A method and system for diagnosing dementia using genetic information data and brain imaging data.

Abstract

One embodiment of the present disclosure provides a dementia diagnosis method, comprising the steps of: generating genetic feature data by comparing and analyzing genetic information of normal individuals and genetic information of dementia patients; encoding the brain image data of the normal individuals and the brain image data of the dementia patients, respectively, and generating brain feature data corresponding to the encoded brain image data of the normal individuals and the dementia patients, respectively; and using data obtained by combining the genetic feature data and the brain feature data in a predetermined manner as training data, generating a dementia diagnosis model trained to determine whether or not a specific individual has dementia based on the genetic information and genetic feature data of the specific individual.

Description

Technical Field

[0001] The present invention relates to a method and system for diagnosing dementia using gene information data and brain image data. More specifically, the gene information of normal people, the gene information of dementia patients, the brain image data of normal people, and the brain image data of dementia patients are analyzed to extract characteristic data respectively, and a method and system for diagnosing dementia based on a model obtained by combining and training the characteristic data are provided.

Background Art

[0002] Dementia is a disease in which the regions of the cerebrum or hippocampus responsible for brain memory decrease due to the accumulation of beta-amyloid.

[0003] According to the statistics of "The Current Situation of Dementia in the Republic of Korea" announced by the Central Dementia Center, the number of dementia patients aged 65 and above was counted as 648,223 (incidence rate 9.8%) based on 2015, and it is estimated that it will exceed 1 million (incidence rate 10.3%) in 2024 and 2 million (incidence rate 12.3%) in 2041.

[0004] In addition, conventional dementia examination tools require at least 30 minutes or more of examination time, and diagnose mild cognitive impairment or dementia based on MRI imaging, questionnaire surveys on memory, and the opinions of doctors. However, the current situation is that there is a shortage of specialized medical personnel such as dementia specialists or neuropsychologists and clinical psychologists who have completed specialized education, and there is also a shortage of hospitals that can diagnose dementia, such as insufficient separate examination places.

[0005] Therefore, there is a need for a new method for diagnosing dementia in addition to the opinions of doctors.

Summary of the Invention

Problems to be Solved by the Invention

[0006] This disclosure aims to solve the problems of the prior art described above, and provides a method and system for diagnosing dementia by analyzing the genetic information of healthy individuals and the genetic information of dementia patients, as well as the brain image data of healthy individuals and the brain image data of dementia patients, extracting feature data from each, and combining the feature data to create a trained model.

[0007] The technical problems that this invention aims to solve are not limited to those described above, and other technical problems may be derived from the following description. [Means for solving the problem]

[0008] As a technical means for solving the technical problems described above, an embodiment relating to the first aspect of this disclosure provides a method for diagnosing dementia. This method includes the steps of: generating genetic feature data by comparing and analyzing the genetic information of a normal person and the genetic information of a dementia patient; encoding brain image data of a normal person and brain image data of a dementia patient, respectively, and generating brain feature data corresponding to the encoded brain image data of a normal person and brain image data of a dementia patient, respectively; and generating a dementia diagnostic model trained to determine whether or not a specific person has dementia based on the genetic information and genetic feature data of that specific person, using data obtained by combining the genetic feature data and the brain feature data according to a predetermined method as training data.

[0009] Furthermore, embodiments relating to a second aspect of this disclosure provide a dementia diagnostic system. The system includes a communication module, at least one processor, and a memory electrically connected to the processor and storing at least one code executed by the processor, the memory storing code that, when executed through the processor, causes the processor to generate genetic feature data by comparing and analyzing the genetic information of a normal person and the genetic information of a dementia patient, encode the brain image data of a normal person and the brain image data of a dementia patient, respectively, generate brain feature data corresponding to the encoded brain image data of a normal person and the brain image data of a dementia patient, respectively, and use the data obtained by combining the genetic feature data and the brain feature data in a predetermined manner as training data to generate a dementia diagnostic model trained to determine whether or not a particular person has dementia based on the genetic information and genetic feature data of that particular person. [Effects of the Invention]

[0010] According to the present invention, the accuracy of dementia diagnosis can be improved by considering both genetic data and brain imaging data.

