first drug and the second drug for each of the patients included in the second category according to the synergy value of the combination of the first drug and the second drug. I can.

As used herein, the term cell refers to a mammalian cell, preferably a human cell. A single cell may be used as well as a plurality of cells according to the teachings of the present invention. Preferably, the plurality of cells comprises 10 or more and 500 or less cells. The cells are preferably present in a single suspension so that the number of cells can be counted even if adherent cells and aggregates are still detected. The plurality of cells can be obtained from all biological samples such as cell lines, primary cultures, and cell samples (e.g., biopsies (surgical biopsies including incisional biopsy or excisional biopsy, fine needle aspiration biopsy, etc.), complete resection or body fluids). Biopsy screening methods are known in the art.

Cells in a biological sample can be analyzed for functional enzymes without any pretreatment. Therefore, it is preferable that the cells in the biological sample are free cells (ie, whole cells) and cells that can grow. However, it is highly appreciated that pretreatment of cells such as cell extracts or generation of non-free cells is also examined by the present invention.

Diseases that can be detected by the method of the present invention include cancer, and cancer is expressed by cancer cells, and non-limiting examples of cancer cells include adenocarcinoma cells, adrenal gland tumor cells, and enamel epithelium. Ameloblastoma cells, anaplastic cells, anaplastic carcinoma of thyroid cells, angiofibroma cells, angioma cells, angiosarcoma cells, afudoma cells (apudoma cell), argentaffinoma cell, arrhenoblastoma cell, ascites tumor cell, ascitic tumor cell, astroblastoma cell, appearance Astrocytoma cells, ataxia-telangiectasia cells, atrial myxoma cells, basal cell carcinoma cells, benign tumor cells, osteosarcoma cells (bone cancer cell), bone tumor cell, brain stem glioma cell, brain tumor cell, breast cancer cell, Burkitt's lymphoma cell, Cancerous cells, carcinoid cells, carcinoma cells, cerebellar astrocytoma cells, cerebellar astrocytoma cells, cervical cancer cells, cherry angioma cells , Cholangiocarcinoma cells, cholangioma cells, chondroblastoma cells, chondroma cells, chondrosarcoma cells, chorioblastoma cells, chorioblastoma cells ( choriocarcinoma cells), colon cancer cells, common acute lymphoblastic leukemia cells, craniopharyngioma cells, cystocarcinoma cells, cystocarcinoma cells, and cystofbroma cells. cell), cystoma cell, cytoma cell, ductal carcinoma in situ cell, ductal papilloma cell, dysgerminoma cell, brain tumor Cells (encephaloma cells), endometrial carcinoma cells (endometrial carcinoma cells), endothelioma cells (endothelioma cells), epithelioma cells (ependymoma cells), epithelioma cells (epithelioma cells), erythroleukemia cells (erythroleukemia cells), Ewing's sarcoma Ewing's sarcoma cells, extra nodal lymphoma cells, feline sarcoma cells, fibro adenoma cells, fibro sarcoma cells, follicular adenocarcinoma cells of the thyroid gland (follicular cancer of the thyroid cell), ganglioglioma cell, gastrinoma cell, polymorphic glioblastoma cell (g lioblastoma multiform cells), glioma cells, gonadoblastoma cells, hemangioblastoma cells, haemangioendothelioblastoma cells, haemangioendothelioma cells, perivascular Hemangiopericytoma cells, hematolymphangioma cells, haemocytoblastoma cells, haemocytoma cells, hairy cell leukemia cells, Hamartoma cells, hepatocarcinoma cells, hepatocellular carcinoma cells, hepatoma cells, histoma cells, Hodgkin's disease cells, adrenal tumors Cells (hypernephroma cells), infiltrating cancer cells (infiltrating cancer cells), infiltrating ductal cell carcinoma cells (infiltrating ductal cell carcinoma cells), insulinoma cells (insulinoma cells), juvenile angioforoma cells (juvenile angioforoma cells), Kaposi's sarcoma Cells (Kaposi sarcoma cells), kidney tumor cells (kidney tumor cells), large cell lymphoma cells (large cell lymphoma cells), leukemia cells (leukemia cells), chronic leukemia cells (chronic leukemia cells), acute leukemia cells (acute leukemia cells) ), lipoma cells, liver cancer cells, liver metastases cell), Luke carcinoma cells, lymphadenoma cells, lymphangioma cells, lymphocytic leukemia cells, lymphocytic lymphoma cells, lymphoid cucumbers Lymphoeytoma cells, lymphoedema cells, lymphoma cells, lung cancer cells, malignant mesothelioma cells, malignant teratoma cells, Mastocytoma cells, medulloblastome cells, melanoma cells, meningioma cells, mesothelioma cells, metastatic cells, metastatic cells ( metastasis cell), metastatic spread cell, Morton's neuroma cell, multiple myeloma cell, myeloblastoma cell, myeloid leukemia cell, bone marrow Myelolipoma cells, myeloma cells, myoblastoma cells, myxoma cells, nasopharyngeal carcinoma cells, neoplastic cells, nephroblastoma cells cell), neuroblastoma cell, neurofibroma cell, neurofibromatosis cell, Glioma cells, neuroma cells, non-Hodgkin's lymphoma cells, oligodendroglioma cells, optic glioma cells, osteochondroma cells ( osteochondroma cells, osteosarcoma cells, osteosarcoma cells, ovarian cancer cells, Paget's disease of the nipple cells, pancoast tumor cells cell), pancreatic cancer cell, pheochromocytoma cell, pheoehromocytoma cell, plasmacytoma cell, primary brain tumor cell, fetal potential Progonoma cells, prolactinoma cells, renal cell carcinoma cells, retinoblastoma cells, rhabdomyosarcoma cells, rhabdosarcoma cells , Solid tumor cell, sarcoma cell, secondary tumor cell, semi-noma cell, skin cancer cell, small cell carcinoma cell ), squamous cell carcinoma cell, strawberry haemangioma cell, T-cell lymphoma cell, teratoma Cells (teratoma cells), testicular cancer cells (testicular cancer cells), thymoma cells (thymoma cells), trophoblastic tumor cells (trophoblastic tumor cells), tumorigenic cells (tumorigenic cells), tumor initiation cells (tumor initiation cells) , Tumor progression cells, vestibular schwannoma cells, Wilm's tumor cells, or a combination thereof.

