WO2022034691A1 - コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置 - Google Patents

コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置 Download PDF

Info

Publication number
WO2022034691A1
WO2022034691A1 PCT/JP2020/030905 JP2020030905W WO2022034691A1 WO 2022034691 A1 WO2022034691 A1 WO 2022034691A1 JP 2020030905 W JP2020030905 W JP 2020030905W WO 2022034691 A1 WO2022034691 A1 WO 2022034691A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
correlation
amyloid
subject
magnetic susceptibility
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2020/030905
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
フェリックス ユリアン ブランデンブルグ
晃裕 奥野
裕紀 青山
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Splink Inc
Original Assignee
Splink Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Splink Inc filed Critical Splink Inc
Priority to US18/020,834 priority Critical patent/US11972564B2/en
Priority to PCT/JP2020/030905 priority patent/WO2022034691A1/ja
Priority to JP2022542567A priority patent/JP7489645B2/ja
Publication of WO2022034691A1 publication Critical patent/WO2022034691A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0033—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room
    • A61B5/004—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
    • A61B5/0042—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part for the brain
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
    • A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
    • 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
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • 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
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • 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
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • 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
    • 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
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2576/00—Medical imaging apparatus involving image processing or analysis
    • A61B2576/02—Medical imaging apparatus involving image processing or analysis specially adapted for a particular organ or body part
    • A61B2576/026—Medical imaging apparatus involving image processing or analysis specially adapted for a particular organ or body part for the brain
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06T2207/30016—Brain
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03—Recognition of patterns in medical or anatomical images

