WO2021225226A1 - 알츠하이머 진단 장치 및 방법 - Google Patents
알츠하이머 진단 장치 및 방법 Download PDFInfo
- Publication number
- WO2021225226A1 WO2021225226A1 PCT/KR2020/011036 KR2020011036W WO2021225226A1 WO 2021225226 A1 WO2021225226 A1 WO 2021225226A1 KR 2020011036 W KR2020011036 W KR 2020011036W WO 2021225226 A1 WO2021225226 A1 WO 2021225226A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- alzheimer
- amyloid
- beta
- image
- tau protein
- Prior art date
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 76
- 238000000034 method Methods 0.000 title claims description 73
- 102000013455 Amyloid beta-Peptides Human genes 0.000 claims abstract description 93
- 108010090849 Amyloid beta-Peptides Proteins 0.000 claims abstract description 93
- 102000013498 tau Proteins Human genes 0.000 claims abstract description 67
- 108010026424 tau Proteins Proteins 0.000 claims abstract description 67
- 230000014509 gene expression Effects 0.000 claims abstract description 49
- 210000001525 retina Anatomy 0.000 claims abstract description 48
- 208000024827 Alzheimer disease Diseases 0.000 claims description 82
- 230000013016 learning Effects 0.000 claims description 50
- 230000004256 retinal image Effects 0.000 claims description 39
- 230000002207 retinal effect Effects 0.000 claims description 37
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 30
- 201000010099 disease Diseases 0.000 claims description 28
- 208000010877 cognitive disease Diseases 0.000 claims description 24
- 208000027061 mild cognitive impairment Diseases 0.000 claims description 24
- 108090000765 processed proteins & peptides Proteins 0.000 claims description 23
- 208000037259 Amyloid Plaque Diseases 0.000 claims description 20
- 230000002787 reinforcement Effects 0.000 claims description 18
- 238000013473 artificial intelligence Methods 0.000 claims description 16
- 230000008499 blood brain barrier function Effects 0.000 claims description 11
- 210000001218 blood-brain barrier Anatomy 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000003384 imaging method Methods 0.000 claims description 10
- 238000004040 coloring Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- VFLDPWHFBUODDF-FCXRPNKRSA-N curcumin Chemical compound C1=C(O)C(OC)=CC(\C=C\C(=O)CC(=O)\C=C\C=2C=C(OC)C(O)=CC=2)=C1 VFLDPWHFBUODDF-FCXRPNKRSA-N 0.000 claims description 6
- 210000001808 exosome Anatomy 0.000 claims description 5
- 230000019612 pigmentation Effects 0.000 claims description 5
- 238000001228 spectrum Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- -1 theracumin Chemical compound 0.000 claims description 4
- AZOPGDOIOXKJRA-UHFFFAOYSA-L chembl1817788 Chemical compound [Na+].[Na+].C1=C(C([O-])=O)C(O)=CC=C1N=NC1=CC=C(C=2C=CC(=CC=2)N=NC=2C=C(C(O)=CC=2)C([O-])=O)C=C1 AZOPGDOIOXKJRA-UHFFFAOYSA-L 0.000 claims description 3
- IQFVPQOLBLOTPF-HKXUKFGYSA-L congo red Chemical compound [Na+].[Na+].C1=CC=CC2=C(N)C(/N=N/C3=CC=C(C=C3)C3=CC=C(C=C3)/N=N/C3=C(C4=CC=CC=C4C(=C3)S([O-])(=O)=O)N)=CC(S([O-])(=O)=O)=C21 IQFVPQOLBLOTPF-HKXUKFGYSA-L 0.000 claims description 3
- 229940109262 curcumin Drugs 0.000 claims description 3
- 235000012754 curcumin Nutrition 0.000 claims description 3
- 239000004148 curcumin Substances 0.000 claims description 3
- VFLDPWHFBUODDF-UHFFFAOYSA-N diferuloylmethane Natural products C1=C(O)C(OC)=CC(C=CC(=O)CC(=O)C=CC=2C=C(OC)C(O)=CC=2)=C1 VFLDPWHFBUODDF-UHFFFAOYSA-N 0.000 claims description 3
- 230000000873 masking effect Effects 0.000 claims 2
- 230000001939 inductive effect Effects 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 16
- 210000004556 brain Anatomy 0.000 description 15
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 102000004169 proteins and genes Human genes 0.000 description 9
- 108090000623 proteins and genes Proteins 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000012014 optical coherence tomography Methods 0.000 description 8
- 238000002600 positron emission tomography Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000009471 action Effects 0.000 description 6
- 230000035945 sensitivity Effects 0.000 description 6
- 208000003098 Ganglion Cysts Diseases 0.000 description 4
- 208000005400 Synovial Cyst Diseases 0.000 description 4
- 238000009825 accumulation Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 4
- 206010012289 Dementia Diseases 0.000 description 3
- 210000004027 cell Anatomy 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000002347 injection Methods 0.000 description 3
- 239000007924 injection Substances 0.000 description 3
- 231100000331 toxic Toxicity 0.000 description 3
- 230000002588 toxic effect Effects 0.000 description 3
- 208000000044 Amnesia Diseases 0.000 description 2
- 206010070834 Sensitisation Diseases 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000012098 association analyses Methods 0.000 description 2
- 239000000090 biomarker Substances 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000012377 drug delivery Methods 0.000 description 2
- 238000013399 early diagnosis Methods 0.000 description 2
- 238000001506 fluorescence spectroscopy Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004254 retinal expression Effects 0.000 description 2
- 239000000523 sample Substances 0.000 description 2
- 230000008313 sensitization Effects 0.000 description 2
- 238000010186 staining Methods 0.000 description 2
- 206010061818 Disease progression Diseases 0.000 description 1
- 208000026139 Memory disease Diseases 0.000 description 1
- 229910019142 PO4 Inorganic materials 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000001574 biopsy Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000024245 cell differentiation Effects 0.000 description 1
- 210000004720 cerebrum Anatomy 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 230000003920 cognitive function Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 239000002872 contrast media Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 238000012631 diagnostic technique Methods 0.000 description 1
- 230000005750 disease progression Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010195 expression analysis Methods 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000003446 ligand Substances 0.000 description 1
- 231100000863 loss of memory Toxicity 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000006984 memory degeneration Effects 0.000 description 1
- 208000023060 memory loss Diseases 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002105 nanoparticle Substances 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
- 210000001328 optic nerve Anatomy 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- NBIIXXVUZAFLBC-UHFFFAOYSA-K phosphate Chemical compound [O-]P([O-])([O-])=O NBIIXXVUZAFLBC-UHFFFAOYSA-K 0.000 description 1
- 239000010452 phosphate Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002250 progressing effect Effects 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 230000010344 pupil dilation Effects 0.000 description 1
- 239000000376 reactant Substances 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 210000000225 synapse Anatomy 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
- A61B5/4088—Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/0016—Operational features thereof
- A61B3/0025—Operational features thereof characterised by electronic signal processing, e.g. eye models
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/14—Arrangements specially adapted for eye photography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0071—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence by measuring fluorescence emission
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2560/00—Constructional details of operational features of apparatus; Accessories for medical measuring apparatus
- A61B2560/02—Operational features
- A61B2560/0204—Operational features of power management
- A61B2560/0209—Operational features of power management adapted for power saving
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10064—Fluorescence image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Definitions
- the present invention relates to an apparatus and method for diagnosing Alzheimer's disease, and more particularly, to identify beta-amyloid and plaque mass accumulated in the brain, and derivatives bound to tau protein in the retinal layer, and based on data learned by artificial intelligence technology
- the present invention relates to an apparatus and method for diagnosing Alzheimer's disease, which can efficiently diagnose the cycle stage (current progress) of Alzheimer's.