6 is a view for explaining the configuration of an apparatus for verifying the effectiveness of a drug acting on a target gene using artificial intelligence according to one disclosure.

According to one disclosure, the drug validity verification apparatus 100 may include a processor 1300 and a memory 1100, but as shown in FIG. 6, it may be implemented with more configurations than essential components.

For example, as shown in FIG. 6, the drug validity verification apparatus 100 according to an embodiment includes a memory 1100, a display unit 1210, a camera 1610 and a processor 1300, an output unit ( 1200), a communication unit 1500, a sensing unit 1400, an A/V input unit 1600, and a user input unit 1700 may be further included.

The memory 1100 may store a program for processing and controlling the processor 1300, and may store an image input to the drug validation device 100 or guide information output from the drug validation device 100. . Also, the memory 1100 may store specific information for determining whether to output the guide information.

The memory 1100 is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (for example, SD or XD memory), and RAM. (RAM, Random Access Memory) SRAM (Static Random Access Memory), ROM (ROM, Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk And at least one type of storage medium among optical disks.

Programs stored in the memory 1100 can be classified into a plurality of modules according to their functions, for example, a UI module 1110, a touch screen module 1120, a notification module 1130, and the like. .

The UI module 1110 may provide a specialized UI, GUI, etc. that are linked with the drug validation device 100 for each application. The touch screen module 1120 may detect a user's touch gesture on the touch screen and transmit information on the touch gesture to the processor 1300. The touch screen module 1120 according to an embodiment may recognize and analyze a touch code. The touch screen module 1120 may be configured as separate hardware including a controller.

The notification module 1130 may generate a signal to notify the occurrence of an event of the drug validity verification device 100. Examples of events that occur in the drug validation device 100 include call signal reception, message reception, key signal input, schedule notification, and the like. The notification module 1130 may output a notification signal in the form of a video signal through the display unit 1210, may output a notification signal in the form of an audio signal through the sound output unit 1220, and the vibration motor 1230 It is also possible to output a notification signal in the form of a vibration signal through. For example, the notification module 1130 may generate a signal for outputting guide information based on the estimated lane information.

The output unit 1200 may output an audio signal, a video signal, or a vibration signal, and the output unit 1200 may include a display unit 1210, an audio output unit 1220, and a vibration motor 1230. have.

The display unit 1210 displays and outputs information processed by the drug validity verification apparatus 100. Specifically, the display unit 1210 may output an image captured by the camera 1610. Also, the display 1210 may synthesize and output the guide information generated by the processor 1300 with the captured image.

In addition, the display 1210 may display a user interface for executing an operation related to the response in response to a user's input.