Definitions

  • the present invention relates to a computer program, an information processing device, an information processing method, a trained model generation method, and a correlated image output device.
  • Alzheimer's disease The main cause of dementia is said to be Alzheimer's disease.
  • the cause of Alzheimer's disease has not yet been elucidated, but as the condition progresses, peculiar lesions are found in the brain.
  • amyloid plaque deposition has been shown to occur in the earliest stages of onset of Alzheimer's disease and begin well before clinical symptoms appear (eg, a dozen years ago).
  • Patent Document 1 discloses a device that injects a drug that binds to amyloid ⁇ in brain tissue into a subject and uses a PET (Positron Emission Tomography) image showing the concentration distribution of the drug on a cut surface crossing the brain. ing.
  • PET Pulsitron Emission Tomography
  • the present invention has been made in view of such circumstances, and is a computer program, an information processing apparatus, an information processing method, a trained model generation method, and a computer program capable of estimating signs of amyloid ⁇ -related disease without using PET images.
  • An object of the present invention is to provide a correlated image output device.
  • the computer program acquires an MRI image of a subject in a computer, and when the MRI image is input, the correlation between the magnetization rate that can be specified based on the MRI image and the amyloid ⁇ is obtained.
  • the acquired MRI image is input to the trained model that outputs the shown correlation image, and the process of outputting the correlation image showing the correlation between the magnetization rate of the subject and the amyloid ⁇ is executed.
  • the information processing apparatus has a correlation between an acquisition unit that acquires an MRI image of a subject, a magnetization rate that can be specified based on the MRI image when the MRI image is input, and amyloid ⁇ .
  • the trained model for outputting the correlation image showing the above is provided with an output unit for inputting the acquired MRI image and outputting the correlation image showing the correlation between the magnetization rate of the subject and the amyloid ⁇ .
  • the information processing method obtains an MRI image of a subject, and when the MRI image is input, the correlation showing the correlation between the magnetization rate that can be specified based on the MRI image and the amyloid ⁇ .
  • the acquired MRI image is input to the trained model that outputs the image, and the correlation image showing the correlation between the magnetization rate of the subject and the amyloid ⁇ is output.
  • an MRI image is acquired, a correlation image showing the correlation between the magnetic susceptibility and amyloid ⁇ is acquired, and the acquired MRI image and the correlation image are used to describe the above.
  • a trained model that outputs a correlated image showing the correlation between the identifiable magnetic susceptibility and amyloid ⁇ based on the MRI image is generated.
  • the correlated image output device outputs a correlated image showing the correlation between the magnetic susceptibility of the subject and amyloid ⁇ when the MRI image of the subject is input.
  • signs of dementia can be estimated without using PET images.
  • FIG. 1 is a schematic diagram showing an example of the configuration of the information processing system of the present embodiment.
  • the information processing system includes a server 50 as an information processing device.
  • the server 50 is connected to the communication network 1.
  • a terminal device 10 used by medical professionals, researchers, and the like is connected to the communication network 1.
  • the terminal device 10 for example, a personal computer, a smartphone, a tablet, or the like can be used. Further, the terminal device 10 can acquire or transfer a medical image from an MRI device or the like (not shown).
  • the server 50 includes a control unit 51, a communication unit 52, a storage unit 53, an output processing unit 54, an estimation unit 55, and a learning processing unit 56 that control the entire server 50.
  • the image DB 61 is connected to the server 50.
  • the control unit 51 can be configured by a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.
  • Each function of the server 50 may be distributed to a plurality of servers.
  • the estimation unit 55 may be provided in the server 50
  • the learning processing unit 56 may be provided in another server.
  • the information processing device can be incorporated in the server 50, it may be incorporated in a device other than the server 50.
  • the communication unit 52 is composed of a required communication module or the like, and provides a communication function with the terminal device 10 via the communication network 1.
  • the communication unit 52 can acquire, for example, a medical image (for example, an MRI image) of a subject, time information, and the like from the terminal device 10. Details of the MRI image and time information will be described later.
  • MRI images are also referred to as MR images.
  • the storage unit 53 can be composed of a hard disk, a semiconductor memory, or the like, and can store required data such as data obtained as a result of processing in the server 50.
  • the output processing unit 54 performs output processing when providing the brain state estimation result to the terminal device 10.
  • the estimation unit 55 has a function of estimating the brain state, and includes an input data generation unit 551 and a model unit 552.
  • the model unit 552 is composed of a semiconductor memory, a hard disk, or the like, and stores a model (learned model) generated by machine learning.
  • the trained model can be constructed, for example, by a neural network.
  • the input data generation unit 551 generates data to be input to the learning model when performing the brain state estimation process.
  • the learning processing unit 56 has a function of generating a trained model by machine learning, and includes a learning data generation unit 561, a model unit 562, and a parameter determination unit 563.
  • the model unit 562 is composed of a semiconductor memory, a hard disk, or the like, and stores a model before machine learning.
  • the trained model generated by performing machine learning in the learning processing unit 56 can be stored in the model unit 552 of the estimation unit 55.
  • the model unit 562 may store a model in the middle of machine learning, a model for re-learning, and a model that has been trained. Further, the learning processing unit 56 is not an indispensable configuration, and may be provided in another server that performs learning processing.
  • the training data generation unit 561 generates training input data and teacher data when generating a trained model.
  • the parameter determination unit 563 adjusts the parameters (for example, weights, biases, etc.) of the neural network when generating the trained model, and finally determines the parameters.
  • the learning processing unit 56 is, for example, a hardware such as a CPU (for example, a multi-processor having a plurality of processor cores mounted), a GPU (Graphics Processing Units), a DSP (Digital Signal Processors), an FPGA (Field-Programmable Gate Arrays), or the like. It can be configured by combining wear.