- Alzheimer's disease is the accumulation of harmful protein residues called beta-amyloid and tau protein in the brain, and the neurons (nerve cells) of the cerebrum involved in learning and memory die, resulting in loss of memory, computational power, language ability, temporal and spatial comprehension, And it is a dementia disease in which thinking skills such as judgment are gradually decreased.
- dementia refers to a state in which brain neurons die or function poorly, resulting in a change in personality as well as memory loss, and the loss of thinking and behavioral abilities, which has reached the point where it interferes with daily life and activities.
- the most common cause is Alzheimer's disease.
- a method for diagnosing Alzheimer's disease and Alzheimer's dementia there is a method of diagnosing it using equipment such as CT, MRI, and PET.
- CT, MRI, PET, etc. take a long time to use due to the characteristics of the equipment and are very expensive, and parts with metal prostheses, head prostheses, etc. are inserted frequently affect the formation of magnetic fields, which may affect the measurement results. have.
- AD Alzheimer's disease
- the retina which is the only optic nerve reactant connected to the brain, is directly connected to the brain during the cell differentiation stage of the human body, derivatives using ligands can be identified by expressing expression in the brain and retina.
- the state-of-the-art equipment that is currently developed and used is not a target for professionally targeting Alzheimer's, but has a public nature, so a method for professionally and quickly identifying Alzheimer's is required.
- An embodiment of the present invention provides a device for identifying beta-amyloid and plaque mass accumulated in the brain, and a derivative bound to tau protein in the retinal layer, and the Alzheimer's cycle stage ( Current progress) can be diagnosed early, and by analyzing image images related to mild cognitive impairment and Alzheimer's dementia expressed in the retina, and detecting results related to geriatric diseases, It is an object of the present invention to provide a device and method for diagnosing Alzheimer's disease that can diagnose the disease, provide information on geriatric diseases using artificial intelligence technology, and continuously update the diagnostic model, thereby accurately and quickly diagnosing the disease.
- an Alzheimer's diagnosis apparatus includes a camera for photographing the retina of a user injected with the derivative, a display for displaying a retina image photographed through the camera, and the photographed through the camera. It may include a diagnostic control unit that analyzes the retinal image, detects the expression of beta-amyloid and tau protein in the retinal image, and outputs Alzheimer's diagnosis information corresponding to the detected expression to the display.
- the camera may photograph the left eye and the right eye of the user, respectively.
- the diagnosis control unit may include an artificial intelligence learning module.
- the diagnostic controller may recognize the proximity of the user through a proximity sensor provided in the camera, and may switch the operation of the camera to a power saving mode if the user is not recognized.
- the diagnostic control unit diagnoses Alzheimer's disease by performing a peptide-based derivative imaging test that has passed through the blood-brain-barrier (BBB), and after the pigmentation of the peptide in the retinal layer is made, the beta-amyloid spectrum and the retina Check the expression value from my specific wavelength band within the retinal layer section, determine the location of a single beta-amyloid plaque by examining the image colored in beta-amyloid, and when the location of the single beta-amyloid plaque is determined, multi-dimensional complex beta-amyloid Estimate the expression level for each frame to estimate the plaque position, determine the final expressed beta-amyloid plaque to estimate the distribution position of beta-amyloid, and after pigmentation of the peptide in the retinal layer, tau protein to check the expression of tau protein Check the spectrum and retinal layer section, examine the image colored by the tau protein to automatically calculate the accumulated tau protein volume before progressing to Alzheimer's, calculate the tau protein volume, and then based on the beta amyloid index level Mild cognitive impairment level can
- the diagnostic control unit generates a beta amyloid segmented image from the retinal image to be diagnosed when the expressed retinal image is input after coloring of the peptide in the retinal layer, and the distribution of beta amyloid in the retina in the segmented image location can be calculated.
- the diagnostic control unit when the retinal image expressed after the peptide coloring in the retinal layer is input, is beta-amyloid learning data or tau protein that has previously learned the position, distribution, and cycle pattern for tracing beta-amyloid expression.
- a reinforcement learning algorithm is performed using tau protein learning data that has previously learned a pattern corresponding to a volume and a position, so that a diagnosis operation can be performed by automatically discriminating a region to be diagnosed.
- the diagnosis control unit masks a disease-related region in the retinal image, converts the masked retinal image into a 3D image including a depth value, and analyzes the converted 3D image to be associated with Alzheimer's disease. pattern can be detected.
- the diagnosis controller may diagnose mild cognitive impairment when the level of mild cognitive impairment is below a certain positive result value, and diagnose Alzheimer's when the level of mild cognitive impairment exceeds a predetermined positive result value.
- the derivative is expressed in the near-infrared region, and may be used by combining any one or more of curcumin, theracumin, Congo red, thioflavin S-T, or chrysamine with an exosome.
