Method and system for diagnosing dementia by using gene information data and brain image data
A system combining genetic and brain image data analysis addresses the inefficiencies of current dementia diagnosis methods by generating a trained model for accurate and convenient dementia diagnosis, enhancing testing efficiency and accessibility.
Patent Information
- Application Number
- PCT/KR2024/005595
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-25
- Filing Date
- 2024-04-25
- Publication Date
- 2024-11-28
AI Technical Summary
Current dementia diagnosis methods are time-consuming and require specialized medical professionals, with a shortage of facilities and resources, necessitating a more efficient and accessible diagnostic approach beyond traditional doctor's opinions.
A method and system that combines genetic information and brain image data analysis to generate feature data, training a model for dementia diagnosis, utilizing a server with a communication module, processor, and memory to compare and encode data, and create a dementia diagnosis model.
Improves diagnostic accuracy and convenience by integrating genetic and brain image data, increasing the efficiency of dementia testing and overcoming resource limitations.
Smart Images

Figure KR2024005595_28112024_PF_FP_ABST
Abstract
Description
Dementia diagnosis method and system using genetic information data and brain image data
[0001] The present invention relates to a method and system for diagnosing dementia using genetic information data and brain image data, and more specifically, to a method and system for diagnosing dementia based on a model that extracts feature data by analyzing genetic information of a normal person and genetic information of a dementia patient and brain image data of a normal person and brain image data of a dementia patient, and combines the feature data to learn the feature data.
[0002] Dementia is a disease in which the area of the brain responsible for memory, the hippocampus or cerebrum, decreases as beta-amyloid accumulates.
[0003] According to the statistics on the 'Status of Dementia in Korea' published by the National Dementia Center, the number of dementia patients aged 65 years or older was 648,223 (prevalence rate of 9.8%) as of 2015, and it is estimated that the number will exceed 1 million (prevalence rate of 10.3%) in 2024 and 2 million (prevalence rate of 12.3%) in 2041.
[0004] Furthermore, existing dementia screening tools require lengthy testing times, at least 30 minutes, and diagnose mild cognitive impairment or dementia based on MRI scans, memory questionnaires, and physician opinions. However, there is a shortage of specialized medical professionals, such as dementia specialists or trained neuropsychologists or clinical psychologists, and a lack of dedicated testing sites, resulting in a shortage of hospitals capable of treating dementia.
[0005] Accordingly, a new dementia diagnosis method other than medical opinion is needed.
[0006] The present disclosure is intended to solve the problems of the prior art described above, and provides a method and system for diagnosing dementia based on a model that extracts feature data by analyzing genetic information of a normal person and genetic information of a dementia patient, brain image data of a normal person and brain image data of a dementia patient, and combines the feature data to learn the feature data.
[0007] The technical problems to be solved by the present invention are not limited to the technical problems described above, and other technical problems of the present invention can be derived from the following description.
[0008] As a technical means for solving the above-described technical problem, an embodiment according to a first aspect of the present disclosure provides a dementia diagnosis method. The method includes a step of generating genetic feature data by comparing and analyzing genetic information of a normal person and genetic information of a dementia patient, a step of encoding brain image data of a normal person and brain image data of a dementia patient respectively, and generating brain feature data for each of the encoded brain image data of the normal person and the brain image data of the dementia patient respectively, and a step of generating a dementia diagnosis model trained to determine whether a specific person has dementia based on the genetic information and genetic feature data of the specific person, using data obtained by combining the genetic feature data and the brain feature data according to a preset method as learning data.
[0009] In addition, an embodiment according to a second aspect of the present disclosure provides a dementia diagnosis 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 to be executed by the processor, wherein the memory stores a code that, when executed through the processor, causes the processor to compare and analyze genetic information of a normal person and genetic information of a dementia patient to generate genetic feature data, encode each of the brain image data of the normal person and the brain image data of the dementia patient, generate brain feature data for each of the encoded brain image data of the normal person and the brain image data of the dementia patient, and generate a dementia diagnosis model trained to determine whether a specific person has dementia based on the genetic information and the genetic feature data of the specific person using data obtained by combining the genetic feature data and the brain feature data according to a preset method as learning data.
