Method for providing clinical decision support information for disease identification and therapy determination based on product-tailored database, and device for performing the same

Abstract

There is provided a method of providing clinical decision support information for disease identification and therapy determination comprising: receiving a test result obtained using one or more diagnostic test reagent products selected based on sample information, from a target terminal; obtaining, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database; and providing the obtained clinical decision support information for disease identification and therapy determination to the target terminal or a terminal associated with the target terminal.

Description

METHOD FOR PROVIDING CLINICAL DECISION SUPPORT INFORMATION FOR DISEASE IDENTIFICATION AND THERAPY DETERMINATION BASED ON PRODUCT-TAILORED DATABASE, AND DEVICE FOR PERFORMING THE SAME

[0001] The present disclosure relates to a method for providing clinical decision support information for disease identification and therapy determination based on a product-tailored database, and a computer device for performing the same.

[0002] In vitro diagnostics is a field that is rapidly growing in the diagnostic market for the early diagnosis of diseases. Among them, molecular diagnostics, based on the accuracy of diagnostic results, is being used to diagnose causative genetic factors of infections by viruses, bacteria, etc. In addition, point-of-care testing, immunochemical diagnostics, and others are widely known as in vitro diagnostic technologies for the rapid diagnosis of diseases. In the case of molecular diagnostics, methods using nucleic acids, in particular, are usefully used for diagnosing causative genetic factors of infections by viruses, bacteria, etc., based on high specificity and sensitivity thereof.

[0003] In most diagnostic methods using nucleic acids, a nucleic acid amplification reaction that amplifies a target nucleic acid (e.g., a viral or bacterial nucleic acid) is used. As a representative example, in the Polymerase Chain Reaction (PCR), a nucleic acid amplification reaction, repeated cycles of double-stranded DNA denaturation, annealing of oligonucleotide primers to the DNA template, and primer extension by DNA polymerase are performed (Mullis et al., U.S. Patent Nos. 4,683,195, 4,683,202, and 4,800,159; Saiki et al., Science 230:1350-1354(1985)).

[0004] Various other methods for amplifying nucleic acids have been proposed, such as Ligase Chain Reaction (LCR), Strand Displacement Amplification (SDA), Nucleic Acid Sequence-Based Amplification (NASBA), Transcription Mediated Amplification (TMA), Recombinase Polymerase Amplification (RPA), Loop-mediated Isothermal Amplification (LAMP), and Rolling-Circle Amplification (RCA).

[0005] Meanwhile, with the development of the internet environment, technologies are being developed to assist medical practices by utilizing various information during a medical professional's diagnosis or therapy determination. Recently, with the advancement of artificial intelligence technology, methods for combining artificial intelligence technology with medical assistance technology are also being attempted. For example, medical assistance technologies such as robots that assist medical practices by applying artificial intelligence technology to patients' medical images are being developed.

[0006] For the diagnosis or treatment of diseases based on the aforementioned molecular diagnostics, a process of interpreting the molecular diagnostic test results is required. For example, even with an infection by the same pathogen (e.g., E. coli), the disease manifested in the patient may differ depending on symptoms, the primary site of pathogen expression, etc., and a different therapy determination is made accordingly. Therefore, an accurate diagnosis of the disease is required through an disease identification that comprehensively considers which pathogen is causing the infection, what diseases are related to that pathogen, and what symptoms the patient is currently experiencing, to enable an accurate therapy determination. This disease identification is typically performed by medical professionals. For example, a specialist in laboratory medicine comprehensively analyzes the results of a molecular diagnostic test performed using a patient's sample such as blood, urine, bodily fluid, or tissue, based on expertise, experience, and know-how of the specialist, which may then lead to disease diagnosis, therapy determination for treatment, and prognosis management.

[0007] However, this conventional technology has a limitation of being dependent on the medical practice of individual medical professionals. For example, in cases such as the emergence of a new infectious disease, difficulty in disease identification by medical professionals due to somewhat unclear molecular diagnostic test results, or the presence of co-infections with multiple pathogens, the disease identification of the disease may be inaccurate due to a lack of experience or know-how on the part of the medical professional. As another example, if the appropriate therapy determination for treating the interpreted disease varies significantly among patients, or if a particular infectious disease has a high risk of side effects, applying a generally known therapy determination may lead to unexpected side effects for the patient or over-therapy determination.

[0008] Accordingly, there is a growing need for the development of technology to assist the medical practice of medical professionals with respect to diagnostic tests.

[0009] A problem to be solved according to one embodiment includes efficiently assisting the medical practice of medical professionals by providing reference information that may be referred to for disease identification and therapy determination related to a diagnostic test, without the need for the medical professional to perform cumbersome searches.

[0010] A problem to be solved according to another embodiment includes supporting medical professionals to refer to more reliable information for the disease identification and therapy determination of test results obtained using a specific product, by providing product-tailored clinical decision support information and reference information for disease identification and therapy determination according to the diagnostic reagent product used in the patient's diagnostic test.

[0011] A problem to be solved according to still another embodiment includes effectively assisting the medical practice of medical professionals by providing clinical decision support information and reference information regarding disease identification and therapy determination that are highly relevant to a patient's diagnostic test results.

[0012] A problem to be solved according to still another embodiment includes overcoming limitations in response reliability, such as hallucination, which frequently occurs in Generative AI (Generative Artificial Intelligence), by efficiently selecting and effectively analyzing optimal references from a vast amount of publicly available data on disease identification and therapy determination to provide clinical decision support information for disease identification and therapy determination based on artificial intelligence that selects reference information suitable for a diagnostic reagent product.

[0013] However, the problems to be solved by the present invention are not limited to those described above, and other problems not explicitly mentioned will be clearly understood by those skilled in the art to which the present invention pertains from the following description.

[0014] In accordance with an aspect of the present disclosure, there is provided a method for providing clinical decision support information for disease identification and therapy determination, performed by a computer device, the method comprising: receiving a test result obtained by testing with one or more diagnostic test reagent products selected based on sample information, from a target terminal; obtaining, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database, wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information supporting the product-specific clinical decision support information, both of which are labeled with an identifier based on a specification of a diagnostic test reagent product whereby the product-specific clinical decision support information and the product-specific reference information can be used as considerations for the disease identification and therapy determination based on the test result using the one or more diagnostic test reagent products; and providing the clinical decision support information to the target terminal or to a terminal associated with the target terminal.

[0015] The one or more product-specific reference information are obtained from one or more data sources, and wherein the one or more data sources comprise at least one selected from the group consisting of: a product developer during a product development stage of the one or more diagnostic test reagent products; an institution associated with regulatory approval of the one or more diagnostic test reagent products; at least one of a disease research institution, a management institution, and a medical institution, based on a test using the one or more diagnostic test reagent products; and a medical expert associated with the one or more diagnostic test reagent products.

[0016] The one or more product-specific reference information include product-specific reference information, and wherein the product-specific reference information comprises at least one selected from the group consisting of: (i) research and development data generated by a researcher or developer during a research and development stage of the one or more diagnostic test reagent products; and (ii) clinical data obtained through clinical trials using the one or more diagnostic test reagent products in a post-development stage.

[0017] The one or more product-specific reference information further include public reference information that is publicly available from an academic institution, a medical institution or a public institution.

[0018] The one or more product-specific reference information comprise reference information selected from one or more data sources based on a specification of any one of the one or more diagnostic test reagent products.

[0019] The specification of any one of the one or more diagnostic test reagent products comprises at least one selected from the group consisting of: (a) an intended use including at least one of target analyte information, pathogen information, pathogenic gene or single nucleotide polymorphism (SNP) information, contamination substance information, sample type, host information, disease name, infectious disease name, and symptom information; (b) a design specification including information about at least one of an analyte panel configuration, an optical channel configuration, an internal control (IC), a positive control (PC), and a negative control (NC); (c) a performance specification including at least one of sensitivity, specificity, turn-around time (TAT), reproducibility, and linearity; and (d) a technical specification including at least one of a sampling technique, an extraction technique, a nucleic acid amplification test (NAAT) reaction setup technique, an amplification technique, a signal analysis technique, and an information display technique.

[0020] The one or more product-specific reference information are selected based on one or more criteria selected from the group consisting of: (a) a specification relevance that quantifies a degree to which the reference information includes content related to the specification of any one of the one or more diagnostic test reagent products; (b) a reliability of a data source; (c) a number of times cited in other clinical decision support information related to the specification of any one of the one or more diagnostic test reagent product; (d) a number of times cited in other reference information; and (e) a recognition of an author or an affiliated institution.

[0021] The selecting of the one or more product-specific reference information is performed using a generative artificial intelligence (AI) model trained to select the product-specific reference information from reference information based on the specification of any one of the one or more diagnostic test reagent product.

[0022] The one or more product-specific reference information correspond to reference information selected for each of the one or more diagnostic test reagent products and are labeled with a corresponding product identifier.

[0023] The one or more product-specific reference information correspond to reference information selected for each of the one or more diagnostic test reagent products and are labeled with metadata corresponding to content related to the specification of any one of the one or more diagnostic test reagent products, wherein the metadata include two or more metadata selected from the group consisting of: a product name, a product identifier, a product version, a catalog number, a product group name, a product group identifier, a data source type, a target analyte name, a pathogen name, pathogenic gene or single nucleotide polymorphism (SNP) information, a sample type, host information, a disease name, an infectious disease name, an author, an affiliated institution, an academic journal, an information disclosure date, and clinical information, wherein the clinical information includes at least one selected from the group consisting of: a subject's symptoms, underlying diseases, country, region, age, gender, a Cycle threshold value (Ct value) of each target analyte, an disease identification result, a therapy determination result, and a clinical date.

[0024] The one or more product-specific reference information comprise one or more selected from the group consisting of: clinical paper information including clinical trial content related to detection or analysis of at least one of a single target analyte or a plurality of target analytes that any one of the one or more diagnostic test reagent products is intended to detect or analyze; clinical paper information including content related to disease identification and / or therapy determination based on a sole detection of any one of the plurality of target analytes; and clinical paper information including content related to disease identification and / or therapy determination based on a simultaneous detection of two or more of the plurality of target analytes.

[0025] The one or more product-specific clinical decision support information comprise one or more selected from the group consisting of: (a) an disease identification result and / or a therapy determination result estimated based on one or more product-specific reference information; (b) summary information generated for each of the one or more product-specific reference information; and (c) a statistical analysis result corresponding to a type of disease identification result and / or a type of therapy determination result included in each of the one or more product-specific reference information.

[0026] The summary information comprises one or more selected from the group consisting of: a title, a clinical date, clinical values, a key sentence included in the one or more product-specific reference information, and data source information for the one or more product-specific reference information.

[0027] The one or more product-specific clinical decision support information are generated in a predefined report format using a generative artificial intelligence (AI) model trained to generate a natural language response to a natural language input.

[0028] The one or more product-specific clinical decision support information are generated based on a prompt that instructs the generative artificial intelligence (AI) model to generate the one or more product-specific clinical decision support information using the one or more product-specific reference information.

[0029] The one or more product-specific clinical decision support information comprise a plurality of product-specific clinical decision support information generated by the generative artificial intelligence (AI) model using each of a plurality of product-specific reference groups, wherein the plurality of product-specific reference groups are obtained by grouping the one or more product-specific reference information based on whether values of at least one piece of metadata among a plurality of metadata correspond to each other, and wherein the plurality of metadata comprise one or more selected from the group consisting of: a data source type, a target analyte name, a pathogen name, pathogenic gene or single nucleotide polymorphism (SNP) information, a sample type, host information, a disease name, an infectious disease name, an author, an affiliated institution, an academic journal, an information disclosure date, and clinical information, the clinical information comprising one or more selected from the group consisting of: a subject's symptoms, underlying diseases, country, region, age, gender, a Cycle threshold value (Ct value) of each target analyte, an disease identification result, a therapy determination result, and a clinical date.

[0030] The method further comprising updating the product-tailored database at a preset cycle, wherein the updating comprises: (a) newly obtaining reference information from one or more data sources including a reference search module based on a specification of any one of the one or more diagnostic test reagent products; (b) additionally selecting product-specific reference information from the newly obtained reference information; and (c) updating the product-specific clinical decision support information using the additionally selected product-specific reference information.

[0031] The obtaining the clinical decision support information comprises: (a) searching the product-tailored database for one or more product-specific clinical decision support information labeled with an identifier corresponding to the one or more diagnostic test reagent products; and (b) obtaining the clinical decision support information based on the searching.

[0032] The obtaining the clinical decision support information comprises: (a) generating a query text from the test result received from the target terminal, the query text including information of the one or more diagnostic test reagent products, a detection result of one or more target analytes targeted by the one or more diagnostic test reagent products, and a sample type according to the sample information; and (b) obtaining the clinical decision support information from the product-tailored database using the query text.

[0033] The detection result of the one or more target analytes comprises one or more selected from the group consisting of: (a) a detection status, a positive / negative result value, and a detected amount for each of the one or more target analytes; and (b) a type of genotype determined based on the detection result of each of the one or more target analytes and whether a mutation is included.

[0034] The information of the one or more diagnostic test reagent products comprises one or more selected from the group consisting of: a product name, a product identifier, a product version, a catalog number, a product group name, and a product group identifier of the one or more diagnostic test reagent products.

[0035] The query text further comprises two or more selected from the group consisting of: (a) a subject's symptoms; (b) a subject's underlying diseases; (c) de-identified patient information comprising one or more selected from the group consisting of: country, region, age, and gender; and (d) a Cycle threshold value (Ct value) of the one or more target analytes.

[0036] The query text further comprises at least a part of raw data obtained from a diagnostic test apparatus, the raw data being included in the test result.

[0037] The query text further comprises information that is manually obtained based on a user input of the target terminal or information that is automatically obtained based on linkage with software executed on the target terminal or with another device.

[0038] The obtaining the clinical decision support information from the product-tailored database using the query text comprises: (a) searching the product-tailored database for one or more product-specific clinical decision support information labeled with an identifier and / or metadata related to the query text; (b) calculating, for each of the searched one or more product-specific clinical decision support information, a query relevance that quantifies a degree to which the product-specific clinical decision support information includes content related to the query text; and (c) determining, as the clinical decision support information, the product-specific clinical decision support information for which the query relevance satisfies a preset condition.

[0039] The obtaining the clinical decision support information from the product-tailored database using the query text comprises performing at least one selected from the group consisting of: (a) searching for the product-specific clinical decision support information using a trained artificial intelligence (AI) model; (b) calculating the query relevance using the trained artificial intelligence (AI) model; and (c) determining the clinical decision support information using the trained artificial intelligence (AI) model.

[0040] The obtaining the clinical decision support information from the product-tailored database using the query text comprises: (a) when a preset reference search condition is satisfied, searching the product-tailored database for one or more product-specific reference information labeled with an identifier and / or metadata related to the query text, (b) calculating a reference score for each of the searched one or more product-specific reference information using one or more selected from the group consisting of a query relevance that quantifies a degree to which it includes content related to the query text, a reliability of a data source, a number of times cited in other clinical decision support information related to the corresponding diagnostic reagent product, a number of times cited in other reference information, and a recognition of an author or an affiliated institution, and (c) generating the clinical decision support information using the product-specific reference information for which the reference score satisfies a preset condition.

[0041] The obtaining the clinical decision support information from the product-tailored database using the query text comprises: generating a result of summarizing and analyzing the one or more product-specific reference information in a predefined report format using a generative artificial intelligence (AI) model trained to generate a natural language response to a natural language input.

[0042] The clinical decision support information comprises one or more selected from the group consisting of: (a) an disease identification result and / or a therapy determination result estimated to have a highest probability value based on one or more product-specific reference information that supports the clinical decision support information; (b) summary information for each of the one or more product-specific reference information; (c) a statistical analysis result for a type of disease identification result and / or a type of therapy determination result included in each of the one or more product-specific reference information; and (d) a statistical analysis result for each of a number of pieces of reference information that do not correspond to the detection result of the one or more target analytes among the one or more product-specific reference information, and a clinical date and item-specific clinical values included in the non-corresponding product-specific reference information.

[0043] The clinical decision support information is generated according to a preset standard format.

[0044] The standard format is a format preset in the computer device by a user account of the target terminal or a terminal associated with the target terminal, or is a user-customized format provided from the target terminal or the terminal associated with the target terminal.

[0045] The providing the clinical decision support information comprises: providing the test result, the clinical decision support information, and one or more product-specific reference information that supports the clinical decision support information together.

[0046] The target terminal or the terminal associated with the target terminal comprises at least one selected from the group consisting of a terminal of a medical professional including a clinician's portable terminal or therapy determination terminal, a server of a medical institution to which the medical professional belongs, a database of the medical institution, at least one terminal of a medical professional connected to the server of the medical institution via a network, and a testing terminal of the medical institution.

[0047] The method further comprising, after the step of providing the clinical decision support information to the target terminal or the terminal associated with the target terminal: receiving individual clinical data from the target terminal or the terminal associated with the target terminal, the individual clinical data including: (a) the test result, (b) user information including one or more selected from the group consisting of a user name, affiliation, position, address, contact information, and country corresponding to an account of the target terminal or the terminal associated with the target terminal, and (c) a user's disease identification and therapy determination result based on the clinical decision support information; and storing the individual clinical data in the product-tailored database.

[0048] Wherein usage rights for the individual clinical data are granted only to user accounts that have participated in providing at least one piece of the plurality of individual clinical data stored in the product-tailored database.

[0049] The method further comprising: updating a weight used for obtaining the clinical decision support information based on the product-tailored database, based on the individual clinical data.

[0050] The obtaining the clinical decision support information comprises: when a Cycle threshold value (Ct value) included in the test result and / or an age of a subject is outside a preset appropriate range, obtaining the clinical decision support information based on a weight assigned to a therapy determination type corresponding to symptomatic therapy.

[0051] The obtaining the clinical decision support information comprises: when an underlying disease of a subject included in the test result corresponds to any one of preset cautionary underlying diseases related to an immunocompromised state, obtaining the clinical decision support information using (a) a weight assigned to the any one cautionary underlying disease and (b) a follow-up history including previously stored test results by test date for the subject and clinical decision support information.

[0052] The obtaining the clinical decision support information comprises: when a positive pathogen according to the test result is any one of preset antibiotic-resistant bacteria, obtaining the clinical decision support information including (a) a message recommending to proceed with an additional test including an antibiotic resistance test for the subject, and (b) therapy determination results for each result of the additional test, obtained based on a weight assigned to the any one antibiotic-resistant bacterium.