- Alzheimer's disease diagnosis method is a photographing step of photographing the retina of a user injected with the derivative using a camera, a diagnostic control unit analyzes the retinal image photographed through the camera, and in the retinal image It may include a detection step of detecting the expression of beta-amyloid and tau protein, and a diagnostic step of outputting Alzheimer's diagnosis information corresponding to the sensed expression to a display.
- the photographing step may include photographing the left eye and the right eye of the user, respectively.
- the photographing step may include recognizing the proximity of the user through a proximity sensor provided in the camera, and switching the operation of the camera to a power saving mode if the user is not recognized.
- the detecting step is to diagnose Alzheimer's disease by performing a peptide-based derivative imaging test that has passed through the BBB (Blood-Brain-Barrier). Checking the expression value from my specific wavelength band within the retinal layer section, examining the image colored on beta-amyloid to determine the location of a single beta-amyloid plaque, When the location of the single beta-amyloid plaque is determined, multi-dimensional complex beta To calculate the expression level for each frame to estimate the position of the amyloid plaque, to determine the final expressed beta-amyloid plaque to determine the distribution position of the beta-amyloid, After the peptide coloring in the retinal layer is made, to check the expression of the tau protein After checking the tau protein spectrum and retinal layer section, and automatically calculating the tau protein volume accumulated before Alzheimer's by examining the image colored on the tau protein, and after calculating the tau protein volume, the beta amyloid index level It may include calculating the mild cognitive impairment level based on the.
- a beta amyloid segmented image is generated from the retinal image to be diagnosed, and beta amyloid distribution in the retina in the segmented image It may include calculating a location.
- the sensing step when the retinal image expressed after the peptide coloring in the retinal layer is input, the beta-amyloid learning data or tau protein learned in advance for the position, distribution, and cycle pattern for tracking beta-amyloid expression. It may include the step of automatically discriminating a region to be diagnosed by performing a reinforcement learning algorithm using tau protein learning data that has previously learned a pattern corresponding to a volume and a position for expression tracking, and performing a diagnostic operation.
- the detecting step masks a disease-related region in the retinal image, converts the masked retinal image into a 3D image including a depth value, and analyzes the converted 3D image to be associated with Alzheimer's. It may include detecting a pattern.
- the diagnosing step may include diagnosing mild cognitive impairment if the mild cognitive impairment level is below a certain positive result value, and diagnosing Alzheimer's if the mild cognitive impairment level exceeds a predetermined positive result value.
- Alzheimer's cycle phase based on data learned by a device for identifying beta-amyloid and plaque mass accumulated in the brain, and derivatives bound to tau protein in the retinal layer, and deep learning-type artificial intelligence technology.
- current status can be diagnosed early, and by analyzing the image images related to mild cognitive impairment and Alzheimer's dementia expressed in the retina and detecting the results related to geriatric diseases, Alzheimer's disease can be cured in a harmless way Early diagnosis is possible, and by providing information on geriatric diseases using artificial intelligence technology and continuously updating the diagnosis model, there is an effect of accurately and rapidly diagnosing diseases.
- FIG. 1 is a diagram illustrating an appearance of an apparatus for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIG. 2 is a block diagram illustrating the configuration of an apparatus for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIG. 3 is a flowchart illustrating a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIG. 4 is a view for explaining a process of determining the location of a plaque mass in the retina and measuring progress in the Alzheimer's disease diagnosis method according to an embodiment of the present invention.
- FIG. 5 is a diagram for explaining a learning model used in a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIG. 6 is a flowchart for explaining a process of diagnosing Alzheimer's disease by analyzing a preprocessed image in the Alzheimer's disease diagnosis method according to an embodiment of the present invention.
- FIG. 7 is a diagram illustrating an ROC curve of a reinforcement learning model in a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIG. 8 is a view for explaining an algorithm for analyzing a data stream in a retinal node in the Alzheimer's disease diagnosis method according to an embodiment of the present invention.
- FIGS. 9 and 10 are diagrams taken by an AD diagnosis apparatus in a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIG. 1 is a view showing the appearance of an apparatus for diagnosing Alzheimer's disease according to an embodiment of the present invention
- FIG. 2 is a block diagram showing the configuration of an apparatus for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- the Alzheimer's disease diagnosis apparatus 100 is configured to photograph the retina of a user such as a normal person or a patient, and includes a proximity sensor 110 , a camera 120 , and a selection input unit. 130 , the positioning unit 140 , the display 150 , the lighting unit 160 , the chin rest unit 170 , the communication unit 180 , and the diagnosis control unit 190 may be included.
- the proximity sensor 110 may detect the proximity of the user when the user approaches the camera 120 to photograph the retina, and may be configured at a position adjacent to the camera 120 .
- the camera 120 is configured to photograph the user's retina, and may be configured to photograph the user's left eye or right eye once.
- the camera 120 is small and includes a color or monochrome CCD, CMOS, etc. sensor, a multi/hyper-spectral image sensor can be added, and a lens, iris and filter can be configured internally.
- the aperture and filter transmit the light reflected from the retina to the lens and sensor, and the filter can be used in the form of improving infrared contrast.
- the lens inside the camera 120 for retinal imaging can be focused automatically or manually through the image output to the display 150 to find an optimal element for close-up photography.
- the lighting unit 160 may be configured to emit light for photographing by the camera 120 , and the brightness and direction of the lighting unit 160 may be adjusted in various forms to obtain an image suitable for diagnosis.
- the problem due to the lack of pupil dilation during retinal imaging is that an image can be output using the lighting unit 160, which prevents the pupil from being narrowed by lighting when taking an image to obtain an image.
- the chin rest 170 adjusts the position of the user's face so that the retina is located in the camera 120 shooting range. It can be connected to the part, and the position can be adjusted by moving up, down, left, and right according to the operation of the position adjusting unit 140 .
- the selection input unit 130 may be configured to receive a selection input such as an operation command from an examinee such as a doctor or a nurse.
- the position adjusting unit 140 may adjust the position of the chin rest unit 170 so that the user's retina is located in the shooting range of the camera 210 , and the brightness of the camera 120 by adjusting the position of the position adjusting unit 140 .
- the brightness of the lighting unit 160 in charge of may be automatically adjusted.
- the position control unit 140 may operate in the form of a lever for inputting a direction
- the selection input unit 130 may include a key or button, and the selection input unit 130 and the position control unit 140 . ), the position of the camera 120 and the direction and brightness of the lighting unit 160 can be controlled.