[0011] Furthermore, according to the present invention, the convenience of the examination can be improved by diagnosing dementia based on genetic data and brain imaging data.

[0012] The effects of the present invention are not limited to those described above, but include all effects that can be understood from the following description. [Brief explanation of the drawing]

[0013] [Figure 1] This is a diagram illustrating a dementia diagnostic system according to one embodiment of the present invention. [Figure 2] Figure 1 is a diagram showing the detailed configuration of the server. [Figure 3] This diagram illustrates an example of genetic feature data. [Figure 4]This diagram illustrates an example of genetic feature data. [Figure 5] This diagram illustrates an example of extracting brain feature data. [Figure 6] This diagram illustrates an example of extracting brain feature data. [Figure 7] This diagram illustrates an example of extracting brain feature data. [Figure 8a] This diagram is shown to illustrate the accuracy of dementia diagnosis results obtained using a dementia diagnostic model. [Figure 8b] This diagram is shown to illustrate the accuracy of dementia diagnosis results obtained using a dementia diagnostic model. [Figure 9] This is a flowchart showing the sequence of a dementia diagnosis method according to another embodiment of the present invention. [Modes for carrying out the invention]

[0014] The present disclosure will be described in detail below with reference to the accompanying drawings. However, the present disclosure may be implemented in various different forms and is not limited to the embodiments described herein. The accompanying drawings are provided to facilitate understanding of the embodiments disclosed herein and do not limit the technical ideas disclosed herein. All terms used herein, including technical and scientific terms, should be interpreted as generally understood by those skilled in the art in which this disclosure pertains. Predefined terms should be interpreted to have further meanings consistent with the relevant technical literature and the content of this disclosure, and should not be interpreted in an overly ideal or restrictive sense unless specifically defined.

[0015] To clearly illustrate this disclosure in the drawings, irrelevant details have been omitted, and the size, form, and shape of each component shown in the drawings are subject to various modifications. Parts identical or similar throughout the specification are denoted by the same or similar reference numerals.

[0016] In the following description, suffixes such as "module" and "section" for components used are given or mixed only for the convenience of preparing the specification, and do not have meanings or roles that are mutually distinguishable by themselves. Further, when explaining the embodiments disclosed in this specification, if it is determined that a specific description of related well-known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed description thereof is omitted.

[0017] Throughout this specification, when a certain part is described as being "connected (connected, contacted, or coupled)" to another part, it includes not only the case where it is "directly connected (connected, contacted, or coupled)", but also the case where it is "indirectly connected (connected, contacted, or coupled)" via other members therebetween. Further, when a certain part is described as "including (comprising, or providing)" a certain component, it means that other components can be further "included (comprising, or providing)" without excluding other components unless otherwise specifically stated to the contrary.

[0018] In this specification, terms indicating ordinal numbers such as "first" and "second" are used only for the purpose of distinguishing one component from another, and do not limit the order or relationship of the components. For example, the first component of the present disclosure may be called the second component, and similarly, the second component may also be called the first component. In this specification, singular expressions used should be interpreted to include plural expressions as well, unless the contrary meaning is clearly indicated.

[0019] FIG. 1 is a drawing shown to explain a dementia diagnosis system according to an embodiment of the present invention. Referring to FIG. 1, the dementia diagnosis system can include a server (100) and a user terminal (200). The server (100) and the user terminal (200) can be communicatively connected to each other via a communication network.

[0020] The server (100) compares and analyzes the genetic information of normal people and the genetic information of dementia patients to generate or extract genetic feature data.

[0021] The server (100) encodes the brain image data of normal people and the brain image data of dementia patients respectively. The server (100) generates or extracts brain feature data corresponding to the encoded brain image data of normal people and the encoded brain image data of dementia patients respectively. For example, the brain image data can be an MRI photo of the brain.

