KR102476603B1 - System for diagnosing gene using self-improving genetic sequensing based on artificial intelligence - Google Patents

Abstract

In a method for diagnosing a gene including a disease gene corresponding to user information using an artificial intelligence-based gene analysis model, user information including profile information including race, ethnicity, and gender of the user is received from a user terminal Receiving step, subdividing the user information into information for each item, and comparing information by matching with the genetic analysis model based on a predetermined item among the subdivided items, disease suspected primarily from the user information Extracting gene information including the gene, and constructing a feedback questionnaire to check whether or not a match is made for an item having a correlation with a gene item including the suspected disease gene based on the gene analysis model, When feedback information on the feedback questionnaire is received from the user terminal, updating gene information including the suspected disease gene from the feedback information, and repeating the feedback questionnaire construction and update to determine the gene including the suspected disease gene and transmitting a report on the gene including the confirmed suspected disease gene to a user terminal, wherein the gene analysis model subdivides and classifies information on a plurality of unspecified subjects, disease information, and gene information. It is characterized by being a model that learns whether or not there is a correlation between items from big data for each item based on artificial intelligence.

Description

Gene diagnosis system through artificial intelligence-based self-improving genome sequencing

The present invention relates to a genetic diagnosis system through artificial intelligence-based self-improving genome sequencing, and more specifically, to a disease corresponding to user information by using artificial intelligence-based self-improving genome sequencing including correlation information between genes. It relates to a genetic diagnosis system that diagnoses and analyzes genes including genes.

The global genetic testing market is growing rapidly along with research and development of genetic testing based on the data of analyzing the genetic information of more than one million people. After genetic testing technology was introduced in Korea, WES (Whole Exome Sequencing) genomic information was turned into big data, and the Personal Genome Map Platform (PMAP) was developed as a technology to integrate and optimize it. This is a supercomputing system that detects disease susceptibility and prevents disease in advance, and is expected to be applicable in various fields such as newborn genetic disease screening, rare disease and cancer genetic testing, and development of new drugs and stem cell treatments.

On the other hand, a disease gene is a gene capable of controlling a disease and is a drug target candidate. Therefore, technology for predicting disease genes is one of the technologies most needed in the pharmaceutical industry for efficient treatment methods and drug development. Recently, with the development of high-throughput screening technology, many studies are underway to discover drug target candidates or disease genes through experiments. However, the excavation work through experiments is time-consuming and costly, and the excavation results far fall short of the expected number of drug target candidates. Recently, a method of predicting disease genes through computational technology and experimentally verifying them to discover drug target candidates at low cost and in a short time has been proposed.

However, there is a limitation in that it is impossible to accurately predict disease genes in a state where genetic data are not secured, and even if disease genes are predicted, they are meaningful disease genes in response to constantly evolving viruses through vast disease gene pools and mutations. There are bound to be limitations in predicting . Therefore, in order to respond to the genetic data that causes increasingly massive diseases, human disease genes are analyzed/predicted more quickly and accurately to form a disease gene information pool of individuals and groups, and eventually a genome map including genetic data correlated with diseases. is required to secure

The present invention is to solve the above problems, by using artificial intelligence-based self-improving genome sequencing that includes correlation information for each gene item, by analyzing/predicting genetic information including human disease genes more quickly and accurately The purpose is to secure a pool of individual and group genetic information, and based on this, to provide a system that diagnoses genes, including disease genes, of users from all information such as behavioral and psychological information, including health information related to genes.

In the method for diagnosing genes including disease genes corresponding to user information using an artificial intelligence-based gene analysis model according to an embodiment for realizing the object of the present invention described above, the race of the user from the user terminal , Receiving user information including profile information including ethnicity and gender, subdividing the user information into information for each item, and matching with the gene analysis model based on a predetermined item among the subdivided items Comparing information, extracting information on a gene including a suspected disease gene primarily from the user information, and based on the gene analysis model, an item whose correlation with the item of the gene including the suspected disease gene is at least a certain value constructing a feedback questionnaire for checking whether a match exists for, when feedback information on the feedback questionnaire is received from the user terminal, updating gene information including a suspected disease gene from the feedback information, and constructing a feedback questionnaire and confirming genes including suspected disease genes by repeating the update, and transmitting a report on the genes including the confirmed suspected disease genes to a user terminal, wherein the gene analysis model is used to determine a number of unspecified genes. It is an artificial intelligence-based model that learns whether or not there is a correlation between items from big data for each item classified by subdividing subject information, disease information, and genetic information.

In one embodiment of the present invention, the user information includes information on the user's disease history and family history, and the step of extracting the gene information including the suspected disease gene is primarily based on the profile information , Secondarily, information on genes including suspected disease genes can be extracted by weighting genetic information based on the disease history and family history.

In one embodiment of the present invention, the user information further includes the user's health checkup information and family health checkup information, and in the step of extracting the gene information including the suspected disease gene, the disease history and Gene information including a suspected disease gene may be extracted by weighting gene information based on the user's health checkup information and family health checkup information together with the family history.