  • a CPU for example, a multi-processor having a plurality of processor cores mounted
  • a GPU Graphics Processing Units
  • DSP Digital Signal Processors
  • FPGA Field-Programmable Gate Arrays
  • the image DB 61 can record various images used in machine learning when generating a trained model. Further, the image DB 61 can record various images related to the brain state estimation result by the server 50.
  • the brain state estimation method by the server 50 estimates the accumulation state of amyloid ⁇ in the brain, that is, the distribution state of amyloid ⁇ .
  • the inventor has a significant correlation between QSM (Quantitative Susceptibility Mapping), which is a method for quantitatively imaging the magnetic susceptibility of living tissue, and amyloid PET, which detects the accumulation of amyloid ⁇ by PET (Positron Emission Tomography).
  • QSM Quantitative Susceptibility Mapping
  • amyloid PET which detects the accumulation of amyloid ⁇ by PET (Positron Emission Tomography).
  • FIG. 2 is a schematic diagram showing a first example of the brain state estimation method by the estimation unit 55.
  • the model unit 552 is a trained model, and the model is composed of a neural network, includes an input layer, an intermediate layer, and an output layer, and the parameters (weight, bias, etc.) of the neural network are determined by machine learning. ..
  • the input data generation unit 551 inputs the MRI image acquired from the terminal device 10 via the communication unit 52 to the model unit 552 as input data.
  • the MRI image includes a T1-weighted image and a T2-weighted image.
  • a complex number image consisting of a real part and an imaginary part can be generated.
  • the intensity image is an image showing the absolute value of the real part and the imaginary part of each pixel.
  • the phase image showing the phase between the real part and the imaginary part of each pixel is an image showing the difference in phase caused by the difference in magnetic susceptibility between living tissues.
  • the intensity image and the phase image are, for example, a T1 weighted image, a T2 weighted image, and the like.
  • the MRI image may include an image generated by a predetermined image processing from the MRI image.
  • a QSM (Quantitative Susceptibility Mapping) image is also referred to as a quantitative magnetic susceptibility mapping image and can be generated from an MRI image.
  • the QSM image is a map obtained by quantitatively obtaining the local magnetic susceptibility from the phase image.
  • Magnetic susceptibility is a physical property value that indicates the likelihood of magnetic polarization (magnetization) that occurs when a substance reacts to an external magnetic field. Since all substances have weak antimagnetism, biological tissues show a slightly negative magnetic susceptibility. , Shows a positive magnetic susceptibility when iron deposition occurs. The change in magnetic susceptibility can also occur due to factors other than iron deposition, such as fibrosis and deoxygenation of hemoglobin.
  • the MRI image may be an image whose magnetic susceptibility can be specified and can be generated from the MRI image. That is, the MRI image includes, in addition to the T1-weighted image, a T2-weighted image and a QSM image generated from the MRI image by a predetermined image processing.
  • the model unit 552 has a function as a correlated image output device, and can output a correlated image showing the correlation between the magnetic susceptibility and amyloid ⁇ when an MRI image relating to the brain is input.
  • the model unit 552 can estimate the current distribution state of amyloid ⁇ , but the model unit 552 determines the distribution state of amyloid ⁇ in the future (for example, one year later, two years later, etc.). Whether to estimate or not can be determined in advance as a specific period. Further, in the case of performing estimation at a plurality of future time points such as one year and two years later, it is sufficient to prepare a model unit 552 for each specific period, which is specialized for each of the plurality of future time points. The details of the method of generating the trained model will be described later.
  • the correlated image is a phase relationship between the magnetic susceptibility and the PET signal of the PET image (for example, SUV (Standardized Uptake Value), SUVR (Standardized Uptake Value Ratio)) for each corresponding pixel (voxel) of the MRI image and the PET image.
  • the number r is specified.
  • Various methods can be used to illustrate (image) the correlation coefficient r. For example, a voxel having a correlation coefficient r equal to or higher than a threshold value may be illustrated, and the magnitude of the correlation coefficient r may be shown. It may be imaged in an identifiable manner like a heat map.
  • the sizes of the MRI image, PET image, and correlation image are the same and can be, for example, 128 x 128 x 64 (1048576 voxels in total), but are not limited to this.
  • SUVR is the sum of the SUVs (Amyloid ⁇ protein accumulation) of the four sites of the cerebral gray matter (prefrontal cortex, anterior and posterior cingulate cortex, parietal lobe, and lateral temporal lobe) in a specific reference region (eg, cerebellum, etc.). ) Can be obtained by dividing by SUV.
  • the correlation coefficient r can be, for example, 0 ⁇ r ⁇ 1, and the closer it is to 1, the greater the correlation.
  • FIG. 3 is a schematic diagram showing an example of the correlation between the MRI image and the PET image when the correlation coefficient r is large.
  • a voxel signal with an MRI image (iron load, magnetization rate, etc. in the case of a QSM image) and a PET signal of a voxel corresponding to the voxel in the PET image (a voxel having the same coordinates on a three-dimensional image) are signaled (iron).
  • the correlation coefficient r is large.
  • the correlation in cluster units can be obtained, and if the number of voxels is less than the predetermined threshold value, the correlation in voxel units can be obtained.
  • the region of interest can be determined in advance, and the correlation of each region of interest can be obtained based on the voxels in the region of interest.
  • FIG. 4 is a schematic diagram showing an example of the correlation between the MRI image and the PET image when the correlation coefficient r is small.
  • a voxel signal (iron load) with an MRI image and a PET signal of the voxel corresponding to the voxel in the PET image (a voxel having the same coordinates on a three-dimensional image) are used as a signal (iron load) and a PET signal (SUV). Map to a fixed two-dimensional coordinate. As shown in FIG. 4, when the data mapped to the two-dimensional coordinates has a weak relationship between the signals, the correlation coefficient r is small.
  • the MRI image of the subject can be input to the model unit 552 to estimate the distribution state of amyloid ⁇ in the subject's brain.
  • Diseases related to amyloid ⁇ include, for example, mild cognitive impairment (MCI), mild cognitive impairment due to Alzheimer's disease (MCIdue to AD), prodromal AD, and pre-onset stage / pre-onset of Alzheimer's disease. It includes neurodegenerative diseases such as clinical AD (preclinical AD), Parkinson's disease, multiple sclerosis, cognitive decline, cognitive impairment, and amyloid positive / negative diseases.