[0053] A type of the product-specific reference information stored in the product-tailored database includes papers and professional books, the clinical decision support information comprises first clinical decision support information for disease identification and therapy determination generated using the product-specific reference information corresponding to the papers and second clinical decision support information for disease identification and therapy determination generated using the product-specific reference information corresponding to the professional books, and the providing the clinical decision support information comprises: transmitting the clinical decision support information to the target terminal or the terminal associated with the target terminal, to control the first clinical decision support information for disease identification and therapy determination and the second clinical decision support information for disease identification and therapy determination to be displayed in a comparable manner on one screen at the target terminal or the terminal associated with the target terminal.

[0054] According to one embodiment of the present disclosure, reference information that can be referred to for disease identification and therapy determination related to a diagnostic test is provided. In one embodiment, for a patient's diagnostic test results obtained using a diagnostic reagent product, clinical decision support information for disease identification and therapy determination that is selected to be product-specific, highly reliable, and highly relevant to the test results may be provided. Accordingly, the medical practice of a medical professional can be performed more effectively by referring to the provided clinical decision support information for disease identification and therapy determination, and user convenience may be improved because the medical professional's cumbersome task of searching for references is omitted.

[0055] According to another embodiment of the present disclosure, product-tailored clinical decision support information and reference information for disease identification and therapy determination may be provided according to the diagnostic reagent product. In particular, a product-tailored database may be provided in which product-specific clinical decision support information for disease identification and therapy determination, selected according to the specification of the diagnostic reagent product, and product-specific reference information that supports it are stored, and clinical decision support information for disease identification and therapy determination may be provided based on the product-tailored database. Accordingly, medical professionals may refer to more reliable clinical decision support information for disease identification and therapy determination for each diagnostic reagent product.

[0056] According to still another embodiment of the present disclosure, reference information obtained from a reliable data source may be provided. For example, for each diagnostic reagent product, development-related data or clinical data using the product may be provided by the product developer or related institutions during the product development process, and based on such data, clinical decision support information for disease identification and therapy determination specific to the diagnostic reagent product may be provided. Accordingly, medical professionals may refer to highly reliable, product-specific clinical decision support information for disease identification and therapy determination.

[0057] According to still another embodiment of the present disclosure, reference information suitable for a patient's molecular diagnostic test results may be selected based on artificial intelligence to provide clinical decision support information for disease identification and therapy determination. In particular, clinical decision support information for disease identification and therapy determination based on a product-tailored database may be provided using Generative AI. Accordingly, optimal references may be efficiently selected and effectively analyzed based on artificial intelligence, and response reliability can be improved by referencing the product-tailored database.

[0058] The effects of the present disclosure are not limited to the above-described effects, and it should be understood that the present disclosure includes all effects that can be inferred from the configurations of the invention described in the detailed description or the claims.

[0059] FIG. 1 is a block diagram illustrating a system for disease identification and therapy determination according to an embodiment.

[0060] FIG. 2 schematically illustrates a block diagram of a computer device according to an embodiment.

[0061] FIG. 3 is a diagram illustrating a modular representation of software implemented by the computer device shown in FIG. 2.

[0062] FIG. 4 illustrates an exemplary flowchart for a computer device to build a product-tailored database according to an embodiment.

[0063] FIG. 5 is a diagram illustrating an exemplary test result according to an embodiment.

[0064] FIG. 6 schematically illustrates a block diagram of a query text according to an embodiment.

[0065] FIG. 7 schematically illustrates a block diagram of clinical decision support information for disease identification and therapy determination according to an embodiment.

[0066] FIG. 8 exemplarily illustrates a conceptual diagram of a process for obtaining clinical decision support information for disease identification and therapy determination according to a first embodiment.

[0067] FIG. 9 exemplarily illustrates a conceptual diagram of a process for obtaining clinical decision support information for disease identification and therapy determination according to a second embodiment.

[0068] FIG. 10 exemplarily illustrates a conceptual diagram of a process for obtaining clinical decision support information for disease identification and therapy determination according to a third embodiment.

[0069] FIG. 11 is a diagram exemplarily illustrating the display of clinical decision support information for disease identification and therapy determination on a screen of a target terminal or a terminal associated with the target terminal according to an embodiment.

[0070] FIG. 12 is a diagram exemplarily illustrating the display of additional information based on a user input related to clinical decision support information for disease identification and therapy determination on a screen of a target terminal or a terminal associated with the target terminal according to an embodiment.

[0071] FIG. 13 schematically illustrates a block diagram of individual clinical data according to an embodiment.

[0072] FIG. 14 illustrates an exemplary flowchart for a computer device to provide clinical decision support information for disease identification and therapy determination according to an embodiment.

[0073] In accordance with an aspect of the present disclosure, there is provided a computer device, comprising: a memory storing a computer program including one or more instructions; and a processor that loads the computer program from the memory and execute the computer program, wherein the one or more instructions, when executed by the processor, cause the processor to: receive a test result obtained using one or more diagnostic test reagent products selected based on sample information, from a target terminal; obtain, based on the test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database, wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information as evidence of the product-specific clinical decision support information, both of which are labeled with an identifier based on a specification of a diagnostic test reagent product, whereby the product-specific clinical decision support information and the product-specific reference information can be used as considerations for the disease identification and therapy determination based on the test result using the one or more diagnostic test reagent products; and provide the obtained clinical decision support information to the target terminal or to a terminal associated with the target terminal.

[0074] In accordance with another aspect of the present disclosure, there is provided A computer-readable non-transitory recording medium storing a computer program, wherein the computer program, when executed by a processor, causes the processor to: receive a test result obtained by testing with one or more diagnostic test reagent products selected based on sample information, from a target terminal; obtain, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database, wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information as evidence of the product-specific clinical decision support information, both of which are labeled with an identifier based on a specification of a diagnostic test reagent product, whereby the product-specific clinical decision support information and the product-specific reference information can be used as considerations for the disease identification and therapy determination based on the test result using the one or more diagnostic test reagent products; and provide the obtained clinical decision support information to the target terminal or to a terminal associated with the target terminal.

[0075] The advantages and features of the embodiments and the methods of accomplishing the embodiments will be clearly understood from the following description taken in conjunction with the accompanying drawings. However, embodiments are not limited to those embodiments described, as embodiments may be implemented in various forms. It should be noted that the present embodiments are provided to make a full disclosure and also to allow those skilled in the art to know the full range of the embodiments. Therefore, the embodiments are to be defined only by the scope of the appended claims.

[0076] Terms used in the present specification will be briefly described, and the present disclosure will be described in detail.

[0077] In terms used in the present disclosure, general terms currently as widely used as possible while considering functions in the present disclosure are used. However, the terms may vary according to the intention or precedent of a technician working in the field, the emergence of new technologies, and the like. In addition, in certain cases, there are terms arbitrarily selected by the applicant, and in this case, the meaning of the terms will be described in detail in the description of the corresponding invention. Therefore, the terms used in the present disclosure should be defined based on the meaning of the terms and the overall contents of the present disclosure, not just the name of the terms.

[0078] When it is described that a part in the overall specification "includes" a certain component, this means that other components may be further included instead of excluding other components unless specifically stated to the contrary.

[0079] In addition, a term such as a "unit" or a "portion" used in the specification means a software component or a hardware component such as FPGA or ASIC, and the "unit" or the "portion" performs a certain role. However, the "unit" or the "portion" is not limited to software or hardware. The "portion" or the "unit" may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example, the "unit" or the "portion" includes components (such as software components, object-oriented software components, class components, and task components), processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, database, data structures, tables, arrays, and variables. The functions provided in the components and "unit" may be combined into a smaller number of components and "units" or may be further divided into additional components and "units".

[0080] Hereinafter, the embodiment of the present disclosure will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art may easily implement the present disclosure. In the drawings, portions not related to the description are omitted in order to clearly describe the present disclosure.

[0081] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0082] Before describing FIG. 1, some terms used herein will be reviewed.

[0083] In this specification, the term "test result" indicates test data obtained from a diagnostic test. The diagnostic test may be a test using at least one diagnostic technique selected from the group consisting of In Vitro Diagnostics (IVD), imaging diagnostics, endoscopic examination, functional testing, tissue examination, and nuclear medicine examination, but is not limited to any one thereof. An in vitro diagnostic test according to one embodiment may be a diagnostic test using at least one diagnostic technique selected from the group consisting of Molecular Diagnostics (MDx), point-of-care testing (POCT), Immunochemistry, self-monitoring of blood glucose, hematology diagnostics, clinical microbiology diagnostics, hemostasis diagnostics, and histopathology diagnostics, but is not limited to any one thereof.

[0084] The term "molecular diagnostic test result" refers to test data obtained from a molecular diagnostic test. A molecular diagnostic test according to one embodiment is a method for detecting a target analyte using a nucleic acid, and may be, for example, a diagnostic test using a nucleic acid amplification reaction that amplifies a target nucleic acid (e.g., a viral or bacterial nucleic acid).

[0085] The aforementioned target analyte, particularly a target nucleic acid molecule, may be amplified by various methods: polymerase chain reaction (PCR) (Mullis et al., U.S. Patent Nos. 4,683,195, 4,683,202, and 4,800,159; Saiki et al., Science 230:1350-1354(1985)), ligase chain reaction (LCR), strand displacement amplification (SDA), transcription-mediated amplification, nucleic acid sequence-based amplification (NASBA), rolling circle amplification (RCA) and Q-Beta Replicase, loop-mediated isothermal amplification (LAMP), recombinase polymerase amplification (RPA), etc.

[0086] The term "target analyte" may refer to various substances (e.g., biological materials and non-biological materials). Specifically, such a target analyte may include at least one of biological materials, more specifically nucleic acid molecules (e.g., DNA and RNA), proteins, peptides, carbohydrates, lipids, amino acids, biological compounds, hormones, antibodies, antigens, metabolites, and cells.

[0087] The term "sample" includes biological samples (e.g., cells, tissues, and bodily fluids) and non-biological samples (e.g., food, water, and soil). Among these, the biological sample may include at least one of, for example, a virus, bacterium, tissue, cell, blood (including whole blood, plasma, and serum), lymph, bone marrow fluid, saliva, sputum, swab, aspiration, milk, urine, stool, ocular fluid, semen, brain extract, cerebrospinal fluid, joint fluid, pleural fluid, bronchial lavage fluid, ascites, and amniotic fluid. Such a sample may or may not include the aforementioned target analyte.

[0088] Meanwhile, when the aforementioned target analyte is or includes a nucleic acid molecule, the sample presumed to contain the target analyte may be subjected to a nucleic acid extraction process known in the art (see: Sambrook, J. et al., Molecular Cloning. A Laboratory Manual, 3rd ed. Cold Spring Harbor Press (2001)). The nucleic acid extraction process may vary depending on the type of sample. In addition, if the extracted nucleic acid is RNA, a reverse transcription process for synthesizing cDNA may be additionally performed (see: Sambrook, J. et al., Molecular Cloning. A Laboratory Manual, 3rd ed. Cold Spring Harbor Press (2001)).

[0089] According to an embodiment, an amplification reaction for amplifying a signal indicating the presence of a target analyte may be performed in a manner in which a signal is also amplified as the target analyte is amplified (e.g., a real-time PCR method). Alternatively, according to one embodiment, the amplification reaction may be performed in a manner in which the target is not amplified, and only the signal indicating the presence of the target is amplified (e.g., CPT method (Duck P, et al., Biotechniques, 9:142-148(1990)), Invader assay (U.S. Patent Nos. 6,358,691 and 6,194,149)). As such, an amplification reaction may be accompanied by a signal change, and therefore, the progress of such an amplification reaction may be evaluated by measuring the signal change. As a means of providing such a signal, a signal-generating composition including a label itself or an oligonucleotide linked to a label may be used. Various methods (e.g., TaqManTM probe method, molecular beacon method, etc.) for generating a signal indicating the presence of a target analyte using a signal-generating composition are known.

[0090] Based on such an amplification reaction, a multiplex diagnostic technique for detecting multiple target nucleic acids in a single tube may be used. For example, there are various multiplex technologies for simultaneously detecting several types of viruses using methods such as the aforementioned PCR and LAMP as examples of nucleic acid amplification reactions.

[0091] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art to which the present disclosure pertains may easily carry out the embodiments. However, the present disclosure may be embodied in many different forms and is not limited to the embodiments described herein.

[0092] FIG. 1 is a block diagram illustrating a system 1000 for disease identification and therapy determination according to an embodiment.

[0093] Referring to FIG. 1, the system 1000 for disease identification and therapy determination may include a product-tailored database 100, a computer device 200, and one or more target terminals 300. At least some of the components included in the system 1000 for disease identification and therapy determination may be connected to each other through a network. In one embodiment, the computer device 200 may be connected to the product-tailored database 100 and the one or more target terminals 300 through a network. Here, the network may be configured through various communication networks such as wired and wireless networks, and may be composed of various communication networks such as a Local Area Network (LAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), etc. In another embodiment, the computer device 200 may mediate data communication between the product-tailored database 100 and the target terminal 300.

[0094] The product-tailored database 100 may store information including content related to disease identification and / or therapy determination related to diagnosis (e.g., molecular diagnosis). In one embodiment, such information may be that which has been collected from existing public information. In one embodiment, such information may be that which has been collected from developers or related institutions during the development process of each product. Specifically, the product-tailored database 100 may store reference information including content related to disease identification and / or therapy determination associated with each of a plurality of diagnostic test reagent products. In other words, the product-tailored database 100 may be a database in which reference information related to each diagnostic reagent product is mapped to and stored with the respective product identifier.

[0095] Here, reference information refers to information including content that may be referred to for disease identification and / or therapy determination related to diagnosis. According to an implementation, reference information may be broadly construed as a concept encompassing various types, such as papers published by various academic or research institutions, specialized books used as textbooks in medical schools, guidelines related to the disease identification or therapy determination of certain diseases announced by various institutions, specialized knowledge, experience, and know-how related to disease identification or therapy determination provided by various medical professionals, development-related materials provided by product developers for each diagnostic reagent product, and clinical data using the product. Reference information according to one embodiment may have various formats including documents, images, and videos, or may be the result of conversion or standardization to have a predefined specific format (e.g., a certain document format) from the various formats of the original.

[0096] The product-tailored database 100 may store reference information selected according to the specification of the product so that it may be referred to for the disease identification and / or therapy determination of test results using the diagnostic reagent product. Specifically, the product-tailored database 100 may be for storing (i) one or more product-specific clinical decision support information for disease identification and therapy determination and (ii) one or more product-specific reference information that supports the corresponding product-specific clinical decision support information. The product-specific clinical decision support information and the product-specific reference information stored in the product-tailored database 100 may be labeled with an identifier based on the specification of the diagnostic reagent product so that they may be referenced for the disease identification and therapy determination of test results using the diagnostic reagent product. For example, the product specification may include target analyte information that the product is intended to detect or analyze (e.g., species of a pathogen), the type of sample available for the product, and host information. Among the collected reference information, reference information describing content related to at least some of this specification information may be selected for the product, and the selected reference information may be labeled with the product identifier (e.g., product code). This will be described in more detail later.

[0097] A product-tailored database 100 according to one embodiment may be one of the components of the computer device 200. For example, the computer device 200 may include the product-tailored database 100 as an entity that performs data storage and management, and the product-tailored database 100 may be included in the computer device 200 or exist under the management of the computer device 200.

[0098] A product-tailored database 100 according to another embodiment may be one of the components of the system 1000 for disease identification and therapy determination, and may exist outside the computer device 200 and be implemented in a form communicable with the computer device 200. For example, the product-tailored database 100 may be managed and controlled by a different server than the computer device 200, or may be at least partially managed by the computer device 200.

[0099] A product-tailored database 100 according to still another embodiment may be at least partially implemented as a cloud or include a separate storage server to provide storage space to the computer device 200.

[0100] The computer device 200 may provide clinical decision support information for disease identification and therapy determination based on the product-tailored database 100. In one embodiment, the computer device 200 may access a pre-configured product-tailored database 100 and may generate clinical decision support information for disease identification and therapy determination using the reference information stored in the product-tailored database 100.

[0101] Here, disease identification is for identifying a disease by interpreting the test results of a subject, and according to embodiments, it may be broadly construed as a concept encompassing a method for disease identification for disease identification, the disease identification result as the outcome of the disease identification, the provision of reference information during the disease identification process, and a series of procedures thereof. Furthermore, a therapy determination refers to information for determining a therapeutic intervention for treating a disease, and according to embodiments, it may be broadly construed as a concept encompassing a method of prescribing medication to treat a disease or alleviate symptoms, a sample therapy determination slip detailing the type and dosage of medication, duration of administration, and other particulars as the result of the therapy determination, the provision of reference information during the prescribing process, and a series of procedures thereof. In one embodiment, other particulars may include precautions for taking the medication, precautions for preventing side effects, precautions for alleviating the disease or symptoms, or reference points for maximizing the effect of the medication, and may include, for example, information on recommended, prohibited, or restricted physical therapy elements (e.g., sauna), dietary elements (e.g., Ca+ intake), and lifestyle elements (e.g., smoking cessation).

[0102] The computer device 200 receives a test result of a subject and may provide clinical decision support information for disease identification and therapy determination based on the product-tailored database 100 using the received test result. In one embodiment, the clinical decision support information for disease identification and therapy determination is information that may be referenced for the disease identification and therapy determination of a subject's test results obtained using a diagnostic reagent product, and may be, in one embodiment, the clinical decision support information with a relatively high relevance to the subject's test results from among the product-specific clinical decision support information stored in the product-tailored database 100, and may include a recommended disease identification result and a recommended therapy determination result estimated for the corresponding test results. Here, a subject according to one embodiment mainly refers to a living organism including an animal (e.g., human, cow, pig, chicken, etc.) or a plant, but is not limited thereto. According to an implementation, a subject may be broadly construed as a concept encompassing inanimate objects such as soil or water, and the clinical decision support information for disease identification and therapy determination may be information regarding disease identification or therapy determination suitable for the type of such a subject according to the diagnostic reagent product.

[0103] The computer device 200 receives the test result of a subject from a target terminal 300 and may provide clinical decision support information for disease identification and therapy determination based on the received test result to the target terminal 300 or a terminal (not shown) associated with the target terminal 300. In one embodiment, the target terminal 300 may be a diagnostic test apparatus where a diagnostic test using a sample collected from a subject is performed, or a user terminal of a medical professional connected to the diagnostic test apparatus via a wired or wireless network, and the target terminal 300 may obtain and store the subject's diagnostic test results generated by the diagnostic test apparatus. In another embodiment, the target terminal 300 may be a server of a medical institution, and may be a server of a medical institution that receives, stores, and manages the diagnostic test results of a subject from the aforementioned diagnostic test apparatus or the user terminal of a medical professional.

[0104] The computer device 200, the target terminal 300, and the device associated with the target terminal 300 according to one embodiment may each be implemented according to the various embodiments described below.