[0022] The server (100) uses the data obtained by combining the genetic feature data and the brain feature data according to a predetermined method as training data to generate a dementia diagnosis model trained to determine the presence or absence of dementia in a specific person based on the genetic information and genetic feature data of the specific person. The server (100) diagnoses dementia corresponding to the input genetic information and brain image data based on the generated dementia diagnosis model.

[0023] The user terminal (200) can receive the dementia diagnosis result from the server (100). In addition, the user terminal (200) can send genetic information and brain image data to the server (100).

[0024] The user terminal (200) can be communicatively connected to the server (100) through a communication network. The user terminal (200) can mean all kinds of handheld wireless communication devices such as a notebook computer equipped with a WEB browser, a desktop, a laptop, a wireless communication device with guaranteed portability and mobility, or a smartphone, a tablet PC, etc.

[0025] Figure 2 is a drawing showing the detailed configuration of the server shown in Figure 1. Referring to Figure 2, the server (100) can include a communication module (110), a processor (120), and a memory (130).

[0026] The communication module (110) may include devices that include hardware and software necessary for sending and receiving signals, such as control signals or data signals, with other network devices via wired or wireless connections.

[0027] The communication module (110) can receive genetic information and brain imaging data from a user terminal for healthy individuals, patients with mild cognitive impairment, and patients with dementia. The communication module (110) can also transmit dementia diagnosis results to the user terminal.

[0028] The processor (120) can include various types of devices that control and process data. The processor (120) may mean a hardware-integrated data processing device that has physically configured circuitry to perform functions expressed in code or instructions contained within a program.

[0029] For example, the processor (120) can be implemented in the form of a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an ASIC (application-specific integrated circuit), an FPGA (field programmable gate array), etc., but the scope of the present invention is not limited to these.

[0030] The processor (120) performs actions according to the code stored in memory (130).

[0031] The memory (130) can store at least one of the following: information and data input through the communication module (110), information and data necessary for functions performed by the processor (120), and data generated by the execution of the processor (120).

[0032] Memory (130) should be interpreted as a general term for non-volatile memory devices that maintain stored information even without a power supply, and volatile memory devices that require power to maintain stored information. In addition to volatile memory devices that require power to maintain stored information, memory (130) may also include magnetic storage media or flash storage media, but the scope of the present invention is not limited thereto.

[0033] Memory (130) is electrically connected to the processor (120) and stores at least one piece of code that is executed by the processor (120). Memory (130) stores code that causes the processor (120) to perform functions and procedures such as the following:

[0034] Memory (130) stores code that triggers the generation of genetic feature data by comparing and analyzing the genetic information of normal individuals and dementia patients. For example, the genetic information may be data representing the composition of the microbiome as sample-specific bacterial abundance data, obtained through the analysis of 16s rRNA sequencing data from fecal samples. Alternatively, the genetic feature data may be vectors generated based on the frequency of occurrence of each bacterium.

[0035] Memory (130) may store code that triggers the clustering of bacterial information contained in the genetic information of normal individuals and dementia patients to generate bacterial data. Memory (130) may also store code that triggers the conversion of the generated bacterial data into numerical data to extract vector-based genetic feature data. For example, based on the stored code, memory (130) can perform ASV clustering and classification annotation (Taxon annotation) on 16s rRNA data.

[0036] For example, memory (130) can convert ASV data into numerical data based on the stored code, perform one-hot encoding on each sample for each ASV, and generate numerical vectors representing the ASV features for each sample. Here, ASV may be taxonomic information matched down to the species level.

[0037] Furthermore, memory (130) may store code that triggers the comparative analysis of genetic information of patients with mild cognitive impairment to extract genetic characteristic data.

[0038] Furthermore, memory (130) may store code that uses a CNN to extract features from brain image data and vectorize the brain image data corresponding to normal individuals, patients with mild cognitive impairment, and patients with dementia, respectively.

[0039] Memory (130) stores codes that trigger the encoding of brain image data from healthy individuals and brain image data from dementia patients, respectively. For example, memory (130) may store codes that trigger the encoding of brain image data from healthy individuals and brain image data from dementia patients, region by region. Here, the brain image data may be brain MRI images, and memory (130) may store codes that trigger the encoding of brain MRI images into brain image data based on a dementia diagnostic model.

