WO2021060462A1 - 画像処理装置、方法およびプログラム、学習装置、方法およびプログラム、並びに導出モデル - Google Patents

画像処理装置、方法およびプログラム、学習装置、方法およびプログラム、並びに導出モデル Download PDF

Info

Publication number
WO2021060462A1
WO2021060462A1 PCT/JP2020/036255 JP2020036255W WO2021060462A1 WO 2021060462 A1 WO2021060462 A1 WO 2021060462A1 JP 2020036255 W JP2020036255 W JP 2020036255W WO 2021060462 A1 WO2021060462 A1 WO 2021060462A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
mri
medical image
tissue
learning
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/JP2020/036255
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
篤志 橘
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.)
Fujifilm Corp
Original Assignee
Fujifilm Corp
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 Fujifilm Corp filed Critical Fujifilm Corp
Priority to JP2021549039A priority Critical patent/JP7334256B2/ja
Priority to EP20867297.2A priority patent/EP4035602A4/en
Publication of WO2021060462A1 publication Critical patent/WO2021060462A1/ja
Priority to US17/699,190 priority patent/US12217852B2/en
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
    • G06T12/00—Tomographic reconstruction from projections
    • G06T12/30—Image post-processing, e.g. metal artefact correction
    • 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G06F17/12—Simultaneous equations, e.g. systems of linear equations
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—Two-dimensional [2D] image generation
    • G06T11/10—Texturing; Colouring; Generation of textures or colours
    • 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • 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
    • G06T2207/10081—Computed x-ray tomography [CT]
    • 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
    • G06T2207/30088—Skin; Dermal
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2211/00—Image generation
    • G06T2211/40—Computed tomography
    • G06T2211/441—AI-based methods, deep learning or artificial neural networks