In one embodiment of the present invention, the step of constructing the feedback questionnaire consists of questions for confirming whether or not an item having a direct correlation with a gene containing a suspected disease gene is matched with a certain value or not, or It can be composed of questions for indirectly confirming whether an item having a correlation higher than a predetermined value is matched with an item having a correlation higher than a predetermined value.

In one embodiment of the present invention, the feedback information may include at least one of additional profile information, physical characteristic information, and disease information of the user.

In one embodiment of the present invention, the information of the gene including the suspected disease gene is extracted based on the probability of suspicion, and the repetition of constructing and updating the feedback questionnaire determines that the probability of suspicion of the suspected gene exceeds a predetermined criterion. can be repeated until

In one embodiment of the present invention, the report is a genome map displaying the correlation between genes including suspected disease genes confirmed based on the gene analysis model and items having a certain or higher correlation with the genes. can

In one embodiment of the present invention, user information and feedback information received from the user terminal may be used as learning data for updating the genetic analysis model.

In one embodiment of the present invention, a predetermined item as a criterion for matching with the gene analysis model may be profile information including the user's race, ethnicity, and gender.

In the system for diagnosing genes including disease genes corresponding to user information using an artificial intelligence-based gene analysis model according to an embodiment for realizing the object of the present invention, a central server and communication, user information including profile information including the user's race, ethnicity, and gender, and feedback information for determining whether the gene including the disease gene has a low probability of suspicion or is a specific gene from the user information A user terminal that transmits to the central server, communicates with the central server through a network, and the correlation between items is obtained from the big data for each item in which each of the received unspecified subject information, disease information, and genetic information is subdivided and classified. A gene analysis server configured to learn a gene analysis model based on artificial intelligence whether or not there is a genetic analysis model, and a central server communicating with the user terminal and the gene analysis server through a network, wherein the central server receives information from the user terminal. A matching comparison unit that subdivides the user information into item-specific information and compares the information by matching the information with the gene analysis model based on a predetermined item among the subdivided items, and a disease gene primarily suspected from the user information extracts gene information including, and when feedback information on the feedback questionnaire is received from the user terminal, updates the gene information including the suspected disease gene from the feedback information, and constructs and updates the feedback questionnaire repeatedly to suspect suspicion. A predictive disease extraction unit for determining genes including disease genes to be identified, and a feedback questionnaire for confirming whether or not matching items having a correlation with items of genes including the suspected disease genes based on the gene analysis model are equal to or higher than a certain value. Includes a feedback questionnaire component constituting.

According to the present invention, by using a model that has learned big data of genetic information including subject information including diseases and diseases of an unspecified number of people secured through artificial intelligence-based health checkups or medical examinations and disease gene information of the subject, , It is not a method of predicting genes including disease genes by directly taking genetic information from a conventional subject, but presenting a disease estimated only with user information and health checkup information, and through the resulting feedback information, genes including disease genes of can be predicted.

In addition, according to the present invention, by using the learned model, it is possible to analyze not only the correlation between diseases, diseases, and disease genes, but also causes and causal relationships caused by non-disease factors, and genes already identified through research. Even if it is not a relationship, it is possible to provide a self-improving model that identifies a user's disease or physical characteristics through correlation analysis between new genes. In addition, not only the correlation between genes, but also the correlation between items of user information or feedback information can be analyzed together.

In addition, by using user information input from the user as learning data, the accuracy of the learned model can be increased, and at the same time, the model can be used for various purposes such as the development of a cure or the healthcare industry.

1 is a configuration diagram for explaining a genetic diagnosis system according to an embodiment of the present invention.
2 is an exemplary view showing a gene analysis model managed by the gene analysis server of FIG. 1 .
FIG. 3 is a block diagram for explaining the configuration of a central server of the genetic diagnosis system of FIG. 1 .
Figure 4 is an exemplary diagram of a genome map as a result provided by the genetic analysis server of Figure 1.
5 is a flowchart illustrating a genetic diagnosis method according to an embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described in detail so that those skilled in the art can easily practice it. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. In addition, in adding reference numerals to components of each drawing, the same reference numerals are given to the same components even though they are displayed on different drawings. In addition, in this specification, when a component is said to be "connected" or "connected" to another component, this is not only when both components are "directly" connected, but also when another component is interposed therebetween. It also includes cases where they are connected by "functional" or "communication" as well as cases where they are "physically" connected. In addition, when a part "includes" a certain component, it means that it may further include other components without excluding other components unless otherwise stated.

1 is a configuration diagram for explaining a genetic diagnosis system according to an embodiment of the present invention.

Referring to FIG. 1, the genetic diagnosis system 1000 of the present invention, when receiving user-specific profile information and health checkup information from a user terminal, converts them into user information and, based on artificial intelligence, pre-learned unspecified numbers Based on the model that learns the correlation between subject information, disease information, and genetic information including disease genes, genes including disease genes matched from user information are traced back, and feedback is received to diagnose the user. is configured to lower

The genetic diagnosis system 1000 learns the correlation between the user terminal 100 and each item of genetic information including unspecified number of subjects information, disease information, and disease genes based on artificial intelligence, and the correlation between the items of information A gene analysis server 300 configured to manage a genetic correlation analysis model that has made a database, and user information including profile information and health checkup information for each user input through the user terminal 100 are converted into data, and the genetic correlation analysis and a central server 500 configured to, based on the model, backtrack genes including matching disease or disease genes from user information and diagnose the user.

