A patient disease diagnostic system through symptom reconstruction.

Abstract

To provide a system for diagnosing a patient's disease which can improve the accuracy of measurement of the similarity between a symptom of a patient and a symptom in a disease.SOLUTION: A system for diagnosing a patient's disease by a symptom reconstruction includes: a disease-symptom collection unit for collecting a group of known symptoms of a disease; a symptom reconstruction unit for generating a symptom reconstruction group by classifying the group of the symptoms of the disease in the same group set in advance; and a symptom similarity measuring unit for measuring a symptom similarity by comparing a symptom of the patient and the symptom reconstruction groups for all diseases.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a patient disease diagnosis system that can diagnose a patient's disease using the symptom similarity between the patient's symptoms and known diseases. More specifically, the present invention relates to a patient disease diagnosis system that can diagnose a patient's disease through symptom reconstruction.

[0002] The present invention is an invention made under the support of the Ministry of Science, ICT and Future Planning (MSIT) of the Republic of Korea, with project number 2022-0-00333. The specialized research management institution for the above project is the Institute for Information & communications Technology Planning and Evaluation (IITP). The name of the research project is "R&D of Source Technologies for the SW Computing Industry (Informatization)", the name of the research topic is "Development of an Integrated SW Solution for Multifaceted Analysis of Rare Pediatric Diseases A", the managing agency is "Slibillion Co., Ltd.", and the research period is from April 1, 2022 to December 31, 2024.

[0003] <00第18]] This application claims priority to Korean Patent Application No. 10-2023-0139926, filed with the Korean Intellectual Property Office on October 18, 2023, and the disclosure of the above patent application is incorporated herein by reference. ]>(END]]

Background Art

[0004] The symptom similarity between a patient's symptoms and known diseases is the most important information in diagnosing a patient's disease. In the case of a patient who has not yet received a diagnosis (rather than a patient who has already received a diagnosis), for all known diseases, after measuring the degree of similarity between the known symptoms of the disease and the patient's symptoms, it is a common procedure to confirm from the diseases with high symptom similarity.

[0005] In order to measure the similarity of diseases, it is necessary to measure the similarity between the patient's symptoms and the symptoms of the disease. However, the symptoms of a disease can vary among patients with the same disease.

[0006] [[ID=》 Furthermore, in the case of genetic diseases, symptoms may be similar among family members who share the same genetic defect. However, even if another independent family has the same genetic disease, the symptoms may differ between the two families. In addition, symptoms may vary depending on the population used in the literature reports that form the basis for the disease symptom information. This is evident from the fact that among the more than 300 diseases registered in OMIM (Online Mendelian Inheritance in Man), clinical differences between individuals with the same disease are explicitly mentioned.

[0007] Figure 1 shows the symptom information for disease ID 620158 in OMIM.

[0008] Referring to Figure 1, the symptom information does not distinguish whether the symptoms appeared in a single individual or in a family. For example, looking at the OMIM information for the disease, among the listed symptoms, dysphagia (difficulty swallowing food) has only been reported in a 47-year-old male patient with the disease. Also, memory impairment has not been reported in a 58-year-old female patient diagnosed with the disease. On the other hand, a 6-year-old girl diagnosed with the disease reported gait disturbance, loss of language ability, and epilepsy, symptoms that have not been reported in patients diagnosed with the disease.

[0009] As mentioned above, even with the same disease, the symptoms expressed by individuals can differ.

[0010] Conventional methods for measuring the similarity between a patient's symptoms and the symptoms included in a disease have the problem of leading to misdiagnosis of the patient's illness by deriving incorrect symptom similarity scores. [Prior art documents] [Patent Documents]

[0011] [Patent Document 1] Republic of Korea Published Patent No. 10-2022-0138327 (Publication Date: October 12, 2022) [Overview of the project] [Problems that the invention aims to solve]

[0012] The technical problem that this invention aims to solve is to provide a patient disease diagnostic system that can improve the accuracy of measuring the degree of symptom similarity between a patient's symptoms and the symptoms included in a disease. [Means for solving the problem]

[0013] To solve these problems, the disease diagnosis system for patients through symptom reconstruction according to an embodiment of the present invention includes: a disease-symptom collection unit that collects a set of symptoms for which a disease is known; a symptom reconstruction unit that generates a set of symptom reconstructions by classifying the set of disease symptoms into already established groups; and a symptom similarity measurement unit that measures the degree of symptom similarity by comparing the patient's symptoms with the set of symptom reconstructions for all diseases.