- the display 150 is configured in various forms such as LCD or LED to output visual information, and is configured to output an image photographed by the camera 120, an image processed through the diagnosis control unit 190, and a diagnosis result. and may include a touch screen.
- the communication unit 180 may transmit/receive data to and from an external server through various wired or wireless communication methods such as Wi-Fi, 3G/4G, and Bluetooth, and may wirelessly connect within a short distance to an external output device such as a printer.
- Wi-Fi Wireless Fidelity
- 3G/4G Fifth Generation
- Bluetooth Wireless Fidelity
- the diagnostic control unit 190 may include various hardware components employed in computing devices such as communication circuits and memory, and may include a proximity sensor 110 , a camera 120 , an input unit 130 , a position control unit 140 , and a display. 150 , the lighting unit 160 , the chin rest unit 170 , and the communication unit 180 may be electrically connected to perform various data processing and calculations.
- the diagnosis control unit 190 may include an artificial intelligence learning (reinforcement learning) module, and the artificial intelligence learning module may be implemented in the above-described hardware configuration, or separate hardware (results are transmitted to the server and processed in the server). ) can also be configured.
- artificial intelligence learning reinforcement learning
- the diagnosis controller 190 may acquire and analyze a retinal image of the user using the camera 120 for Alzheimer's diagnosis, and may provide the diagnosis result to the display 150 or an external server.
- the diagnostic control unit 190 may recognize the proximity of the user to the camera 120 using the proximity sensor 110 when the camera 120 is photographed, and when the user is not close, including the camera 120 , Alzheimer's disease Power consumption may be reduced by switching the diagnostic apparatus 100 to the automatic power saving mode.
- the diagnostic control unit 190 may control at least some of the direction or brightness of the lighting unit 160 when the camera 120 is photographed to obtain a retina image suitable for analysis, and may be combined with an artificial intelligence module for artificial intelligence learning and analysis. Diagnosis function can be performed through the connection of
- the diagnostic control unit 190 analyzes the beta-amyloid plaque image and the tau protein plaque image expressed in the retinal ganglion of the user in the photographed retinal image, and generates data by imaging it, thereby detecting a disease-related pattern in the expressed retinal image.
- the diagnosis control unit 190 may acquire a plurality of images using the camera 120 , and may acquire a retina image by synthesizing the plurality of images, and an image obtained after performing a plurality of images along the frame flow of the image The back can be synthesized to determine the location of the plaque within the retinal ganglion.
- the diagnosis controller 190 may detect a disease-related pattern by converting a two-dimensional retinal image into a three-dimensional image including a depth value and analyzing the converted three-dimensional image in order to derive a more accurate analysis result.
- the diagnosis controller 190 may mask a disease-related region in the retinal image in order to reduce the amount of computation for 3D transformation, and may convert the masked region into a 3D image.
- the diagnostic control unit 190 uses the preprocessed 3D image as a reinforcement learning input value for an analysis model of the retina image, so that the accuracy of analysis may be improved.
- the diagnostic control unit 190 may analyze the retinal image using the PET information control and the retina image learning model that are related to Alzheimer's, and the patterns included in the retinal image and the pre-stored patterns (learned patterns) , and if the pattern included in the retinal image corresponds to a pre-stored pattern, the pattern may be detected as a disease-related pattern.
- the diagnostic control unit 190 may detect a pattern, volume, and distribution associated with a portion of beta-amyloid and tau proteins that progress to mild cognitive impairment (MCI) or Alzheimer's in the retinal expression image.
- MCI mild cognitive impairment
- the diagnostic control unit 190 performs OCT (Optical Coherence Tomography) in a specific wavelength band in which beta-amyloid or amyloid-containing plaques in the retina can be seen, and at least some of the patterns of spectral components corresponding to amyloid in the retinal layer. can detect OCT (Optical Coherence Tomography) in a specific wavelength band in which beta-amyloid or amyloid-containing plaques in the retina can be seen, and at least some of the patterns of spectral components corresponding to amyloid in the retinal layer. can detect OCT (Optical Coherence Tomography) in a specific wavelength band in which beta-amyloid or amyloid-containing plaques in the retina can be seen, and at least some of the patterns of spectral components corresponding to amyloid in the retinal layer. can detect OCT (Optical Coherence Tomography) in a specific wavelength band in which beta-amyloid or amyloid-containing plaques in the retina can be seen, and at least some of the patterns of spectral components
- the diagnostic control unit 190 may detect a portion of the accumulation level (amount) within 10 years before the onset of a disease leading to Alzheimer's by obtaining the tau protein volume from the retinal expression image.
- the diagnosis controller 190 may output disease information corresponding to the detected pattern using the display 150 .
- the disease information may include, for example, information related to an observation about a disease, a disease progression, or a part of specificity and sensitivity by artificial intelligence.
- the diagnostic control unit 190 may output a retinal image together with disease information using the display 150 , and outputs a retinal image including a mark for identifying a region in which a pattern associated with a disease is detected for user convenience. can do.
- the diagnosis control unit 190 may transmit a retina image to an external server using the communication unit 180 , and the retina image provided to the server may be used as training data for an analysis model of the retina image.
- the diagnosis control unit 190 can receive an analysis model of the retina image from the server using the communication unit 180, and provides learning data for the analysis model to the server, and updates the analysis model provided from the server. Accuracy can be improved.
- FIG. 3 is a flowchart illustrating a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- FIGS. 1 and 2 perform the process of FIG. 3 .
- a process of injecting a peptide-based derivative to a user by oral administration or injection is first performed.
- Alzheimer's disease is known to be caused by excessive increase of beta-amyloid protein in the brain, and when the concentration of beta-amyloid increases, nerve cells in the brain are destroyed and memory is eventually erased.
- beta-amyloid when diagnosing Alzheimer's disease by brain biopsy or positron emission tomography (PET) that can confirm protein distribution, beta-amyloid can be used as a major measure of AD diagnosis, that is, as a biomarker.
- beta-amyloid protein accumulates in the body and cuts synapses in the brain, so derivatives are used to measure the expression of beta-amyloid and tau protein early.
- FIG. 3 after administration of the derivative, it is a flowchart including the step of measuring the expression (light-up) in the brain and retina.