The user terminal 100 may be implemented as a computer (pc) or a portable terminal that can be accessed through the network 10, and is configured to transmit and receive data by accessing the central server 500 through the network 10. do.

Specifically, the user terminal 100 is a computer or portable terminal owned by a user, such as an individual who wants to be diagnosed with a disease or a gene including a disease gene through the gene diagnosis system 1000, and the user can access the web through the user terminal. A user interface (UI) that connects to the central server 500 through the network 10 in the form of a web, application, or web app to perform a new subscription or login, and interwork with the central server 500 ) Through the input of their own health checkup information and profile information, additionally, if there is a feedback request from the central server 500, it is configured to be able to input feedback information as a response to this.

The computer includes, for example, a desktop, a laptop, a tablet PC, etc. equipped with a web browser, and a portable terminal, for example, ensures portability and mobility. As a wireless communication device that becomes a smartphone (smartphone), PCS (Personal Communication System), GSM (Global System for Mobile communication), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) devices, etc. A communication device may be included.

In addition, the network 10 connecting the user terminal 100, the genetic analysis server 300 and the central server 500 includes a local area network (LAN), a wide area network (WAN), Value Added Network (VAN), Personal Area Network (PAN), mobile radio communication network, Wireless Broadband Internet (Wibro), Mobile WiMAX, High Speed Downlink Packet Access (HSDPA), or It can be implemented in all types of wired/wireless networks such as satellite communication networks.

The gene analysis server 300 learns the correlation between each item of genetic information including unspecified number of subject information, disease information, and disease genes based on artificial intelligence, and the correlation between the information items is databased. It is configured to manage the correlation analysis model. In addition, if there is a request from the central server 500 through the network 10, the genetic analysis server 300 provides the genetic correlation analysis model managed and provides user information from the central server 500. Upon receipt, it is configured to update the genetic correlation model.

For example, the genetic analysis server 300 subdivides and classifies a plurality of unspecified subject information, disease information, and genetic information stored in a DB or provided from an external server into different items, and It is configured to learn whether or not there is a correlation between items from big data for each item obtained from a plurality of subjects. For example, based on artificial intelligence, information on the correlation between different items within profile information among subject information, for example, ethnic information and gender information, as well as between heterogeneous information, for example, profile information and genetic information It is configured to learn the correlation for each item of

For example, in the case of learning the correlation between a disease and a gene, based on artificial intelligence, the correlation between all functional groups in which each gene is correlated with a disease is calculated. On the other hand, in the process of calculating the correlation between a disease and a gene, if there is already known disease information about each gene or expressed from a specific gene through academia, a prior correlation can be calculated through the corresponding data. In this case, the genetic analysis server 300 may assign a correlation between nodes for each item in the network of the genetic correlation analysis model by weighted summing a pre-correlation to a correlation calculated between a disease and a gene. .

On the other hand, the correlation is displayed on a network connecting nodes to nodes when creating a genome map (see FIG. 2). That is, when referring to FIG. 2, each item is represented by a node, and although not shown for convenience, an arbitrary node forms a network with all nodes except itself, and only when it has a digitized correlation of a certain value. It can also be connected to a network. The degree of correlation may indicate a correlation between items themselves, or may indicate a correlation between specific information within items. This is determined by the degree to which the item is subdivided. For example, when the information within an item is determined in a dichotomous manner such as yes or no, it may indicate a self-correlation between the items. Meanwhile, the size of the node may be determined according to the relative occurrence frequency of the corresponding item, and FIG. 2 is an exemplary diagram and does not reflect the size of the node.

The genetic analysis server 300 is configured to update an item having a significant network correlation based on an arbitrary item (node) to a specific item, and to feed back in a learning cycle of the next data. That is, the genetic analysis server 300 recalculates the correlation indicating the correlation with an arbitrary item taken as a criterion by using the updated specific item in the next data learning cycle, and the network connected to the specific item It is configured to continuously search for another particular item that is meaningful.

If no other specific item is discovered because there is no item having a significant network correlation, the discovery procedure is terminated. For example, when an arbitrary item is referred to as a specific disease, a disease item having a significant network correlation with the specific disease item may be grouped, and a disease gene item having a significant network correlation with the disease item may be grouped. In this case, a correlation between a disease-disease-disease gene having a certain degree of correlation may be discovered.