  • FIG. 5 is a schematic diagram showing the first example of the brain state estimation result by the estimation unit 55.
  • the model unit 552 can output a correlation image showing the correlation between the magnetic susceptibility and amyloid ⁇ when an MRI image relating to the brain is input. Since the correlation image is an image in which the correlation coefficient between the magnetic susceptibility and amyloid ⁇ is specified for each voxel, the value of each voxel represents a value related to the amount of amyloid ⁇ . As a result, if an MRI image of the brain at a certain point in time is obtained, it becomes possible to estimate the distribution state of amyloid ⁇ in the brain.
  • the correlation image is schematically shown for each cluster unit, voxel unit, and interest region unit.
  • An image as shown in FIG. 5 can be processed by the output processing unit 54 and displayed on the terminal device 10, for example.
  • the points in the correlation image represent the location of the deposition of amyloid ⁇ .
  • values related to the amount of amyloid ⁇ can be represented by voxel values (eg, may be expressed in terms of degree of lightness and darkness).
  • the output processing unit 54 can superimpose and display the correlation image output by the model unit 552 and another MRI image (for example, a T1-weighted image).
  • another MRI image for example, a T1-weighted image
  • FIG. 5 an image in which a correlation image and another MRI image are superimposed for each cluster unit, voxel unit, and region of interest unit is schematically represented.
  • a T1-weighted image having a feature that the structure of the brain is easy to see can be used. This makes it possible to easily determine which part of the brain the distribution state of amyloid ⁇ shown in the correlated image corresponds to.
  • FIG. 6 is a schematic diagram showing an example of the distribution state of amyloid ⁇ according to cognitive impairment.
  • FIG. 6 schematically illustrates the distribution of amyloid ⁇ in the brains of subjects with NC (healthy elderly), MCI (mild cognitive impairment), and AD (Alzheimer's disease). In addition, it may be different from the actual distribution state of amyloid ⁇ . In the case of NC, the distribution of amyloid ⁇ is hardly seen. On the other hand, in the case of AD, it can be seen that amyloid ⁇ is deposited at a plurality of sites in the brain.
  • VBM Vehicle Based Morphometry
  • T1-weighted images T1-weighted images
  • the deposition of amyloid plaque by amyloid ⁇ is considered to be a pathological change that occurs from the earliest stage of the onset of Alzheimer's disease, and is said to begin ten years before the onset of clinical symptoms.
  • FIG. 6 in the case of MCI, it can be seen that the distribution of amyloid ⁇ is more prominent than that of NC.
  • the brain state estimation method of the present embodiment it is possible to diagnose the onset at an extremely early stage in the onset process of Alzheimer's disease, and it is only useful for early detection of cognitive impairment in the subject. It is useful for the study of preclinical Alzheimer's disease.
  • FIG. 7 is a schematic diagram showing a second example of the brain state estimation result by the estimation unit 55.
  • FIG. 7 shows the distribution of amyloid ⁇ in the brain of a subject at present, 1 year and 2 years later.
  • a time-series model constructed with a specific period of one year may be used.
  • a model (learned model) can be generated by machine learning.
  • a time-series model constructed with a specific period of 2 years may be used.
  • the estimation unit 55 determines the neurodegenerative disease including future dementia of the subject based on the correlation image showing the correlation between the magnetic susceptibility of the subject's brain and amyloid ⁇ , which is output by the model unit 552.
  • the subject is currently diagnosed with MCI, but can be predicted at this time that he may be diagnosed with AD two years later.
  • the example of FIG. 7 is an example, and is not limited to one year and two years later.
  • MCI mimild cognitive impairment
  • FIG. 8 is a schematic diagram showing a second example of the brain state estimation method by the estimation unit 55.
  • the configuration of the model unit 552 is the same as that of the first example illustrated in FIG.
  • the input data generation unit 551 inputs time information for specifying the period to the model unit 552 as input data.
  • the specified period is a period (specific period) for specifying how much time has passed from the state of the brain at a certain point in time to estimate the state of the brain, and may be one year, two years, or the like. can.
  • the model unit 552 can output a correlation image showing the correlation between the magnetic susceptibility of the brain after the period and the amyloid ⁇ shown by the MRI image. For example, if the specific period is one year, the model unit 552 outputs a correlation image showing the correlation between the magnetic susceptibility of the brain one year later and amyloid ⁇ , and if the specific period is two years, the model unit 552. Can output a correlated image showing the correlation between the magnetic susceptibility of the brain and amyloid ⁇ after 2 years.
  • the model unit 552 by inputting the time information for specifying the period into the model unit 552, it is possible to estimate the distribution state of amyloid ⁇ in the brain of the subject after the period specified by the time information. Further, if the time information for specifying the required period is input to the model unit 552, only the MRI image of the subject's current brain state can be obtained, and the amyloid ⁇ in the subject's brain after the required period has elapsed from the present. Distribution state can be estimated.
  • FIG. 9 is a schematic diagram showing a third example of the brain state estimation method by the estimation unit 55.
  • the configuration of the model unit 552 is the same as that of the first example illustrated in FIG.
  • the input data generation unit 551 inputs the T1-weighted image and the T2-weighted image of the subject's brain to the model unit 552.
  • the T1-weighted image and the T2-weighted image are directly input to the model unit 552 is shown, but the present invention is not limited to this, and the T1-weighted image and the T2-weighted image are images.
  • the image after image processing may be input to the model unit 552 including the step of processing.
  • the estimation unit 55 acquires a T1-weighted image and a T2-weighted image based on the MRI image of the subject and inputs the T1-weighted image and the T2-weighted image, the magnetization rate that can be specified based on the T2-weighted image and the amyloid
  • the acquired T1-weighted image and T2-weighted image can be input to the model unit 552 that outputs the correlation image showing the correlation with ⁇ , and the correlation image showing the correlation between the magnetization rate of the subject and the amyloid ⁇ can be output. ..
  • the T2-weighted image can be used to detect iron deposition.
  • the T1-weighted image can emphasize substances other than water and blood, and for example, the thickness of the cortex can be observed from the neocortex in the brain to the entire area of the cerebral cortex. It is known that the amyloid ⁇ -induced amyloid plaque begins to accumulate from the base of the neocortex and then spreads to the entire area of the cerebral cortex. It is also known that in Alzheimer-type dementia, the atrophy rate in each cortex such as the lateral temporal lobe cortex and the retrosplenial cortex of the corpus callosum gradually increases.