[0040] Memory (130) stores code that triggers the generation of brain feature data corresponding to encoded brain image data of normal individuals and brain image data of dementia patients, respectively. Memory (130) may also store code that triggers the extraction of brain feature data corresponding to brain image data of patients with mild cognitive impairment.

[0041] For example, memory (130) may store coded brain image data of a normal person, brain image data of a dementia patient, and brain image data of a patient with mild cognitive impairment, as well as code that causes the generation of vector-format brain feature data based on a predetermined feature extraction scheme.

[0042] Memory (130) stores code that triggers the generation of a dementia diagnostic model trained to determine whether a particular person has dementia based on their genetic information and genetic feature data, using data obtained by combining genetic feature data, brain feature data of normal individuals, and brain feature data of dementia patients according to a predetermined method as training data. For example, memory (130) may store code that triggers the generation of an input vector by concatenating genetic feature data and brain feature data. For example, based on the stored code, memory (130) can generate an input vector by concatenating brain feature data obtained by a CNN with transformed genetic feature data.

[0043] Memory (130) may store code that uses the generated input vectors to train a dementia diagnostic model and triggers the generation of the dementia diagnostic model. For example, based on the stored code, Memory (130) can input the combined input vectors into the dementia diagnostic model, train the model, and run it. Memory (130) can calculate the predicted values ​​and loss for dementia or normal. Here, the dementia diagnostic model may be a model generated using Keras.

[0044] Memory (130) stores code that, based on a dementia diagnostic model, triggers a diagnosis of dementia in response to input genetic information and brain imaging data. For example, memory (130) stores code that, using a dementia diagnostic model, receives the genetic information and brain imaging data of a specific person as input and triggers a determination that the person's condition is at least one of mild cognitive impairment, dementia, or normal.

[0045] Figures 3 and 4 are diagrams shown to illustrate an example of gene feature data. Referring to Figures 3 and 4, the genetic feature data can include bacterial-specific ASV IDs, taxonomic information data about ASVs, and sample-specific bacterial information for samples from normal individuals and dementia patients. For example, the genetic feature data can include information about OUT ID, kingdom, phylum, class, order, family, genus, species, AD01, and C88.

[0046] Genetic feature data can be extracted in the form of vectors based on the frequency of occurrence of each ASV, such as AD1=[1,0,0,1] and C88=[0,1,1,0].

[0047] The dementia diagnostic system can convert each ASV into a vector (410) where each feature has a unique value. For example, if there are four ASVs, the dementia diagnostic system can perform one-hot encoding, resulting in ASV1:[1,0,0,0], ASV2:[0,1,0,0], ASV3:[0,0,1,0], and ASV4:[0,0,0,1].

[0048] The dementia diagnostic system can generate transformed data (420) for a sample. For example, the dementia diagnostic system can generate transformed data for sample 1, which contains ASV1, and for sample 2, which contains ASV2 and ASV3, such as sample 1: [1,0,0,0] and sample 2: [0,1,1,0].

[0049] The dementia diagnostic system can extract vector-based gene feature data (440) based on the number of occurrences (430) of each ASV. For example, if ASV1 appears 5 times and ASV3 appears 3 times in Sample 1, and ASV1 appears 2 times and ASV3 appears 7 times in Sample 2, the dementia diagnostic system can extract gene feature data (440) as follows: Sample 1: [5,3,0,0], Sample 2: [2,0,7,0].

[0050] Figures 5 to 7 are diagrams illustrating an example of extracting brain feature data. Referring to Figures 5 to 7, the dementia diagnostic system can encode MRI images of normal individuals, dementia patients, and patients with mild cognitive impairment by region. The dementia diagnostic system can combine the region-coded brain image data (610) and feature extraction data (620) to generate a feature map (710).

[0051] The dementia diagnostic system can generate layers relating to the features of brain image data by repeatedly applying max pooling and flattening to a feature map (710), and extract brain feature data in vector format. Based on the brain feature data, the dementia diagnostic system can analyze newly input brain image data and genetic information to diagnose the presence or absence of dementia.