The sound output unit 1220 outputs audio data received from the communication unit 1500 or stored in the memory 1100. In addition, the sound output unit 1220 outputs sound signals related to functions (eg, a call signal reception sound, a message reception sound, and a notification sound) performed by the drug validity verification apparatus 100. For example, the sound output unit 1220 may output guide information generated as a signal from the notification module 1130 as an sound signal under the control of the processor 1300.

The processor 1300 typically controls the overall operation of the drug validation device 100. For example, the processor 1300, by executing programs stored in the memory 1100, the user input unit 1700, the output unit 1200, the sensing unit 1400, the communication unit 1500, the A/V input unit 1700 ) And so on. In addition, the processor 1300 may perform a function of the drug validity verification apparatus 100 by executing programs stored in the memory 1100.

The sensing unit 1400 may sense a state of the drug validation device 100 or a state around the drug validation device 100 and transmit the detected information to the processor 1300.

The sensing unit 1400 includes a magnetic sensor 1410, an acceleration sensor 1420, a temperature/humidity sensor 1430, an infrared sensor 1440, a gyroscope sensor 1450, and a position sensor. (For example, a GPS) 1460, an atmospheric pressure sensor 1470, a proximity sensor 1480, and an RGB sensor 1490 may be included, but the present invention is not limited thereto. Since the function of each sensor can be intuitively inferred by a person skilled in the art from its name, a detailed description will be omitted.

According to an embodiment, the sensing unit 1400 may measure a distance between the vehicle and at least one object determined from the captured image.

The communication unit 1500 may include one or more components that allow the drug validation device 100 to communicate with other devices (not shown) and servers (not shown). Another device (not shown) may be a computing device such as the drug validation device 100 or a sensing device, but is not limited thereto. For example, the communication unit 1500 may include a short range communication unit 1510, a mobile communication unit 1520, and a broadcast reception unit 1530.

The short-range wireless communication unit 1510 includes a Bluetooth communication unit, a Bluetooth Low Energy (BLE) communication unit, a near field communication unit, a WLAN (Wi-Fi) communication unit, a Zigbee communication unit, and an infrared ( IrDA, infrared data association) communication unit, WFD (Wi-Fi Direct) communication unit, UWB (ultra wideband) communication unit, Ant+ communication unit, etc. may be included, but is not limited thereto.

The mobile communication unit 1520 transmits and receives a radio signal with at least one of a base station, an external terminal, and a server on a mobile communication network. Here, the wireless signal may include a voice call signal, a video call signal, or various types of data according to transmission/reception of text/multimedia messages.

The broadcast receiver 1530 receives a broadcast signal and/or broadcast-related information from outside through a broadcast channel. Broadcast channels may include satellite channels and terrestrial channels. According to an implementation example, the drug validity verification apparatus 100 may not include the broadcast reception unit 1530.

The A/V (Audio/Video) input unit 1600 is for inputting an audio signal or a video signal, and may include a camera 1610 and a microphone 1620. The camera 1610 may obtain an image frame such as a still image or a moving image through an image sensor in a video call mode or a photographing mode. The image captured through the image sensor may be processed through the processor 1300 or a separate image processing unit (not shown).

The microphone 1620 receives an external sound signal and processes it as electrical voice data. For example, the microphone 1620 may receive an acoustic signal from an external device or a user. The microphone 1620 may receive a user's voice input. The microphone 1620 may use various noise removal algorithms to remove noise generated in the process of receiving an external sound signal.

The user input unit 1700 refers to a means for a user to input data for controlling the drug validation device 100. For example, the user input unit 1700 includes a key pad, a dome switch, a touch pad (contact type capacitance method, pressure type resistive film method, infrared detection method, surface ultrasonic conduction method, integral type). Tension measurement method, piezo effect method, etc.), a jog wheel, a jog switch, and the like, but are not limited thereto.

FIG. 7 is a diagram illustrating a configuration of a processor that generates an artificial intelligence learning model through data learning according to one disclosure.

The processor 1300 according to an embodiment may include a data learning unit 1310 and a data recognition unit 1320.

The data learning unit 1310 may learn a gene expression state according to an action of a drug. The data learning unit 1310 may learn a criterion regarding which data to use and whether to expand the relationship graph by using the data in order to increase the accuracy of the relationship graph. The data learning unit 1310 acquires data to be used for learning and applies the acquired data to a data recognition model to be described later, thereby learning a criterion for forming an appropriate relationship network according to learning of the relationship graph.