[0105] The computer device 200 may be implemented as one or more computers that operate via a computer program for realizing the functions described in this specification. The computer device 200 may include any type of server and / or any type of user terminal. A server may include, for example, any type of computing system or computing device such as a microprocessor, mainframe computer, digital processor, portable device, and device controller. A user terminal may include any type of terminal capable of interacting with a server or another computing device. User terminals may include, for example, mobile phones, smart phones, desktop computers, laptop computers, personal digital assistants (PDAs), slate PCs, tablet PCs, and ultra-books.

[0106] In one embodiment, the computer device 200 is a server such as a computer operated by a service provider for disease identification and therapy determination based on the product-tailored database 100, and may be implemented, in one embodiment, to be communicable with the target terminal 300 and / or a terminal associated with the target terminal 300 via a network. In another embodiment, the computer device 200 is a server associated with the medical institution of the target terminal 300, and may be, for example, a server of a medical institution implemented in an on-premise environment communicable with the target terminal 300 and / or a terminal associated with the target terminal 300 via an internal network.

[0107] The target terminal 300 and the device associated with the target terminal 300 may each include any type of server and / or any type of user terminal associated with a medical institution or a medical professional. In one embodiment, the target terminal 300 and the device associated with the target terminal 300 are each a computing terminal owned or used by a client (e.g., a medical institution, a medical professional), and may be, for example, any type of user terminal associated with a medical professional or a server associated with a medical institution. The target terminal 300 and the device associated with the target terminal 300 according to one embodiment may include at least one selected from the group consisting of a terminal of a medical professional including a clinician's mobile terminal or therapy determination terminal, a terminal of a medical professional including a testing apparatus or a laboratory physician's terminal, a server of the medical institution to which the medical professional belongs (not shown), a database of the medical institution (not shown), and at least one terminal of a medical professional connected to the server of the medical institution via a network. In one embodiment, the aforementioned server or database of the medical institution may include one or more devices among a LIS (Laboratory Information System), an EMR (Electronic Medical Record), and an HIS (Hospital Information System) of the corresponding medical institution, or may be implemented to operate while connected to the one or more devices.

[0108] According to one implementation, the computer device 200 may receive a test result from the target terminal 300 and provide clinical decision support information for disease identification and therapy determination based on the test result to the target terminal 300. In one implementation, the target terminal 300 is a terminal of a medical professional, including a clinician's mobile terminal or therapy determination terminal, and the computer device 200 may receive a test result from the medical professional's terminal and provide clinical decision support information regarding it to the corresponding medical professional's terminal. In one implementation, the target terminal 300 is a device such as a medical institution's LIS, and the computer device 200 may receive a test result from the medical institution's device and provide clinical decision support information regarding it to the corresponding medical institution's device. Accordingly, the corresponding medical institution's device may store the provided clinical decision support information in association with the corresponding test result, so that when the test result is provided to a medical professional's terminal, the corresponding clinical decision support information may be provided together.

[0109] According to another implementation, the computer device 200 may receive a test result from the target terminal 300 and provide clinical decision support information for disease identification and therapy determination based on the test result to a device associated with the target terminal 300. In another implementation, the target terminal 300 is a device such as a medical institution's LIS, and the device associated with the target terminal 300 may be a terminal of a medical professional, including a clinician's mobile terminal or therapy determination terminal. For example, the computer device 200, as a server of a medical institution, may receive a test result from a device such as the medical institution's LIS and provide clinical decision support information regarding it to the terminal of the corresponding medical professional related to the test result. In another implementation, the target terminal 300 is a terminal of one medical professional, including a testing apparatus or a laboratory physician's terminal, and the device associated with the target terminal 300 may be a device such as a medical institution's LIS or a terminal of another medical professional, including a clinician's mobile terminal or therapy determination terminal related to the test result. In another implementation, the device associated with the target terminal 300 may be a terminal designated by the target terminal 300. For example, when a test result is received from the target terminal 300, information about the terminal to which the clinical decision support information is to be transmitted may be received together, or it may be determined according to internal policy information as the terminal of the physician in charge corresponding to the user account of the target terminal 300 and the test result.

[0110] However, the aforementioned implementations are merely examples, and the embodiments of the present disclosure are not limited thereto and may be implemented in various modified forms.

[0111] Meanwhile, the transmission of the test result by the target terminal 300 described above may be performed manually upon a request made by a user input at the target terminal 300, or may be performed automatically when a predetermined condition (e.g., generation or storage of a test result) is met. In one embodiment, the request may include a request for clinical decision support information for disease identification and therapy determination based on the subject's test result. In another embodiment, the request may include an input of one or more search terms related to the test result and a request for the provision of clinical decision support information for disease identification and therapy determination based on the search terms. In this case, the aforementioned search terms may include information about the diagnostic reagent product (e.g., product identifier, product name, product code, etc.).

[0112] An application that is predefined and distributed may be installed on the target terminal 300 and / or the device associated with the target terminal 300 according to one embodiment. When the application is executed on the target terminal 300, the target terminal 300 may connect to the computer device 100 and transmit / receive data with the computer device 100. The application may be a program module capable of communicating with an external device, and such a program module may be included in the terminal or another device communicable therewith in the form of an operating system, an application program module, and other program modules, and may be physically stored on various known storage devices. For example, the application may be a web-based application program module that operates on the operating system of a computer such as a PC, or a mobile application program module that operates on the operating system of a smartphone or tablet PC.

[0113] The target terminal 300 and / or the device 400 associated with the target terminal 300 according to another embodiment may receive the following information through a web-based service or a cloud-based service provided by the computer device 200, or may receive the following information while communicating with the computer device 200 on its own.

[0114] Meanwhile, in addition to the components shown in FIG. 1, other components may be further included in the system 1000 for disease identification and therapy determination. In some embodiments, the system 1000 for disease identification and therapy determination may further include one or more data providing servers (not shown) that are sources for collecting reference information, a server of a medical institution (not shown), a database of a medical institution (not shown), etc. Alternatively, according to another embodiment, some of the components shown in FIG. 1 may be omitted.

[0115] FIG. 2 schematically illustrates a block diagram of a computer device 200 according to an embodiment.

[0116] Referring to FIG. 2, the computer device 200 may include a memory 210, a communication unit 220, and a processor 230. The configuration of the computer device 200 shown in FIG. 2 is merely a simplified example. In one embodiment, the computer device 200 may include other components for performing the computing environment of the computer device 200, and only some of the disclosed components may constitute the computer device 200.

[0117] The memory 210 may store at least one instruction that is executable by the processor 230. In one embodiment, the memory 210 may store any form of information generated or determined by the processor 230 and any form of information received by the computer device 200. In one embodiment, the memory 210 may be a storage medium that stores a computer program that causes the processor 230 to perform operations according to the embodiments of the present disclosure. Therefore, the memory 210 may refer to computer-readable media for storing software code, data targeted for execution of the code, and execution results of the code, which are necessary for carrying out the embodiments of the present disclosure.

[0118] In one embodiment, the memory 210 may refer to any type of storage medium. For example, the memory 210 may include at least one type of storage medium from among flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), Random Access Memory (RAM), Static Random Access Memory (SRAM), Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Programmable Read-Only Memory (PROM), magnetic memory, magnetic disk, and optical disk. The computer device 200 may also operate in conjunction with web storage that performs the storage function of the memory 210 on the internet. The foregoing description of the memory is merely an example, and the memory 210 in the present disclosure is not limited thereto.

[0119] The communication unit 220 may be configured regardless of its communication mode, such as wired and wireless, and may be configured as various communication networks such as a Personal Area Network or a Wide Area Network. In addition, the communication unit 220 may operate based on the well-known World Wide Web, and may also use wireless transmission technologies used for short-range communication, such as Infrared Data Association (IrDA) or Bluetooth. In one embodiment, the communication unit 220 may be responsible for transmitting and receiving data necessary for performing the techniques according to the present disclosure.

[0120] The processor 230 may control the overall operation of the computer device 200 and may perform a series of operations for providing clinical decision support information for disease identification and therapy determination based on test results. In one embodiment, the processor 230 may be composed of at least one core, and may include a processor for data analysis and / or processing, such as a central processing unit (CPU), a graphics processing unit (GPU), a micro controller unit (MCU), a general purpose graphics processing unit (GPGPU), or a tensor processing unit (TPU) of the computer device 200.

[0121] The processor 230 may read a computer program stored in the memory 210 to provide clinical decision support information for disease identification and therapy determination using an Artificial Intelligence (AI) model, according to an embodiment of the present disclosure. In one embodiment, such an AI model may include a Generative AI model. In one embodiment, the generative AI model includes a natural language processing-based language model that is trained to generate natural language responses to natural language inputs using natural language, and may include, for example, GPT (Generative Pre-trained Transformer) or BERT (Bidirectional Encoder Representations from Transformers). In one embodiment, the AI model may be a generative AI model trained using at least one selected from the group consisting of a Large Language Model (LLM), Generative Adversarial Networks (GAN), a Variational Auto-Encoder (VAE), and a transformer. As an example, a language model may be a model that has learned human-used natural language using a large amount of text data and is trained to perform various tasks based on the natural language to generate results suitable for a given task in natural language. The basic structure and learning method of such an AI model will be described again later.

[0122] The processor 230 according to one embodiment may perform operations for training a neural network. The processor 230 may perform calculations for training the neural network, such as processing input data for training in deep learning (DL), extracting features from the input data, calculating errors, and updating the weights of the neural network using backpropagation. At least one of the CPU, GPGPU, and TPU of the processor 230 may process the training of the network function. For example, the CPU and GPGPU may together process the training of the network function and the classification of data using the network function. In addition, in one embodiment of the present disclosure, the processors of a plurality of computing devices may be used together to process the training of the network function and the classification of data using the network function. Furthermore, a computer program executed on the computing device according to an embodiment of the present disclosure may be a CPU, GPGPU, or TPU executable program.

[0123] The processor 230, by executing at least one instruction stored in the memory 210, may perform the technical features according to the embodiments of the present disclosure. Various technical features performed by the processor 230 will be described with reference to FIG. 3.

[0124] FIG. 3 is a diagram illustrating a modular representation of software implemented by the computer device 200 shown in FIG. 2.

[0125] Referring to FIG. 3, the software implemented as the hardware processor 230 executes at least one instruction stored in the memory 210 may be modularized into at least one of a DB construction unit 211, an disease identification and therapy determination unit 212, and a learning unit 213. In one embodiment, each of the DB construction unit 211, the disease identification and therapy determination unit 212, and the learning unit 213 may be implemented as a computer program, and the instructions and data for their execution may be stored in the memory 210 and executed by the processor 230, but it is not limited thereto.

[0126] According to one embodiment of the present disclosure, the DB construction unit 211 may be implemented to build the product-tailored database 100. Specifically, the DB construction unit 211 may build the product-tailored database 100 to store reference information associated with each of one or more diagnostic test reagent products, manage the data stored in the product-tailored database 100, and control the product-tailored database 100 so that the data stored therein is used for providing the clinical decision support information for disease identification and therapy determination described below.

[0127] More specifically, the DB construction unit 211 collects reference information including content related to disease identification and / or therapy determination associated with each of one or more diagnostic test reagent products, and may select reference information from the collected reference information according to the specification of each product so that it may be referred to for the disease identification and / or therapy determination of test results using each of the one or more diagnostic test reagent products. In addition, the DB construction unit 211 performs a database conversion process on the reference information selected for each product, and may store the processed result in the product-tailored database 100. The database conversion process may include, for example, data processing tasks that create a database in a searchable format by applying indexing, metadata setting, labeling, etc., based on predefined indexing items or metadata items for the target data. For example, the DB construction unit 211 may label the reference information selected for each product with an identifier (e.g., a product identifier) based on the specification of each diagnostic reagent product. According to an implementation, the database conversion process may also include data processing tasks that encode a document in a manner suitable for searching and create a database in a searchable format based on a high-dimensional vector that compactly represents the meaning according to the document encoding.

[0128] FIG. 4 illustrates an exemplary flowchart for the computer device 200 to build the product-tailored database 100 according to an embodiment.

[0129] Referring to FIG. 4, in operation S410, the DB construction unit 211 may obtain reference information including content related to disease identification and / or therapy determination from one or more data sources, based on information of a target diagnostic reagent product that is the subject of database creation. In one embodiment, the target diagnostic reagent product may be any one of the one or more diagnostic test reagent products mentioned above.

[0130] Here, a diagnostic reagent refers to a reagent used at least partially in the process of a diagnostic test (e.g., a molecular diagnostic test). In one embodiment, the diagnostic reagent may include a nucleic acid amplification reagent (e.g., a CR reagent) and / or a nucleic acid extraction reagent. For example, a nucleic acid amplification reagent may include a set of oligonucleotides including a primer and a probe for the detection of a target analyte, and a reaction medium used for the nucleic acid amplification reaction. The reaction medium may include pH-related substances (e.g., tris buffer, ethylene-diamine-tetraacetic acid (EDTA), etc.), ionic strength-related substances (e.g., ionic substances), enzymes (e.g., RTase, DNA polymerase, etc.), and enzyme stabilization-related substances (e.g., sugars such as sucrose). In addition, a diagnostic reagent product is a product developed for a certain intended use, and may be, for example, a Multiplex real-time PCR diagnostic reagent product developed by a specific product developer (e.g., Seegene, Inc.) and for simultaneously testing for 4 types of SARS-CoV-2 target genes.

[0131] The information of the target diagnostic reagent product may include at least a part of the specification of the corresponding product. The specification of a diagnostic reagent product according to an embodiment may include, but is not limited to, at least one selected from the group consisting of (a) intended use, (b) design specification, (c) performance specification, and (d) technical specification. Throughout the specification, product refers to a diagnostic reagent product, and product and diagnostic reagent product may be used interchangeably.

[0132] The intended use may include at least one of target analyte information that the product is intended to detect or analyze, pathogen information, pathogenic gene or single nucleotide polymorphism (SNP) information, contamination substance information, sample type, host information, disease name, infectious disease name, and symptom information. For example, in the case of a Multiplex real-time PCR diagnostic reagent product, the target analyte information according to the intended use includes 4 types of SARS-CoV-2 target genes to be simultaneously detected, the infectious disease name includes SARS-CoV-2, the host information includes human, the disease information includes acute respiratory disease, and the symptom information may include lethargy, high fever of 37.5 degrees or higher, cough, sore throat, sputum, muscle pain, headache, dyspnea, and pneumonia. In addition, the sample type may include nasopharyngeal swabs and oropharyngeal swabs.

[0133] The design specification may include information about at least one of an Analyte panel configuration, Optical Channel configuration, internal control (IC), positive control (PC), and negative control (NC) for the product. For example, the design specification may be a configuration in which a total of 26 types of tests are performed in 4 panels for Respiratory.

[0134] The performance specification may include at least one of the sensitivity, specificity, Turn Around Time (TAT), Reproducibility, and Linearity of each product. Here, reproducibility indicates the consistency of molecular diagnostic test results when the same sample is tested at different times or in different places, and linearity indicates how accurate the test result is as the sample concentration increases. Furthermore, the performance specification may further include the limit of detection (LoD), which indicates the minimum detectable sample concentration, the range of inclusive and exclusive strains, In-silico analysis results, and feasibility analysis results.

[0135] The technical specification may include at least one of a sampling technique, extraction technique, nucleic acid amplification test (NAAT) reaction setup technique, amplification technique, signal analysis technique, and information display technique used in the research and development of each product. The sampling technique includes the technique used to collect a sample from a host (e.g., nasopharyngeal swab method), the extraction technique includes the specimen collection technique and / or nucleic acid extraction technique (e.g., bead transfer type, liquid transfer type), the NAAT reaction setup technique includes the technique used for the setup of the nucleic acid amplification reaction (e.g., liquid handler operation, pipetting type, well plate info., setup device), the amplification technique includes the technique used for nucleic acid amplification (e.g., type of nucleic acid amplification reaction (e.g., real-time PCR), PCR instrument, type (e.g., primer only, primer-probe set) or structure of oligonucleotide), the signal analysis technique includes the technique used for the analysis of the signal indicating the nucleic acid amplification result (e.g., target signal generation mechanism for amplification, amplification graph analysis algorithm, noise processing algorithm, etc.), and the information display technique may include the technique used for the visual display of the test result from the signal analysis (e.g., viewing method of the amplification graph). Furthermore, the technical specification may further include information about an automated research and development system.

[0136] In one embodiment, the reference information obtained from the one or more data sources may include at least one of product-specific reference information and public reference information.

[0137] In one embodiment, the product-specific reference information may include at least one of (i) research and development data generated by a researcher or developer during the product research and development phase and (ii) clinical data obtained through clinical trials using the product in the post-development phase. For example, the former research and development data includes one or more of laboratory notes, design documents, and reference literature, and the latter clinical data may include clinical data after product approval or authorization, experimental verification data by an internal or external evaluator, etc. For example, the aforementioned development-related phases include a feasibility test phase to confirm whether a product design or idea is technically feasible, a verification phase to experimentally confirm whether the design output meets the product requirements, a validation phase to confirm the clinical performance of the product in an actual use environment, and an after phase to collect results on the clinical performance of the product after its launch. According to an implementation, the product-specific reference information may be a reference created by a person involved in development during the product's development-related phases, or it may be one or more references directly or indirectly selected or recommended by a person involved in development from among numerous publicly available public reference information.

[0138] As an example, the product-specific reference information may be a reference including an analysis of the clinical significance related to disease identification and / or therapy determination based on PCR data obtained using the corresponding diagnostic reagent product. For example, even when targeting the same target analyte, the clinical significance for each range of the Ct value may be interpreted slightly differently for each diagnostic reagent product according to the specification of each product (e.g., design specification, etc.). The product-specific reference information may include the clinical significance for one or more value ranges of the Ct value in the test results using the product (e.g., the grey zone range of the Ct value and disease identification opinions thereon). In this way, by utilizing internally produced clinical data from clinical trials using the product by the developer or an affiliated analysis institution, it becomes possible to provide more product-specific reference information as literature that forms the basis for a medical professional's judgment on the test results using the product.

[0139] In one embodiment, the public reference information includes reference information disclosed by at least one of academic institutions, medical institutions, and public institutions, and may, for example, include external academic clinical papers that are disclosed with limited or no restrictions. In one embodiment, the DB construction unit 211 may determine a search query based on information of the target diagnostic reagent product and collect search data from one or more data sources through an information search using the determined search query. This search data may be used as the target for selecting reference information suitable for the product in a later step. For example, the product-tailored database 100 stores information of the target diagnostic reagent product provided by the product developer, and the DB construction unit 211 may set a search query to include at least part of the intended use according to the product's specification. For example, in the case of the aforementioned PCR diagnostic reagent product, the DB construction unit 211 sets a search query that includes at least some of Multiplex real-time PCR, SARS-CoV-2, human, acute respiratory disease, and nasopharyngeal swab under an AND condition, and may collect the public reference information as search data by performing a literature information search on a pre-set web search or web public database according to the search query. Alternatively, in a similar manner, product-specific reference information that may be used directly or indirectly for disease identification and therapy determination for the product may be collected as search data by performing a search on an internal database where a large amount of product-specific reference information related to product development is stored.