  • the model unit 552 can output a correlation image showing the correlation between the magnetic susceptibility that can be specified based on the T2-weighted image and the amyloid ⁇ .
  • the model unit 552 can estimate the current distribution state of amyloid ⁇ , but the model unit 552 determines the distribution state of amyloid ⁇ in the future (for example, one year later, two years later, etc.). Whether to estimate or not can be determined in advance as a specific period.
  • the model unit 552 is trained using T1-weighted images, the future distribution of amyloid ⁇ in the subject's brain is estimated more accurately in consideration of the spread and atrophy rate of the amyloid plaque in the brain. be able to. This makes it possible to more accurately estimate the signs of amyloid ⁇ -related diseases including dementia without using PET images.
  • the time information as illustrated in FIG. 8 may be further input.
  • the model 552 can more accurately estimate the distribution state of amyloid ⁇ in the subject's brain at a (future) time point after a specific period specified by the input time information.
  • FIG. 10 is a schematic diagram showing a first example of a method of generating a trained model in the learning processing unit 56.
  • the model unit 562 can be configured by, for example, a neural network.
  • the learning data generation unit 561 acquires an MRI image of the brain as learning input data, and acquires a correlation image showing the correlation between the magnetic susceptibility and amyloid ⁇ as training data and gives it to the model unit 562.
  • the parameter determination unit 563 so that the correlation image output by the model unit 562 approaches the correlation image as the teacher data (so that the value of the loss function is minimized).
  • the trained model can be generated by adjusting the parameters of the model unit (for example, weight wij, bias bkl) and finally determining.
  • the correlation image output by the model unit 562 and the correlation image as teacher data are obtained when a specific period (for example, one year, two years, etc.) has passed from the brain at the time when the magnetic susceptibility indicated by the MRI image was obtained.
  • the correlation between the magnetic susceptibility of the brain and amyloid ⁇ is shown.
  • FIG. 11 is a schematic diagram showing a second example of a method of generating a trained model in the learning processing unit 56.
  • the model unit 562 is the same as the first example shown in FIG.
  • the learning data generation unit 561 acquires time information for specifying a period as input data for learning in addition to an MRI image about the brain, and acquires a correlation image showing the correlation between the magnetic susceptibility and amyloid ⁇ as teacher data. It is given to the model unit 562.
  • time information for specifying the period is further input to the model unit 562, and the correlation image output by the model unit 562 shows the correlation between the magnetic susceptibility of the brain after the period and the amyloid ⁇ shown by the MRI image. You can learn that it is an image.
  • the model unit 552 shown in FIG. 8 can be generated.
  • a T1-weighted image and a T2-weighted image may be input to the model unit 562 shown in FIG. That is, the T1-weighted image and the T2-weighted image are input to the model unit 562, and the parameters of the model unit 562 (for example, the weight wij, so that the correlated image output by the model unit 562 approaches the correlated image as the teacher data).
  • the bias bkl By adjusting the bias bkl), a trained model can be generated.
  • the model unit 562 shown in FIG. 10 has a T1-weighted image, a T2-weighted image, and time information. Just enter.
  • the learning of the model 562 can be performed by individually using the correlation images of each cluster unit, voxel unit, or region of interest unit, which is a set of voxels.
  • a voxel is the smallest building block of a 3D image and is a small volume cube with a scalar or vector value.
  • a cluster is a three-dimensional area composed of a plurality of voxels.
  • a region of interest (ROI: Region Of Interest) is a specific narrowed area for observation or measurement.
  • FIG. 12 is a flowchart showing an example of the procedure of the brain state estimation process.
  • the control unit 51 acquires an MRI image of the subject (S11).
  • the MRI image may be, for example, a T2-weighted image, a QSM image, a combination of a T2-weighted image and a T1-weighted image, or a combination of a QSM image and a T1-weighted image.
  • the control unit 51 inputs the acquired MRI image to the trained model (S12).
  • the control unit 51 displays the correlation image output by the trained model on the terminal device 10 (S13), and determines whether or not to superimpose it on the MRI image (for example, a T1 weighted image) (S14). Whether or not the correlated image and the MRI image are superimposed can be determined based on the instruction from the terminal device 10. When not superimposed on the MRI image (NO in S14), the control unit 51 ends the process. When superimposing on the MRI image (YES in S14), the control unit 51 superimposes the correlation image and the MRI image and displays them on the terminal device 10 (S15), and ends the process.
  • the control unit 51 displays the correlation image output by the trained model on the terminal device 10 (S13), and determines whether or not to superimpose it on the MRI image (for example, a T1 weighted image) (S14). Whether or not the correlated image and the MRI image are superimposed can be determined based on the instruction from the terminal device 10.
  • the control unit 51 ends the process.
  • the control unit 51
  • FIG. 13 is a flowchart showing an example of the procedure of the trained model generation process.
  • the control unit 51 reads out the model (S21) and sets the initial values of the parameters of the neural network (S22).
  • the control unit 51 acquires an MRI image (S23) and acquires a correlation image as teacher data (S24).
  • the MRI image may be, for example, a T2-weighted image, a QSM image, a combination of a T2-weighted image and a T1-weighted image, or a combination of a QSM image and a T1-weighted image.
  • the control unit 51 inputs the MRI image to the model, and adjusts the neural network parameters so that the value of the loss function based on the correlation image output by the model and the correlation image acquired as the teacher data is minimized (S25). ..
  • the control unit 51 determines whether or not the value of the loss function is within the permissible range (S26), and if the value of the loss function is not within the permissible range (NO in S26), the processing after step S25 is continued. When the value of the loss function is within the allowable range (YES in S26), the control unit 51 stores the generated trained model (S27) and ends the process.
  • the server 50 can also be realized by using a computer equipped with a CPU (processor), RAM, and the like.
  • a computer program (which can be recorded on a recording medium) that defines the processing procedure as shown in FIGS. 12 and 13 is read by a recording medium reading unit provided in the computer, the read computer program is loaded into the RAM, and the computer program is loaded.