[0010] According to the present invention, accuracy can be improved by diagnosing dementia by considering both genetic data and brain image data.
[0011] In addition, according to the present invention, the convenience of examination can be increased by diagnosing dementia based on genetic data and brain image data.
[0012] The effects of the present invention are not limited to the effects described above, and include all effects understood from the following description.
[0013] FIG. 1 is a diagram illustrating a dementia diagnosis system according to one embodiment of the present invention.
[0014] Figure 2 is a drawing showing the detailed configuration of the server illustrated in Figure 1.
[0015] Figures 3 and 4 are diagrams illustrating examples of genetic feature data.
[0016] Figures 5 to 7 are diagrams illustrating examples of extracting brain feature data.
[0017] Figures 8(a) and 8(b) are diagrams illustrating the accuracy of dementia diagnosis results according to the dementia diagnosis model.
[0018] Figure 9 is a flowchart illustrating the sequence of a dementia diagnosis method according to another embodiment of the present invention.
[0019] Hereinafter, the present disclosure will be described in detail with reference to the attached drawings. However, the present disclosure may be implemented in various different forms and is not limited to the embodiments described herein. In addition, the attached drawings are only intended to facilitate understanding of the embodiments disclosed in the present specification, and the technical concepts disclosed in the present specification are not limited by the attached drawings. All terms, including technical and scientific terms, used herein should be interpreted as having meanings generally understood by a person of ordinary skill in the technical field to which the present disclosure pertains. Terms defined in the dictionary should be interpreted as having additional meanings consistent with the relevant technical literature and the present disclosure, and shall not be interpreted in an extremely ideal or restrictive sense unless otherwise defined.
[0020] To clearly explain the present disclosure in the drawings, parts irrelevant to the description have been omitted, and the size, shape, and appearance of each component shown in the drawings may vary. Identical / similar parts throughout the specification are designated with identical / similar drawing reference numerals.
[0021] The suffixes "module" and "part" used in the following description for components are assigned or used interchangeably solely for the convenience of writing the specification, and do not in themselves have distinct meanings or roles. Furthermore, in describing the embodiments disclosed herein, detailed descriptions of related known technologies have been omitted if they are deemed to obscure the gist of the embodiments disclosed herein.
[0022] Throughout the specification, when a part is said to be "connected (connected, in contact with, or coupled)" to another part, this includes not only cases where it is "directly connected (connected, in contact with, or coupled)" but also cases where it is "indirectly connected (connected, in contact with, or coupled)" with another member in between. Furthermore, when a part is said to "include (have or provide)" a certain component, this does not mean that it excludes other components, but rather that it may "include (have or provide)" other components, unless otherwise specifically stated.
[0023] As used herein, ordinal terms such as "first," "second," etc., are used solely to distinguish 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 referred to as the "second component," and similarly, the second component may also be referred to as the "first component." As used herein, singular forms should be construed to include plural forms, unless explicitly stated otherwise.
[0024] FIG. 1 is a diagram illustrating a dementia diagnosis system according to one embodiment of the present invention.
[0025] Referring to FIG. 1, the dementia diagnosis system may include a server (100) and a user terminal (200). The server (100) and the user terminal (200) may be connected to each other through a communication network.
[0026] The server (100) compares and analyzes genetic information of a normal person and genetic information of a dementia patient to generate or extract genetic feature data.
[0027] The server (100) encodes brain image data of a normal person and brain image data of a dementia patient, respectively. The server (100) generates or extracts brain feature data for each of the encoded brain image data of a normal person and brain image data of a dementia patient. For example, the brain image data may be an MRI image of the brain.
[0028] The server (100) uses data obtained by combining genetic feature data and brain feature data according to a preset method as learning data, and generates a dementia diagnosis model trained to determine whether a specific person has dementia based on the genetic information and genetic feature data of the specific person. The server (100) diagnoses dementia based on the input genetic information and brain image data based on the generated dementia diagnosis model.