Definitions

  • the present disclosure relates to image processing devices, methods and programs, learning devices, methods and programs, and derivation models.
  • the acquired three-dimensional image has a different expression format depending on the device, that is, the modality. For example, even if the texture is the same, the density may be different or the contrast of the entire image may be different in the expression format. For this reason, lesions that were not visible in the 3D image acquired by one modality may be visible in the 3D image acquired by another modality. Therefore, the accuracy of diagnosis can be improved by acquiring three-dimensional images having different expression formats depending on a plurality of types of modality.
  • Non-Patent Document 1 a method of converting a CT image into an MRI image using a conversion model constructed by machine learning has been proposed (see Non-Patent Document 1 below).
  • the method described in Non-Patent Document 1 constructs a conversion model by machine learning using a CT image and a T2-weighted image of MRI as teacher data, and uses the constructed conversion model to convert a CT image into T2 of MRI. This is a method of converting to a weighted image.
  • T1 weighted by calculating the tissue eigenvalues derived by the methods described in JP-A-2019-005557, JP-A-2015-525604 and Non-Patent Document 2 by arithmetic formulas using various parameters.
  • MRI images in any representation format such as images, T2-weighted images and diffusion-weighted images can be derived. Therefore, it is not necessary to take a plurality of times to acquire MRI images having different expression formats, so that the burden on the patient and the medical cost can be reduced.
  • Non-Patent Document 1 Deep CT to MR Synthesis Using Paired and Unpaired Data, Sensors 2019.19 (10), 2361
  • Non-Patent Document 2 Rapid magnetic resonance quantification on the brain: Optimization for clinical usage, Magn Reson Med 2008.60 (2), 320-329
  • JP-A-2019-005557, JP-A-2015-525604 and Non-Patent Document 2 need to be photographed by an MRI apparatus in order to derive tissue eigenvalues.
  • MRI imaging is contraindicated in patients with implantable pacemakers and claustrophobic patients. Therefore, for such a patient, it is not possible to acquire an MRI image or acquire a tissue eigenvalue to derive an arbitrary type of MRI image.
  • a CT image can be converted into an MRI image by using the method described in Non-Patent Document 1.
  • the transformation model generated by learning can only derive MRI images in a single representation format. Therefore, in order to derive an MRI image having a plurality of expression formats from a CT image, it is necessary to construct a conversion model for each expression format of the MRI image, so that the load for constructing the conversion model is large. It becomes.
  • the present disclosure has been made in view of the above circumstances, and an object of the present disclosure is to make it possible to easily obtain an MRI image in a desired expression format.
  • the image processing apparatus includes an image acquisition unit that acquires at least one target medical image having an expression format different from that of an MRI image, and an image acquisition unit.
  • an image acquisition unit that acquires at least one target medical image having an expression format different from that of an MRI image
  • an image acquisition unit When at least one medical image having a different representation format from the MRI image is input, it is constructed by machine learning using a plurality of teacher data so as to output the tissue-specific value of MRI for the input medical image.
  • an organization-specific value derivation unit that has a derivation model, inputs a target medical image into the derivation model, and derives an organization-specific value for the target medical image.
  • the target medical image and the medical image may be CT images.
  • the target medical image and the medical image may be a plurality of CT images of the same subject acquired by using radiation having different energy distributions.
  • a plurality of CT images having different energy distributions which are acquired by photographing a subject using radiation having different energy distributions, can be acquired by, for example, a dual energy CT apparatus or a photon counting CT apparatus.
  • the derived model is a machine using a plurality of teacher data including a learning medical image including a specific structure of the subject and a learning tissue specific value for the same subject as the subject. It may be constructed by performing learning.
  • the image processing apparatus may further include an MRI image deriving unit that derives a predetermined type of MRI image by using the tissue eigenvalues.
  • the image processing apparatus may further include a display control unit that displays an MRI image on the display unit.
  • the display control unit may further display the target medical image on the display unit.
  • the learning apparatus is a machine using a plurality of teacher data so as to output an MRI tissue-specific value for a medical image when at least one medical image having a representation format different from that of the MRI image is input. It is equipped with a learning unit that builds a derived model by performing learning.
  • the derived model according to the present disclosure is a machine using a plurality of teacher data so that when at least one medical image having a representation format different from that of the MRI image is input, the tissue-specific value of MRI for the medical image is output. It is built by learning.
  • the image processing method acquires at least one target medical image having a representation format different from that of an MRI image, and obtains an image for medical use.
  • At least one medical image having a different representation format from the MRI image is input, it is constructed by machine learning using a plurality of teacher data so as to output the tissue-specific value of MRI for the input medical image. It has a derivation model, and the target medical image is input to the derivation model to derive the tissue-specific value for the target medical image.
  • the learning method according to the present disclosure is a machine using a plurality of teacher data so that when at least one medical image having a representation format different from that of the MRI image is input, the tissue-specific value of MRI for the medical image is output.