[0140] In another embodiment, the DB construction unit 211 may determine one or more data sources based on the information of the target diagnostic reagent product and collect reference information from the determined data sources. For example, the DB construction unit 211 may store and manage information on a plurality of data sources that provide reliable reference information on the disease identification and / or therapy determination of molecular diagnostic test results in the product-tailored database 100. When a target diagnostic reagent product is determined, it determines one or more data sources suitable for the product's specification from among the plurality of data sources, and may collect reference information from the data sources by performing a search with the search query.

[0141] In one embodiment, the one or more data sources may include at least one selected from the group consisting of a first data source obtained from the product developer during the product development stage of the diagnostic reagent product, a second data source obtained from an institution associated with the approval of the diagnostic reagent product, a third data source obtained from at least one of a disease research institution, a management institution, and a medical institution according to tests using the diagnostic reagent product, and a fourth data source obtained from a medical expert associated with the diagnostic reagent product. If the data source according to the embodiment described above included data sources based on whether they were product-specific, the data source according to the embodiment described above may include various data sources based on the provider of the reference information.

[0142] The first data source may include cases where the reference information is directly obtained from the product developer, and may include, for example, cases where reliable literature information that may be referred to for disease identification or therapy determination related to the product is provided from the product developer's terminal. In addition, the first data source may include cases where source information of reliable references is obtained from the product developer, and may include, for example, cases where the name of the relevant institution or a webpage address where guideline information for disease identification or therapy determination related to the product is disclosed is provided from the product developer's terminal.

[0143] The second data source may include product approval-related institutions in each country, and may include, for example, the Ministry of Food and Drug Safety of Korea, the Food and Drug Administration (FDA) of the United States, etc. The third data source may include the National Institute of Health, which researches diseases related to the product, the Korea Disease Control and Prevention Agency, and specialists or medical institutions with considerable recognition regarding the treatment of the disease. In the fourth data source, a medical expert may be understood as a concept encompassing medical experts related to animals, plants, and other fields. For example, a medical expert may be a medical professional with experience in clinical trials using the product or clinical trials highly relevant to the product, such as a laboratory physician or clinician, or may be a veterinarian for animals, a plant researcher, a soil researcher, a water quality researcher, etc. However, these multiple data sources are not limited to the embodiments described above, and according to implementations, may further include one or more data providing devices (not shown) that provide public or private medical data on disease disease identification and / or therapy determination. The data providing device may, for example, further include an academic institution server (e.g., Nature server, Science server), a public medical data providing server (e.g., MEDLINE), a private medical data providing server based on a license (e.g., Micromedex, Lexicomp, Epocrates, etc.), an information search (e.g., Google scholar) using a search engine (e.g., Open Access Search Engine), and web crawling.

[0144] In operation S420, the DB construction unit 211 may select product-specific reference information from the obtained reference information based on the specification of the target diagnostic reagent product. In one embodiment, the DB construction unit 211 may store in the product-tailored database 100 the product-specific reference information selected based on the specification of the product so that it may be referred to for the disease identification and / or therapy determination of test results using the target diagnostic reagent product. Throughout the specification, product-specific reference information refers to reference information selected for the diagnostic reagent product and stored in the product-tailored database 100.

[0145] In one embodiment, the product-specific reference information selected based on the specification of the target diagnostic reagent product may be one or more pieces of reference information that at least partially include content related to the specification of the product, from among the reference information or search data collected in operation S410. For example, the DB construction unit 211, as the intended use of the product's specification, selects reference information including the content of each of the target analyte information that the product is intended to detect or analyze, sample type, host information, disease name, infectious disease name, and symptom information, and may perform a database conversion process as product-specific reference information by mapping the identifier of the diagnostic reagent product to the selected reference information.

[0146] In one embodiment, the product-specific reference information selected based on the specification of the target diagnostic reagent product may be that which has been selected based on a weight assigned to a specification relevance, which is a quantified degree of inclusion of content related to the specification of the product. For example, the DB construction unit 211 calculates the specification relevance for each piece of reference information or search data collected in operation S410 by counting whether each contains the content of the target analyte information, sample type, host information, disease name, infectious disease name, and symptom information, or quantitatively calculates the specification relevance using a predetermined mathematical formula in which the inclusion (or level) of each of the contents is an independent variable and the resulting specification relevance is a dependent variable. For example, a relatively high priority may be assigned to the contents. In the aforementioned selection process, the DB construction unit 211 may select a certain number of top-ranking reference information in order of high specification relevance, or select one or more pieces of reference information with a specification relevance equal to or greater than a predefined threshold value.

[0147] In one embodiment, the product-specific reference information selected based on the specification of the target diagnostic reagent product may be that which has been selected based on a weight assigned to at least one selected from the group consisting of (a) specification relevance, (b) reliability of the data source, (c) the number of times cited in other clinical decision support information for disease identification and therapy determination related to the diagnostic reagent product, (d) the number of times cited in other reference information, and (e) the recognition of the author or affiliated institution. In one embodiment, a different predefined mathematical formula may be used to calculate each of the aforementioned factors (a) to (e), and a ranking for each piece of reference information may be calculated by computing the scores in a predetermined manner. In one embodiment, a higher reliability may be assigned to product-specific reference information than to public reference information, and accordingly, product-specific reference information and public reference information may be included in the product-specific reference information in an appropriate ratio.

[0148] In one embodiment, the DB construction unit 211 may select the product-specific reference information by at least partially using a trained AI model. Specifically, the trained AI model may be a model trained to select product-specific reference information including content related to the disease identification and / or therapy determination related to the product's specification from the reference information when the collected reference information is provided. For example, the trained AI model may be a model trained to calculate the ranking of reference information based on at least one of the factors (a) to (e) mentioned above and to select the product-specific reference information based on the ranking. The AI model may include natural language processing and a large language model, and various embodiments thereof will be described again later.

[0149] In one embodiment, the product-specific reference information selected based on the specification of the target diagnostic reagent product may include at least one selected from the group consisting of clinical paper information including clinical trial content related to one or more of a single target analyte or a plurality of target analytes that each product is intended to detect or analyze, clinical paper information including content related to disease identification and / or therapy determination following the sole detection of one of the plurality of target analytes, and clinical paper information including content related to disease identification and / or therapy determination following the simultaneous detection of two or more of the plurality of target analytes. For example, clinical papers related to all target analytes in the test items of the product, clinical papers including at least one target analyte, and clinical papers consistent with co-infection of at least two target analytes may be selected.

[0150] In one embodiment, the DB construction unit 211 may perform a database conversion process on the selected one or more product-specific reference information and store the processed result in the product-tailored database 100. In one embodiment, the database conversion process may include a task of extracting significant information from the selected product-specific reference information and a task of labeling the reference information with metadata based on the extracted information.

[0151] The DB construction unit 211 performs text analysis on each of the selected product-specific reference information and may extract data that meets specific criteria from within each piece of reference information. Specifically, the DB construction unit 211 analyzes whether data of predefined significant items is included in each piece of product-specific reference information, and if it is included, may extract the data of the significant items.

[0152] In one embodiment, the significant items may include two or more selected from the group consisting of target analyte information, pathogen information, pathogenic gene or SNP information, sample type, host information, disease name, infectious disease name, author, affiliation, academic journal, information disclosure date, and clinical information (including at least one of a subject's symptoms, underlying diseases, country, region, age, gender, Ct value (Cycle threshold value) of each target analyte, disease identification result, therapy determination result, and clinical date). In one embodiment, the clinical information may include at least one of a subject's symptoms, underlying diseases, country, region, age, gender, Ct value of each target analyte, disease identification result, therapy determination result, and clinical date. In one embodiment, the significant items may further include one or more key sentences and / or a conclusion included in each piece of reference information, which may be processed in natural language through content and meaning analysis of the document for each piece of reference information, or processed in a form where a predefined classification value for the corresponding item is assigned.

[0153] The DB construction unit 211 may label the selected reference information with metadata corresponding to the content related to the specification of the product included in the reference information. The DB construction unit 211 may label each of the selected product-specific reference information with at least one piece of metadata from among the product name, product identifier, product version, catalog number, product group name associated with the product, and product group identifier. Here, a product group is the result of grouping a plurality of diagnostic test reagent products, and for example, a plurality of product groups may be set according to the type of disease or symptom. For example, the plurality of product groups may include Respiratory infections, Gastrointestinal tract infection, and Human Papillomavirus, and each product may be classified into at least one of the plurality of product groups based on its intended use.

[0154] In one embodiment, the DB construction unit 211 may further label each of the selected product-specific reference information with the two or more pieces of metadata extracted from the reference information as described above. For example, the DB construction unit 211 may label the reference information with two or more pieces of metadata from the group consisting of target analyte information, pathogen information, pathogenic gene or SNP information, sample type, host information, disease name, infectious disease name, author, affiliation, academic journal, information disclosure date, and clinical information (including at least one of a subject's symptoms, underlying diseases, country, region, age, gender, Ct value of each target analyte, disease identification result, therapy determination result, and clinical date) included in each piece of reference information. In one embodiment, this data may be processed into forms such as numerical data or categorical data, predefined based on each data attribute.

[0155] As described above, the DB construction unit 211 may store the product-specific reference information selected for the target diagnostic reagent product in the product-tailored database 100. The DB construction unit 211 may build the product-tailored database 100, in which product-specific reference information is stored for each product, by performing operations S410 to S420 for each diagnostic reagent product. According to an implementation, the process of selecting reference information based on metadata may be performed after the database conversion process, such as metadata labeling, has been performed, but it is not limited thereto.

[0156] In operation S430, the DB construction unit 211 may generate, for each product, one or more product-specific clinical decision support information for disease identification and therapy determination that may be referred to for the disease identification and / or therapy determination of test results using the product, by using the product-specific reference information selected for each product. Here, the clinical decision support information for disease identification and therapy determination may refer to information for assisting, guiding, or supporting the disease identification and therapy determination actions of a medical professional regarding the test results using the product.

[0157] In one embodiment, the disease identification and therapy determination unit 212 may generate product-specific clinical decision support information for disease identification and therapy determination that includes an estimated disease identification result and therapy determination result based on the disease identification results and / or therapy determination results included in the product-specific reference information selected for each product. In another embodiment, the DB construction unit 211 classifies each piece of product-specific reference information selected for each product into one or more of a plurality of classification items based on metadata, and the disease identification and therapy determination unit 212 may generate product-specific clinical decision support information for disease identification and therapy determination for each classification item as described above, using the reference information classified according to each classification item. The plurality of classification items may include, for example, at least one selected from the group consisting of biological classification of the target analyte, sample type, biological classification of the host, disease type, infectious disease type, type of the subject's symptoms, country classification, region classification, age group classification, gender classification, and Ct value category classification. The classification item may also be set as a combination of two or more of the above classifications, and may be set by an administrator to include the components of a query text to be described later, or appropriate classification items may be determined for each product by the DB construction unit 211 analyzing the metadata of the reference information for each product. For example, if a first classification item is set where the target analyte is SARS-CoV-2, the sample is a nasopharyngeal swab, and the patient's regional classification is Asian, reference information including this content is classified as the first classification item, and clinical decision support information for disease identification and therapy determination customized for the first classification item may be generated using the classified reference information.

[0158] Product-specific clinical decision support information for disease identification and therapy determination (hereinafter, product-specific clinical decision support information) according to one embodiment may include at least one selected from the group consisting of (a) an disease identification result and / or therapy determination result estimated to have the highest probability value based on one or more product-specific reference information; (b) summary information for each of the one or more product-specific reference information; and (c) a statistical analysis result for the type of disease identification result and / or therapy determination result included in each of the one or more product-specific reference information (see FIG. 7).

[0159] The estimated disease identification result and / or therapy determination result includes the disease identification result and / or therapy determination result estimated based on the corresponding product-specific reference information, and may be, for example, the disease identification result and therapy determination result estimated to have the highest probability, or a probability value estimated for each disease identification result and therapy determination result included in the product-specific reference information. Such disease identification results and therapy determination results may be processed as a grouped result for disease identification types or therapy determination types that are similar above a certain level.

[0160] The summary information for the one or more product-specific reference information may be summary information for each piece of product-specific reference information, or it may be summary information for one or more pieces of reference information that support the disease identification result and / or therapy determination result estimated to have the highest probability among the product-specific reference information. In one embodiment, the summary information may include at least one selected from the group consisting of a title, clinical date, clinical value, key sentence included in the one or more product-specific reference information, and data source information for the one or more product-specific reference information, and may include, for example, a list of related papers, the title and clinical date of the paper, sentences expressed in the paper, and information regarding the data source of the paper, such as the source name (e.g., institution name) or source link.

[0161] The statistical analysis result may be a statistical analysis result for the type of disease identification result and / or therapy determination result of each piece of product-specific reference information. For example, it may include probability values per disease identification result (or per type) and probability values per therapy determination result (or per type), statistically analyzed from the disease identification results and / or therapy determination results (or their types) included in the product-specific reference information.

[0162] The disease identification and therapy determination unit 212 according to one embodiment may generate the product-specific clinical decision support information using a trained AI model. In one embodiment, the AI model may be a generative AI model trained to generate a natural language response to a natural language input. The disease identification and therapy determination unit 212 may perform a summary and analysis of one or more product-specific reference information using the generative AI model, generate the summary and analysis result as product-specific clinical decision support information in a predefined report format, and may store the generated product-specific clinical decision support information in the product-tailored database 100 in association with the corresponding product-specific reference information.

[0163] In one embodiment, the generative AI model may be a general-purpose Large Language Model. The disease identification and therapy determination unit 212 generates a prompt instructing the generation of one or more product-specific clinical decision support information using one or more product-specific reference information, and by providing the generated prompt to the generative AI model, it may obtain the product-specific clinical decision support information generated by the generative AI model.

[0164] In another embodiment, the generative AI model may be a model trained to generate a plurality of different product-specific clinical decision support information using the product-specific reference information stored in the product-tailored database 100. In one embodiment, the disease identification and therapy determination unit 212 may group one or more product-specific reference information into a plurality of product-specific reference groups (e.g., single / co-infection group, age-specific group, combination group including underlying diseases, etc.) based on whether the values of at least one piece of metadata among a plurality of pieces of metadata correspond to each other. The generative AI model may generate a plurality of product-specific clinical decision support information using each of the plurality of product-specific reference groups. Alternatively, the generative AI model may be a model trained to perform the aforementioned grouping as well.

[0165] In operation S440, the DB construction unit 211 may store in the product-tailored database 100 the generated one or more product-specific clinical decision support information for disease identification and therapy determination and the product-specific reference information that supports the product-specific clinical decision support information. In one embodiment, the product-specific clinical decision support information and the product-specific reference information stored in the product-tailored database 100 may be labeled with an identifier based on the specification of the corresponding diagnostic reagent product, so that they may be referred to for the disease identification and therapy determination of test results using the diagnostic reagent product. In one embodiment, the identifier based on the specification includes a product identifier, and may further include a specification classification identifier determined based on the specification (e.g., intended use, design specification, performance specification, technical specification).

[0166] The DB construction unit 211 may control the provision of clinical decision support information for disease identification and therapy determination that may be referred to for the disease identification and / or therapy determination of diagnostic test results using the target diagnostic reagent product, based on the product-tailored database 100.

[0167] The DB construction unit 211 may update the product-tailored database 100 at a preset interval. The update of the product-tailored database 100 according to one embodiment may include new acquisition of reference information (e.g., updating with the latest clinical papers) from one or more data sources including a reference search module (e.g., a paper search engine) based on the specification of the diagnostic reagent product, additional selection of product-specific reference information from the newly obtained reference information, and updating of the product-specific clinical decision support information by further using the additionally selected product-specific reference information. In addition, the DB construction unit 211 may expand the products applicable to the product-tailored database 100 by performing the above operations S410 to S430 for a new product based on information of the newly developed product received from a new product developer.

[0168] Meanwhile, a product according to one embodiment refers to a single product, and reference information may be selected and created into a database for each product. A product according to another embodiment refers to a product group in which two or more products are grouped based on the product's specification, and reference information may also be selected and created into a database for each product group. Accordingly, in the product-tailored database 100, reference information may be searched based on product information (e.g., product identifier) or product group information (product group identifier).

[0169] The product-tailored database 100 according to one embodiment may include a search engine for information retrieval. In one embodiment, the product-tailored database 100 may include a first search engine (not shown) for searching reference information and / or a second search engine (not shown) for searching clinical decision support information for disease identification and therapy determination. In one embodiment, the first search engine may be implemented to provide search results based on a predefined keyword search method (e.g., regular expressions, SQL-based pattern matching, etc.) or a semantic search method (e.g., similarity comparison in a vector space, etc.). For this purpose, in the product-tailored database 100, the reference information may be previously converted into a database for each diagnostic reagent product by applying indexing, metadata setting, and encoding methods corresponding to the keyword search method or semantic search method. In another embodiment, the second search engine may be implemented to provide search results for clinical decision support information for disease identification and therapy determination based on a keyword search method. For this purpose, each piece of clinical decision support information for disease identification and therapy determination may be converted into a database for each product by applying indexing, metadata setting, etc., corresponding to the keyword search method. Meanwhile, according to implementations, various known database techniques may be applied to the product-tailored database 100, and it is not limited to any one thereof.

[0170] As described above, using the method for building the product-tailored database 100 described in this specification, it is possible to build the product-tailored database 100 in which product-specific clinical decision support information for disease identification and therapy determination for a diagnostic reagent product and the product-specific reference information that supports it are stored. According to an implementation, this technical feature may also be used independently without being combined with the technical feature of providing clinical decision support information for disease identification and therapy determination below.

[0171] The disease identification and therapy determination unit 212 may be implemented to provide clinical decision support information for disease identification and therapy determination based on the product-tailored database 100. Specifically, the disease identification and therapy determination unit 212 receives a test result tested with one or more diagnostic test reagent products from the target terminal 300, and may provide clinical decision support information for disease identification and therapy determination applicable to the diagnostic reagent product from the product-tailored database 100 based on the received test result.

[0172] In one embodiment, the test result is test data obtained by a diagnostic test (e.g., a molecular diagnostic test), and may be a test result obtained using one or more diagnostic test reagent products selected based on sample information. For example, a test order including sample information (e.g., sample ID, sample type, sample container type, test code, etc.) is transmitted to a testing apparatus, a test is performed with a diagnostic reagent product determined based on the sample information in the testing apparatus, and the test result generated by the testing apparatus may be stored in the target terminal 300. As exemplified earlier, the target terminal 300 may be the testing apparatus, a medical professional's terminal connected to the testing apparatus via a network, a server of a medical institution, etc. This test result is one performed using one or more diagnostic test reagent products for which product-specific clinical decision support information is built in the product-tailored database 100, and may be one performed targeting one or more target analytes presumed to be present in the subject's sample.