The genetic analysis server 300 is configured to store and manage information on items for which correlations are output in a DB using the genetic correlation analysis model. The DB configuration includes disease, gene, genetic influence information, risk factor, allele, definitive genotype, predicted genotype, relative risk, correlation rate, crossover rate (can calculate double crossover rate), group information (variability) and standard according to the country. It may contain genetic information (constancy). For example, when a meaningful network correlation between a disease and a gene is discovered, the gene analysis server 300 can store and manage disease and gene information as shown in [Table 1] below.

disease gene risk factor allele genotype result relative
risk liver cancer STAT4 G T TG caution 1.39 HLA-DRB1-LOC107986589 A G GA KIF1B A G AA breast cancer CASC16 T C CC Attention 0.9 CCDC170 - ESR1 A G GA colorectal cancer C5orf66 A C CC Good 0.69 MYRF G T GT CASC8, CCAT2 G T TT stomach cancer PSCA T C CT Good 0.67 PRKAA1 C T TT lung cancer TERT G A AA Good 0.34 BPTF A G GG stroke PITX2 - MIR297 G A GG Attention 0.99 SPSB4 G A GA Alzheimer's CLU T C TT Attention 0.92 SORL1 T C TC type 2 diabetes LOC105375716, SLC30A8 C T CT Attention 0.89 KCNQ1 C T CT myocardial infarction AP3D1 - DOT1L C A CA Attention 0.88 High blood pressure UMOD A G AA Attention 0.81 FGF5 C T TT

On the other hand, the relative risk increases as the number of predicted risk factors increases, and in calculating the relative risk, the degree of network correlation from an arbitrary node serving as a reference may have an effect. Meanwhile, the algorithm for calculating the relative risk is outside the scope of the present invention, and detailed description thereof will be omitted.

In this way, since the genetic analysis server 300 repeatedly updates the correlation of the network between items, it is possible to discover correlations that were not known between items of information, and furthermore, not only the correlations linking disease-disease-disease genes. In addition, when other items that are highly correlated with them, for example, specific items related to physical characteristics that are revealed in appearance are discovered, disease genes can be more efficiently discovered from this information, and diseases can be predicted and diagnosed. On the other hand, the high correlation or degree of correlation of a specific item means that it is higher than other items or has a correlation or degree of correlation higher than a certain value, which is a predetermined criterion.

Hereinafter, the central server 500 will be described in detail with reference to FIGS. 3 and 4 .

FIG. 3 is a block diagram for explaining the configuration of a central server of the genetic diagnosis system of FIG. 1 . Figure 4 is an exemplary diagram of a genome map as a result provided by the genetic analysis server of Figure 1.

1 to 4, the central server 500 is configured to communicate with the user terminal 100 and the gene analysis server 300 through the network 10, and through the user terminal 100 When profile information and health checkup information for each user are input, it is converted into user information in a form comparable to the DB, and based on the DB information based on the gene analysis model received from the gene analysis server 300, each information It is configured to output suspected diseases to the user terminal 100 in order of high probability according to the matching rate. Thereafter, when information on an item having a high correlation with the suspected disease is fed back from the user terminal 100, based on this, a suspected disease gene is determined based on a disease with a high probability, and the disease is caused by the disease gene. It is configured to present possible additional diseases and conditions to the user terminal 100 .

Specifically, the central server 500 includes an information reception unit 510, a user information management unit 520, a matching comparison unit 530, a predicted disease extraction unit 540, a feedback questionnaire construction unit 550, and an information transmission unit. 560, and an update unit 570.

The information receiving unit 510 is configured to receive user information including user profile information and health checkup information from the user terminal 100, and additionally provides feedback according to the questionnaire contents generated from the feedback questionnaire configuration unit 550. configured to receive information. In addition, the information receiving unit 510 is configured to receive the genetic analysis model in which the correlation between each item of genetic information is databased based on artificial intelligence from the genetic analysis server 300 .

The user information management unit 520 is configured to register at least one user as a customer in the genetic diagnosis system 1000 and to match and manage information input by the user with customer information.

Specifically, the user information management unit 520 stores and manages user information including basic information such as the user's name, contact information, and e-mail address input from the user terminal 100, and when the user signs up as a member, A user identification code is assigned to each user, matched with the entered user information, and stored and managed in the customer DB.

The user information includes, in addition to the basic information, user profile information and health checkup information for diagnosing diseases or genes including disease genes.

Specifically, the profile information includes user group information including the user's gender, disease history, family history, country, ethnicity, and race, body information such as weight and height, current residence, work status, and working hours. , average sleep time, etc., and current status information of individual users are included. The profile information includes not only information for which statistical genetic information is known, but also information for which it is unclear whether or not it is correlated with genetic information. For example, ethnic information is information in which 50% of Koreans have pure Korean blood, 25% of Japanese blood, and 18% of Chinese blood. In this case, the corresponding gene information can be predicted by the ratio of each lineage from the ethnic information input by the user. Meanwhile, the profile information may be sequentially input from the user terminal 100 from general information to special information, or information having the greatest influence on disease prediction may be input with priority for each category.

The health checkup information includes not only the health checkup information of the user himself, but also the health checkup information of his family. Accordingly, it is possible to grasp family genotype disease information inherited from the father or mother, and to predict gene-related information. On the other hand, health checkup information may be configured to receive information directly from the user terminal 100, obtain information by parsing from a checkup result table, or be connected to a server of the National Health Insurance Corporation, which is an external server, to receive the information upon request. there is.

In addition, the user information manager 520 is configured to store and manage feedback information input from the user terminal 100 .