According to an embodiment, when there are a plurality of pre-built data recognition models, the data learning unit 1310 may determine a data recognition model having a high correlation between input training data and basic training data as a data recognition model to be trained. have. In this case, the basic training data may be pre-classified by data type, and the data recognition model may be pre-built for each data type. For example, basic training data is classified based on various criteria such as the region where the training data was created, the time when the training data was created, the size of the training data, the genre of the training data, the creator of the training data, and the type of objects in the training data. Can be.

The data recognition unit 1320 may increase the accuracy of which drugs should be used to treat diseases by studying the expression states of drugs and target genes and other genes. The data recognition unit 1320 may increase the accuracy of recommending a new drug taking method by acquiring predetermined data according to a preset criterion by learning and using the data recognition model using the acquired data as an input value. . Also, a result value output by the data recognition model using the acquired data as an input value may be used to update the data recognition model.

At least one of the data learning unit 1310 and the data recognition unit 1320 may be manufactured in the form of at least one hardware chip and mounted on a device. For example, at least one of the data learning unit 1310 and the data recognition unit 1320 may be manufactured in the form of a dedicated hardware chip for artificial intelligence (AI), or an existing general-purpose processor (eg, a CPU Alternatively, it may be manufactured as a part of an application processor) or a graphics dedicated processor (eg, a GPU) and mounted on the aforementioned various devices.

In this case, the data learning unit 1310 and the data recognition unit 1320 may be mounted on one device or may be mounted on separate devices. For example, one of the data learning unit 1310 and the data recognition unit 1320 may be included in the device, and the other may be included in the server. In addition, the data learning unit 1310 and the data recognition unit 1320 may provide model information built by the data learning unit 1310 to the data recognition unit 1320 through wired or wireless communication, or the data recognition unit ( Data input to 1320 may be provided to the data learning unit 1310 as additional learning data.

Meanwhile, at least one of the data learning unit 1310 and the data recognition unit 1320 may be implemented as a software module. When at least one of the data learning unit 1310 and the data recognition unit 1320 is implemented as a software module (or a program module including an instruction), the software module is a computer-readable non-transitory It may be stored in a non-transitory computer readable media. In addition, in this case, at least one software module may be provided by an operating system (OS) or a predetermined application. Alternatively, some of the at least one software module may be provided by an operating system (OS), and the remaining part may be provided by a predetermined application.

In addition, the data recognition unit 1320 may train a data recognition model used to determine a situation by using the training data. In this case, the data recognition model may be a pre-built model. For example, the data recognition model may be a model built in advance by receiving basic training data (eg, sample images, etc.).

The data recognition model may be constructed in consideration of the application field of the recognition model, the purpose of learning, or the computer performance of the device. The data recognition model may be, for example, a model based on a neural network. For example, a model such as a deep neural network (DNN), a recurrent neural network (RNN), or a bidirectional recurrent deep neural network (BRDNN) may be used as a data recognition model, but is not limited thereto.

According to various embodiments, when there are a plurality of pre-built data recognition models, the data learning unit 1310 may determine a data recognition model having a high correlation between input training data and basic training data as a data recognition model to be trained. have. In this case, the basic training data may be pre-classified by data type, and the data recognition model may be pre-built for each data type. For example, basic training data is classified based on various criteria such as the region where the training data was created, the time when the training data was created, the size of the training data, the genre of the training data, the creator of the training data, and the type of objects in the training data. Can be.

An embodiment may also be implemented in the form of a recording medium including instructions executable by a computer, such as a program module executed by a computer. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transmission mechanism, and includes any information delivery medium.

Further, in the present specification, the "unit" may be a hardware component such as a processor or a circuit, and/or a software component executed by a hardware configuration such as a processor.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that other specific forms can be easily modified without changing the technical spirit or essential features of the present invention will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (10)