[0173] FIG. 5 is a diagram illustrating an exemplary test result 10 according to an embodiment.

[0174] Referring to FIG. 5, the test result 10 may include data for each item among at least some of a predefined set of items. These items may include at least one selected from the group consisting of information of one or more target analytes targeted by the diagnostic test, detection result, basis of detection, time of test, information of one or more diagnostic test reagent products used in the diagnostic test, and sample type. For example, these items are basic items and may be raw data obtained from a diagnostic testing apparatus (e.g., a PCR instrument) that performs the diagnostic test, or essential items of a test report generated as a result of analyzing the raw data.

[0175] The information of the one or more target analytes may include a name of the one or more target analytes. In one embodiment, the name of the target analyte may be expressed based on a biological classification system, and for example, may be expressed as a scientific name of a specific level (e.g., species, genus, family, order, etc.) located in a pre-stored taxonomy.

[0176] The detection result of the one or more target analytes may include at least one selected from the group consisting of (a) a detection result of each of the one or more target analytes, such as whether each target analyte is detected, a positive / negative result value, and a detected amount, and (b) a test result determined based on the detection result of each of the one or more target analytes, such as a type of genotype and whether a mutation is included. In one embodiment, the type of genotype may include Genotyping and Haplotyping. Furthermore, the detection basis may be a basis used to determine whether a detection is made in each detection result, and may include, for example, a Ct value of each target analyte, and according to embodiments, may be broadly construed as a concept encompassing a signal value, a cycle value, a quantitative value, and the like related to the detection of the target analyte.

[0177] The information of the one or more diagnostic test reagent products used may include at least one selected from the group consisting of a product name, a product identifier, a product version, a catalog number, a product group name, and a product group identifier for any one of the diagnostic test reagent products used. For example, the information of the one or more diagnostic test reagent products used may include a product name such as AllplexTMSARS-CoV-2 Assay, and a manufacturing number, serial number, etc., as a product identifier pre-assigned to the product.

[0178] In one embodiment, the type of sample may include any one of virus, bacterium, tissue, cell, blood, lymph, bone marrow fluid, saliva, sputum, swab, aspirate, milk, urine, feces, ocular fluid, semen, brain extract, cerebrospinal fluid, synovial fluid, thymic fluid, bronchial lavage fluid, ascites, and amniotic fluid, but is not limited thereto. In another embodiment, the type of sample may include one or more super-categories for the aforementioned types or one or more sub-categories refined for each type, and may, for example, further include sub-categories of swabs (e.g., nasopharyngeal, oropharyngeal, etc.).

[0179] In one embodiment, the aforementioned items may further include at least two selected from the group consisting of a subject's symptoms, a subject's underlying diseases, a subject's de-identified patient information such as the subject's country, region, age, and gender, the subject's biological classification, an extraction technology, an amplification preparation technology, an amplification technology, a signal analysis technology, and an information display technology used for the molecular diagnostic test, and a model name, version, and manufacturer of a test apparatus or analysis software used for the molecular diagnostic test. The biological classification of the subject may also be expressed as a scientific name of a specific level located in a taxonomy, and for example, may be included only in the case of a non-human. As an example, these items are optional items, are not included in raw data obtained from a molecular diagnostic test apparatus, and may be selectively added to the test result 10 based on user input or the like.

[0180] In one embodiment, the test result 10 may include raw data obtained from a diagnostic test apparatus. For example, among the aforementioned items, the detection result of one or more target analytes, the information of any one diagnostic reagent product used, and the type of sample are basic items and may be included in the test result 10 as the raw data.

[0181] In one embodiment, the test result 10 may further include information manually obtained based on a user input of a target terminal 300 and / or information automatically obtained based on software executed on the target terminal 300 or in conjunction with another device. For example, among the aforementioned items, the subject's symptoms, underlying diseases, region, age, gender, etc., are optional items, and may be input by a user (e.g., a doctor) upon a request for related information provision through a corresponding application on the target terminal 300, or may be reflected from a patient's medical records through a server of a medical institution at the target terminal 300 and added to the test result 10.

[0182] In one embodiment, the test result 10 may be a test result generated using at least one of a diagnostic reagent from the same reagent manufacturer, the same quantitative real-time PCR (qPCR) device, the same extraction method, the same target signal generation mechanism, and the same polymerase master mix. Here, being identical may comprehensively mean that the subjects are the same, the types of the subjects are the same, or the subject or the type of the subject belongs to a preset certain category.

[0183] In one embodiment, a computer device 200 may receive the test result 10 based on a user input for requesting clinical decision support information for disease identification and therapy determination based on the test data, through software installed on a target terminal 300 (e.g., analysis software used for analysis of test data obtained by a diagnostic test apparatus and for information display). In another embodiment, the computer device 200 may receive the test result 10 of a subject through an interactive U / I provided via an application installed on the target terminal 300.

[0184] In one embodiment, an disease identification and therapy determination unit 212 may search a product-tailored database 100 for product-specific clinical decision support information that is labeled with an identifier corresponding to the one or more diagnostic test reagent products used, and obtain it as clinical decision support information for disease identification and therapy determination tailored to the corresponding product. For example, the disease identification and therapy determination unit 212 may search the product-tailored database 100 for clinical decision support information for disease identification and therapy determination that is labeled with an identifier (e.g., a product code) of the diagnostic reagent product used in the test result 10.

[0185] In another embodiment, the disease identification and therapy determination unit 212 may generate a query text from the test result 10 and use the query text to obtain clinical decision support information for disease identification and therapy determination customized for the corresponding product from the product-tailored database 100. The query text is a text for information search in the product-tailored database 100, and may be data extracted from the test result 10 or data modified, processed, or converted from the data into a form suitable for DB search.

[0186] FIG. 6 schematically illustrates a block diagram of a query text 20 according to one embodiment.

[0187] Referring to FIG. 6, the query text 20 may include a detection result 21 of one or more target analytes, information 22 on one or more diagnostic test reagent products used, and a type of sample 23, and each may be understood by referring to the above-described embodiments. For example, the detection result 21 of one or more target analytes may be a positive / negative result value for each of Sarbecovirus (E gene), SARS-CoV-2 (N gene), SARS-CoV-2 (RdRP gene), and SARS-CoV-2 (S gene), or may be a name or identifier of a target analyte that is positive among them. Furthermore, the information 22 on the one or more diagnostic test reagent products used may be a diagnostic reagent product name or product identifier included in the test result 10, or a corresponding management number, which may be an identification number preset for each product for DB search in advance.

[0188] In one embodiment, the query text 20 may further include at least two selected from the group consisting of (a) a subject's symptoms, (b) a subject's underlying diseases, (c) a subject's de-identified patient information such as the subject's country, region, age, and gender, and (d) a Ct value of one or more target analytes. In one embodiment, each of these may be original text extracted from the test result 10, or a predefined identifier or a fixed text expression corresponding to the content, type, or value range of the text.

[0189] In one embodiment, as described above, the query text 20 may further include at least a part of raw data obtained from the corresponding diagnostic test apparatus and included in the test result 10, for example, the detection result 21 of one or more target analytes, the information 22 of any one diagnostic reagent product used, the type of sample 23, and the Ct value of each target analyte. Furthermore, the query text 20 may further include (i) information manually obtained based on a user input of the target terminal 300, or (ii) information automatically obtained based on software executed on the target terminal 300 or in conjunction with another device, for example, the subject's symptoms, underlying diseases, and de-identified patient information. In one embodiment, a part of the query text 20 may be obtained by the former method, and another part may be obtained by the latter method.

[0190] In one embodiment, the disease identification and therapy determination unit 212 may use the query text 20 to search the product-tailored database 100 for at least one piece of product-specific reference information, and may generate clinical decision support information for disease identification and therapy determination using the searched at least one piece of product-specific reference information. For example, in the product-tailored database 100, product-specific reference information is selected for each product and is periodically updated, and the disease identification and therapy determination unit 212 may, upon a request from the target terminal 300, perform a real-time search for product-specific reference information associated with the query text 20 from the product-tailored database 100 to generate clinical decision support information for disease identification and therapy determination.

[0191] In another embodiment, the disease identification and therapy determination unit 212 may use the query text 20 to search the product-tailored database 100 for clinical decision support information for disease identification and therapy determination suitable for the query text 20. For example, as described above, product-specific reference information and product-specific clinical decision support information are prepared for each product in the product-tailored database 100, and according to embodiments, this information may be distributed as an information set during the product development process. Furthermore, the disease identification and therapy determination unit 212 may search for clinical decision support information for disease identification and therapy determination that is customized for a classification item corresponding to the query text 20 from among the product-specific clinical decision support information.

[0192] Specifically, the disease identification and therapy determination unit 212 may search the product-tailored database 100 for one or more product-specific clinical decision support information labeled with an identifier and / or metadata related to the query text 20. For example, among the product-specific clinical decision support information labeled with the corresponding product identifier, the disease identification and therapy determination unit 212 may search for product-specific clinical decision support information in which the content of each item of the query text 20 and the value of the corresponding metadata item at least partially correspond, and may determine the product-specific clinical decision support information with the most corresponding items as the clinical decision support information for disease identification and therapy determination applicable to the corresponding product. The above-mentioned information search may be performed based on search query relevance.

[0193] In the former embodiment, for each of at least one piece of product-specific reference information for the corresponding product used, the disease identification and therapy determination unit 212 may calculate a search query relevance that quantifies the degree to which it contains content related to the query text 20, and may determine the product-specific clinical decision support information that satisfies a preset condition (e.g., weighting the search query relevance, being above a threshold value, a specific number of top results, etc.) as the clinical decision support information for disease identification and therapy determination suitable for the test result 10.

[0194] In the latter embodiment, for each of one or more pieces of pre-generated product-specific clinical decision support information for the corresponding product used, the disease identification and therapy determination unit 212 may calculate a search query relevance that quantifies the degree to which each piece of product-tailored clinical decision support information or the product-specific reference information used to generate the information contains content related to the query text 20, and may determine the product-specific clinical decision support information that satisfies a preset condition (e.g., weighting the search query relevance, being above a threshold value, a specific number of top results, etc.) as the clinical decision support information for disease identification and therapy determination suitable for the test result 10.

[0195] In one embodiment, when a preset reference search condition is met (e.g., the test result 10 includes a special underlying disease, the search query relevance is below a set value), the disease identification and therapy determination unit 212 may search the product-tailored database 100 for one or more product-specific reference information labeled with an identifier and / or metadata related to the query text 20. Furthermore, for each of the searched one or more product-specific reference information, the disease identification and therapy determination unit 212 may calculate a reference score using at least one selected from the group consisting of a search query relevance that quantifies the degree to which it contains content related to the query text 20, a reliability of a data source, the number of times it has been cited in other clinical decision support information for disease identification and therapy determination related to the diagnostic reagent product, the number of times it has been cited in other reference information, and the recognition of the author or affiliated institution. Furthermore, the disease identification and therapy determination unit 212 may generate clinical decision support information for disease identification and therapy determination suitable for the test result 10 using the product-specific reference information whose calculated reference score satisfies a preset condition. For example, when the detailed information included in the test result 10 corresponds to a special case (e.g., an underlying disease, etc.) and it is necessary to refine the product-specific clinical decision support information pre-stored in the product-tailored database 100, the disease identification and therapy determination unit 212 may use the query text 20 to select, in real time, reference information that better matches the query text 20 from among the product-specific reference information, and may generate, in real time, clinical decision support information for disease identification and therapy determination suitable for the corresponding test result 10 using the selected reference information.

[0196] The above-mentioned information search may be performed based on a reference score.

[0197] In the former embodiment, for each of at least one piece of product-specific reference information searched from the product-tailored database 100 using the query text 20, the disease identification and therapy determination unit 212 may calculate a reference score using at least one selected from the group consisting of (a) the specification relevance, (b) a search query relevance that quantifies the degree to which it contains content related to the query text 20, (c) a test result relevance that quantifies the degree to which it contains content related to the test result 10, (d) a reliability of the reference information, (e) the number of times it has been cited in other clinical decision support information for disease identification and therapy determination related to any one of the diagnostic test reagent products used, (f) the number of times it has been cited in other reference information, and (g) the recognition of the author or affiliated institution. The disease identification and therapy determination unit 212 may generate clinical decision support information for disease identification and therapy determination by selectively using the product-specific reference information whose reference score satisfies a preset condition (e.g., above a threshold value, a specific number of top results) from among the searched product-specific reference information.

[0198] In the latter embodiment, for each of one or more product-specific clinical decision support information searched using the query text 20, the disease identification and therapy determination unit 212 may calculate a reference score using at least some of the elements (a) to (g) above for the product-specific reference information used to generate each piece of product-specific clinical decision support information, and may determine the product-specific clinical decision support information whose calculated reference score satisfies a preset condition as the clinical decision support information for disease identification and therapy determination suitable for the test result 10.

[0199] FIG. 7 schematically illustrates a block diagram of clinical decision support information 30 for disease identification and therapy determination according to one embodiment.

[0200] Referring to FIG. 7, the clinical decision support information 30 for disease identification and therapy determination may include at least one selected from the group consisting of an estimated disease identification result and / or therapy determination result 31, summary information 32 for one or more pieces of reference information, and a statistical analysis result 33 based on one or more pieces of reference information. Overlapping descriptions will be omitted.

[0201] The estimated disease identification result and / or therapy determination result 31 may include an disease identification result and / or a therapy determination result estimated based on one or more pieces of reference information searched from the product-tailored database 100 using the query text 20. Furthermore, the summary information 32 for the one or more pieces of reference information may be summary information for the searched one or more pieces of reference information, or may be summary information for one or more pieces of reference information that support the disease identification result and / or therapy determination result estimated to have the highest probability as described above, among the searched one or more pieces of reference information. Furthermore, the statistical analysis result 33 may be a statistical analysis result for the type of disease identification result and / or therapy determination result of each of the searched one or more pieces of reference information.

[0202] In one embodiment, the clinical decision support information 30 for disease identification and therapy determination may further include at least one selected from the group consisting of the number of pieces of reference information that do not correspond to the detection result 21 of one or more target analytes among the searched one or more pieces of reference information, and a statistical analysis result for each of a clinical date and item-specific clinical values included in the corresponding non-corresponding reference information. For example, when the searched reference information includes a therapy determination paper that is inconsistent with the positive result in the test result 10, the clinical decision support information 30 for disease identification and therapy determination may further include at least one of a result statistically processed based on the number of papers and a result of statistically processing all clinical data published in the corresponding papers regardless of the number of papers.

[0203] In one embodiment, the clinical decision support information 30 for disease identification and therapy determination may further include at least one of the aforementioned search query relevance and reference score for each piece of reference information used to generate the clinical decision support information 30 for disease identification and therapy determination. For example, the clinical decision support information 30 for disease identification and therapy determination may further include a specification relevance, a search query relevance, a test result relevance, etc., as a reference score for each of the papers used to generate the clinical decision support information 30 for disease identification and therapy determination.

[0204] The disease identification and therapy determination unit 212 according to an embodiment may acquire the clinical decision support information 30 for disease identification and therapy determination using a trained AI model. In one embodiment, at least one of searching for product-specific clinical decision support information from the product-tailored database 100, calculating a search query relevance, and determining clinical decision support information for disease identification and therapy determination based on the search query relevance, as described above, may be performed using the trained AI model.

[0205] In one embodiment, the AI model may include a generative AI model trained to generate a natural language response to a natural language input. According to the various embodiments below, clinical decision support information for disease identification and therapy determination may be obtained using an AI model, and this will be described with reference to FIGS. 8 to 10. The embodiments of the AI model described below may at least partially correspond to the AI model used in the process of building the product-tailored database 100 described above, and it may be understood that at least some of the following embodiments may be applied in a similar manner in the process of building the product-tailored database 100. Similarly, overlapping descriptions will be omitted.

[0206] FIG. 8 illustratively shows a conceptual diagram of a process for obtaining clinical decision support information 30 for disease identification and therapy determination according to a first embodiment.

[0207] Referring to FIG. 8, the disease identification and therapy determination unit 212 according to the first embodiment may generate the clinical decision support information 30 for disease identification and therapy determination using a trained AI model 40. In one embodiment, the AI model 40 may be a generative AI model trained to acquire a search result 50 including one or more pieces of reference information from the product-tailored database 100 using a training text, and to generate the aforementioned clinical decision support information for disease identification and therapy determination using the obtained search result. In other words, the AI model 40 may be trained to perform both a DB search function and a function of generating clinical decision support information for disease identification and therapy determination based on the search, using the training text as training data. Embodiments related to the training of the AI model 40 will be described later in the section on a training unit 213.

[0208] In the first embodiment, the disease identification and therapy determination unit 212 may provide the query text 20 to the AI model 40 as input data, and may acquire the clinical decision support information 30 for disease identification and therapy determination generated in response to the query text 20 from the AI model 40. The AI model 40 may use the query text 20 to acquire a search result 50 including one or more product-specific reference information from the product-tailored database 100, and may generate clinical decision support information 30 for disease identification and therapy determination suitable for the test result 10 using the search result 50. In one embodiment, the AI model 40 may be implemented through a specific training process performed by the computer device 200. The disease identification and therapy determination unit 212 may receive the AI model 40 trained by the computer device 200 and store it in a memory 210, and execute the AI model 40 to prepare for obtaining clinical decision support information for disease identification and therapy determination. As described above, the AI model 40 may include a transformer-based language model such as GPT, BERT, or a derivative model thereof. Furthermore, the process of providing input data according to one embodiment may further include a preprocessing process for converting it into a form suitable for use as input in the AI model 40, and may, for example, further include a process for text preprocessing such as data tokenization, cleaning, encoding, and embedding. These processes may be performed with reference to known technologies, and a description thereof will be omitted.

[0209] FIG. 9 illustratively shows a conceptual diagram of a process for obtaining clinical decision support information 30 for disease identification and therapy determination according to a second embodiment.