The feedback information is additional information, such as when the probability of predicting a disease or a gene including a disease gene is low from information input by a user, or when a confirmation procedure is required to determine whether a specific disease or disease gene or a specific gene is required. This is information input from the user according to the contents of the questionnaire generated by the feedback questionnaire configuration unit 350 to be described later, in order to obtain . The feedback information may be sequentially provided one time or a plurality of times.

The feedback information may include additional profile information or physical characteristic information of the user. The physical characteristic information, for example, in the case of hair, baldness, baldness, whether or not the head is empty, hair thickness, hair color, etc. Conditions related to traits determined by genetic factors It includes all information related to psychological genes, such as information, parting direction, and hair style. In addition, information about the user's behavioral psychology is included. For example, whether the user is extroverted or introverted, intuitive or rational, and whether the user is spontaneous or planned, the information is directly input from the user or can be inferred from a feedback questionnaire to determine this. may be

Also, the feedback information includes disease information. For example, information about a disease or symptom that the user currently suffers from or has a history of suffering from may be included. For example, "Does your stomach feel bloated after eating?", "Do you fart often?", "Do you burp often?" It can be done in a yes or no way to questions about various symptoms, such as

In addition, the feedback information may further include a disease history, a BMI index, and the like.

The matching comparison unit 530 determines whether information on each item of user information input by the user matches, based on the gene analysis model provided from the gene analysis server 300 .

The matching comparison unit 530 converts user information into data corresponding to the gene analysis model in order to match the user information with the gene analysis model. That is, after subdividing user information into information for each item, data conversion is performed.

For example, the matching comparison unit 530 primarily extracts item-specific information about country, ethnicity, race, gender, etc. from profile information among user information input by the user, designates a specific item among them as a standard item, , Information can be compared while sequentially matching other items based on the control genotype information of the gene analysis model for each information on the reference item. The criterion item may be the user's race and country information extracted primarily. At this time, if the user is Asian and Korean, other information is based on a genetic analysis model based on the same Asian and Korean genotype information as the user. items can be compared.

Meanwhile, when there is no information on the reference item or there is no matching result from the information on the reference item, items specified in a predetermined sequence may be matched and compared. For example, if there is no information about a specific ethnicity or race input from the user, or if there is no matching information, other items are compared and judged based on the control genotype information of the genetic analysis model by gender information based on gender information can do.

The predicted disease extraction unit 540 primarily determines whether information having a correlation with genetic information including a specific disease or disease gene is matched from user information matched to the gene analysis model by the matching comparison unit 530. It is configured to extract information of genes including suspected disease or disease genes.

In addition, on the premise that the information is provided from the user, the predicted disease extractor 540 secondarily extracts the user's family history, disease history information, and health checkup data results from profile information among user information, and confirms therefrom. It is possible to determine whether the disease is matched with the information of the disease or gene including the disease gene extracted from the gene analysis model by weighting the disease. For example, when there is a specific cancer in the family history, for example, gastric cancer, information on gastric cancer is added to a second suspected disease in addition to a disease primarily extracted based on a genetic analysis model matched from input user information. can do.

In addition, the genetic information including the suspected disease or disease gene from the user information is updated according to the user's updated feedback information on the feedback questionnaire in the feedback questionnaire component 550 to be described later, and group statistical information and correlation information Through correlation, it can be finally formed as a user's genetic map.

That is, by continuously updating feedback information including physical characteristic information or disease information, focusing on items that have a high correlation with genetic information on genes, including primarily suspected diseases or disease genes, Determine the genetic information for a gene. That is, when an item highly correlated with a specific disease is matched from the user's feedback information, the probability of suspecting the disease increases. For example, when the primary suspected disease is gastric cancer, based on item information highly correlated with gastric cancer, for example, a disease such as heartburn or feedback information such as weight loss as a result of physical characteristics, gastric cancer disease can be determined. Meanwhile, the feedback information may be updated until the probability of suspicion of a suspected specific disease or specific disease gene or specific gene exceeds a predetermined criterion.

In particular, genetic information prediction including disease genes presents genetic information including disease genes that are highly correlated with predicted diseases, or family health examination data based on the genetic analysis model (risk factors inherited from family members can be predicted) ) and family history information, the user's relative risk for each disease is calculated by matching information including the appearance frequency information of the disease related to the gene, and the risk factors and alleles' stochastic genotypes are predicted accordingly.

For example, the relative risk is predicted by calculating the occurrence frequency information of the user's family members with gastric cancer, the crossover rate of gastric cancer gene correlation, group data, and health checkup data, and the risk of the PSCA gene correlated with gastric cancer according to the calculated relative risk result. Among the number of cases of arrangement of factor T and random alleles, the nucleotide sequence with the highest probability is determined when compared with the genetic analysis model. For many diseases related to genes, the above method is used to predict other disease genes and synthesized to form genetic information necessary for developing a disease genetic map.

The feedback questionnaire component 550 checks whether an item having a high correlation with the disease or gene including the disease gene is matched in order to determine the disease or gene including the disease gene predicted from the user information based on the probability. It is configured to fill out the contents of the questionnaire for The survey content may be configured in the form of an electronic questionnaire consisting of a plurality of questions.