  • a derived model is constructed by training.
  • image processing method and the learning method according to the present disclosure may be provided as a program for executing the computer.
  • image processing devices include a memory for storing an instruction to be executed by a computer and a memory.
  • the processor comprises a processor configured to execute a stored instruction.
  • Acquire at least one target medical image having a representation format different from that of the MRI image When at least one medical image having a different representation format from the MRI image is input, it is constructed by machine learning using a plurality of teacher data so as to output the tissue-specific value of MRI for the input medical image. It has a derivation model, inputs the target medical image to the derivation model, and executes the process of deriving the tissue-specific value for the target medical image.
  • Other learning devices include a memory for storing instructions to be executed by a computer and a memory.
  • the processor comprises a processor configured to execute a stored instruction.
  • a derived model is performed by performing machine learning using a plurality of teacher data so as to output the tissue-specific value of MRI for the medical image. Executes the process of constructing.
  • a plurality of types of MRI images for the patient can be easily acquired without using a plurality of conversion models.
  • Schematic block diagram showing the configuration of the image processing apparatus according to the present embodiment Conceptual diagram of machine learning performed in this embodiment
  • the figure which shows the display screen of an MRI image Flowchart showing processing performed in this embodiment
  • FIG. 1 is a hardware configuration diagram showing an outline of a diagnostic support system to which the image processing apparatus according to the embodiment of the present disclosure is applied.
  • the image processing device 1, the modality 2, and the image storage server 3 according to the present embodiment are connected in a communicable state via the network 4.
  • Modality 2 is a device that generates a three-dimensional image showing the site by photographing the site to be diagnosed of the subject. Specifically, a CT device, an MRI device, and PET (Positron Emission Tomography). Equipment, etc.
  • the three-dimensional image composed of a plurality of slice images generated by the modality 2 is transmitted to the image storage server 3 and stored.
  • the modality 2 includes the CT device 2A and the MRI device 2B.
  • the CT device 2A is a dual energy CT device or a photon counting CT device capable of photographing a subject with radiation having a different energy distribution. It is assumed that the MRI apparatus 2B can derive the tissue eigenvalue of MRI by photographing the subject described in, for example, Japanese Patent Publication No. 2015-525604 and Non-Patent Document 2 described above.
  • the image storage server 3 is a computer that stores and manages various data, and is equipped with a large-capacity external storage device and database management software.
  • the image storage server 3 communicates with another device via a wired or wireless network 4 to send and receive image data and the like.
  • various data including image data of medical images generated by modality 2 are acquired via a network and stored in a recording medium such as a large-capacity external storage device for management.
  • the storage format of the image data and the communication between the devices via the network 4 are based on a protocol such as DICOM (Digital Imaging and Communication in Medicine).
  • DICOM Digital Imaging and Communication in Medicine
  • the image storage server 3 also stores and manages a plurality of teacher data described later.
  • the image processing device 1 of the present embodiment is a computer in which the image processing program and the learning program of the present embodiment are installed.
  • the computer may be a workstation or personal computer operated directly by the diagnosing doctor, or it may be a server computer connected to them via a network.
  • the image processing program and the learning program are stored in the storage device of the server computer connected to the network or the network storage in a state of being accessible from the outside, and are downloaded and installed on the computer used by the doctor upon request. ..
  • it is recorded and distributed on a recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory), and installed on a computer from the recording medium.
  • FIG. 2 is a diagram showing a schematic configuration of an image processing device realized by installing an image processing program and a learning program on a computer.
  • the image processing device 1 includes a CPU (Central Processing Unit) 11, a memory 12, and a storage 13 as a standard workstation configuration. Further, the image processing device 1 is connected to a display unit 14 such as a liquid crystal display and an input unit 15 such as a keyboard and a mouse.
  • a display unit 14 such as a liquid crystal display
  • an input unit 15 such as a keyboard and a mouse.
  • the storage 13 is composed of a hard disk drive or the like, and stores CT images to be processed, a plurality of teacher data acquired from the image storage server 3 via the network 4, and various information including information necessary for processing.
  • the image processing program and the learning program are stored in the memory 12.
  • the image processing program includes an image acquisition process for acquiring a target medical image having a representation format different from that of the MRI image, an tissue unique value derivation process for deriving an MRI tissue unique value for the target medical image, and an tissue unique value.
  • the learning program that defines the MRI image derivation process for deriving the MRI image in the desired expression format using the above and the display control process for displaying the derived MRI image on the display unit 14 is a process to be executed by the CPU 11.
  • the image acquisition unit 21 acquires a target medical image having an expression format different from that of the MRI image from the image storage server 3 via an interface (not shown) connected to the network.
  • a plurality of teacher data used for learning are acquired from the image storage server 3.