- BBB blood brain barrier
- the low water solubility means that sufficient elements for light-up may not be obtained even when a certain amount is eaten or used as an injection, and in some cases, curcumin, theracumin, Congo red, thioflavin ST, chrysamine, etc. can be labeled by linking (binding) with exosomes.
- Exosome refers to a drug delivery platform technology and plays an important role in transferring information between cells as nanoparticles with a size of 60 to 100 nm secreted by cells.
- the exosome is used as a drug delivery system and a derivative is loaded thereto and used as a diagnostic reagent.
- a diagnostic test is performed by performing a peptide-based derivative imaging test that penetrates the BBB, which starts through peptide sensitization, Precisely, sensitization can be used to identify plaques in multiple retinal layers (within multiple phosphate layers of OCT) in a retinal image.
- the increased/decreased discrete values can be compared with previous data to predict where the peptide is colored by first predicting the level based on the recognition of plaque in the retinal layer.
- amyloid spectral signature of the fluorescence dataset can check the expression value from a specific wavelength band in the retina within the retinal layer section and examine the image colored with beta-amyloid.
- the fluorescence data set learned in advance by the artificial intelligence learning method may be loaded into the memory, and data set classification/comparison may be performed by the artificial intelligence learning classification method.
- This method can perform an algorithm to determine the location of plaque masses in the retina in combination with OCT while isolating the light-up values (to avoid confusion with other cell expression), and this process is a single beta-amyloid. It can be a criterion for determining the location of plaque.
- the value to be calculated is based on the value obtained by the repeated reinforcement learning operation of the learning layer learned by artificial intelligence learning. It is possible to estimate a quantity, or a distribution position.
- an attempt for tau protein analysis may be sequentially made, or may be made simultaneously with beta-amyloid expression (light-up) analysis.
- tau protein spectral signature and retinal layer section of the fluorescence data set can be checked to confirm tau protein expression (light-up).
- the identified section may include the case of varying the wavelength of the OCT system to obtain the tau protein volume, where the tunable case consists of a near-infrared region similar to the beta amyloid wavelength band and the tau protein wavelength band, but the characteristics of the two toxic proteins It is possible to change the wavelength band as it includes a part to optimize for it.
- the changed wavelength band it is possible to automatically calculate the accumulated volume from 10 years before the progression to Alzheimer's in the pigmented ganglion (Nervous system) of tau protein expression (light-up). By transforming from 2D to 3D, it can be used to obtain a three-dimensional volume.
- the amount of volume can be calculated using data previously learned by artificial intelligence using tau protein expression comparison data.
- the current diagnosis of Alzheimer's disease using beta-amyloid may include some opinions on the results.
- the AD diagnosis apparatus 100 since beta-amyloid toxic protein is detected even in a normal person, the AD diagnosis apparatus 100 according to the present invention takes into consideration both the beta-amyloid and the secondary-generated tau protein index generated as a primary result to derive a result based on the doctor's opinion. The accuracy of diagnosis can be greatly improved compared to existing equipment.
- the level of mild cognitive impairment may be calculated based on the previously calculated beta-amyloid index level.
- Amnestic mild cognitive impairment which is a pre-dementia stage, is a mild cognitive impairment if the values of the beta-amyloid index level and the tau protein index level do not exceed the positive result value based on the doctor's opinion, and the result is a positive result value Beyond this, you can create an Alzheimer's index level.
- FIG. 4 is a view for explaining a process of determining the location of a plaque mass in the retina and measuring progress in the Alzheimer's disease diagnosis method according to an embodiment of the present invention.
- an image entering the OCT area may be a single data or a stream of multiple data. may be
- Model-1 can be used to segment specific proteins (beta-amyloid, tau) by applying the minimum critical technique to the image. It can be used because you can see the characteristics of mixed features and invisible parts.
- the segmented protein image can be used as an input to Model-2.
- Model-2 refers to a post-processing step, in which an algorithm (refer to FIG. 3) to determine the location of a plaque mass in the retina is driven, and the amount or distribution of beta-amyloid can be determined after this step.
- an algorithm (refer to FIG. 3) to determine the location of a plaque mass in the retina is driven, and the amount or distribution of beta-amyloid can be determined after this step.
- Model-1 The reason for dividing Model-1 and Model-2 is to use them separately because the preprocessing of Model-1 is related to the imaging equipment or methodology in the OCT system.
- the minimum critical technique can be turned on/off, and in some cases, the image layer can be created by distinguishing the OCT retinal layer. This is because the image segmentation step may be reduced accordingly.
- beta-amyloid and tau protein spectral signatures can be obtained through spectral analysis imaging and image processing, and beta-amyloid index generation and tau protein index generation are two signals generated by expression as a derivative (contrast agent). They can be used separately in a way to obtain a final result by combining them.
- FIG. 5 is a diagram for explaining a learning model used in a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- the reinforcement learning model may be used for clustering, association analysis, and communication unit analysis.
- a reinforcement learning model is a method in which an agent defined in an environment recognizes the current state and selects an action or action sequence that maximizes a reward among selectable actions.
- An action can be determined by referring to the meaning of the association analysis.
- Information representing the current image domain situation is defined as a state, and the environment can be configured to compose a layer (a learning stage - a discrimination layer), and the agent receives a reward. can be derived as a result.
- reinforcement learning can be said to be a very useful tool in analyzing medical images, and when the above-described process is repeated, a reward generated as an output forms a virtuous cycle structure by feedback and can be recycled as an input.
- FIG. 6 is a flowchart for explaining a process of diagnosing Alzheimer's disease by analyzing a preprocessed image in the Alzheimer's disease diagnosis method according to an embodiment of the present invention.
- the AD diagnosis apparatus 100 finds a semantically related part of an image in a learned model, an indexing step of grouping and classifying similar items, and high-speed detection of protein patterns corresponding to diseases in similar items A step, a step of applying a backtracking technique to backtrack the accuracy of the primarily detected pattern, and a post-processing step of feeding back the input from the probabilistic conclusion may be performed.
- the AD diagnosis apparatus 100 may perform a reinforcement learning algorithm in step 602 .
- the AD diagnosis apparatus 100 may perform a reinforcement learning algorithm using the learning data including the beta-amyloid learning data and the tau protein learning data in step 603 , and generate a derivative diagnosis domain learning model in step 604 . .
- the AD diagnosis apparatus 100 may generate a model (eg, a reinforcement learning model) by pre-learning a diagnosis region of a retinal image in advance to perform a diagnosis operation by automatically discriminating a region to be diagnosed.