The questionnaire contents of the electronic questionnaire consist of questions to confirm matching of items that have a high direct correlation with a suspected disease or gene including a disease gene, or have a high correlation with an item with a high correlation. It may be composed of questions for checking whether an item, that is, an item having an indirectly high correlation, is matched.

That is, when information on a disease or gene including a disease gene predicted from the prediction disease extraction unit 540 is extracted, the feedback questionnaire component 550 based on the gene analysis model, including the corresponding disease or disease gene After constructing questionnaire contents to check whether items having a high correlation with genes match, and then receiving feedback information about the questionnaire contents from the user terminal 100, the predicted disease extraction unit 540 performs the feedback information Updates and extracts information on a disease or gene including a disease gene that is strongly or newly predicted from the above, and the feedback questionnaire component 550 checks whether an item having a high correlation with the disease or gene including the disease gene is matched. Perform repetitive tasks such as constructing survey contents for The iterative task may be repeated until the suspected disease or disease gene or the probability of suspicion of the gene exceeds a predetermined criterion.

The information transmission unit 560 is configured to transmit the contents of the feedback questionnaire generated by the feedback questionnaire configuration unit 550 to the user terminal 100 and to transmit a report on a predicted disease or disease gene. In addition, the information transmission unit 560 further converts user information and feedback information to match the DB format of the gene analysis server 300 to update the gene analysis model, and the gene analysis server 300 It is configured to transmit to

The report is based on the user information and the feedback information, and the gene including the suspected disease gene determined based on the gene analysis model and the genome displaying the correlation between items having a certain or higher correlation with the gene It may be a map (see FIG. 4).

Referring to FIG. 4, based on ethnicity, race, and gender information, which are Asian, Korean, and male groups, which are standard items, based on profile information, health checkup information, and feedback information, suspected diseases and disease gene information are transmitted to nodes and them. A network of connections is represented. In the network, the degree of correlation between items is digitized, and the size of a node is expressed differently according to a relative frequency of occurrence, for example, the number of matching information for each item with high correlation.

The update unit 570 converts data of information into a DB format of the gene analysis server 300 in order to update (re-learn) the gene analysis model using all information input from the user terminal 100. is configured to do Data conversion for the update may be the time when user information is initially input and data is converted to match the genetic analysis model, or at the time of report creation in consideration of user feedback information provided until immediately before the final report. Conversion may be performed.

On the other hand, even if all information of the user input to the central server 500 does not contain accurate genetic information, it is possible to analyze the correlation for each item, so that not only the user's disease, but also physical characteristics or the relationship between them can be analyzed. It can be used to provide a model of self-improving to identify.

In addition, the update unit 570, in addition to all information of the user input to the central server 500, if there is genetic test data performed for detailed examination after the user checks the report, the external server Data conversion may be performed by matching corresponding data information with user information.

Meanwhile, the gene diagnosis system according to an embodiment of the present invention has been described as an example in which the gene analysis server 300 and the central server 500 are separated and communicate with each other, but it is not limited thereto. This is only in consideration of the efficient aspect of learning the genetic analysis model, and on the other hand, it may consist of one server.

According to the present invention, gene analysis, which is a model that learns big data of genetic information including subject information including diseases and diseases of a large number of unspecified persons secured through artificial intelligence-based health checkups or medical interviews, and genetic information of the subject's disease Using the model, genetic information is directly taken from a conventional subject to predict disease genes, rather than a method of predicting disease genes, and presents a disease estimated only with user information and health checkup information, and genes including disease genes through the resulting feedback information of can be predicted.

In addition, by using the genetic analysis model, it is possible to analyze not only the correlation between diseases, diseases and disease genes, but also the causes and causal relationships caused by non-disease factors, and even if the genetic relationship is not already revealed through research, it is possible to analyze new Through correlation analysis between genes, it is possible to provide a self-improving model that identifies a user's disease or physical characteristics.

In addition, by using user information input from the user as learning data, the accuracy of the learned model can be increased, and at the same time, the model can be used for various purposes such as the development of a cure or the healthcare industry.

FIG. 5 is a flow chart illustrating the disease gene diagnosis method of FIG. 1 .

1 to 5, a method for providing a genetic diagnosis system according to an embodiment of the present invention includes the steps of at least one user accessing a central server through a user terminal, registering as a member, and inputting user information. (S100), matching user information to a pre-learned genetic analysis model (S200), extracting genetic information on genes including primarily suspected diseases or disease genes (S300), disease or disease genes Constructing a feedback questionnaire to check whether items having a high correlation with the included genes are matched (S400), updating genes including suspected diseases or disease genes through feedback information input from the user (S500), Confirmation of suspected diseases or genes including disease genes by repeating configuration and updating of feedback questionnaires (S600), generating a report on genes including confirmed suspected diseases or disease genes, and providing the report to a user terminal (S700). ), and using the input user information and feedback information to update the genetic analysis model (S800).

In the step (S100) of the user subscribing and inputting user information, the user may connect to the central server 300 through the app, web or web app of the user terminal and perform new subscription, and through the user terminal This is done by entering user information. The user information includes the user's profile information and health examination information for diagnosing diseases or genes including disease genes, in addition to the user's basic information.