[0052] Figures 8a and 8b are diagrams shown to illustrate the accuracy of dementia diagnosis results obtained using dementia diagnostic models. Referring to Figure 8a, it can be confirmed that when a dementia diagnostic model was generated by combining MRI images and 16s rRNA data, the result of diagnosing dementia was output with a loss of 0.203 relative to the test set and an accuracy of 0.91. At this point, the dementia diagnostic model can be trained over 10 epochs.

[0053] On the other hand, referring to Figure 8b, it can be confirmed that when a dementia diagnostic model was generated using only 16s rRNA data without MRI images, the result of diagnosing dementia was a loss of 0.356 relative to the test set and an accuracy of 0.80.

[0054] This confirms that when a dementia diagnostic model is generated by combining MRI images and 16S rRNA data and used to diagnose dementia, the accuracy improves to 0.91 and the loss decreases by 0.153 compared to when a dementia diagnostic model is generated and used only 16S rRNA data.

[0055] Figure 9 is a flowchart showing the procedure for a dementia diagnosis method according to another embodiment of the present invention. The dementia diagnostic method described below can be performed by the dementia diagnostic system and server described above with reference to Figures 1 to 8. Therefore, the details of the embodiments of this disclosure described above with reference to Figures 1 to 8 can also be applied to the embodiments described below, and any content that overlaps with the above description will be omitted below. The steps described below do not necessarily have to be performed in order, the order of the steps can be set in various ways, and the steps may be performed almost simultaneously.

[0056] Referring to Figure 9, the dementia diagnostic method includes a gene feature data extraction stage (S100), a brain feature data extraction stage (S200), and a dementia diagnostic stage (S300) based on the gene feature data and brain feature data.

[0057] The gene feature data extraction stage (S100) is a stage in which gene feature data is extracted by comparing and analyzing the gene information of normal individuals and the gene information of dementia patients. For example, in the gene feature data extraction stage (S100), bacterial data can be generated by clustering the bacterial information of the gene information of normal individuals and the gene information of dementia patients, and this bacterial data can be converted into numerical data to extract gene feature data in vector format.

[0058] The brain feature data extraction stage (S200) involves encoding brain image data from both healthy individuals and dementia patients, and extracting brain feature data corresponding to the encoded brain image data from both individuals. For example, in the brain feature data extraction stage (S200), brain image data from both healthy individuals and dementia patients can be encoded region by region, and vector-format brain feature data can be extracted based on the encoded brain image data and predetermined feature extraction data.

[0059] The dementia diagnosis stage (S300) based on genetic feature data and brain feature data is a stage in which a dementia diagnosis model is generated, which is trained to determine whether or not a specific person has dementia based on the genetic information and genetic feature data of that specific person, using data obtained by combining genetic feature data and brain feature data according to a predetermined method as training data. For example, in the dementia diagnosis stage (S300) based on genetic feature data and brain feature data, a single input vector is generated by concatenating genetic feature data and brain feature data, a dementia diagnosis model is trained using this input vector, and dementia can be diagnosed based on the dementia diagnosis model in accordance with the input genetic information and brain image data.

[0060] A person skilled in the art to which this disclosure belongs will understand, based on the above description, that it is possible to readily modify this disclosure into other specific forms without departing from the technical idea or essential features of this disclosure. Therefore, the above-described embodiments are illustrative in all respects and should not be construed as limiting. The scope of this disclosure is indicated by the claims set forth below, and all changes or modifications derived from the meaning and scope of the claims and the concept of equivalents thereto should be construed as being included within the scope of this disclosure. The scope of this application is indicated by the claims set forth below, rather than the above-described detailed description, and all changes or modifications derived from the meaning and scope of the claims and the concept of equivalents thereto should be construed as being included within the scope of this application. [Industrial applicability]

[0061] This invention has industrial applicability because it can be used in technologies for diagnosing dementia.