[0029] The user terminal (200) can receive dementia diagnosis results from the server (100). In addition, the user terminal (200) can transmit genetic information and brain image data to the server (100).
[0030] The user terminal (200) can be connected to the server (100) through a communication network. The user terminal (200) may refer to any type of handheld-based wireless communication device, such as a notebook, desktop, laptop, or wireless communication device with portability and mobility, equipped with a web browser, or a smartphone or tablet PC.
[0031] Figure 2 is a drawing showing the detailed configuration of the server illustrated in Figure 1.
[0032] Referring to FIG. 2, the server (100) may include a communication module (110), a processor (120), and a memory (130).
[0033] The communication module (110) may include a device including hardware and software necessary to transmit and receive signals such as control signals or data signals through wired or wireless connections with other network devices.
[0034] The communication module (110) can receive genetic information and brain image data of normal individuals, patients with mild cognitive impairment, and dementia patients from user terminals. In addition, the communication module (110) can transmit dementia diagnosis results to the user terminal.
[0035] The processor (120) may include various types of devices that control and process data. The processor (120) may refer to a data processing device built into hardware that has a physically structured circuit to perform functions expressed in code or instructions included in a program.
[0036] In one example, the processor (120) may be implemented in the form of a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc., but the scope of the present invention is not limited thereto.
[0037] The processor (120) performs operations according to the code stored in the memory (130).
[0038] The memory (130) can store at least one of information and data input to the communication module (110), information and data required for functions performed by the processor (120), and data generated according to the execution of the processor (120).
[0039] Memory (130) should be interpreted as a general term for a non-volatile storage device that maintains stored information even when no power is supplied and a volatile storage device that requires power to maintain stored information. Memory (130) may include magnetic storage media or flash storage media in addition to a volatile storage device that requires power to maintain stored information, but the scope of the present invention is not limited thereto.
[0040] The memory (130) is electrically connected to the processor (120) and stores at least one code to be executed by the processor (120). The memory (130) stores code that, when executed by the processor (120), causes the processor (120) to perform the following functions and procedures.
[0041] The memory (130) stores a code that causes the generation of genetic feature data by comparing and analyzing genetic information of a normal person and genetic information of a dementia patient. For example, the genetic information may be data on the composition of the microbiota as bacterial abundance data for each sample through analysis of 16s rRNA sequencing data for fecal samples. In addition, the genetic feature data may be a vector generated based on the number of occurrences of each bacteria.
[0042] The memory (130) may store code that causes the genetic information of a normal person to cluster the bacterial information of each dementia patient's genetic information to generate bacterial data. The memory (130) may also store code that causes the generated bacterial data to be converted into numerical data to extract vector-type genetic feature data. For example, the memory (130) may perform ASV clustering and taxonomic annotation on 16s rRNA data based on the stored code.
[0043] For example, the memory (130) may convert ASV data into numerical data based on the stored code, perform one-hot encoding on each ASV sample, and generate a numerical vector representing ASV characteristics for each sample. Here, the ASV may be taxonomic information matched to a species.
[0044] Additionally, the memory (130) may store a code that causes genetic information of a patient with mild cognitive impairment to be compared and analyzed to extract genetic feature data.
[0045] Additionally, the memory (130) may store a code that extracts features for brain image data using CNN and causes brain image data for a normal person, a patient with mild cognitive impairment, and a patient with dementia to be vectorized.
[0046] The memory (130) stores a code that causes the brain image data of a normal person and the brain image data of a dementia patient to be encoded, respectively. For example, the memory (130) may store a code that causes the brain image data of a normal person and the brain image data of a dementia patient to be encoded, region by region. Here, the brain image data may be a brain MRI image, and the memory (130) may store a code that causes the brain MRI image to be converted into brain image data based on a dementia diagnosis model.
[0047] The memory (130) stores codes that cause the generation of brain feature data for each of coded brain image data of a normal person and brain image data of a dementia patient. In addition, the memory (130) may store codes that cause the extraction of brain feature data for brain image data of a patient with mild cognitive impairment.