The profile information includes information such as race, country, gender, family history, and disease history. The health checkup information includes not only the user but also information about the user's family.

In the step of matching to the gene analysis model (S200), item-specific information on country, ethnicity, race, gender, etc. is primarily extracted from profile information among user information input by the user, and specific items among them are designated as standard items. , Information can be compared while sequentially matching other items based on the control genotype information of the gene analysis model for each information on the reference item. The criterion item may be the user's race and country information extracted primarily. At this time, if the user is Asian and Korean, other information is based on a genetic analysis model based on the same Asian and Korean genotype information as the user. items can be compared.

Meanwhile, when there is no information on the reference item or there is no matching result from the information on the reference item, items specified in a predetermined sequence may be matched and compared. For example, if there is no information about a specific ethnicity or race input from the user, or if there is no matching information, other items are compared and judged based on the control genotype information of the genetic analysis model by gender information based on gender information can do.

In the step of extracting genetic information on genes including suspected diseases or disease genes (S300), information correlated with genetic information on genes including specific diseases or disease genes from user information matched to the genetic analysis model. It is configured to extract genetic information, including suspected diseases or disease genes, depending on whether the

On the premise that information is provided from the user, the user's family history, disease history information, and health checkup data are secondarily extracted from the user's profile information, and the disease identified therefrom is weighted to determine the genetic analysis model. It is possible to check whether the disease is matched with the information of the gene including the disease or disease gene extracted from . For example, when there is a specific cancer in the family history, for example, gastric cancer, information on gastric cancer is added to a second suspected disease in addition to a disease primarily extracted based on a genetic analysis model matched from input user information. can do.

In the step of constructing a feedback questionnaire to check whether items having a high correlation with genes including disease or disease genes are matched (S400), the disease or genes predicted from user information or genes including disease genes are determined based on probability. In order to do this, it is configured to fill out a questionnaire to check whether items that have a high correlation with the disease or gene including the disease gene are matched. The survey content may be configured in the form of an electronic questionnaire consisting of a plurality of questions.

Thereafter, when feedback information input from the user is input, the suspected disease or gene including the disease gene is updated through the corresponding information (S500). That is, when information on a disease or gene including a disease gene predicted from the prediction disease extraction unit 540 is extracted, the feedback questionnaire component 550 based on the gene analysis model, including the corresponding disease or disease gene After constructing questionnaire contents to check whether items having a high correlation with genes match, and then receiving feedback information about the questionnaire contents from the user terminal 100, the predicted disease extraction unit 540 performs the feedback information to update and extract strongly or newly predicted disease or disease gene information, and again, the feedback questionnaire component 550 checks whether items having a high correlation with genes including the disease or disease gene are matched Perform repetitive tasks such as configuring (S600).

Meanwhile, the iterative task may be repeated until the probability of suspecting a suspected disease or a gene including a disease gene exceeds a predetermined criterion.

When the suspected disease or gene including the disease gene is confirmed, a report is generated and provided to the user terminal (S700). The report may be a genome map in which genes including suspected diseases or disease genes are determined based on the gene analysis model based on the user information and the feedback information (see FIG. 4 ). Referring to FIG. 4, based on ethnicity, race, and gender information, which are Asian, Korean, and male groups, which are standard items, based on profile information, health checkup information, and feedback information, suspected diseases and disease gene information are transmitted to nodes and them. A network of connections is represented. In the network, the degree of correlation between items is digitized, and the size of a node is expressed differently according to a relative frequency of occurrence, for example, the number of matching information for each item with high correlation.

Additionally, the input user information and feedback information are used as learning materials for updating the gene analysis model (S800). Specifically, in order to update (re-learn) the gene analysis model using all information input from the user terminal 100, it is configured to perform data conversion of information in the DB format of the gene analysis server 300. . Data conversion for the update may be the time when user information is initially input and data is converted to match the genetic analysis model, or at the time of report creation in consideration of user feedback information provided until immediately before the final report. Conversion may be performed.

On the other hand, even if all information of the user input to the central server 500 does not contain accurate genetic information, it is possible to analyze the correlation for each item, so that not only the user's disease, but also physical characteristics or the relationship between them can be analyzed. It can be used to provide a model of self-improving to identify.

Matters not described for the method for diagnosing disease genes of FIG. 5 are the same as those previously described for the genetic diagnosis system through FIGS. 1 to 4 or can be easily inferred from the description, so the description below will be omitted. .

The disease gene diagnosis method according to an embodiment described with reference to FIG. 5 may be implemented in the form of a recording medium including instructions executable by a computer, such as an application or 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. Also, computer readable media may include all computer storage 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.

The disease gene diagnosis method according to an embodiment of the present invention described above can be executed by an application basically installed in a terminal (this may include a program included in a platform or operating system, etc. may be executed by an application (that is, a program) directly installed in the master terminal through an application providing server such as an application store server, an application or a web server related to the corresponding service. In this sense, the method for diagnosing disease genes according to an embodiment of the present invention described above is implemented as an application (i.e., a program) that is basically installed in a terminal or directly installed by a user, and is implemented as a computer-readable recording medium such as a terminal. can be recorded in

The above description of the present invention is for illustrative purposes, and those skilled in the art can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, the embodiments described above should be understood as 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 distributed may be implemented in a combined form.