  • the server 50 can be realized on a computer by executing the program on a CPU (processor).
  • the accumulation of amyloid ⁇ in the brain has been mainly described, but the present embodiment is not limited to the site in the brain, and the amyloid ⁇ in a site other than the brain. If the accumulation is associated with some disease, this embodiment can also be applied to the site.
  • the computer program of the present embodiment acquires an MRI image of a subject in a computer, and when the MRI image is input, a correlation image showing the correlation between the magnetization rate that can be specified based on the MRI image and the amyloid ⁇ .
  • the acquired MRI image is input to the trained model that outputs the above, and the process of outputting the correlation image showing the correlation between the magnetization rate of the subject and the amyloid ⁇ is executed.
  • the information processing apparatus of the present embodiment has a correlation between an acquisition unit that acquires an MRI image of a subject, a magnetic susceptibility that can be specified based on the MRI image when the MRI image is input, and an amyloid ⁇ .
  • the trained model that outputs an image is provided with an output unit that inputs an acquired MRI image and outputs a correlated image showing the correlation between the magnetic susceptibility of the subject and the amyloid ⁇ .
  • the information processing method of the present embodiment acquires an MRI image of a subject, and when the MRI image is input, outputs a correlation image showing the correlation between the magnetization rate that can be specified based on the MRI image and the amyloid ⁇ .
  • the acquired MRI image is input to the trained model to be used, and a correlation image showing the correlation between the magnetization rate of the subject and the amyloid ⁇ is output.
  • an MRI image is acquired, a correlation image showing the correlation between the magnetic susceptibility and amyloid ⁇ is acquired, and the acquired MRI image and the correlation image are used in the MRI image.
  • a trained model that outputs a correlated image showing the correlation between the susceptible susceptibility that can be specified based on the amyloid ⁇ and the amyloid ⁇ is generated.
  • the correlated image output device of the present embodiment outputs a correlated image showing the correlation between the magnetic susceptibility of the subject and amyloid ⁇ when the MRI image of the subject is input.
  • the computer program acquires the MRI image of the subject.
  • MRI images include T1-weighted images and T2-weighted images.
  • MR Magnetic Resonance
  • the intensity image is an image showing the absolute value of the real part and the imaginary part of each pixel.
  • the phase image showing the phase between the real part and the imaginary part of each pixel is an image showing the difference in phase caused by the difference in magnetic susceptibility between living tissues.
  • the intensity image and the phase image are, for example, a T1 weighted image, a T2 weighted image, and the like.
  • the MRI image may include an image generated by a predetermined image processing from the MRI image.
  • a QSM (Quantitative Susceptibility Mapping) image is also referred to as a quantitative magnetic susceptibility mapping image and can be generated from an MRI image.
  • the QSM image is a map obtained by quantitatively obtaining the local magnetic susceptibility from the phase image.
  • Magnetic susceptibility is a physical property value that indicates the likelihood of magnetic polarization (magnetization) that occurs when a substance reacts to an external magnetic field. Since all substances have weak antimagnetism, biological tissues show a slightly negative magnetic susceptibility. , Shows a positive magnetic susceptibility when iron deposition occurs.
  • the trained model can output a correlation image showing the correlation between the magnetic susceptibility that can be specified based on the MRI image and amyloid ⁇ when the MRI image is input.
  • the magnetic susceptibility that can be specified based on the MRI image is, for example, a magnetic susceptibility that can be specified based on an MRI image such as a T2-weighted image or a QSM image or an image obtained by image processing from the MRI image. That is, the MRI image can include an image in which the magnetic susceptibility can be specified and can be generated from the MRI image.
  • the correlation image is a correlation coefficient between the magnetic susceptibility and the PET signal (for example, SUVR (Standardized Uptake Value Ratio)) of the PET image for each corresponding pixel (voxel) of the MRI image (for example, a QSM image) and the PET image.
  • r is specified.
  • the correlation coefficient r can be, for example, 0 ⁇ r ⁇ 1, and the closer it is to 1, the greater the correlation.
  • the pre-learning model (also simply referred to as a "model") can be configured with a neural network, for example.
  • An MRI image is acquired as learning input data, and a correlation image showing a correlation between the magnetic susceptibility that can be specified based on the MRI image and amyloid ⁇ is acquired as training data.
  • a trained model is generated by inputting an MRI image into the model and adjusting the model parameters (for example, weights and biases) so that the correlated image output by the model approaches the correlated image as teacher data. be able to.
  • the acquired MRI image of the subject can be input to the trained model to estimate the distribution state of the amyloid ⁇ of the subject. This makes it possible to estimate signs of amyloid ⁇ -related disease without using PET images.
  • the computer program of the present embodiment acquires a T1-weighted image and a T2-weighted image based on the MRI image of the subject and inputs the T1-weighted image and the T2-weighted image to the computer, the computer program specifies the T2-weighted image based on the T2-weighted image.
  • the acquired T1-weighted image and T2-weighted image are input to the trained model that outputs a correlation image showing the correlation between the possible magnetization rate and the amyloid ⁇ , and the correlation between the magnetization rate of the subject and the amyloid ⁇ is shown. Output the correlation image and execute the process.
  • the computer program acquires a T1-weighted image and a T2-weighted image based on the MRI image of the subject.
  • T2-weighted images can be used to detect iron deposition.
  • the T1-weighted image can emphasize substances other than water and blood, and for example, the thickness of the cortex can be observed from the neocortex in the brain to the entire area of the cerebral cortex. It is known that the amyloid ⁇ -induced amyloid plaque begins to accumulate from the base of the neocortex and then spreads to the entire area of the cerebral cortex. It is also known that in Alzheimer-type dementia, the atrophy rate in each cortex such as the lateral temporal lobe cortex and the retrosplenial cortex of the corpus callosum gradually increases.
  • the trained model can output a correlation image showing the correlation between the identifiable magnetic susceptibility and the amyloid ⁇ based on the T2-weighted image.