[0014] The symptom reconstruction unit can generate the symptom reconstruction set such that it satisfies the following equation 1.

[0015] <Expression 1> JPEG0007862812000001.jpg7170

[0016] Here, S is the set of all known symptoms for the disease, s i This is a set of symptom reconstructions obtained by reconstructing S.

[0017] The symptom reconstruction set (s i All symptoms belonging to ) should all belong within the single group (G) diagnosed with the disease, and all symptom reconfigurations (s i A diagnosis can consist of three or more symptoms.

[0018] The three or more symptoms included in the symptom reconstruction set (s i ) may not be in a hierarchical relationship with each other in the hierarchical relationship of the HPO.

[0019] The standard of the same group can be any one of an individual, monozygotic twins, siblings, a single family line, a single group reported in the literature, and a racial group with a similar genetic composition.

[0020] It can further include a disease discrimination unit that discriminates the disease of a patient by using the symptom similarity with respect to the symptom reconstruction set.

[0021] The disease discrimination unit can select the one with the highest symptom similarity among the symptom reconstruction sets for a disease as the representative symptom similarity for the disease, and assign a ranking according to the representative symptom similarity for each disease.

[0022] The disease discrimination unit can select the maximum symptom similarity with the highest symptom similarity, and discriminate the disease to which the symptom reconstruction set having the maximum symptom similarity belongs as the disease of the patient.

[0023] In addition to the technical problems of the present invention described above, other features and advantages of the present invention will be described below, or those having ordinary knowledge in the technical field to which the present invention belongs should be clearly understood from such technology and description.

Effect of the Invention

[0024] According to the present invention as described above, it has the following effects.

[0025] The present invention classifies the set of symptoms of a disease into the same group to generate a symptom reconstruction set, and then measures the symptom similarity between the symptom reconstruction set and the symptoms of the patient, so that the accuracy of the symptom similarity can be improved compared with the prior art that measures the symptom similarity between the symptom information of the disease and the symptoms of the patient.

[0026] Furthermore, other features and advantages of the present invention can be further understood through the embodiments of the present invention. [Brief explanation of the drawing]

[0027] [Figure 1] This figure shows symptom information for OMIM disease ID 620158. [Figure 2] This is a schematic diagram of a patient disease diagnosis system through symptom reconstruction according to one embodiment of the present invention. [Figure 3] This is a schematic diagram illustrating the function of a patient disease diagnosis system through symptom reconstruction according to one embodiment of the present invention. [Figure 4] This is an illustrative diagram illustrating how the disease discrimination unit according to the present invention discriminates a disease. [Figure 5] This is another illustrative diagram illustrating how the disease discrimination unit according to the present invention discriminates a disease. [Figure 6] This figure shows the ratio of the group morphology that yielded the highest similarity according to the symptom similarity measurement method of the present invention. [Modes for carrying out the invention]

[0028] In this specification, when assigning reference numbers to the components of each figure, it should be noted that, to the extent possible, the same component will have the same number, even if it is shown in other figures.

[0029] On the other hand, the meanings of the terms used herein should be understood as follows:

[0030] A singular expression should be understood to include multiple expressions unless it is clearly defined otherwise in the context.

[0031] Terms such as "includes" or "possesses" should be understood as not preemptively excluding the existence or possibility of adding one or more other features, figures, steps, actions, components, parts, or combinations thereof.

[0032] The following describes in detail a desirable embodiment of the invention devised to solve the above-mentioned problems, with reference to the attached diagram.