1000: genetic diagnosis system
100: user terminal
300: gene analysis server
500: central server
510: information receiver
520: user information management unit
530: matching comparison unit
540: predicted disease extraction unit
550: feedback questionnaire component
560: information transmission unit
570: update unit

Claims (10)

For diagnosing genes, including disease genes corresponding to user information, using an artificial intelligence-based gene analysis model in a central server including an information receiving unit, a matching comparison unit, a feedback questionnaire component unit, a prediction disease extraction unit, and an information transmission unit. in the method,
receiving user information including profile information including race, ethnicity, and gender of the user from the user terminal in the information receiving unit;
subdividing the user information into information for each item in the matching comparison unit, and comparing the information by matching with the gene analysis model based on a predetermined item among the subdivided items;
In the feedback questionnaire constructing unit, information on a gene including a suspected disease gene is primarily extracted from the user information, and an item whose correlation with the item of the gene including the suspected disease gene is higher than a certain value based on the gene analysis model. Configuring a feedback questionnaire to check whether or not a match is made for;
When the predicted disease extraction unit receives feedback information on the feedback questionnaire from the user terminal, the gene information including the suspected disease gene is updated from the feedback information, and the feedback questionnaire is repeatedly constructed and updated to repeat the suspected disease gene. Determining genes including; and
Transmitting, by the information transmission unit, a report on genes including the determined suspected disease genes to a user terminal;
The gene analysis model is a model that learns whether or not there is a correlation between items from big data for each item classified by subdividing and classifying a plurality of unspecified subject information, disease information, and genetic information, based on artificial intelligence. method for diagnosing. According to claim 1,
The user information includes information on the user's disease history and family history,
The step of extracting the gene information including the suspected disease gene is based primarily on the profile information, and secondarily weighting the gene information based on the disease history and family history to obtain information on the gene including the suspected disease gene. A method for diagnosing a gene characterized by extraction. According to claim 2,
The user information further includes the user's health checkup information and family health checkup information,
In the step of extracting gene information including the suspected disease gene, secondarily, the suspected disease gene is determined by weighting gene information based on the user's health checkup information and family health checkup information along with the disease history and family history. A method for diagnosing a gene characterized by extracting information of a gene including According to claim 1,
The step of constructing the feedback questionnaire is composed of questions to confirm whether a gene containing a suspected disease gene is matched with a gene having a correlation of at least a predetermined value, or indirectly, when the correlation is at least a predetermined value. A method for diagnosing a gene, characterized in that it consists of questions to check whether an item having a correlation with the item is a match of a certain value or more. According to claim 1,
The method for diagnosing a gene, characterized in that the feedback information includes at least one of the user's additional profile information, physical characteristic information, and disease information. According to claim 1,
Extraction of gene information including the suspected disease gene is based on the probability of suspicion,
The method for diagnosing a gene, characterized in that the repetition of constructing and updating the feedback questionnaire is repeated until the probability of suspicion of the suspected gene exceeds a predetermined criterion. According to claim 1,
The report is for diagnosing a gene, characterized in that it is a genome map displaying a correlation between a gene including a suspected disease gene determined based on the gene analysis model and items having a certain or higher correlation with the gene. method. According to claim 1,
The method for diagnosing genes, characterized in that the user information and feedback information received from the user terminal are used as learning data for updating the gene analysis model. According to claim 1,
The method for diagnosing genes, characterized in that the predetermined item as a criterion for matching with the genetic analysis model is profile information including race, ethnicity, and gender of the user. In a system for diagnosing genes including disease genes corresponding to user information using an artificial intelligence-based genetic analysis model,
Communicates with the central server through the network, and user information including profile information including the user's race, ethnicity, and gender, and whether the probability of suspicion is low to predict genes including disease genes from user information, or whether it is a specific gene A user terminal for transmitting feedback information for determining the to the central server;
Communicates with the central server through a network, and analyzes genes based on artificial intelligence to determine whether or not there is a correlation between items from big data for each item classified by subdividing and classifying a plurality of unspecified subject information, disease information, and genetic information provided. Gene analysis server configured to learn the model; and
And a central server communicating with the user terminal and the gene analysis server through a network,
The central server,
a matching comparison unit that subdivides the user information received from the user terminal into information for each item, matches the information with the gene analysis model based on a predetermined item among the subdivided items, and compares the information;
Extracting information on a gene including a suspected disease gene primarily from the user information, and updating information on a gene including a suspected disease gene from the feedback information when feedback information for a feedback questionnaire is received from the user terminal, a predictive disease extraction unit that confirms genes including suspected disease genes by repeating feedback questionnaire construction and updating; and
A system for diagnosing genes including a feedback questionnaire component configuring a feedback questionnaire to determine whether or not a matching item having a correlation with a gene item including the suspected disease gene is equal to or higher than a certain value based on the gene analysis model. .

Images