- a model eg, a reinforcement learning model
- the learning model can be created based on the derivative image for learning and the image of the diagnostic site, and the position, distribution, and cycle pattern are learned in advance for tracing beta-amyloid expression, and the pattern corresponding to the volume and position is generated for tracing the expression of tau protein. can learn
- the AD diagnosis apparatus 100 may classify (index) the semantic associations of the retina images.
- the agent constituting the state becomes the analysis data image and derives the reward as a result, and when the data of the pre-trained model is not enough, the Q-Value is calculated based on the result (reward) by the decision (action). However, even when a certain probability is reached, Q-Value and pixel classification can be derived.
- the AD diagnosis apparatus 100 may detect beta-amyloid deposits in retinal ganglion tissue, a toxic protein pattern generated by Alzheimer's disease, and an expression level of tau protein using the indexing result.
- the AD diagnosis apparatus 100 attempts to analyze the pattern by stacking the pattern layers, for example, 200 layers stacking may be attempted.
- the pattern layer of 150 times showed a result of about 0.576%, which is half of the Q-value used in the ROC (receiver operating characteristic) curve.
- the AD diagnosis apparatus 100 can apply the pattern accuracy backtracking technique, which is performed to add reliability to the result after the pattern analysis is primarily completed.
- the accuracy of the detected pattern is improved. and backtracking for verification.
- the sensitivity is the rate at which a person with a disease is judged to have a “disease”
- Specificity may be the rate at which a person who does not have the disease is judged to be "disease-free" when tested.
- the AD diagnosis apparatus 100 may update the model through feedback, for example, the AD diagnosis apparatus 100 may reuse a Q-value derived as a final result as an input.
- the AD diagnosis apparatus 100 may display the derivative expression analysis diagnosis result and image, transmit the two expression data analyzed in step 611 to the server, and in step 612 use a printer to print the results. can be printed out.
- FIG. 7 is a diagram illustrating an ROC curve of a reinforcement learning model in a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- the graph may include a curve in which sensitivity and specificity are visualized simultaneously, and the performance of a diagnostic technique may be quantified by calculating an area under the curve (AUC).
- AUC area under the curve
- the AUC value is 1, it is a 100% perfect judgment method, and as a result of analyzing the expression patterns of the two current beta-amyloid and tau proteins, the result was derived as a value exceeding 0.9%.
- tau protein is accumulated in an accurate amount every year from 10 years before the onset of AD. Therefore, it relates to a method for automatically obtaining its volume, and to a method for measuring the expression distribution of beta-amyloid.
- Results of comparing with PET-based data by finding a cycle (repetition over time) based on beta-amyloid computational power, finding a pattern (distribution position that is constantly generated according to the flow of space), and calculating sensitivity and specificity can be rough
- FIGS. 9 and 10 are diagrams taken by an AD diagnosis apparatus in a method for diagnosing Alzheimer's disease according to an embodiment of the present invention.
- tau protein in normal people and AD can be clearly distinguished, and according to this result, it can be used as an auxiliary tool for diagnosis in determining the opinion of a doctor.
- amnestic mild cognitive impairment is defined as a prognostic stage of Alzheimer's dementia disease.
- PET scan PET CT scan to confirm the accumulation of beta-amyloid, a pathological finding of Alzheimer's disease, in the brain cortex
- beta-amyloid index index accumulated in the retina were measured, respectively, and a learning model was created using the sample.
- a composition for detecting or treating beta-amyloid plaque which is a near-infrared-based fluorescence-expressing derivative, and diagnosing or treating Alzheimer's disease (application number: 10-2014-0147501), and the ability to selectively detect tau fibroprotein that can be used for early diagnosis of Alzheimer's disease
- a distribution control group with each aMCI patient was created, trained by reinforcement learning, and the beta-amyloid index index of new patients (normal, aMCI patients, Alzheimer's patients) was measured. can be verified.