Claims

1. In dementia diagnostic methods performed by dementia diagnostic systems, a) A step of generating genetic characteristic data by comparing and analyzing the genetic information of normal individuals and patients with dementia, b) The steps of encoding brain image data of a normal person and brain image data of a dementia patient, respectively, and generating brain feature data corresponding to the encoded brain image data of the normal person and the brain image data of the dementia patient, c) A step of generating a dementia diagnostic model that is trained to determine whether or not a specific person has dementia based on the genetic information and genetic feature data of a specific person, using data obtained by combining the genetic feature data and the brain feature data according to a predetermined method as training data. Includes, A dementia diagnosis method comprising: in step a) above, clustering bacterial information from the genetic information of a normal person and the genetic information of a dementia patient to generate bacterial data, and converting the bacterial data into numerical data to generate the genetic feature data in vector format.

2. A dementia diagnosis method according to claim 1, wherein the gene feature data is vector data generated based on the frequency of occurrence of each bacterium.

3. A dementia diagnosis method according to claim 1, wherein the genetic information includes 16s rRNA data.

4. A dementia diagnosis method according to claim 1, wherein in step b), brain image data of a normal person and brain image data of a dementia patient are coded region by region, and brain feature data in vector format is generated based on the coded brain image data and a predetermined feature extraction method.

5. A dementia diagnosis method according to claim 1, wherein in step c), the gene feature data and the brain feature data are linked to generate a single input vector, and the dementia diagnosis model is trained based on the input vector.

6. In the dementia diagnosis method according to claim 1, step a) includes a step of generating the genetic characteristic data by comparing and analyzing the genetic information of a patient with mild cognitive impairment with the genetic information of a normal person, A dementia diagnostic method, wherein step b) includes the step of comparing and analyzing brain image data of a patient with mild cognitive impairment with brain image data of a normal person to generate brain characteristic data.

7. A dementia diagnosis method according to claim 1, wherein step c) includes a step of using the dementia diagnosis model to receive the genetic information and brain image data of a specific person as input, and determining the condition of the specific person as at least one of mild cognitive impairment, dementia, and normal.

8. Communication module and At least one processor, The system includes a memory electrically connected to the processor, in which at least one code executed by the processor is stored. The memory stores code that, when executed through the processor, causes the processor to compare and analyze the genetic information of a normal person and the genetic information of a dementia patient to generate genetic feature data, encode the brain image data of a normal person and the brain image data of a dementia patient, respectively, generate brain feature data corresponding to the encoded brain image data of a normal person and the brain image data of a dementia patient, and use the data obtained by combining the genetic feature data and the brain feature data according to a predetermined method as training data to generate a dementia diagnostic model that is trained to determine whether or not a particular person has dementia based on the genetic information and genetic feature data of that particular person. A dementia diagnostic system, wherein the memory stores code that causes the processor to cluster bacterial information from the genetic information of a normal person and the genetic information of a dementia patient to generate bacterial data, convert the bacterial data into numerical data, and extract the genetic feature data in vector form.

9. A dementia diagnostic system according to claim 8, wherein the gene feature data is vector data generated based on the frequency of occurrence of each bacterium.

10. A dementia diagnosis system according to claim 8, wherein the memory stores code that causes the processor to encode the brain image data of a normal person and the brain image data of a dementia patient region by region, and to extract the brain feature data in vector form based on the encoded brain image data and predetermined feature extraction data.

11. A dementia diagnosis system according to claim 8, wherein the memory stores code that causes the processor to generate an input vector by concatenating the gene feature data and the brain feature data, to train the dementia diagnosis model using the input vector, and to diagnose dementia based on the dementia diagnosis model.

12. A dementia diagnosis system according to claim 8, wherein the memory stores code that causes the processor to generate the gene characteristic data by comparing and analyzing the gene information of a patient with mild cognitive impairment with the gene information of a normal person, and to generate the brain characteristic data by comparing and analyzing the brain image data of the patient with mild cognitive impairment with the brain image data of a normal person.

13. A dementia diagnosis system according to claim 8, wherein the memory stores code that causes the processor to receive the genetic information and brain image data of a specific person as input using the dementia diagnosis model, and to determine the state of the specific person as at least one of mild cognitive impairment, dementia, and normal.

Images