[0033] Figure 2 is a schematic diagram of a patient disease diagnosis system through symptom reconstruction according to one embodiment of the present invention, and Figure 3 is a schematic illustrative diagram for explaining what the patient disease diagnosis system through symptom reconstruction according to one embodiment of the present invention does.

[0034] Referring to Figure 2, a patient disease diagnosis system 1000 through symptom reconstruction according to one embodiment of the present invention includes a disease-symptom collection unit 100, a symptom reconstruction unit 200, a symptom similarity measurement unit 300, and a disease discrimination unit 400.

[0035] The disease-symptom collection unit 100 can collect a set of symptoms associated with a known disease.

[0036] The disease-symptom collection unit 100 can collect a set of symptoms for a disease from a database 2000 that organizes and stores known symptoms for diseases, such as OMIM (Online Mendelian Inheritance in Man), D0 (Disease Ontology), and HPO (Human Phenotype Ontology).

[0037] OMIM is a continuously updated database of human genetic disorders and genes. OMIM includes information such as genes linked to specific diseases, clinical characteristics of the diseases, and references to related research.

[0038] D0 provides a hierarchical and standardized ontology (classification and structuring of information) for human diseases. Using D0, information related to complex disease classification systems can be organized and integrated with other databases.

[0039] The Human Phenomena Oddity (HPO) provides a standardized ontology for phenomenological abnormalities in humans. It offers a standard language for describing the clinical characteristics of genetic disorders and can be used in the diagnosis and research of these disorders. The HPO structures specific symptoms and phenomena related to disease, using a hierarchical structure that shows the relationships between interrelated symptoms.

[0040] The symptom reconstruction unit 200 can generate a symptom reconstruction set by classifying a set of disease symptoms into the same group that has already been established.

[0041] The symptom reconstruction unit 200 generates a set of symptom reconstructions that satisfy the following equation 1, where S is the set of all known symptoms for the disease, and s i This is a set of symptom reconstructions obtained by reconstructing S.

[0042] <Expression 1> JPEG0007862812000002.jpg6170

[0043] In this case, the symptom reconstruction set (s i All symptoms belonging to ) should all belong within the single group (G) diagnosed with the disease, and all symptom reconfigurations (s i ) can consist of three or more symptoms. Here, the symptom reconfiguration set (s i Three or more symptoms included in ) may not necessarily be in a hierarchical relationship with each other within the HPO hierarchy.

[0044] The symptom reconstruction unit 200 can reconstruct disease symptom information by using a set of symptoms observed simultaneously within the same population for known symptoms of the disease.

[0045] Actual examples are shown in Table 1 below. In this case, the criteria for "same group" may include individuals, identical twins, siblings, a single family, a single group reported in the literature, or racial groups with similar genetic makeup. In this invention, as one example, only individuals, family lines, and single groups reported in the literature are given as examples of "same group," but the criteria for "same group" are not limited to these three.

[0046] [Table 1]

[0047] The symptoms are displayed as HPO, and the disease recognizer is an OMIM ID. The forms used for the same group are individual, family, and group, and all have a group recognizer. Since further individual, family, and group information may be added to the database in the future, even if there is only one form of the same group, there will always be a group recognizer (similar to the GROUP example in 620481). It can also be seen that there is an "ALL" category that integrates all of these symptoms and constitutes the symptoms in the same way as the existing system.

[0048] The symptom similarity measurement unit 300 can measure the degree of symptom similarity by comparing the patient's symptoms with a set of symptom reconstructions for all diseases.

[0049] The symptom similarity measurement unit 300 can use a variety of known methods with respect to the patient's symptoms and a reconstructed set of symptoms for the disease.

[0050] Unlike conventional techniques in which symptom similarity is calculated using disease IDs and a set of symptom IDs corresponding to the disease IDs, the symptom similarity measurement unit 300 according to the present invention can calculate symptom similarity when both the disease ID and the ID of the reconstructed symptom reconstruction group (symptom reconstruction set) are specified.