[0048] For example, the memory (130) may store a code that causes the brain image data of a normal person, the brain image data of a dementia patient, and the brain image data of a patient with mild cognitive impairment to be generated in vector form based on a preset feature detection method.
[0049] In the memory (130), a code is stored that causes a dementia diagnosis model trained to determine whether a specific person has dementia based on the genetic information and genetic feature data of a specific person, using data that combines genetic feature data, brain feature data of a normal person, and brain feature data of a dementia patient according to a preset method as learning data. For example, the memory (130) may store a code that causes a single input vector to be generated by linking the genetic feature data and brain feature data. For example, the memory (130) may link brain feature data acquired from a CNN and converted genetic feature data based on the stored code to generate a single input vector.
[0050] The memory (130) may store code that causes a dementia diagnosis model to be generated by learning the generated input vector. For example, the memory (130) may input the combined input vector into the dementia diagnosis model based on the stored code, and learn and execute the dementia diagnosis model. The memory (130) may produce a prediction value and loss for dementia or normal. Here, the dementia diagnosis model may be a model generated using Keras.
[0051] The memory (130) stores a code that causes a diagnosis of dementia based on input genetic information and brain image data based on a dementia diagnosis model. For example, the memory (130) stores a code that causes a diagnosis of dementia based on input genetic information and brain image data of a specific person using a dementia diagnosis model and determines the condition of the specific person as at least one of mild cognitive impairment, dementia, and normal.
[0052] Figures 3 and 4 are diagrams illustrating examples of genetic feature data.
[0053] Referring to FIGS. 3 and 4 , the genetic feature data may include ASV IDs for each bacterium, taxonomic information data for ASVs, and bacterial information for each sample for samples from normal individuals and dementia patients. For example, the genetic feature data may include information on OUT ID, kingdom, phylum, class, order, family, genus, species, AD01, and C88.
[0054] Gene feature data can be extracted in the form of a vector based on the number of occurrences per ASV, such as AD1=[1,0,0,1] and C88=[0,1,1,0].
[0055] The dementia diagnosis system can convert each ASV into a vector (410) in which each feature has a unique value. For example, if there are four ASVs, the dementia diagnosis system can perform one-hot encoding as follows: ASV1:[1,0,0,0], ASV2:[0,1,0,0], ASV3:[0,0,1,0], and ASV4:[0,0,0,1].
[0056] The dementia diagnosis system can generate transformation data (420) for the samples. For example, the dementia diagnosis system can generate transformation data for sample 1 in which ASV1 exists and sample 2 in which ASVs 2 and 3 exist, such as sample 1:[1,0,0,0], sample 2:[0,1,1,0].
[0057] The dementia diagnosis system can extract genetic feature data (440) in vector form based on the number of occurrences (430) of each ASV. For example, if ASV1 appears 5 times and ASV3 3 times in sample 1, and ASV1 appears 2 times and ASV3 7 times in sample 2, the dementia diagnosis system can extract genetic feature data (440) such as Sample 1: [5,3,0,0] and Sample 2: [2,0,7,0].
[0058] Figures 5 to 7 are diagrams illustrating examples of extracting brain feature data.
[0059] Referring to Figures 5 to 7, the dementia diagnosis system can encode each region of MRI images of a normal person, a dementia patient, and a patient with mild cognitive impairment. The dementia diagnosis system can then combine the region-specific coded brain image data (610) and feature detection data (620) to generate a feature map (710).
[0060] The dementia diagnosis system can extract brain feature data in vector form by repeatedly performing Max pooling and flattening on the feature map (710) to create layers of features of brain image data. The dementia diagnosis system can diagnose dementia by analyzing newly input brain image data and genetic information based on the brain feature data.
[0061] Figures 8(a) and 8(b) are diagrams illustrating the accuracy of dementia diagnosis results according to the dementia diagnosis model.