  • the target medical image is a CT image acquired by the CT device 2A.
  • the learning unit 25 performs machine learning for constructing the derivation model 30.
  • the learning unit 25 constructs a derivation model 30 by performing machine learning using teacher data composed of a combination of a learning CT image and a learning tissue eigenvalue.
  • the derivation model 30 is a convolutional neural network (hereinafter referred to as CNN (Convolutional Neural Network)), which is one of multi-layer neural networks in which a plurality of processing layers are hierarchically connected and deep learning is performed. ) And) shall be constructed by machine learning.
  • CNN Convolutional Neural Network
  • the CNN consists of a plurality of convolutional layers and pooling layers.
  • the convolutional layer performs convolutional processing using various kernels on the input image, and outputs a feature amount map consisting of feature amount data obtained by the convolutional processing.
  • the convolutional layer applies the kernel to the entire input image or the feature map output from the processing layer in the previous stage while shifting the attention pixels of the kernel. Further, the convolutional layer applies an activation function such as a sigmoid function to the convolutional value, and outputs a feature map.
  • the pooling layer reduces the amount of data in the feature map by pooling the feature map output by the convolutional layer, and outputs the feature map with the reduced amount of data.
  • the tissue eigenvalues for each pixel of the input CT image are output from the final layer of the CNN.
  • FIG. 3 is a conceptual diagram of machine learning performed in this embodiment. Although machine learning using two CT images will be described here, one CT image may be used or three or more CT images may be used.
  • the teacher data 40 is obtained by irradiating the subject with radiation having different energy distributions in the CT apparatus 2A, and the learning CT images 41A and 41B having different energy distributions and the MRI apparatus 2B.
  • the same subject as the subject from which the learning CT images 41A and 41B were acquired was photographed and acquired by the methods described in JP-A-2019-005557, JP-A-2015-525604, Non-Patent Document 2 and the like. It consists of a combination with tissue-specific values (that is, T1 value, T2 value and PD value) Us.
  • the learning unit 25 inputs the learning CT images 41A and 41B included in the teacher data to the CNN, and outputs the tissue-specific values Ux for the learning CT images 41A and 41B from the CNN.
  • the learning unit 25 derives the loss L0 based on the difference between the output tissue eigenvalue Ux and the learning tissue eigenvalue Us.
  • the loss L0 is the difference between the T1 value, T2 value, and PD value of the output tissue eigenvalue Ux and the T1 value, T2 value, and PD value of the learning tissue eigenvalue Us.
  • the learning unit 25 learns the CNN using a large amount of teacher data so that the loss L0 is equal to or less than a predetermined threshold value.
  • the number of convolutional layers constituting the CNN, the number of pooling layers, the coefficient of the kernel in the convolutional layer, the size of the kernel, etc. are derived so that the loss L0 is equal to or less than a predetermined threshold value.
  • CNN learning is performed.
  • the CNN outputs the tissue eigenvalues for each pixel of the CT image Ci.
  • the learning unit 25 may perform learning a predetermined number of times instead of learning so that the loss L0 is equal to or less than a predetermined threshold value.
  • the learning unit 25 performs machine learning of CNN, and when the CT image Ci is input, a derivation model 30 that outputs the tissue-specific value for each pixel of the CT image Ci is constructed.
  • the MRI image deriving unit 23 derives an MRI image using the tissue eigenvalues derived by the organization eigenvalue deriving unit 22.
  • the MRI image has various expression formats such as a T1-weighted image, a T2-weighted image, a fat-suppressed image, and a diffusion-weighted image, and the contrast, that is, the appearance differs depending on the expression format.
  • T1-weighted images predominantly adipose tissue appears white, water, humoral components and cysts appear black, and tumors appear slightly black.
  • T2-weighted images not only adipose tissue, but also water, humoral components and cysts appear white.
  • These various representations of MRI images can be derived by calculating tissue eigenvalues, i.e. T1 values, T2 values and PD values using predetermined parameters. Specifically, parameters such as repetition time TI (Inversion Time), echo time TE (Echo Time), and repetition time TR (Repetition Time) according to the expression format were applied to the T1 value, T2 value, and PD value. By performing the calculation according to a predetermined calculation formula, it is possible to generate an MRI image in a desired expression format.
  • a table and an arithmetic expression that define the relationship between the expression formats of various MRI images and various parameters are stored in the storage 13.
  • the MRI image derivation unit 23 reads out from the storage 13 parameters corresponding to the expression format of the MRI image input in advance from the input unit 15. Then, the MRI image M0 in the input expression format is derived by performing an operation on the T1 value, the T2 value, and the PD value using the read parameters.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Radiology & Medical Imaging (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Public Health (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Pathology (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Veterinary Medicine (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Operations Research (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
PCT/JP2020/036255 2019-09-27 2020-09-25 画像処理装置、方法およびプログラム、学習装置、方法およびプログラム、並びに導出モデル Ceased WO2021060462A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2021549039A JP7334256B2 (ja) 2019-09-27 2020-09-25 画像処理装置、方法およびプログラム、学習装置、方法およびプログラム、並びに導出モデル
EP20867297.2A EP4035602A4 (en) 2019-09-27 2020-09-25 IMAGE PROCESSING DEVICE, METHOD AND PROGRAM, LEARNING DEVICE, METHOD AND PROGRAM, AND DERIVATION MODEL
US17/699,190 US12217852B2 (en) 2019-09-27 2022-03-21 Image processing device, image processing method, image processing program, learning device, learning method, learning program, and derivation model