- the present invention provides a device for identifying beta-amyloid and plaque mass accumulated in the brain, and derivatives bound to tau protein in the retinal layer, and based on data learned by deep learning artificial intelligence technology, Alzheimer's cycle phase (currently Progression status) can be diagnosed early, and it has the effect of being easy to use, inexpensive, and quick to use.
Abstract
Description
AUC(area under curve) | 민감도(T=0.5) | 특이도(T=0.5) | |
베타아밀로이드 | 0.93 | 93.6% | 95.1% |
타우 단백질 | 0.94 | 94.2% | 94.9% |
Claims (19)
- 유도체가 주입된 사용자의 망막을 촬영하는 카메라;상기 카메라를 통하여 촬영한 망막 이미지를 표시하는 디스플레이; 및상기 카메라를 통하여 촬영한 상기 망막 이미지를 분석하고, 상기 망막 이미지에서 베타아밀로이드와 타우 단백질의 발현을 감지하고, 감지된 발현에 대응하는 알츠하이머 진단 정보를 상기 디스플레이로 출력하도록 하는 진단제어부;를 포함하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 1에 있어서,상기 카메라는,상기 사용자의 좌안 및 우안을 각각 촬영하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 1에 있어서,상기 진단제어부는,인공지능 학습 모듈을 구비한 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 1에 있어서,상기 진단제어부는,상기 카메라에 구비된 근접 센서를 통하여 사용자의 근접을 인식하며, 사용자가 인식되지 않으면 상기 카메라의 동작을 절전 모드로 전환시키는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 1에 있어서,상기 진단제어부는,BBB(Blood-Brain-Barrier)를 투과한 펩타이드 기반 유도체 영상 검사를 실시하여 알츠하이머 진단을 하는 것으로, 망막층 내 펩타이드의 착색이 이루어진 후, 베타아밀로이드 스펙트럼과 망막 내 특정 파장대에서 나오는 발현 값을 망막층 구간 내에서 확인하고, 베타아밀로이드에 착색된 영상을 검사하여 단일 베타아밀로이드 플라크의 위치를 판별하고;상기 단일 베타아밀로이드 플라크의 위치가 판별되면, 다차원 복합 베타아밀로이드 플라크 위치를 산정하기 위해 프레임 별 발현 레벨을 산정하고, 최종 발현된 베타아밀로이드 플라크를 확인하여 베타아밀로이드의 분포 위치를 산정하고;망막층 내 펩타이드 착색이 이루어진 후, 타우 단백질 발현을 확인하기 위해 타우 단백질 스펙트럼과 망막층 구간을 확인하고, 타우 단백질에 착색된 영상을 검사하여 알츠하이머로 진행되기 이전부터 쌓인 타우 단백질 체적을 자동계산하고;상기 타우 단백질 체적 양을 계산한 후, 베타아밀로이드 인덱스 레벨을 기초로 경도인지장애 레벨을 산정하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 5에 있어서,상기 진단제어부는,망막층 내 펩타이드의 착색이 이루어진 후, 발현된 망막 이미지가 입력되면, 진단할 망막 이미지에서 베타아밀로이드 분할 영상을 생성하고, 상기 분할 영상에서 망막 내의 베타아밀로이드 분포 위치를 산정하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 5에 있어서,상기 진단제어부는,망막층 내 펩타이드 착색이 이루어진 후 발현된 망막 이미지가 입력되면, 베타아밀로이드 발현 추적을 위해 위치, 분포도, 주기 패턴을 사전에 학습한 베타아밀로이드 학습 데이터 또는 타우 단백질 발현 추적을 위해 체적, 위치에 해당하는 패턴을 사전에 학습한 타우 단백질 학습 데이터를 이용하여 강화 학습 알고리즘을 수행함으로써 진단하고자 하는 영역을 자동으로 구별하여 진단 동작을 수행하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 5에 있어서,상기 진단제어부는,상기 망막 이미지 중 질병과 관련된 영역을 마스킹하고, 마스킹된 망막 이미지를 깊이 값을 포함하는 3차원 이미지로 변환하고, 변환된 3차원 이미지를 분석하여 알츠하이머와 연관된 패턴을 감지하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 5에 있어서,상기 진단제어부는,상기 경도인지장애 레벨이 일정 양성 결과 값 이하이면 경도인지장애로 진단하고, 경도인지장애 레벨이 일정 양성 결과 값 초과이면 알츠하이머로 진단하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 청구항 1에 있어서,상기 유도체는,근적외선 영역에서 발현되는 것으로, 커큐민, 테라큐민, 콩고 레드, 티오플라빈 S-T 또는 크리사민 중 어느 하나 이상을 엑소좀(exosome)과 결합하여 사용하는 것을 특징으로 하는 알츠하이머 진단 장치.
- 카메라를 이용하여 유도체가 주입된 사용자의 망막을 촬영하는 촬영단계;진단제어부에서 상기 카메라를 통하여 촬영한 상기 망막 이미지를 분석하고, 상기 망막 이미지에서 베타아밀로이드와 타우 단백질의 발현을 감지하는 감지단계; 및감지된 발현에 대응하는 알츠하이머 진단 정보를 디스플레이로 출력하도록 하는 진단단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 11에 있어서,상기 촬영단계는,상기 사용자의 좌안 및 우안을 각각 촬영하는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 11에 있어서,상기 촬영단계는,상기 카메라에 구비된 근접 센서를 통하여 사용자의 근접을 인식하는 단계; 및 사용자가 인식되지 않으면 상기 카메라의 동작을 절전 모드로 전환시키는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 11에 있어서,상기 감지단계는,BBB(Blood-Brain-Barrier)를 투과한 펩타이드 기반 유도체 영상 검사를 실시하여 알츠하이머 진단을 하는 것으로, 망막층 내 펩타이드의 착색이 이루어진 후, 베타아밀로이드 스펙트럼과 망막 내 특정 파장대에서 나오는 발현 값을 망막층 구간 내에서 확인하고, 베타아밀로이드에 착색된 영상을 검사하여 단일 베타아밀로이드 플라크의 위치를 판별하는 단계;상기 단일 베타아밀로이드 플라크의 위치가 판별되면, 다차원 복합 베타아밀로이드 플라크 위치를 산정하기 위해 프레임 별 발현 레벨을 산정하고, 최종 발현된 베타아밀로이드 플라크를 확인하여 베타아밀로이드의 분포 위치를 산정하는 단계;망막층 내 펩타이드 착색이 이루어진 후, 타우 단백질 발현을 확인하기 위해 타우 단백질 스펙트럼과 망막층 구간을 확인하고, 타우 단백질에 착색된 영상을 검사하여 알츠하이머로 진행되기 이전부터 쌓인 타우 단백질 체적을 자동계산하는 단계; 및상기 타우 단백질 체적 양을 계산한 후, 베타아밀로이드 인덱스 레벨을 기초로 경도인지장애 레벨을 산정하는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 14에 있어서,상기 감지단계는,망막층 내 펩타이드의 착색이 이루어진 후, 발현된 망막 이미지가 입력되면, 진단할 망막 이미지에서 베타아밀로이드 분할 영상을 생성하고, 상기 분할 영상에서 망막 내의 베타아밀로이드 분포 위치를 산정하는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 14에 있어서,상기 감지단계는,망막층 내 펩타이드 착색이 이루어진 후 발현된 망막 이미지가 입력되면, 베타아밀로이드 발현 추적을 위해 위치, 분포도, 주기 패턴을 사전에 학습한 베타아밀로이드 학습 데이터 또는 타우 단백질 발현 추적을 위해 체적, 위치에 해당하는 패턴을 사전에 학습한 타우 단백질 학습 데이터를 이용하여 강화 학습 알고리즘을 수행함으로써 진단하고자 하는 영역을 자동으로 구별하여 진단 동작을 수행하는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 14에 있어서,상기 감지단계는,상기 망막 이미지 중 질병과 관련된 영역을 마스킹하고, 마스킹된 망막 이미지를 깊이 값을 포함하는 3차원 이미지로 변환하고, 변환된 3차원 이미지를 분석하여 알츠하이머와 연관된 패턴을 감지하는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 14에 있어서,상기 진단단계는,상기 경도인지장애 레벨이 일정 양성 결과 값 이하이면 경도인지장애로 진단하고, 경도인지장애 레벨이 일정 양성 결과 값 초과이면 알츠하이머로 진단하는 단계를 포함하는 것을 특징으로 하는 알츠하이머 진단 방법.