[0062] Referring to Figure 8(a), a dementia diagnosis model was created by combining MRI images and 16s rRNA data, and the results of the dementia diagnosis showed a loss of 0.203 for the test set and an accuracy of 0.91. At this time, the dementia diagnosis model can be trained for 10 epochs.
[0063] On the other hand, referring to Fig. 8(b), the dementia diagnosis model was created using only 16s rRNA data without MRI images, and the results of diagnosing dementia showed that the loss for the test set was 0.356 and the accuracy was 0.80.
[0064] Accordingly, when a dementia diagnosis model was created by combining MRI images and 16s rRNA data and dementia was diagnosed, it was confirmed that the accuracy increased to 0.91 and the loss was reduced by 0.153 compared to when a dementia diagnosis model was created and dementia was diagnosed using only 16s rRNA data.
[0065] Figure 9 is a flowchart illustrating the sequence of a dementia diagnosis method according to another embodiment of the present invention.
[0066] The dementia diagnosis method described below can be performed by the dementia diagnosis system and server described above with reference to FIGS. 1 to 8. Accordingly, the content of the embodiments of the present disclosure described above with reference to FIGS. 1 to 8 can be equally applied to the embodiments described below, and any content overlapping with the above description will be omitted. The steps described below do not necessarily have to be performed in order, and the order of the steps can be set in various ways, and the steps can be performed almost simultaneously.
[0067] Referring to FIG. 9, the dementia diagnosis method includes a gene feature data extraction step (S100), a brain feature data extraction step (S200), and a dementia diagnosis step (S300) based on the gene feature data and brain feature data.
[0068] The genetic feature data extraction step (S100) is a step for extracting genetic feature data by comparing and analyzing genetic information of normal individuals and dementia patients. For example, in the genetic feature data extraction step (S100), bacterial information from each of the genetic information of normal individuals and dementia patients is clustered to generate bacterial data, and the bacterial data is converted into numerical data to extract vector-type genetic feature data.
[0069] The brain feature data extraction step (S200) is a step of encoding brain image data of a normal person and brain image data of a dementia patient, respectively, and extracting brain feature data for each of the encoded brain image data of a normal person and brain image data of a dementia patient. For example, in the brain feature data extraction step (S200), brain image data of a normal person and brain image data of a dementia patient are encoded by region, and brain feature data in vector form can be extracted based on the encoded brain image data and preset feature detection data.
[0070] The dementia diagnosis step (S300) based on genetic feature data and brain feature data is a step of generating a dementia diagnosis model trained to determine whether a specific person has dementia based on the genetic information and genetic feature data of a specific person, using data that combines genetic feature data and brain feature data according to a preset method as learning data. For example, in the dementia diagnosis step (S300) based on genetic feature data and brain feature data, the genetic feature data and brain feature data are connected to generate a single input vector, the dementia diagnosis model is trained using the input vector, and dementia can be diagnosed for the input genetic information and brain image data based on the dementia diagnosis model.
[0071] Those skilled in the art will appreciate that the present disclosure can be easily modified into other specific forms based on the above description without changing the technical spirit or essential characteristics of the present disclosure. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present disclosure is indicated by the following claims, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present disclosure. The scope of the present application is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present application.
[0072] The form for carrying out the invention is substantially the same as the best form for carrying out the invention.
[0073] The present invention can be used in a technology for diagnosing dementia, and thus has industrial applicability.
Claims
1. In a dementia diagnosis method performed by a dementia diagnosis system, a) A step of generating genetic feature data by comparing and analyzing genetic information of a normal person and genetic information of a dementia patient; b) a step of encoding brain image data of a normal person and brain image data of a dementia patient, respectively, and generating brain feature data for each of the encoded brain image data of the normal person and the brain image data of the dementia patient; and c) A dementia diagnosis method, comprising the step of creating a dementia diagnosis model trained to determine whether a specific person has dementia based on genetic information and genetic characteristic data of the specific person, using data obtained by combining the genetic characteristic data and the brain characteristic data according to a preset method as learning data.