[0210] Referring to FIG. 9, the disease identification and therapy determination unit 212 according to the second embodiment may use the query text 20 to acquire a search result 50 including one or more product-specific reference information from the product-tailored database 100, provide a prompt 60 to the AI model 40 that instructs to generate clinical decision support information for disease identification and therapy determination with reference to the search result 50, and acquire the clinical decision support information 30 for disease identification and therapy determination generated in response to the prompt 60 from the AI model 40. A prompt 60 according to one embodiment may include information other than the input to the AI model 40, which is trained to respond to a given input, and / or specific requests to instruct the generation of a response using the corresponding information. In one embodiment, the disease identification and therapy determination unit 212 may generate the above-mentioned prompt using a prompt template corresponding to a request for generating clinical decision support information for disease identification and therapy determination based on pre-stored query text components, and various known algorithms for prompt operations may be used for prompt generation. Furthermore, the above-mentioned operation may be performed based on Retrieval-Augmented Generation (RAG). In one embodiment, the AI model 40 may be implemented through a general-purpose generative AI model (e.g., ChatGPT, Bing, Bard) that has been developed to provide detailed responses corresponding to prompts in all possible cases.

[0211] In the second embodiment, the computer device 200 may be connected via a network to a generative AI server (not shown) that provides the AI model 40, and the disease identification and therapy determination unit 212 may access a user interface capable of exchanging text, images, voice, etc., with the AI model 40 using an API provided by the generative AI server, input the prompt 60 to the generative AI through the accessed user interface, and acquire the clinical decision support information 30 for disease identification and therapy determination as a response corresponding to the prompt.

[0212] Alternatively, in another embodiment, the disease identification and therapy determination unit 212 may generate a prompt that instructs to generate clinical decision support information for disease identification and therapy determination related to the query text 20 with reference to the product-tailored database 100, provide the generated prompt to the AI model 40, and acquire the clinical decision support information 30 for disease identification and therapy determination generated in response to the prompt from the AI model 40. For example, access of the AI model 40 to the product-tailored database 100 is permitted, and the disease identification and therapy determination unit 212 inputs a prompt including a query requesting a response with reference to the reliable product-tailored database 100 and an access link to the product-tailored database 100 to the AI model 40, and the AI model 40 may generate the clinical decision support information 30 for disease identification and therapy determination as a response corresponding to the prompt. Similarly, this operation may be performed based on Retrieval-Augmented Generation (RAG). In one embodiment, the AI model 40 may be implemented through the aforementioned general-purpose generative AI model.

[0213] FIG. 10 illustratively shows a conceptual diagram of a process for obtaining clinical decision support information 30 for disease identification and therapy determination according to a third embodiment.

[0214] Referring to FIG. 10, the AI model 40 according to the third embodiment may be a model pre-trained on product-specific reference information stored in the product-tailored database 100. Furthermore, the disease identification and therapy determination unit 212 may provide the query text 20 to the AI model 40, and acquire the clinical decision support information 30 for disease identification and therapy determination output from the AI model 40 corresponding to the query text 20. In one embodiment, the AI model 40 may be implemented through a specific training process performed by the computer device 200. For example, a model pre-trained using semi-supervised learning to understand the content of each piece of reference information for each product-specific classification item stored in the product-tailored database 100 is prepared, and the AI model 40 may be a model to which supervised learning-based fine-tuning is applied to the pre-trained model and may be additionally trained to generate clinical decision support information for disease identification and therapy determination as a natural language response when a predetermined natural language input is entered. Accordingly, the AI model 40 may be implemented as a medical assistant that has learned the content of product-specific reference information, and the AI model 40 may estimate clinical decision support information for disease identification and therapy determination based on the learned content.

[0215] According to an embodiment, the disease identification and therapy determination unit 212 may acquire clinical decision support information for disease identification and therapy determination using an AI model 40 that includes one or more modularized models. For example, the one or more modularized models may include at least one selected from the group consisting of a first model trained to adaptively determine the query text 20 based on the test result 10, a second model trained to perform a reference information search based on the query text 20, a third model trained to estimate clinical decision support information for disease identification and therapy determination based on a search result, and a fourth model trained to visualize the estimation result in various ways.

[0216] Meanwhile, according to an embodiment, the AI model 40 may be trained to perform the calculation of the aforementioned search query relevance and / or reference score, and the generation of clinical decision support information for disease identification and therapy determination based on it. Alternatively, the disease identification and therapy determination unit 212 may be implemented to provide a prompt that instructs the AI model 40 to perform the corresponding operation. Alternatively, the disease identification and therapy determination unit 212 may also be implemented to perform the corresponding operation at least partially.

[0217] In one embodiment, the clinical decision support information 30 for disease identification and therapy determination may be generated according to a preset standard format. For example, the disease identification and therapy determination unit 212 may set the clinical decision support information 30 for disease identification and therapy determination to be generated in a report format according to a specific standard format set by an administrator, and may request or control the AI model 40 to generate the clinical decision support information for disease identification and therapy determination according to the standard format during the generation process of the clinical decision support information for disease identification and therapy determination.

[0218] In one embodiment, the standard format may be either a format preset in the computer device 200 by a user account of the target terminal 300 or a terminal associated with the target terminal 300, or a user-customized format provided from the target terminal 300 or a terminal associated with the target terminal 300. For example, a basic format by an administrator is set as the standard format, but when a user-customized format is provided from the target terminal 300, the user-customized format may be applied in the generation process of the clinical decision support information for disease identification and therapy determination. For example, a graphic U / I (user interface) that enables visualization of information about the format and editing of the format based on user input is provided to the target terminal 300 through the aforementioned application, and the disease identification and therapy determination unit 212 receives information about the user-customized format through the graphic U / I, and may set the clinical decision support information 30 for disease identification and therapy determination to be generated according to the user-customized format.

[0219] The disease identification and therapy determination unit 212 may provide the obtained clinical decision support information 30 for disease identification and therapy determination to the target terminal 300 or a terminal associated with the target terminal 300. As described above, the target terminal 300 or the terminal associated with the target terminal 300 may be at least one selected from the group consisting of a terminal of a medical professional including a clinician's mobile terminal or therapy determination terminal, a server of a medical institution to which the medical professional belongs, a database of the medical institution, at least one terminal of a medical professional connected to the server of the medical institution via a network, and a test terminal of the medical institution. In one embodiment, in the process of receiving the test result 10, information about a target to which the corresponding clinical decision support information 30 for disease identification and therapy determination is to be provided may be received together, and the disease identification and therapy determination unit 212 may provide the clinical decision support information 30 for disease identification and therapy determination to the target terminal 300 or the terminal associated with the target terminal 300, which corresponds to the target to be provided.

[0220] In one embodiment, the disease identification and therapy determination unit 212 may provide the clinical decision support information 30 for disease identification and therapy determination to the target terminal 300. For example, the target terminal 300 may be a terminal of a medical professional including a clinician's mobile terminal or therapy determination terminal, or a terminal such as a LIS of a medical institution. In another embodiment, the disease identification and therapy determination unit 212 may provide the clinical decision support information 30 for disease identification and therapy determination together with the corresponding test result 10 to a terminal associated with the target terminal 300. For example, the target terminal 300 may be a device such as a LIS of a medical institution or a terminal of a medical professional including a test apparatus or a terminal of a medical technologist, and the terminal associated with the target terminal 300 may be a terminal of a medical professional including a clinician's mobile terminal or therapy determination terminal.

[0221] As an example, when the target is the target terminal 300, the disease identification and therapy determination unit 212 transmits the clinical decision support information 30 for disease identification and therapy determination to the target terminal 300, and the clinical decision support information 30 for disease identification and therapy determination may be displayed through a corresponding application on the target terminal 300. As another example, when the target is a terminal of an external user, the disease identification and therapy determination unit 212 may provide a message including the clinical decision support information 30 for disease identification and therapy determination or an access link thereto to the terminal. For example, the message may be provided based on at least one of Short Message Service (SMS), Long Message Service (LMS), and Multimedia Message Service (MMS), and the access link may be connected to a cloud-based visualization application for displaying the clinical decision support information 30 for disease identification and therapy determination, but is not limited thereto.

[0222] FIG. 11 illustratively shows a view of clinical decision support information 30 for disease identification and therapy determination being displayed on a screen of a target terminal 300 or a terminal associated with the target terminal 300 according to one embodiment.

[0223] As shown in FIG. 11, the clinical decision support information 30 for disease identification and therapy determination may include an disease identification result and a therapy determination result having the highest probability estimated based on related reference information, and may include a table that analyzes and summarizes therapy determination results based on references related to the product. Furthermore, the clinical decision support information 30 for disease identification and therapy determination may include, for each of several therapy determinations described in related papers, the therapy determination content, the characteristics of the patient group considered for the therapy determination (e.g., age, symptoms, underlying diseases, etc.), summaries of related reference papers, data source information, etc. Furthermore, the clinical decision support information 30 for disease identification and therapy determination may include statistics for each therapy determination described in related reference papers (e.g., the number of related papers), statistics on the number of final citations in other clinical decision support information for disease identification and therapy determination related to the product for each reference paper, etc. However, the configuration of the clinical decision support information 30 for disease identification and therapy determination and the arrangement of each area on the screen shown in FIG. 11 are merely examples, and may be implemented in various modified forms.

[0224] Accordingly, a medical professional may intuitively check a suitable disease identification method, a therapy determination method for his / her patient, the content of reference papers that support them, related statistics, etc., through the screen, so user convenience and satisfaction may be improved.

[0225] The disease identification and therapy determination unit 212 may provide the clinical decision support information 30 for disease identification and therapy determination through an interactive U / I, receive a user's query related to the clinical decision support information 30 for disease identification and therapy determination through the interactive U / I, and provide a response corresponding to the query through the interactive U / I. In one embodiment, this operation may be performed using the AI model 40. In one embodiment, the AI model 40 includes a GPT-based interactive artificial intelligence chatbot, and the disease identification and therapy determination unit 212 provides the clinical decision support information 30 for disease identification and therapy determination through the interactive U / I using the AI model 40, receives user input related to the clinical decision support information 30 for disease identification and therapy determination through the interactive U / I, and may provide a response corresponding to the received user input to the target terminal 300 or a terminal associated with the target terminal 300.

[0226] FIG. 12 illustratively shows a view of additional information being displayed on a screen of a target terminal 300 or a terminal associated with the target terminal 300 based on user input related to clinical decision support information 30 for disease identification and therapy determination according to one embodiment.

[0227] As shown in FIG. 12A, the disease identification and therapy determination unit 212 may communicate with the target terminal 300 through an interactive U / I such as a chat room, and may receive the test result 10 of a subject in a predetermined file format through the interactive U / I. When the received test result 10 does not include de-identified patient information, the disease identification and therapy determination unit 212 may request additional provision of the information, and may provide the clinical decision support information 30 for disease identification and therapy determination in a conversational manner with the user based on the response of the target terminal 300.

[0228] As shown in FIG. 12B, the disease identification and therapy determination unit 212 may receive a user's query about the clinical decision support information 30 for disease identification and therapy determination through the interactive U / I, and may provide an appropriate answer to the received query using the AI model 40. Furthermore, the disease identification and therapy determination unit 212 may receive a request for reflecting additional information about the clinical decision support information 30 for disease identification and therapy determination through the interactive U / I, and may update or additionally generate and provide the clinical decision support information 30 for disease identification and therapy determination by additionally applying the additional information (e.g., cough and low-grade fever as patient symptoms). In one embodiment, these operations may be performed in a manner where the disease identification and therapy determination unit 212 mediates communication between the target terminal 300 and the AI model 40 based on an interactive UI provided by a generative AI server, or in a manner of communicating with the target terminal 300 based on an interactive UI provided by the computer device 200 while using the output from the AI model 40.

[0229] However, the screen shown in FIG. 12 is merely an example, and may be implemented in various modified forms. According to an embodiment, the disease identification and therapy determination unit 212 may receive the test result 10 through a medical professional's terminal on which analysis software (e.g., a viewer program) that provides test data is installed, and the clinical decision support information 30 for disease identification and therapy determination may be visualized through the analysis software.

[0230] Meanwhile, the disease identification and therapy determination unit 212 according to an embodiment may, before providing the clinical decision support information 30 for disease identification and therapy determination, provide analysis information about one or more product-specific reference information searched using the query text 20, and may generate the clinical decision support information 30 for disease identification and therapy determination using product-specific reference information corresponding to a user's selection input for at least one of the one or more product-specific reference information. In one embodiment, the analysis information includes at least one of summary information, a search query relevance, and a reference score for each piece of reference information, and according to an embodiment, may further include an disease identification result and / or a therapy determination result estimated using the reference information. For example, summary information of reference information based on a search and / or clinical decision support information for disease identification and therapy determination using them may be provided in the first place, and the clinical decision support information 30 for disease identification and therapy determination may be generated using one or more pieces of reference information finally selected by the user among them. In this case, the summary information of the reference paper finally selected by the user includes a paper title, clinical date, sentences expressed in the paper, etc., and the summary information for each piece of reference information provided in the first place may include less information than the summary information of the finally selected reference paper.

[0231] The disease identification and therapy determination unit 212 according to an embodiment may, before providing the obtained clinical decision support information 30 for disease identification and therapy determination, perform a verification of the obtained clinical decision support information 30 for disease identification and therapy determination using the AI model 40, and may provide the verified clinical decision support information 30 for disease identification and therapy determination.

[0232] Specifically, the disease identification and therapy determination unit 212 generates a prompt, and the prompt may be for instructing the AI model 40 to perform the above-mentioned verification based on (a) generation of clinical decision support information for verification related to the query text 20 using one or more search engines and / or (b) a comparison result between the obtained clinical decision support information 30 for disease identification and therapy determination and the clinical decision support information for verification. For example, the search engine is a search engine connected via a network or provided in the AI model 40 or the disease identification and therapy determination unit 212, and may be implemented to perform a search on the web or a specific pre-set database (e.g., a database of an academic institution). The disease identification and therapy determination unit 212 provides the generated prompt to the AI model 40, and may acquire a verification result corresponding to the prompt from the AI model 40. In one embodiment, the verification result may include at least one of whether the clinical decision support information 30 for disease identification and therapy determination and the clinical decision support information for verification match, the degree of mismatch if they do not match, the mismatched content, and reference literature information related to the mismatched content. In one embodiment, the verified clinical decision support information for disease identification and therapy determination may include the verification result, or may be at least partially updated based on the verification result, and for example, may be updated to also include summary information about the reference literature information related to the mismatched content.

[0233] The disease identification and therapy determination unit 212 according to an embodiment may receive individual clinical data from the target terminal 300 and store it in the product-tailored database 100. In one embodiment, after providing the clinical decision support information 30 for disease identification and therapy determination, the disease identification and therapy determination unit 212 may receive individual clinical data based on the clinical decision support information 30 for disease identification and therapy determination from the target terminal 300. In another embodiment, the disease identification and therapy determination unit 212 may receive individual clinical data created by a user from the target terminal 300, regardless of the provision of the clinical decision support information 30 for disease identification and therapy determination. As an example, such individual clinical data may be understood as a kind of disclosure of information about individual clinical data provided by an individual user.

[0234] FIG. 13 schematically illustrates a block diagram of individual clinical data 70 according to one embodiment.

[0235] Referring to FIG. 13, the individual clinical data 70 may include a test result 71, user information 72, and a result 73 regarding the user's disease identification and therapy determination.

[0236] The test result 71 may be obtained from the test result 10, and for example, includes at least a part of the test result 10. Alternatively, the test result 71 may be manually input based on a user's input of components of the diagnostic test result to the target terminal 300.

[0237] The user information 72 may include at least one selected from the group consisting of a user name, affiliation, position, address, contact information, and country corresponding to the account of the target terminal 300. For example, it may include basic user information required for publication, such as name, affiliation, position, address, email, and country.

[0238] The result 73 regarding the user's disease identification and therapy determination may include a result regarding the user's disease identification and therapy determination based on the clinical decision support information 30 for disease identification and therapy determination. For example, it may be the disease identification result and therapy determination result finally selected by the user from among the disease identification results and therapy determination results included in the clinical decision support information 30 for disease identification and therapy determination provided to the target terminal 300. In one embodiment, the result 73 regarding the user's disease identification and therapy determination may further include user feedback on the finally selected disease identification result and therapy determination result or the clinical decision support information 30 for disease identification and therapy determination, and may, for example, further include a change in the patient's symptoms or disease state according to the finally selected disease identification result and therapy determination result, and other opinions that may be referenced.

[0239] In one embodiment, the individual clinical data 70 may be processed according to a preset format. For example, in the process of data reception or storage, the individual clinical data 70 in which the diagnostic test result and the result of disease identification and therapy determination based on the product-tailored database 100 are described may be registered according to a fixed specific format excluding the patient's personal information.

[0240] In one embodiment, usage rights for the individual clinical data 70 may be granted only to a user account that has participated in providing at least one of a plurality of pieces of individual clinical data stored in the product-tailored database 100. For example, the individual clinical data 70 may be granted the authority to share information among user accounts of autonomous participants who have provided other individual clinical data in the same way. Furthermore, in one embodiment, the user account of the target terminal 300 that provided the individual clinical data 70 may be granted authority to restrict, cancel, and delete the sharing of information about the individual clinical data 70.

[0241] In one embodiment, in the process where the individual clinical data 70 is stored in the product-tailored database 100, a metadata labeling task may be performed in the same or a similar manner as a database processing of reference information. According to an embodiment, in the process of obtaining any clinical decision support information for disease identification and therapy determination performed later, the individual clinical data 70 stored in the product-tailored database 100 may also be used together with the above-mentioned product-specific reference information. Furthermore, in this case, the individual clinical data in the product-tailored database 100 may be mutually used only by users who have provided the individual clinical data as described above, or the use may be restricted based on information sharing restrictions set by the users.

[0242] In one embodiment, based on the individual clinical data 70, a weight used for obtaining clinical decision support information for disease identification and therapy determination based on the product-tailored database 100 may be updated. For example, the disease identification and therapy determination unit 212 may perform additional training for the AI model 40 at a preset cycle to generate clinical decision support information for disease identification and therapy determination using the individual clinical data stored in the product-tailored database 100. As another example, in the generation of subsequent clinical decision support information for disease identification and therapy determination, the weight may be updated so that a higher weight is assigned to the disease identification type or therapy determination type of the result 73 regarding the user's disease identification and therapy determination of the individual clinical data 70 related to the product.

[0243] The disease identification and therapy determination unit 212 according to an embodiment may, when a Ct value and / or a subject's age included in the test result 10 deviates from a preset appropriate range, acquire the clinical decision support information 30 for disease identification and therapy determination based on a weight assigned to a therapy determination type corresponding to symptomatic therapy. For example, when the result value is positive but the Ct value (e.g., Ct = 32) is within a certain range that is relatively high, or when the subject's age is in a relatively young or relatively old age group, the disease identification and therapy determination unit 212 assigns a relatively high weight to product-specific reference information that includes therapy determination content related to symptomatic therapy among the product-specific reference information searched based on the query text 20, and may generate the clinical decision support information 30 for disease identification and therapy determination by selecting product-specific reference information based on the weight. Accordingly, considering symptomatic therapy, a relatively higher weight may be assigned to therapy determinations of symptom-relieving drugs rather than antibiotic therapy determinations, and as a result, excessive antibiotic therapy determinations can be improved and better therapy determinations can be presented to patients.