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2019177220 2019-09-27
JP2019-177220 2019-09-27

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/699,190 Continuation US12217852B2 (en) 2019-09-27 2022-03-21 Image processing device, image processing method, image processing program, learning device, learning method, learning program, and derivation model

Publications (1)

Publication Number Publication Date
WO2021060462A1 true WO2021060462A1 (ja) 2021-04-01

Family

ID=75165241

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/036255 Ceased WO2021060462A1 (ja) 2019-09-27 2020-09-25 画像処理装置、方法およびプログラム、学習装置、方法およびプログラム、並びに導出モデル

Country Status (4)

Country Link
US (1) US12217852B2 (https=)
EP (1) EP4035602A4 (https=)
JP (1) JP7334256B2 (https=)
WO (1) WO2021060462A1 (https=)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20230163210A (ko) * 2022-05-23 2023-11-30 메디컬아이피 주식회사 Ct 영상을 이용한 지방량 측정 방법 및 그 장치

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE112021003110T5 (de) * 2020-06-04 2023-03-23 Fujifilm Corporation Bilderzeugungsvorrichtung, bilderzeugungsverfahren,bilderzeugungsprogramm, lernvorrichtung, lernverfahren und lernprogramm
EP4336452A1 (en) * 2022-09-09 2024-03-13 Koninklijke Philips N.V. Computer-implemented method for processing spectral computed tomography (spectral ct) data, computer program and spectral ct system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015525604A (ja) 2012-07-02 2015-09-07 シンセティックエムアール・エイビイ 改善された磁気共鳴収集のための方法およびシステム
JP2017086337A (ja) * 2015-11-06 2017-05-25 東芝メディカルシステムズ株式会社 磁気共鳴イメージング装置及び画像生成方法
JP2018505705A (ja) * 2014-12-10 2018-03-01 コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． 機械学習を用いた医用イメージングの変換のためのシステムおよび方法
JP2018192264A (ja) * 2017-05-18 2018-12-06 キヤノンメディカルシステムズ株式会社 医用画像処理装置
JP2019005557A (ja) 2017-06-22 2019-01-17 キヤノンメディカルシステムズ株式会社 画像処理装置、磁気共鳴イメージング装置及び画像処理プログラム

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10073156B2 (en) 2012-07-02 2018-09-11 Syntheticmr Ab Methods and systems for improved magnetic resonance acquisition
WO2018048507A1 (en) * 2016-09-06 2018-03-15 Han Xiao Neural network for generating synthetic medical images
EP3475916B1 (en) * 2016-11-30 2019-10-23 Koninklijke Philips N.V. Bone and hard plaque segmentation in spectral ct
JP6979151B2 (ja) * 2017-05-26 2021-12-08 富士フイルムヘルスケア株式会社 磁気共鳴イメージング装置及び磁気共鳴画像処理方法
US10769823B2 (en) 2017-06-22 2020-09-08 Canon Medical Systems Corporation Image processing apparatus, magnetic resonance imaging apparatus, and storage medium
US20210225491A1 (en) * 2017-11-17 2021-07-22 Young Saem AHN Diagnostic image converting apparatus, diagnostic image converting module generating apparatus, diagnostic image recording apparatus, diagnostic image converting method, diagnostic image converting module generating method, diagnostic image recording method, and computer recordable recording medium
JP2021533451A (ja) * 2018-07-29 2021-12-02 ゼブラ メディカル ビジョン リミテッド 医療画像内の視覚的オブジェクトの自動検出のためのシステムおよび方法
WO2020054803A1 (ja) * 2018-09-12 2020-03-19 株式会社Ｓｐｌｉｎｋ 診断支援システムおよび方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015525604A (ja) 2012-07-02 2015-09-07 シンセティックエムアール・エイビイ 改善された磁気共鳴収集のための方法およびシステム
JP2018505705A (ja) * 2014-12-10 2018-03-01 コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． 機械学習を用いた医用イメージングの変換のためのシステムおよび方法
JP2017086337A (ja) * 2015-11-06 2017-05-25 東芝メディカルシステムズ株式会社 磁気共鳴イメージング装置及び画像生成方法
JP2018192264A (ja) * 2017-05-18 2018-12-06 キヤノンメディカルシステムズ株式会社 医用画像処理装置
JP2019005557A (ja) 2017-06-22 2019-01-17 キヤノンメディカルシステムズ株式会社 画像処理装置、磁気共鳴イメージング装置及び画像処理プログラム