  • Training is performed by inputting a T1-weighted image and a T2-weighted image into the model and adjusting model parameters (for example, weights and biases) so that the correlation image output by the model approaches the correlation image as teacher data. You can generate a finished model.
  • the trained model is trained using T1-weighted images, the distribution state of amyloid ⁇ of the subject can be estimated more accurately in consideration of the spread of the amyloid plaque and the atrophy rate. This makes it possible to more accurately estimate the signs of amyloid ⁇ -related diseases without using PET images.
  • the trained model outputs a correlation image showing the correlation between the magnetic susceptibility after a specific period of the brain and amyloid ⁇ when an MRI image of the brain is input.
  • the trained model outputs a correlation image showing the correlation between the magnetic susceptibility of the brain after a specific period and amyloid ⁇ when an MRI image of the brain is input.
  • the trained model can be generated as follows. As learning input data, an MRI image of the brain is acquired, and as teacher data, a correlation image showing the correlation between the magnetic susceptibility of the brain after a specific period and amyloid ⁇ , which is shown by the MRI image, is acquired.
  • a trained model is generated by inputting an MRI image into the model and adjusting the model parameters (for example, weights and biases) so that the correlated image output by the model approaches the correlated image as teacher data. be able to.
  • the correlation image output by the model and the correlation image as teacher data are the brain when a specific period (for example, one year, two years, etc.) has passed from the brain at the time when the magnetic susceptibility shown in the MRI image was obtained.
  • the correlation between the magnetic susceptibility of and amyloid ⁇ is shown.
  • the computer program of the present embodiment can be specified based on the MRI image after the period of the brain when the computer acquires the time information for specifying the period and further inputs the time information for specifying the period.
  • the acquired time information is input to the trained model that outputs a correlation image showing the correlation between the magnetic susceptibility and the amyloid ⁇ , and the correlation between the magnetic susceptibility of the subject's brain after the period and the amyloid ⁇ is shown.
  • Output the correlation image execute the process.
  • the computer program acquires time information that specifies the period.
  • the specified period is a period for specifying how much time has passed from the state of the brain at a certain point in time to estimate the state of the brain.
  • the trained model can output a correlated image showing the correlation between the susceptible magnetic susceptibility that can be specified based on the MRI image and amyloid ⁇ after the period of the brain when further input time information for specifying the period is input. can.
  • the time information that specifies the period is further input to the model, and the correlation image output by the model shows the correlation between the magnetic susceptibility that can be specified based on the MRI image and the amyloid ⁇ after the period in the brain. You can learn to be.
  • time information that specifies the period By inputting time information that specifies the period into the trained model, it is possible to estimate the distribution state of amyloid ⁇ in the brain of the subject after the period specified by the time information. Further, if the time information is used as the time information for specifying the required period, for example, only by obtaining an MRI image of the subject's current brain state, the amyloid ⁇ in the subject's brain after the required period has elapsed from the present. Distribution state can be estimated.
  • the computer program of the present embodiment includes the future dementia of the subject based on the correlation image showing the correlation between the magnetic susceptibility of the subject's brain and amyloid ⁇ , which is output to the computer by the trained model. Perform processing to estimate neurodegenerative disease.
  • the computer program can estimate the subject's future neurodegenerative diseases including dementia based on the correlation image showing the correlation between the magnetic susceptibility of the subject's brain and amyloid ⁇ output by the trained model. For example, depending on the magnitude of the rate of change in the accumulation (accumulation degree) of amyloid ⁇ , it is possible to predict whether or not the subject develops a disease related to the accumulation of amyloid ⁇ such as dementia.
  • the correlation image shows the correlation between the magnetic susceptibility and amyloid ⁇ in a cluster unit, a voxel unit, or a region of interest unit, which is a set of voxels.
  • the correlation image can show the correlation between the magnetic susceptibility and amyloid ⁇ in each cluster unit, voxel unit, or region of interest unit, which is a set of voxels.
  • a voxel is the smallest building block of a 3D image and is a small volume cube with a scalar or vector value.
  • a cluster is a three-dimensional area composed of a plurality of voxels.
  • a region of interest ROI: Region Of Interest
  • the training of the model can be performed by individually using the correlation images of each cluster unit, voxel unit, or region of interest unit, which is a set of voxels.
  • the computer program of the present embodiment causes a computer to execute a process of superimposing and displaying a correlation image output by the trained model and another MRI image.
  • the computer program can superimpose and display the correlation image output by the trained model and other MRI images.
  • MRI images for example, a T1-weighted image having a feature that the structure of the brain is easy to see can be used. This makes it possible to easily determine which part of the brain the distribution state of amyloid ⁇ shown in the correlated image corresponds to.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Epidemiology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Primary Health Care (AREA)
  • Pathology (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Neurology (AREA)
  • Biophysics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Physiology (AREA)
  • Psychiatry (AREA)
  • Quality & Reliability (AREA)
  • Computing Systems (AREA)
  • Psychology (AREA)
  • Child & Adolescent Psychology (AREA)
  • Developmental Disabilities (AREA)
  • Hospice & Palliative Care (AREA)
  • High Energy & Nuclear Physics (AREA)
  • General Business, Economics & Management (AREA)
  • Software Systems (AREA)
  • Neurosurgery (AREA)
PCT/JP2020/030905 2020-08-14 2020-08-14 コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置 Ceased WO2022034691A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US18/020,834 US11972564B2 (en) 2020-08-14 2020-08-14 Recording medium, information processing device, information processing method, trained model generation method, and correlation image output device
PCT/JP2020/030905 WO2022034691A1 (ja) 2020-08-14 2020-08-14 コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置
JP2022542567A JP7489645B2 (ja) 2020-08-14 2020-08-14 コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/030905 WO2022034691A1 (ja) 2020-08-14 2020-08-14 コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置