[0244] The disease identification and therapy determination unit 212 according to an embodiment may, when a subject's underlying disease included in the test result 10 corresponds to any one of preset cautionary underlying diseases related to an immunocompromised state, acquire the clinical decision support information 30 for disease identification and therapy determination using (a) a weight assigned to the any one cautionary underlying disease and (b) a follow-up history of the subject. Here, the follow-up history of the subject may include diagnostic test results by test date and clinical decision support information 30 for disease identification and therapy determination, which are pre-stored for the subject. For example, when the subject is an immunocompromised patient whose immune system is weakened due to cancer, HIV, organ transplant, or other diseases according to the subject's underlying disease, there may be a risk of exposure to bacteria that may cause lung infection. The disease identification and therapy determination unit 212 loads a pre-stored correlation between a positive pathogen and the subject's underlying disease, assigns a relatively high weight to product-specific reference information that includes therapy determination content related to the correlation, and may generate the clinical decision support information 30 for disease identification and therapy determination by selecting product-specific reference information based on the weight. For example, such correlations may include the use of a specific range of antibiotics rather than broad-spectrum antibiotics, discontinuation of therapy determinations of drugs that suppress the immune system, therapy determinations of drugs that help improve the immune system, and simultaneous therapy determination of drugs for inflammation improvement. Furthermore, when the disease identification and therapy determination history of the subject is searched, the disease identification and therapy determination unit 212 provides the clinical decision support information 30 for disease identification and therapy determination using the history regarding disease identification and therapy determination together, and when not, it may control the history regarding disease identification and therapy determination to be used at the time of the subject's next test by storing and managing the history regarding disease identification and therapy determination including the subject's test result 10 and the clinical decision support information 30 for disease identification and therapy determination. Accordingly, a more desirable treatment method may be provided for immunocompromised patients, and a treatment method may be provided based on follow-up observation considering the patient's specific underlying disease.

[0245] The disease identification and therapy determination unit 212 according to an embodiment may, when a positive pathogen according to the test result 10 is any one of preset antibiotic-resistant bacteria, acquire the clinical decision support information 30 for disease identification and therapy determination including (a) a message recommending proceeding with an additional test including an antibiotic resistance test for the subject, and (b) therapy determination results for each additional test result obtained based on a weight assigned to the any one antibiotic-resistant bacterium. For example, the antibiotic-resistant bacteria include preset risk-resistant bacteria such as MRSA, VRSA, VRE, MRPA, MRAB, and CRE, and when the positive pathogen is one of them, the disease identification and therapy determination unit 212 may provide a message recommending proceeding with an antibiotic resistance test and therapy determination results for each expected test result according to the antibiotic resistance test (e.g., therapy determination of an antibiotic to which MRSA is susceptible) as the clinical decision support information 30 for disease identification and therapy determination. In one embodiment, the disease identification and therapy determination unit 212 may generate the clinical decision support information 30 for disease identification and therapy determination based on weights assigned to factors such as the subject's personal characteristics (e.g., history of side effects, kidney function, liver function, etc.) and the local and natural resistance of the antibiotic-resistant bacterium. Accordingly, optimized therapy determination information for antibiotic-resistant bacteria infection may be provided, and therapy determination may be supported in a manner optimized for the patient according to additional test results.

[0246] The disease identification and therapy determination unit 212 according to an embodiment may generate clinical decision support information 30 for disease identification and therapy determination including first clinical decision support information for disease identification and therapy determination generated using product-specific reference information corresponding to papers, and second clinical decision support information for disease identification and therapy determination generated using product-specific reference information corresponding to specialized books. The disease identification and therapy determination unit 212 may transmit the clinical decision support information 30 for disease identification and therapy determination to the target terminal 300 or a terminal associated with the target terminal 300, and may control the first and second clinical decision support information for disease identification and therapy determination to be displayed in a comparable manner on one screen at the target terminal 300 or the device associated with the target terminal 300. For example, when the second clinical decision support information for disease identification and therapy determination may present a textbook therapy determination method or a general therapy determination method applied to a typical patient, the first clinical decision support information for disease identification and therapy determination may present a paper-based therapy determination method that considers recent research achievements or research trends, or a case-specific therapy determination method applicable to a patient with other considerations such as an underlying disease.

[0247] The disease identification and therapy determination unit 212 according to an embodiment may provide therapy determination records of similar patients based on their Electronic Medical Record (EMR) data. Specifically, the target terminal 300 may be connected via a network to a server of a medical institution corresponding to the user account of the target terminal 300 or a database of the medical institution, and the database of the medical institution may store EMR data of patients of the medical institution. The disease identification and therapy determination unit 212 may acquire therapy determination records of similar patients that at least partially correspond to the test result 10 of the subject from the database of the medical institution. In one embodiment, the therapy determination records of similar patients may include de-identified patient information, medical records, and therapy determination results of the similar patients.

[0248] For example, the disease identification and therapy determination unit 212 may request the server of the medical institution to provide therapy determination records of similar patients that (a) include the same content as the detection result of each target analyte, the type of sample, and the subject's symptoms included in the test result 10, and (b) include some of the content of the subject's underlying diseases, country, region, age, and gender included in the test result 10, and may receive the therapy determination records of similar patients from the server of the medical institution as a response to the request. As another example, the disease identification and therapy determination unit 212 may be granted authority to interoperate with the database of the medical institution from the server of the medical institution, and may search the database of the medical institution for the aforementioned therapy determination records of similar patients based on the authority, and may generate the therapy determination records of similar patients based on personal information de-identification processing of the search result.

[0249] In one embodiment, when providing the clinical decision support information 30 for disease identification and therapy determination, the disease identification and therapy determination unit 212 may provide the obtained therapy determination records of similar patients together with the clinical decision support information 30 for disease identification and therapy determination. For example, the therapy determination records of similar patients may include a therapy determination result with the highest number of therapy determinations and summary information about the therapy determination result. Furthermore, the therapy determination records of similar patients may include multiple therapy determination types for several similar patients, and may further include a statistical analysis result for the number of therapy determinations for each of the multiple therapy determination types. Furthermore, for each therapy determination type included in the clinical decision support information 30 for disease identification and therapy determination, therapy determination records of similar patients related to the type, the number of therapy determinations, related statistical analysis results, etc., may be provided.

[0250] Accordingly, while checking product-specific reference papers that may be referred to for the disease identification and therapy determination of the test result 10, a medical professional may also check the therapy determinations actually made for patients similar to the test result 10 and the progress results according to the therapy determinations, so user satisfaction may be greatly improved.

[0251] A training unit 213 may be implemented to perform training for the AI model 40. Training according to one embodiment may mean pre-training the AI model 40 before the start of a service that provides clinical decision support information for disease identification and therapy determination by the computer device 200.

[0252] In one embodiment, the training unit 213 may acquire a pre-trained model that generates a natural language response to a natural language input using a large amount of training text data. For example, the training unit 213 may perform pre-training on an initial language model through a task of predicting the next token based on cross entropy for a large amount of training text data, or may receive an interactive artificial intelligence chatbot model (e.g., a transformer-based large language model) trained to have a natural language conversation with a user by an external device.

[0253] In one embodiment, the training unit 213 may acquire a plurality of training datasets based on reference information stored in the product-tailored database 100, and may acquire the AI model 40 by applying fine-tuning using the plurality of training datasets to the pre-trained model. In one embodiment, a pre-trained generative AI model may be brought in and modified to suit a task for searching for reference information and / or generating clinical decision support information for disease identification and therapy determination, and an error may be calculated by comparing the output of the pre-trained model for training input data with the corresponding training ground truth data, and the parameters of the model may be updated according to backpropagation to reduce the error. For example, each training dataset may include training input data including one or more search queries, and training ground truth data including clinical decision support information for disease identification and therapy determination using reference information associated with the one or more search queries in the product-tailored database 100. As another example, each training dataset may include training input data including one or more search queries, and training ground truth data including reference information (e.g., metadata) associated with the one or more search queries in the product-tailored database 100. As yet another example, each training dataset may include training input data including one or more pieces of reference information, and training ground truth data including clinical decision support information for disease identification and therapy determination using the one or more pieces of reference information. For training according to one embodiment, Large Language Models (LLM), Variational AutoEncoder (VAE), Generative Adversarial Network (GAN), RAG, etc., may be used, and in addition, various known algorithms as learning methods for language models or generative AI models may be used.

[0254] In another embodiment, the training unit 213 may acquire a pre-trained model by a self-supervised learning method using reference information stored in the product-tailored database 100, and may implement the AI model 40 by fine-tuning the pre-trained model in a supervised learning manner using a plurality of training datasets (e.g., a question about product-based disease identification and therapy determination as training input data, and a product-specific reference information-based answer according to the question as training ground truth data) preprocessed as training data based on the product-tailored database 100. For example, by the above-mentioned training, the AI model 40 may be implemented as an interactive artificial intelligence chatbot model trained to have a conversation using reference information stored in the product-tailored database 100, so that it may provide answers at the level of a medical expert. For example, for the same question content, different answers may be generated depending on the type of product used for the test.

[0255] In one embodiment, the AI model 40 may be a model trained by a transfer learning method. Here, transfer learning refers to a learning method of obtaining a pre-trained model having a first task by pre-training a large amount of unlabeled training data in a semi-supervised or self-supervised learning manner, and implementing a target model by fine-tuning the pre-trained model to be suitable for a second task and learning labeled training data in a supervised learning manner. As an example, this fine-tuning may include a process of updating the parameters of the pre-trained model by additionally training the pre-trained model with a specific training dataset for generating an intended response.

[0256] Meanwhile, the AI model 40 in this specification may mean any form of computer program that operates based on a network function, an artificial neural network, and / or a neural network. Throughout this specification, model, neural network, network function, and neural network may be used interchangeably. The AI model 40 includes a deep neural network (DNN), and the deep neural network may include a convolutional neural network (CNN), a recurrent neural network (RNN), an autoencoder, a Generative Adversarial Network (GAN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a Q-network, a U-network, a Siamese network, a Generative Adversarial Network (GAN), a transformer, etc., but is not limited thereto.

[0257] The AI model 40 may be trained in at least one manner of supervised learning, unsupervised learning, semi-supervised learning, self-supervised learning, or reinforcement learning, and various known learning algorithms may be used.

[0258] The AI model 40 may borrow at least a part of a transformer. The transformer may be composed of an encoder that encodes embedded data and a decoder that decodes the encoded data. For the transformer to encode and decode a series of data, the encoders and decoders within the transformer may be processed using an attention algorithm. The transformer may also include additional components other than the attention algorithm, such as embedding, normalization, and softmax. A method of configuring a transformer using an attention algorithm may include the method disclosed in Vaswani et al., Attention Is All You Need, 2017 NIPS, which is incorporated herein by reference.

[0259] FIG. 14 illustrates an exemplary flowchart for a computer device 200 to provide clinical decision support information 30 for disease identification and therapy determination according to one embodiment. FIG. 14 may be understood with reference to all the embodiments described above. In one embodiment, the steps of FIG. 14 may be implemented by a single entity, such as in a manner performed by a server.

[0260] Referring to FIG. 14, in operation S1410, the computer device 200 may receive a test result 10 tested with one or more diagnostic test reagent products selected based on sample information from a target terminal 300.

[0261] In operation S1420, the computer device 200 may, based on the received test result 10, acquire clinical decision support information 30 for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database 100. In one embodiment, the product-tailored database 100 may be configured to store (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information that supports the product-specific clinical decision support information. The product-specific clinical decision support information and the product-specific reference information may be labeled with an identifier based on a specification of the diagnostic reagent product so that they may be referenced for disease identification and therapy determination of a test result using the one or more diagnostic test reagent products.

[0262] In operation S1430, the computer device 200 may provide the obtained clinical decision support information 30 for disease identification and therapy determination to the target terminal 300 or a terminal associated with the target terminal 300.

[0263] According to one embodiment of the present disclosure, as the clinical decision support information 30 for disease identification and therapy determination based on the product-tailored database 100 is provided, a user's cumbersome reference search task is omitted, so user convenience is improved, and a user may refer to more reliable clinical decision support information for disease identification and therapy determination for each diagnostic reagent product, so user satisfaction is enhanced.

[0264] As described above, using the method of providing clinical decision support information for disease identification and therapy determination described in this specification, clinical decision support information for disease identification and therapy determination based on a product-tailored database may be provided using a subject's test results. This technical feature may be used independently without being combined with the technical feature of building the product-tailored database described above.

[0265] Meanwhile, a system 1000 for disease identification and therapy determination according to one embodiment may be a system built based on a global network environment associated with a plurality of medical institutions. In one embodiment, the computer device 200 is connected to a plurality of target terminals 300 via a network, and a user account of each target terminal 300 may be a terminal of any one of a plurality of medical institution accounts, or a terminal of a medical professional belonging to the medical institution account. The computer device 200 may check the products available at each medical institution based on pre-stored medical institution management information, and may grant different usage rights for reference information selected for the product in the product-tailored database 100 for each medical institution according to the check result. For example, the AI model 40 is a general-purpose generative AI model, and the disease identification and therapy determination unit 212 may mediate communication between the AI model 40 and the target terminal 300 based on the usage rights granted to the user account of the target terminal 300, and may control the environment so that the various functions described above are performed. As another example, the computer device 200 includes an AI model 40 trained to provide clinical decision support information for disease identification and therapy determination based on the product-tailored database 100, and may control the AI model 40 to generate clinical decision support information for disease identification and therapy determination using reference information selected for a product corresponding to the usage rights granted to the user account of the target terminal 300 among the product-tailored database 100.

[0266] A system 1000 for disease identification and therapy determination according to another embodiment may be a system built based on a local network environment associated with a single medical institution or a group of medical institutions. In one embodiment, the computer device 200 is a server of the single medical institution or group of medical institutions, the target terminal 300 is a terminal of a medical professional belonging to the medical institution, and the product-tailored database 100 may be an on-premise type private database of the medical institution in which reference information selected for products permitted for the medical institution is stored. Furthermore, the AI model 40 may be installed and executed on the computer device 200 or a terminal of the medical institution, and may be implemented to provide clinical decision support information for disease identification and therapy determination based on the product-tailored database 100 while communicating with the computer device 200 via an internal network of the medical institution. In this case, in the process of providing the therapy determination records of similar patients described above, it may be interoperated with a database of the medical institution via the internal network, and a result in which patient information is de-identified from EMR data of patients may be provided as the therapy determination records of similar patients. The above-described embodiments are merely examples, are not limited thereto, and may be implemented in various modified forms.

[0267] Furthermore, at least some of the components included in the system 1000 for disease identification and therapy determination according to one embodiment may be implemented to operate in a cloud environment. For example, the product-tailored database 100 may be implemented as a cloud database, include one or more computer resources functioning as storage for data storage and one or more servers for managing them, and the server may manage the computer resources to perform computing operations as a cloud database. As another example, the computer device 200 may perform computing operations to provide a cloud computing service including providing the clinical decision support information for disease identification and therapy determination described above, and may include one or more computer resources such as one or more servers for this. In one embodiment, there is no limit to the number of the servers, and each server may have a separate operating system individually or share one. According to an embodiment, such a server may be implemented as a web server, and may provide a web page for providing the clinical decision support information for disease identification and therapy determination described above to the target terminal 300, and may provide a cloud computing service through the web page.

[0268] Various embodiments of the method of providing clinical decision support information for disease identification and therapy determination can be implemented by combining the various technical features described in this specification.

[0269] According to one aspect of the present disclosure, a computer device includes a memory that stores at least one instruction; and a processor to execute the at least one instruction stored in the memory, wherein the at least one instruction, when executed by the processor, causes the processor to perform a method, the method comprising: receiving, from a target terminal, a test result tested with one or more diagnostic test reagent products selected based on sample information; obtaining, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database; wherein the product-tailored database stores: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information that supports the product-specific clinical decision support information; wherein the product-specific clinical decision support information and the product-specific reference information are labeled with an identifier based on a specification of the diagnostic reagent product so that they may be referenced for disease identification and therapy determination of a test result using the one or more diagnostic test reagent products; and providing the obtained clinical decision support information for disease identification and therapy determination to the target terminal or a terminal associated with the target terminal.

[0270] Since each component described in the one aspect of the present disclosure described above overlaps with the method of providing clinical decision support information for disease identification and therapy determination described with reference to FIGS. 1 to 14, a description thereof will be omitted.

[0271] According to one aspect of the present disclosure, a computer program is stored on a computer-readable non-transitory recording medium, and the computer program is programmed to perform each step included in a method; the method comprising: receiving, from a target terminal, a test result tested with one or more diagnostic test reagent products selected based on sample information; obtaining, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database; wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information that supports the product-specific clinical decision support information; wherein the product-specific clinical decision support information and the product-specific reference information are labeled with an identifier based on a specification of the diagnostic reagent product so that they may be referenced for disease identification and therapy determination of a test result using the one or more diagnostic test reagent products; and providing the obtained clinical decision support information for disease identification and therapy determination to the target terminal or a terminal associated with the target terminal.

[0272] Since each component described in the one aspect of the present disclosure described above overlaps with the method of providing clinical decision support information for disease identification and therapy determination described with reference to FIGS. 1 to 14, a description thereof will be omitted.

[0273] According to one aspect of the present disclosure, a computer program stored on a computer-readable non-transitory recording medium is programmed to perform each step included in a method; the method comprising: receiving, from a target terminal, a test result tested with one or more diagnostic test reagent products selected based on sample information; obtaining, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database; wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information that supports the product-specific clinical decision support information; wherein the product-specific clinical decision support information and the product-specific reference information are labeled with an identifier based on a specification of the diagnostic reagent product so that they may be referenced for disease identification and therapy determination of a test result using the one or more diagnostic test reagent products; and providing the obtained clinical decision support information for disease identification and therapy determination to the target terminal or a terminal associated with the target terminal.

[0274] Since each component described in the one aspect of the present disclosure described above overlaps with the method of providing clinical decision support information for disease identification and therapy determination described with reference to FIGS. 1 to 14, a description thereof will be omitted.

[0275] Similarly, various embodiments of a method for building a product-tailored database may be implemented by combining the various technical features described in this specification.