- 청구항 11항 내지 청구항 18항 중 어느 한 항의 알츠하이머 진단 방법을 실행하는 프로그램이 기록되어 컴퓨터로 읽을 수 있는 기록매체.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2020-0053853 | 2020-05-06 | ||
KR1020200053853A KR102378306B1 (ko) | 2020-05-06 | 2020-05-06 | 알츠하이머 진단 장치 및 방법 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021225226A1 true WO2021225226A1 (ko) | 2021-11-11 |
Family
ID=78468135
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2020/011036 WO2021225226A1 (ko) | 2020-05-06 | 2020-08-19 | 알츠하이머 진단 장치 및 방법 |
Country Status (2)
Country | Link |
---|---|
KR (1) | KR102378306B1 (ko) |
WO (1) | WO2021225226A1 (ko) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102642020B1 (ko) * | 2021-12-24 | 2024-02-29 | 주식회사 에이아이플랫폼 | 망막 안저 카메라 환경에서의 특정 객체(치매유발물질) 추적 방법 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7641343B1 (en) * | 2007-07-26 | 2010-01-05 | Motamedi Manouchehr E | Method and apparatus for early diagnosis of Alzheimer's using non-invasive eye tomography by terahertz |
KR20140116353A (ko) * | 2010-10-25 | 2014-10-02 | 스티븐 버두너 | 알츠하이머 병, 외상성 뇌손상, 시력감퇴, 신경병 및 안질환의 진단과 예측을 위해 망막내 아밀로이드를 감지하는 장치와 방법 |
JP2016538986A (ja) * | 2013-11-12 | 2016-12-15 | コグノプティックス, インコーポレイテッド | 眼組織において蛍光を測定する方法 |
JP2019526055A (ja) * | 2016-08-01 | 2019-09-12 | コグノプティックス, インコーポレイテッド | 眼組織におけるタウタンパク質を検出するためのシステムおよび方法 |
KR102075503B1 (ko) * | 2017-06-21 | 2020-02-10 | 주식회사 데이터사이언스랩 | 치매 예측 시스템 및 그 구동방법 |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102051666B1 (ko) | 2017-12-07 | 2019-12-04 | 사회복지법인 삼성생명공익재단 | 아밀로이드 페트(pet) 영상을 이용한 알츠하이머 치매의 병기 측정 방법 및 장치 |
-
2020
- 2020-05-06 KR KR1020200053853A patent/KR102378306B1/ko active IP Right Grant
- 2020-08-19 WO PCT/KR2020/011036 patent/WO2021225226A1/ko active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7641343B1 (en) * | 2007-07-26 | 2010-01-05 | Motamedi Manouchehr E | Method and apparatus for early diagnosis of Alzheimer's using non-invasive eye tomography by terahertz |
KR20140116353A (ko) * | 2010-10-25 | 2014-10-02 | 스티븐 버두너 | 알츠하이머 병, 외상성 뇌손상, 시력감퇴, 신경병 및 안질환의 진단과 예측을 위해 망막내 아밀로이드를 감지하는 장치와 방법 |
JP2016538986A (ja) * | 2013-11-12 | 2016-12-15 | コグノプティックス, インコーポレイテッド | 眼組織において蛍光を測定する方法 |
JP2019526055A (ja) * | 2016-08-01 | 2019-09-12 | コグノプティックス, インコーポレイテッド | 眼組織におけるタウタンパク質を検出するためのシステムおよび方法 |
KR102075503B1 (ko) * | 2017-06-21 | 2020-02-10 | 주식회사 데이터사이언스랩 | 치매 예측 시스템 및 그 구동방법 |
Also Published As
Publication number | Publication date |
---|---|
KR20210135778A (ko) | 2021-11-16 |
KR102378306B1 (ko) | 2022-03-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020242239A1 (ko) | 앙상블 학습 알고리즘을 이용한 인공지능 기반 진단 보조 시스템 | |
WO2019132168A1 (ko) | 수술영상데이터 학습시스템 | |
WO2021049729A1 (ko) | 인공지능 모델을 이용한 폐암 발병 가능성 예측 방법 및 분석 장치 | |
WO2020251217A1 (ko) | 사용자 맞춤형 피부 진단 시스템 및 방법 | |
WO2021040258A1 (ko) | 안구 영상 내 혈관 분할을 이용한 자동 질환 판단 장치 및 그 방법 | |
WO2021025461A1 (ko) | 초음파 영상 기반의 기계 학습을 통한 관상동맥 병변 진단 시스템 및 이의 진단 방법 | |
JPH11244261A (ja) | アイリス認識方法及び装置、データ変換方法及び装置 | |
WO2021225226A1 (ko) | 알츠하이머 진단 장치 및 방법 | |
WO2021261808A1 (ko) | 병변 판독 결과 표시 방법 | |
WO2020076133A1 (ko) | 암 영역 검출의 유효성 평가 장치 | |
CN109661668A (zh) | 用于虹膜识别的图像处理方法和系统 | |
WO2021132813A1 (ko) | 딥러닝 모델을 이용한 통증 평가 방법 및 분석 장치 | |
WO2020180135A1 (ko) | 뇌 질환 예측 장치 및 방법, 뇌 질환을 예측하기 위한 학습 장치 | |
WO2019045385A1 (ko) | 영상 얼라인먼트 방법 및 그 장치 | |
Mikołajczyk et al. | Intelligent system supporting diagnosis of malignant melanoma | |
WO2019221586A1 (ko) | 의료영상 관리 시스템, 방법 및 컴퓨터 판독 가능한 기록매체 | |
WO2020222555A1 (ko) | 영상 분석 장치 및 방법 | |
CN112907571A (zh) | 基于多光谱图像融合识别的目标判定方法 | |
CN109711306B (zh) | 一种基于深度卷积神经网络获取面部特征的方法及设备 | |
KR20110092848A (ko) | 얼굴 인증 및 등록방법 | |
KR102322531B1 (ko) | 착용형 아밀로이드-베타 플라크 검사 장치 및 이를 포함하는 알츠하이머병 진단 시스템 | |
Patel et al. | Glaucoma detection using mask region-based convolutional neural networks | |
WO2023017875A1 (ko) | 안저영상을 기초로 생체 계측치를 예측하는 장치 및 방법 | |
WO2020226455A1 (ko) | 안저 이미지를 이용한 시신경병증 예측 장치 및 시신경병증 예측 결과 제공 방법 | |
JP3731672B2 (ja) | パターン抽出装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20934466 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20934466 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 19.04.2023) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20934466 Country of ref document: EP Kind code of ref document: A1 |