[0051] In other words, the symptom similarity measurement unit 300 according to the present invention can construct information from <disease ID, morphology of the same group, group ID, {symptom ID}>.

[0052] Table 2 below shows disease-symptom information of the prior art, and Table 3 below shows an example of disease-symptom information according to the present invention.

[0053] [Table 2]

[0054] [Table 3]

[0055] The symptom similarity measurement unit 300 can measure the symptom similarity between a patient and all identical population forms of the disease using disease-symptom information that is organized by dividing the same population into categories. Of these, the highest symptom similarity can be used as the symptom similarity between the patient and the disease.

[0056] The disease discrimination unit 400 can determine a patient's disease using a reconstructed set of disease symptoms and the patient's symptoms, along with a degree of symptom similarity.

[0057] Figure 4 is an illustrative diagram illustrating how the disease discrimination unit according to the present invention performs disease discrimination.

[0058] Referring to Figure 4, the disease discrimination unit 400 according to one embodiment of the present invention can select the one with the highest symptom similarity from a set of reconstructed symptoms for a disease as the representative symptom similarity for that disease, assign a ranking according to the representative symptom similarity for each disease, and determine the patient's disease.

[0059] Figure 5 is another illustrative diagram illustrating how the disease discrimination unit according to the present invention discriminates a disease.

[0060] Referring to Figure 5, the disease discrimination unit 400 according to another embodiment of the present invention can select the symptom with the highest symptom similarity and determine that the disease to which the symptom reconstruction set having the highest symptom similarity belongs is the patient's disease.

[0061] To confirm the effectiveness of the present invention, the symptoms diagnosed to 1633 patients diagnosed with a total of 309 diseases were reconstructed according to the present invention, and then the symptom similarity was checked. Symptom similarity was determined according to the method described in the literature "Clinical Diagnostics in Human Genetics with Semantic Similarity Searches in Ontologies." While the paper used Mutual Information for each symptom node, the present invention used the depth of the symptom node (shortest distance to the top-level node).

[0062] As a result, when using conventional methods, i.e., when symptom information is used without classifying individuals by morphology within the same group, the highest degree of similarity was found in only 34% of cases.

[0063] Figure 6 shows the proportion of the group morphology that yielded the highest degree of similarity according to the symptom similarity measurement method of the present invention.

[0064] Referring to Figure 6, when measuring the similarity of symptoms for a disease diagnosed by a patient according to the present invention, by classifying the symptoms of the disease into individual, family, and group categories, the highest symptom similarity was measured from the group category, at 43% of the total.

[0065] The success rate was 34% for conventional methods, 15% for individuals, and an additional 8% for families.

[0066] When the symptoms of the group and the symptoms of the patient were most similar, the difference in scores was 4.61 points on average using the conventional method. However, when the similarity was calculated by classifying the symptoms of the group according to the present invention, the difference was 5.18 points, a difference of 0.57 points on average. This is a single-tailed paired t-test p-value < 0.001, indicating a very significant difference.

[0067] When individual symptoms were categorized and calculated, the average symptom similarity score for the patient's symptoms was 4.62 points when using the conventional method, compared to 5.06 points when following the method according to the present invention. This represents a difference of 0.44 points on average, and these differences are very significant, as the p-value of a single-sided paired t-test was < 0.001.

[0068] When symptoms were categorized and calculated separately for each family lineage, the average symptom similarity score was 3.90 points when using the conventional method compared to 4.28 points when following the method according to the present invention. This represents a difference of 0.38 points on average, and these differences are very significant, as the p-value of a single-sided paired t-test was < 0.001.

[0069] When comparing the conventional method with the method according to the present invention, the largest difference in scores is as shown in Table 4 below.

[0070] [Table 4]

[0071] Comparing the conventional method, which calculates similarity using the overall symptoms of the disease without group differentiation, with the method according to the present invention, which calculates symptom similarity with patients by differentiating symptoms for each group, it was found that a particularly large difference in scores occurs, mainly when calculating by differentiating symptoms for individuals or families.