Publications (1)

Publication Number Publication Date
WO2022034691A1 true WO2022034691A1 (ja) 2022-02-17

Family

ID=80247100

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/030905 Ceased WO2022034691A1 (ja) 2020-08-14 2020-08-14 コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置

Country Status (3)

Country Link
US (1) US11972564B2 (https=)
JP (1) JP7489645B2 (https=)
WO (1) WO2022034691A1 (https=)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023167157A1 (ja) * 2022-03-01 2023-09-07 株式会社Splink コンピュータプログラム、情報処理装置及び情報処理方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118691906B (zh) * 2024-07-18 2025-09-26 首都医科大学附属北京友谊医院 认知状态分类方法、装置、设备和存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013047583A1 (ja) * 2011-09-28 2013-04-04 国立大学法人熊本大学 画像解析装置、画像解析方法及び画像解析プログラム
WO2018064715A1 (en) * 2016-10-03 2018-04-12 Crc For Mental Health Ltd Method for predicting or diagnosing cognitive deterioration
JP2020503991A (ja) * 2016-12-06 2020-02-06 ダルミヤン，インク． 脳障害を特定するための方法及びシステム

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014034724A1 (ja) 2012-08-30 2014-03-06 株式会社島津製作所 脳断層動態画像解析装置
US10779762B2 (en) * 2015-10-20 2020-09-22 Washington University MRI method for in vivo detection of amyloid and pathology in the Alzheimer brain
WO2018176082A1 (en) * 2017-03-28 2018-10-04 Crc For Mental Health Ltd Predicting progression of cognitive deterioration

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013047583A1 (ja) * 2011-09-28 2013-04-04 国立大学法人熊本大学 画像解析装置、画像解析方法及び画像解析プログラム
WO2018064715A1 (en) * 2016-10-03 2018-04-12 Crc For Mental Health Ltd Method for predicting or diagnosing cognitive deterioration
JP2020503991A (ja) * 2016-12-06 2020-02-06 ダルミヤン，インク． 脳障害を特定するための方法及びシステム

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
KUDO, KOHSUKE: "MRI diagnosis of Alzheimer's disease using QSM", NUCLEAR MEDICINE TECHNOLOGY, vol. 38, 30 September 2018 (2018-09-30), pages 395, XP009534844, ISSN: 0289-100X *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023167157A1 (ja) * 2022-03-01 2023-09-07 株式会社Splink コンピュータプログラム、情報処理装置及び情報処理方法

Also Published As

Publication number Publication date
US11972564B2 (en) 2024-04-30
JP7489645B2 (ja) 2024-05-24
JPWO2022034691A1 (https=) 2022-02-17
US20230230233A1 (en) 2023-07-20

Similar Documents

Publication Publication Date Title
Schoonhoven et al. Tau protein spreads through functionally connected neurons in Alzheimer’s disease: a combined MEG/PET study
Miller et al. The diffeomorphometry of temporal lobe structures in preclinical Alzheimer's disease
US9940712B2 (en) Quantitating disease progression from the MRI images of multiple sclerosis patients
EP2790575B1 (en) Method and apparatus for the assessment of medical images
Imms et al. Navigating the link between processing speed and network communication in the human brain
Tian et al. Improving in vivo human cerebral cortical surface reconstruction using data-driven super-resolution
US20170238879A1 (en) Method of Analyzing the Brain Activity of a Subject
WO2018098213A1 (en) Medical image analysis using mechanical deformation information
CN111212600B (zh) 帕金森病诊断装置
CN116369891B (zh) 轻度认知障碍发展进程预测方法、装置和计算机设备
WO2009065079A2 (en) Longitudinal registration of anatomy in magnetic resonance imaging
EP1595205A1 (en) Tissue disorder imaging analysis
JP2014042684A (ja) 医用画像処理装置、およびプログラム
JP2022508684A (ja) ヒトの脳におけるドーパミン機能の非侵襲的代理測定としての神経メラニン感受性磁気共鳴画像法のためのシステム、方法及びコンピュータアクセス可能媒体
Goñi et al. Robust estimation of fractal measures for characterizing the structural complexity of the human brain: Optimization and reproducibility
WO2022054711A1 (ja) コンピュータプログラム、情報処理装置、端末装置、情報処理方法、学習済みモデル生成方法及び画像出力装置
Ono et al. Mesial temporal lobe epilepsy: Revisiting the relation of hippocampal volumetry with memory deficits
JP7489645B2 (ja) コンピュータプログラム、情報処理装置、情報処理方法、学習済みモデル生成方法及び相関画像出力装置
Wang et al. Semi-automatic segmentation of the fetal brain from magnetic resonance imaging
Nygaard et al. Spatiotemporal alterations of gray matter microstructure in newly diagnosed relapsing-remitting multiple sclerosis patients: A longitudinal diffusion kurtosis MRI study
Karahan et al. Individual variability in the human connectome maintains selective cross-modal consistency and shares microstructural signatures
Stangl et al. Population-level analysis of human grid cell activation
Nanayakkara et al. Increased brain volumetric measurement precision from multi-site 3D T1-weighted 3 T magnetic resonance imaging by correcting geometric distortions
Merhof et al. Non-linear integration of DTI-based fiber tracts into standard 3D MR data
CN119864156B (zh) 一种脑代谢网络震中检测方法及装置

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: 20949553

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022542567

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20949553

Country of ref document: EP

Kind code of ref document: A1