2. In paragraph 1, In step a) above, A dementia diagnosis method comprising: generating bacterial data by clustering bacterial information of the genetic information of the normal person and the genetic information of the dementia patient, and converting the bacterial data into numerical data to generate the genetic feature data in vector form.
3. In paragraph 2, A method for diagnosing dementia, wherein the above genetic feature data is vector data generated based on the number of appearances of each bacteria.
4. In paragraph 1, A method for diagnosing dementia, wherein the genetic information includes 16s rRNA data.
5. In paragraph 1, In step b) above, A dementia diagnosis method comprising: coding the brain image data of the normal person and the brain image data of the dementia patient by region, and generating the brain feature data in vector form based on the coded brain image data and a preset feature detection method.
6. In paragraph 1, In step c) above, A dementia diagnosis method comprising: generating one input vector by connecting the genetic feature data and the brain feature data, and training the dementia diagnosis model based on the input vector.
7. In paragraph 1, The above step a) includes a step of generating the genetic feature data by comparing and analyzing the genetic information of a patient with mild cognitive impairment and the genetic information of a normal person, A method for diagnosing dementia, wherein step b) comprises a step of generating the brain feature data by comparing and analyzing brain image data of a patient with mild cognitive impairment with brain image data of a normal person.
8. In paragraph 1, Step c) above, A dementia diagnosis method, comprising a step of inputting the genetic information and brain image data of a specific person using the dementia diagnosis model and determining the condition of the specific person as at least one of mild cognitive impairment, dementia, and normal.
9. Communication module; at least one processor; and A memory electrically connected to the processor and storing at least one code to be executed by the processor, A dementia diagnosis system, wherein the above memory stores a code that causes the processor, when executed through the processor, to compare and analyze genetic information of a normal person and genetic information of a dementia patient to generate genetic feature data, to encode each of the brain image data of the normal person and the brain image data of the dementia patient, to generate brain feature data for each of the encoded brain image data of the normal person and the brain image data of the dementia patient, and to generate a dementia diagnosis model trained to determine whether a specific person has dementia based on the genetic information and genetic feature data of the specific person using data obtained by combining the genetic feature data and the brain feature data according to a preset method as learning data.
10. In paragraph 9, The above memory causes the processor to: A dementia diagnosis system, wherein a code is stored that causes the genetic information of the normal person and the genetic information of the dementia patient to cluster bacterial information to generate bacterial data, converts the bacterial data into numerical data, and extracts the genetic feature data in vector form.
11. In paragraph 9, A dementia diagnosis system wherein the above genetic feature data is vector data generated based on the number of appearances per bacteria.
12. In paragraph 9, The above memory causes the processor to: A dementia diagnosis system, wherein a code is stored that causes the brain image data of the normal person and the brain image data of the dementia patient to be coded by region, and the brain feature data in vector form to be extracted based on the coded brain image data and preset feature detection data.
13. In paragraph 9, The above memory causes the processor to: A dementia diagnosis system, which connects the genetic feature data and the brain feature data to generate one input vector, trains the dementia diagnosis model using the input vector, and stores a code that causes the dementia to be diagnosed based on the dementia diagnosis model.
14. In paragraph 9, The above memory causes the processor to: A dementia diagnosis system, wherein a code is stored that causes the genetic characteristic data to be generated by comparing and analyzing the genetic information of a patient with mild cognitive impairment and the genetic information of a normal person, and the brain characteristic data to be generated by comparing and analyzing the brain image data of the patient with mild cognitive impairment and the brain image data of a normal person.
15. In paragraph 9, The above memory causes the processor to: A dementia diagnosis system, which stores a code that causes a specific person's genetic information and brain image data to be input using the above dementia diagnosis model and to determine the specific person's condition as at least one of mild cognitive impairment, dementia, and normal.
Citation Information
Patent Citations
Information processing device, program, trained model, diagnostic support device, learning device, and prediction model generation method
JP7170145B2
Method and apparatus for deciding Alzheimer''s disease based on machine learning
KR1020180078033A
Battery device
KR1020240077888A
KR20200066578A