Periodically acquiring treatment data of patients who have received the first drug manufactured to cure disease;
Classifying treatment data of patients whose disease has been cured after the first drug administration among the treatment data as a first category, and treatment data of patients whose disease has not been cured after the first drug administration as a second category;
Deriving a target gene from the genes of patients included in the first category by using a first artificial intelligence learning model that has learned the correlation between the gene expression distribution due to the disease and the gene expression distribution by the component of the first drug ;
Analyzing the expression state of the target gene and the expression state of other genes by the first drug from the gene expression information of the patients included in the second category to obtain a biomarker profile; And
Based on the biomarker profile, a second artificial intelligence learning model that learns the relationship between the target gene and the component and content of the first drug is used to treat diseases of patients included in the second category. 2 A method of verifying the effectiveness of a drug acting on a target gene using artificial intelligence, comprising the step of determining whether or not to administer the drug and the content of the second drug. The method of claim 1,
The first artificial intelligence learning model,
It is learned to derive the target gene of the disease, and the change in the expression state of the gene due to the onset of the disease, cell function, the change in the expression state of the gene due to the first drug, and the change in the expression state of the other gene due to the first drug. A method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence, characterized in that it is obtained based on a result of unsupervised learning. The method of claim 1,
The second artificial intelligence learning model,
When the disease is not treated by the first drug, it is learned to derive other treatment methods, and changes in the expression state of the target gene according to changes in the components and contents of the first drug, and the composition and content of the second drug. An apparatus for verifying the effectiveness of a drug acting on a target gene using artificial intelligence, characterized in that it is obtained based on a result of learning the relationship between the change in the expression state of the target gene according to the change. The method of claim 1,
The second artificial intelligence learning model extracts a pathway in which the first drug acts on the target gene, and activates and inhibits other genes included in the pathway according to a change in the content of the first drug. By analyzing (inhibition) and analyzing the active response and inhibitory response of other genes included in the pathway according to the change in the content of the second drug, it is possible to determine a new treatment method for patients included in the second category. Characterized in, a method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence. The method of claim 1,
The second artificial intelligence learning model examines a tissue sample from a patient diagnosed with the disease and compares the state of the protein of the target gene of the first category to AGER, THBS2, and the protein of the target gene of the second category. CA3, MMP12, MMP-1, MMP-7, MMP-9, MMP-13, MMP-8, MMP-10, MMP-2, PIGR, DCN, PGAM1, CD36, FABP, ACP5, CCDC80, PPBP, LYVE1, STC1, SPON1, IL17RC, MMP1, CA1, SERPINC1, TPSB2, CKB/CKBM, NAMPT/PBEF, PPBP/CTAPIII, F9, DCTPP1, F5, SPOCK2, CAT, PF4, MDK, BGN, CKM, POSTN, PGLYRP1 and CXCL12 among Analyzing the altered expression state of the selected at least one protein, increasing or decreasing the content of the first drug, or administering at least one other drug according to the fold difference of the at least one protein having the altered expression state A method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence, characterized in that to determine. The method of claim 1,
After administering a new drug by the second artificial intelligence learning model, checking the disease treatment status of patients included in the second category; And
According to the second artificial intelligence learning model re-learned by reflecting the disease treatment state, the component and content of the first drug, the component and content of the second drug, and the component and content of other drugs are re-determined and A method for verifying the effectiveness of a drug acting on a target gene using artificial intelligence, comprising the step of determining a drug for treatment of patients. The method of claim 1,
Analyzing the expression state of the target gene of each of the patients included in the first category according to the component ratio, dose, number of doses, dose time, and dosage method of the first drug to calculate the utility value of the first drug Step to; and
Effectiveness of a drug acting on a target gene using artificial intelligence, comprising the step of determining a personalized dose method of the first drug for each of the patients included in the first category according to the utility value of the first drug How to verify it. The method of claim 1,
Analyzing the expression state of the target gene of each of the patients included in the second category according to the combination of the first drug and the second drug, and calculating a synergy value of the combination of the first drug and the second drug; And
According to the synergy value of the combination of the first drug and the second drug, for each of the patients included in the second category, comprising the step of determining a personalized dose method of the first drug and the second drug, artificial intelligence A method of verifying the effectiveness of a drug that acts on a target gene by using. Processor; and
Including; a memory for storing executable instructions,
The processor,
By executing the above commands,
Periodically acquiring medical data of patients who have been administered the first drug manufactured to cure disease,
Among the medical data, medical data of patients whose disease has been cured after administration of the first drug are classified as a first category, and medical data of patients whose disease has not been cured after administration of the first drug are classified as a second category,
Using a first artificial intelligence learning model that learned the association between the gene expression distribution due to the disease and the gene expression distribution by the component of the first drug, a target gene is derived from the genes of patients included in the first category,
From the gene expression information of the patients included in the second category, by analyzing the expression state of the target gene and the expression state of other genes by the first drug to obtain a biomarker profile,
Based on the biomarker profile, a second artificial intelligence learning model that learns the relationship between the target gene and the component and content of the first drug is used to treat diseases of patients included in the second category. 2 A device for verifying the effectiveness of a drug acting on a target gene using artificial intelligence, which determines whether or not to administer the drug and the content of the second drug. A computer-readable non-transitory recording medium on which a program for implementing the method of any one of claims 1 to 8 is recorded.

Images