[0276] According to one aspect of the present disclosure, a computer device includes a memory that stores at least one instruction; and a processor to execute the at least one instruction stored in the memory, wherein the at least one instruction, when executed by the processor, causes the processor to perform a method, the method comprising: obtaining, based on information of a target diagnostic reagent product, reference information including content related to disease identification and / or therapy determination from one or more data sources; selecting, from the obtained reference information, one or more product-specific reference information based on a specification of the target diagnostic reagent product; generating one or more product-specific clinical decision support information for disease identification and therapy determination using the selected one or more product-specific reference information; and storing the product-specific clinical decision support information and the selected one or more product-specific reference information in a product-tailored database; wherein the product-specific clinical decision support information and the product-specific reference information are labeled with an identifier based on a specification of the diagnostic reagent product so that they may be referenced for disease identification and therapy determination of a test result using the one or more diagnostic test reagent products; and controlling the provision of clinical decision support information for disease identification and therapy determination that can be referenced for disease identification and / or therapy determination of a test result using the target diagnostic reagent product based on the product-tailored database.

[0277] Since each component described in the one aspect of the present disclosure described above overlaps with the method for building a product-tailored database described with reference to FIGS. 1 to 14, a description thereof will be omitted.

[0278] The steps illustrated above are merely examples, and may be implemented in various forms with addition, omission, or modification of the order, combination, branching, function, and the performing subject within a scope that does not depart from the essential characteristics of each technical feature described throughout the specification.

[0279] Combinations of steps in each flowchart attached to the present disclosure may be executed by computer program instructions. Since the computer program instructions can be mounted on a processor of a general-purpose computer, a special purpose computer, or other programmable data processing equipment, the instructions executed by the processor of the computer or other programmable data processing equipment create a means for performing the functions described in each step of the flowchart. The computer program instructions can also be stored on a computer-usable or computer-readable storage medium which can be directed to a computer or other programmable data processing equipment to implement a function in a specific manner. Accordingly, the instructions stored on the computer-usable or computer-readable recording medium can also produce an article of manufacture containing an instruction means which performs the functions described in each step of the flowchart. The computer program instructions can also be mounted on a computer or other programmable data processing equipment. Accordingly, a series of operational steps are performed on a computer or other programmable data processing equipment to create a computer-executable process, and it is also possible for instructions to perform a computer or other programmable data processing equipment to provide steps for performing the functions described in each step of the flowchart.

[0280] In addition, each step may represent a module, a segment, or a portion of codes which contains one or more executable instructions for executing the specified logical function(s). It should also be noted that in some alternative embodiments, the functions mentioned in the steps may occur out of order. For example, two steps illustrated in succession may in fact be performed substantially simultaneously, or the steps may sometimes be performed in a reverse order depending on the corresponding function.

[0281] The above description is merely exemplary description of the technical scope of the present disclosure, and it will be understood by those skilled in the art that various changes and modifications can be made without departing from original characteristics of the present disclosure. Therefore, the embodiments disclosed in the present disclosure are intended to explain, not to limit, the technical scope of the present disclosure, and the technical scope of the present disclosure is not limited by the embodiments. The protection scope of the present disclosure should be interpreted based on the following claims and it should be appreciated that all technical scopes included within a range equivalent thereto are included in the protection scope of the present disclosure.

[0282] According to the present invention, by selecting reference information suitable for a diagnostic test reagent product based on artificial intelligence and providing clinical decision support information for disease identification and therapy determination, the invention is expected to efficiently select and effectively analyze optimal references from vast publicly available data on disease identification and therapy determination, while overcoming limitations in response reliability, such as hallucinations, which frequently occur in generative artificial intelligence.

Claims

1.A method for providing clinical decision support information for disease identification and therapy determination, performed by a computer device, the method comprising:receiving a test result obtained by testing with one or more diagnostic test reagent products selected based on sample information, from a target terminal;obtaining, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database, wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information as evidence of the product-specific clinical decision support information, both of which are labeled with an identifier based on a specification of a diagnostic test reagent product whereby the product-specific clinical decision support information and the product-specific reference information can be used as considerations for the disease identification and therapy determination based on the test result using the one or more diagnostic test reagent products; andproviding the obtained clinical decision support information to the target terminal or to a terminal associated with the target terminal.2.The method of claim 1, wherein the one or more product-specific reference information are obtained from one or more data sources, andwherein the one or more data sources comprise at least one selected from the group consisting of:a product developer during a product development stage of the one or more diagnostic test reagent products; an institution associated with regulatory approval of the one or more diagnostic test reagent products; at least one of a disease research institution, a management institution, and a medical institution, based on a test using the one or more diagnostic test reagent products; and a medical expert associated with the one or more diagnostic test reagent products.3.The method of claim 1, wherein the one or more product-specific reference information include product-specific reference information,and wherein the product-specific reference information comprises at least one selected from the group consisting of: (i) research and development data generated by a researcher or developer during a research and development stage of the one or more diagnostic test reagent products; and (ii) clinical data obtained through clinical trials using the one or more diagnostic test reagent products in a post-development stage.4.The method of claim 3, wherein the one or more product-specific reference information further include public reference information that is publicly available from an academic institution, a medical institution, or a public institution.5.The method of claim 1, wherein the one or more product-specific reference information comprise reference information selected from one or more data sources based on a specification of any one of the one or more diagnostic test reagent products.6.The method of claim 5, wherein the specification of any one of the one or more diagnostic test reagent products comprises at least one selected from the group consisting of: (a) an intended use including at least one of target analyte information, pathogen information, pathogenic gene or single nucleotide polymorphism (SNP) information, contamination substance information, sample type, host information, disease name, infectious disease name, and symptom information; (b) a design specification including information about at least one of an analyte panel configuration, an optical channel configuration, an internal control (IC), a positive control (PC), and a negative control (NC); (c) a performance specification including at least one of sensitivity, specificity, turn-around time (TAT), reproducibility, and linearity; and (d) a technical specification including at least one of a sampling technique, an extraction technique, a nucleic acid amplification test (NAAT) reaction setup technique, an amplification technique, a signal analysis technique, and an information display technique.7.The method of claim 6, wherein the one or more product-specific reference information are selected based on one or more criteria selected from the group consisting of: (a) a specification relevance that quantifies a degree to which the reference information includes content related to the specification of any one of the one or more diagnostic test reagent products; (b) a reliability of a data source; (c) a number of times cited in other clinical decision support information related to the specification of any one of the one or more diagnostic test reagent product; (d) a number of times cited in other reference information; and (e) a recognition of an author or an affiliated institution.8.The method of claim 7, wherein the selecting of the one or more product-specific reference information is performed using a generative artificial intelligence (AI) model trained to select the product-specific reference information from reference information based on the specification of any one of the one or more diagnostic test reagent product.9.The method of claim 5, wherein the one or more product-specific reference information correspond to reference information selected for each of the one or more diagnostic test reagent products and are labeled with a corresponding product identifier.10.The method of claim 5, wherein the one or more product-specific reference information correspond to reference information selected for each of the one or more diagnostic test reagent products and are labeled with metadata corresponding to content related to the specification of any one of the one or more diagnostic test reagent products, wherein the metadata include two or more metadata selected from the group consisting of: a product name, a product identifier, a product version, a catalog number, a product group name, a product group identifier, a data source type, a target analyte name, a pathogen name, pathogenic gene or single nucleotide polymorphism (SNP) information, a sample type, host information, a disease name, an infectious disease name, an author, an affiliated institution, an academic journal, an information disclosure date, and clinical information, wherein the clinical information includes at least one selected from the group consisting of: a subject's symptoms, underlying diseases, country, region, age, gender, a Cycle threshold value (Ct value) of each target analyte, an disease identification result, a therapy determination result, and a clinical date.11.The method of claim 1, wherein the one or more product-specific reference information comprise one or more selected from the group consisting of: clinical paper information including clinical trial content related to detection or analysis of at least one of a single target analyte or a plurality of target analytes that any one of the one or more diagnostic test reagent products is intended to detect or analyze; clinical paper information including content related to disease identification and / or therapy determination based on a sole detection of any one of the plurality of target analytes; and clinical paper information including content related to disease identification and / or therapy determination based on a simultaneous detection of two or more of the plurality of target analytes.12.The method of claim 1, wherein the one or more product-specific clinical decision support information comprise one or more selected from the group consisting of: (a) an disease identification result and / or a therapy determination result estimated based on one or more product-specific reference information; (b) summary information generated for each of the one or more product-specific reference information; and (c) a statistical analysis result corresponding to a type of disease identification result and / or a type of therapy determination result included in each of the one or more product-specific reference information.13.The method of claim 12, wherein the summary information comprises one or more selected from the group consisting of: a title, a clinical date, clinical values, a key sentence included in the one or more product-specific reference information, and data source information for the one or more product-specific reference information.14.The method of claim 12, wherein the one or more product-specific clinical decision support information are generated in a predefined report format using a generative artificial intelligence (AI) model trained to generate a natural language response to a natural language input.15.The method of claim 14, wherein the one or more product-specific clinical decision support information are generated based on a prompt that instructs the generative artificial intelligence (AI) model to generate the one or more product-specific clinical decision support information using the one or more product-specific reference information.16.The method of claim 14, wherein the one or more product-specific clinical decision support information comprise a plurality of product-specific clinical decision support information generated by the generative artificial intelligence (AI) model using each of a plurality of product-specific reference groups,wherein the plurality of product-specific reference groups are obtained by grouping the one or more product-specific reference information based on whether values of at least one piece of metadata among a plurality of metadata correspond to each other,and wherein the plurality of metadata comprise one or more selected from the group consisting of: a data source type, a target analyte name, a pathogen name, pathogenic gene or single nucleotide polymorphism (SNP) information, a sample type, host information, a disease name, an infectious disease name, an author, an affiliated institution, an academic journal, an information disclosure date, and clinical information, the clinical information comprising one or more selected from the group consisting of: a subject's symptoms, underlying diseases, country, region, age, gender, a Cycle threshold value (Ct value) of each target analyte, an disease identification result, a therapy determination result, and a clinical date.17.The method of claim 1, further comprising updating the product-tailored database at a preset cycle, wherein the updating comprises:(a) newly obtaining reference information from one or more data sources including a reference search module based on a specification of any one of the one or more diagnostic test reagent products;(b) additionally selecting product-specific reference information from the newly obtained reference information; and(c) updating the product-specific clinical decision support information using the additionally selected product-specific reference information.18.The method of claim 1, wherein the obtaining the clinical decision support information comprises:(a) searching the product-tailored database for one or more product-specific clinical decision support information labeled with an identifier corresponding to the one or more diagnostic test reagent products; and(b) obtaining the clinical decision support information based on the searching.19.The method of claim 1, wherein the obtaining the clinical decision support information comprises:(a) generating a query text from the test result received from the target terminal, the query text including information of the one or more diagnostic test reagent products, a detection result of one or more target analytes targeted by the one or more diagnostic test reagent products, and a sample type according to the sample information; and(b) obtaining the clinical decision support information from the product-tailored database using the query text.20.The method of claim 19, wherein the detection result of the one or more target analytes comprises one or more selected from the group consisting of: (a) a detection status, a positive / negative result value, and a detected amount for each of the one or more target analytes; and (b) a type of genotype determined based on the detection result of each of the one or more target analytes and whether a mutation is included.21.The method of claim 19, wherein the information of the one or more diagnostic test reagent products comprises one or more selected from the group consisting of: a product name, a product identifier, a product version, a catalog number, a product group name, and a product group identifier of the one or more diagnostic test reagent products.22.The method of claim 19, wherein the query text further comprises two or more selected from the group consisting of: (a) a subject's symptoms; (b) a subject's underlying diseases; (c) de-identified patient information comprising one or more selected from the group consisting of: country, region, age, and gender; and (d) a Cycle threshold value (Ct value) of the one or more target analytes.23.The method of claim 22, wherein the query text further comprises at least a part of raw data obtained from a diagnostic test apparatus, the raw data being included in the test result.24.The method of claim 22, wherein the query text further comprises information that is manually obtained based on a user input of the target terminal or information that is automatically obtained based on linkage with software executed on the target terminal or with another device.25.The method of claim 19, wherein the obtaining the clinical decision support information from the product-tailored database using the query text comprises:(a) searching the product-tailored database for one or more product-specific clinical decision support information labeled with an identifier and / or metadata related to the query text;(b) calculating, for each of the searched one or more product-specific clinical decision support information, a query relevance that quantifies a degree to which the product-specific clinical decision support information includes content related to the query text; and(c) determining, as the clinical decision support information, the product-specific clinical decision support information for which the query relevance satisfies a preset condition.26.The method of claim 25, wherein the obtaining the clinical decision support information from the product-tailored database using the query text comprises performing at least one selected from the group consisting of:(a) searching for the product-specific clinical decision support information using a trained artificial intelligence (AI) model;(b) calculating the query relevance using the trained artificial intelligence (AI) model; and(c) determining the clinical decision support information using the trained artificial intelligence (AI) model.27.The method of claim 19, wherein the obtaining the clinical decision support information from the product-tailored database using the query text comprises:(a) when a preset reference search condition is satisfied, searching the product-tailored database for one or more product-specific reference information labeled with an identifier and / or metadata related to the query text,(b) calculating a reference score for each of the searched one or more product-specific reference information using one or more selected from the group consisting of a query relevance that quantifies a degree to which it includes content related to the query text, a reliability of a data source, a number of times cited in other clinical decision support information related to the corresponding diagnostic reagent product, a number of times cited in other reference information, and a recognition of an author or an affiliated institution, and(c) generating the clinical decision support information using the product-specific reference information for which the reference score satisfies a preset condition.28.The method of claim 27, wherein the obtaining the clinical decision support information from the product-tailored database using the query text comprises:generating a result of summarizing and analyzing the one or more product-specific reference information in a predefined report format using a generative artificial intelligence (AI) model trained to generate a natural language response to a natural language input.29.The method of claim 19, wherein the clinical decision support information comprises one or more selected from the group consisting of: (a) an disease identification result and / or a therapy determination result estimated to have a highest probability value based on one or more product-specific reference information that supports the clinical decision support information; (b) summary information for each of the one or more product-specific reference information; (c) a statistical analysis result for a type of disease identification result and / or a type of therapy determination result included in each of the one or more product-specific reference information; and (d) a statistical analysis result for each of a number of pieces of reference information that do not correspond to the detection result of the one or more target analytes among the one or more product-specific reference information, and a clinical date and item-specific clinical values included in the non-corresponding product-specific reference information.30.The method of claim 1, wherein the clinical decision support information is generated according to a preset standard format.31.The method of claim 30, wherein the standard format is a format preset in the computer device by a user account of the target terminal or a terminal associated with the target terminal, or is a user-customized format provided from the target terminal or the terminal associated with the target terminal.32.The method of claim 1, wherein the providing the clinical decision support information comprises:providing the test result, the clinical decision support information, and one or more product-specific reference information that supports the clinical decision support information together.33.The method of claim 1, wherein the target terminal or the terminal associated with the target terminal comprises at least one selected from the group consisting of a terminal of a medical professional including a clinician's portable terminal or therapy determination terminal, a server of a medical institution to which the medical professional belongs, a database of the medical institution, at least one terminal of a medical professional connected to the server of the medical institution via a network, and a testing terminal of the medical institution.34.The method of claim 1, further comprising, after the step of providing the clinical decision support information to the target terminal or the terminal associated with the target terminal:receiving individual clinical data from the target terminal or the terminal associated with the target terminal, the individual clinical data including: (a) the test result, (b) user information including one or more selected from the group consisting of a user name, affiliation, position, address, contact information, and country corresponding to an account of the target terminal or the terminal associated with the target terminal, and (c) a user's disease identification and therapy determination result based on the clinical decision support information; andstoring the individual clinical data in the product-tailored database.35.The method of claim 34, wherein usage rights for the individual clinical data are granted only to user accounts that have participated in providing at least one piece of the plurality of individual clinical data stored in the product-tailored database.36.The method of claim 34, further comprising: updating a weight used for obtaining the clinical decision support information based on the product-tailored database, based on the individual clinical data.37.The method of claim 1, wherein the obtaining the clinical decision support information comprises:when a Cycle threshold value (Ct value) included in the test result and / or an age of a subject is outside a preset appropriate range, obtaining the clinical decision support information based on a weight assigned to a therapy determination type corresponding to symptomatic therapy.38.The method of claim 1, wherein the obtaining the clinical decision support information comprises:when an underlying disease of a subject included in the test result corresponds to any one of preset cautionary underlying diseases related to an immunocompromised state, obtaining the clinical decision support information using (a) a weight assigned to the any one cautionary underlying disease and (b) a follow-up history including previously stored test results by test date for the subject and clinical decision support information.39.The method of claim 1, wherein the obtaining the clinical decision support information comprises:when a positive pathogen according to the test result is any one of preset antibiotic-resistant bacteria, obtaining the clinical decision support information including (a) a message recommending to proceed with an additional test including an antibiotic resistance test for the subject, and (b) therapy determination results for each result of the additional test, obtained based on a weight assigned to the any one antibiotic-resistant bacterium.40.The method of claim 1, wherein a type of the product-specific reference information stored in the product-tailored database includes papers and professional books,the clinical decision support information comprises first clinical decision support information for disease identification and therapy determination generated using the product-specific reference information corresponding to the papers and second clinical decision support information for disease identification and therapy determination generated using the product-specific reference information corresponding to the professional books,and the providing the clinical decision support information comprises:transmitting the clinical decision support information to the target terminal or the terminal associated with the target terminal, to control the first clinical decision support information for disease identification and therapy determination and the second clinical decision support information for disease identification and therapy determination to be displayed in a comparable manner on one screen at the target terminal or the terminal associated with the target terminal.41.A computer device, comprising:a memory storing a computer program including one or more instructions; anda processor that loads the computer program from the memory and execute the computer program,wherein the one or more instructions, when executed by the processor, cause the processor to:receive a test result obtained using one or more diagnostic test reagent products selected based on sample information, from a target terminal;obtain, based on the test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database, wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information as evidence of the product-specific clinical decision support information, both of which are labeled with an identifier based on a specification of a diagnostic test reagent product, whereby the product-specific clinical decision support information and the product-specific reference information can be used as considerations for the disease identification and therapy determination based on the test result using the one or more diagnostic test reagent products; andprovide the obtained clinical decision support information to the target terminal or to a terminal associated with the target terminal.42.A computer-readable non-transitory recording medium storing a computer program, wherein the computer program, when executed by a processor, causes the processor to:receive a test result obtained by testing with one or more diagnostic test reagent products selected based on sample information, from a target terminal;obtain, based on the received test result, clinical decision support information for disease identification and therapy determination applicable to the one or more diagnostic test reagent products from a product-tailored database, wherein the product-tailored database is configured to store: (i) one or more product-specific clinical decision support information for disease identification and therapy determination, and (ii) one or more product-specific reference information as evidence of the product-specific clinical decision support information, wherein both of which are labeled with an identifier based on a specification of a diagnostic test reagent product whereby the product-specific clinical decision support information and the product-specific reference information can be used as considerations for the disease identification and therapy determination based on the test result using the one or more diagnostic test reagent products; andprovide the obtained clinical decision support information to the target terminal or to a terminal associated with the target terminal.

Images