[0072] In other words, even among those with the same disease, symptoms can differ depending on the individual, family, or group. It was found that symptoms appearing from individuals or families, which are biologically more homogeneous groups, tend to be more similar to the symptoms of patients than symptoms categorized by group.

[0073] Through this, it was confirmed that when applying the disease diagnosis system for patients through symptom reconstruction according to the present invention, the degree of symptom similarity between the symptoms of a diagnosed patient and the disease to which the patient has been diagnosed can be measured at an even higher level.

[0074] The above-described objectives, other objectives, features, and advantages of the present invention should be readily apparent through the above preferred embodiments relating to the accompanying figures. However, the present invention is not limited to the embodiments described herein and can be embodied in other forms. Rather, the embodiments presented herein are provided to ensure that the disclosed content is thorough and complete, and that the idea of ​​the invention is fully conveyed to a person of the ordinary skill. Where it is referred to herein that a component is on another component, it means that it can be formed directly on the other component, or that a third component can be placed between them.

[0075] Furthermore, when it is mentioned that an element (or component) operates or runs on (ON) another element (or component), it should be understood that the element (or component) operates or runs in the environment in which the other element (or component) operates or runs, or operates or runs through direct or indirect interaction with the other element (or component).

[0076] When an element, component, device, or system is mentioned as containing a component consisting of a program or software, it should be understood, even without explicit mention, that element, component, device, or system also contains hardware (e.g., memory, CPU, etc.) and other programs or software (e.g., operating systems and drivers necessary to drive the hardware) necessary for that program or software to run or operate.

[0077] Furthermore, unless otherwise specified, it should be understood that an element (or component) can be embodied in the form of software, hardware, or both. [Explanation of Symbols]

[0078] 100... Disease-Symptom Collection Department 200...Symptom Reconstruction Department 300...Symptom similarity measurement part 400...Disease Identification Department 1000... A patient disease diagnostic system through symptom reconstruction

Claims

1. The Disease-Symptom Collection Unit collects a set of known symptoms of a disease, A symptom reconstruction unit generates a symptom reconstruction set, which is a subset of the symptom set of the disease, by associating the symptoms belonging to the set of symptoms of the disease with the forms of a group that have already been established. A symptom similarity measurement unit measures the degree of symptom similarity by comparing the patient's symptoms with a reconstructed set of symptoms for all diseases. Includes, The aforementioned group morphology is one of the following: an individual, identical twins, siblings, a single family, a group used in literature reports that form the basis for disease symptom information, or a racial group with a similar genetic makeup. A disease diagnosis system for patients through symptom reconstruction.

2. The patient disease diagnosis system through symptom reconstruction according to claim 1, characterized in that the symptom reconstruction unit generates the symptom reconstruction set so that it satisfies the following formula 1. <Formula 1> Here, S is the set of all known symptoms for the disease, s ｉ This is a set of symptom reconstructions created by reconstructing S.

3. The symptom reconstruction set (s ｉ All symptoms belonging to ) should all belong within the single group (G) diagnosed with the disease, and all symptom reconstruction sets (s ｉ The patient disease diagnosis system through symptom reconstruction according to claim 2, characterized in that the symptoms should consist of three or more symptoms.

4. The symptom reconstruction set (s ｉ A patient disease diagnosis system through symptom reconstruction according to claim 3, characterized in that three or more symptoms included in ) are not in a hierarchical relationship with each other in the HPO hierarchy.

5. The patient disease diagnosis system through symptom reconstruction according to claim 1, further comprising a disease determination unit that determines the patient's disease using the symptom similarity to the symptom reconstruction set.

6. The disease discrimination unit is characterized in that it selects the one with the highest symptom similarity from the set of reconstructed symptoms for a disease as the representative symptom similarity for the disease, and assigns a ranking according to the representative symptom similarity for each disease, as described in claim 5.

7. The disease discrimination unit selects the symptom with the highest symptom similarity and identifies the disease to which the symptom reconstruction set having the highest symptom similarity belongs as the patient's disease, as described in claim 5.

Images