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
ANDREW PALMERA LEYNES, JAEWON YANG, DATTESH D SHANBHAG, SANDEEP S KAUSHIK, FLORIAN WIESINGER, YOUNGHO SEO, THOMAS ARMSTRONG HOPE, : "Direct Pseudo-CT Image Synthesis Using Deep Learning for Pelvis PET/MR Attenuation Correction", PROCEEDINGS OF THE ISMRM 25TH ANNUAL MEETING & EXHIBITION; APRIL 22-27, 2017; HONOLULU, HI, USA, vol. 25, 30 November 2016 (2016-11-30), pages 5661, XP009534675 *
See also references of EP4035602A4
YICHENG CHEN ET AL.: "MR-based PET Attenuation Correction for Brain PET-MR Using Support Vector Machines", PROC. INTL. SOC. MAG. RESON. MED., vol. 23, 2015, XP055813116 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20230163210A (ko) * 2022-05-23 2023-11-30 메디컬아이피 주식회사 Ct 영상을 이용한 지방량 측정 방법 및 그 장치
KR102853901B1 (ko) * 2022-05-23 2025-09-02 메디컬아이피 주식회사 Ct 영상을 이용한 지방량 측정 방법 및 그 장치

Also Published As

Publication number Publication date
JP7334256B2 (ja) 2023-08-28
JPWO2021060462A1 (https=) 2021-04-01
EP4035602A1 (en) 2022-08-03
US20220215936A1 (en) 2022-07-07
US12217852B2 (en) 2025-02-04
EP4035602A4 (en) 2022-11-23

Similar Documents

Publication Publication Date Title
US12125175B2 (en) Methods and system for selective removal of streak artifacts and noise from images using deep neural networks
KR101659578B1 (ko) 자기 공명 영상 처리 방법 및 장치
US12217852B2 (en) Image processing device, image processing method, image processing program, learning device, learning method, learning program, and derivation model
JP7129869B2 (ja) 疾患領域抽出装置、方法及びプログラム
JP7559135B2 (ja) データ処理装置及び方法
JP7170897B2 (ja) 学習装置、方法およびプログラム、画像生成装置、方法およびプログラム、並びに画像生成モデル
Roy et al. Fetal XCMR: a numerical phantom for fetal cardiovascular magnetic resonance imaging
US20190378271A1 (en) Medical information processing apparatus, magnetic resonance imaging apparatus, and medical information processing method
US20220248973A1 (en) Deep learning of eletrical properties tomography
JP2021087629A (ja) 医用データ処理装置
US11176413B2 (en) Apparatus, method, and program for training discriminator discriminating disease region, discriminator discriminating disease region, disease region discrimination apparatus, and disease region discrimination program
WO2020054188A1 (ja) 医用画像処理装置、方法およびプログラム
Oved et al. Deep learning of personalized priors from past MRI scans enables fast, quality-enhanced point-of-care MRI with low-cost systems
JP2022039197A (ja) データ再構成装置、データ再構成方法およびデータ再構成プログラム
JP2020006163A (ja) 医用情報処理装置、方法及びプログラム
KR102090690B1 (ko) 인공신경망을 이용한 자기 공명 영상의 영상 프로토콜 선택 장치와 방법 및 프로그램이 기록된 컴퓨터 판독 가능한 기록매체
CN112292067B (zh) 电导率变化的电学属性断层摄影图
US20220301687A1 (en) Uncertainty maps for deep learning eletrical properties tomography
Netreba et al. Correlation reconstructed spine and time relaxation spatial distribution of atomic systems in MRI
US12217488B2 (en) Learning image generation device, learning image generation method, learning image generation program, learning method, learning device, and learning program
KR102840437B1 (ko) 컴퓨팅 장치가 이미지 데이터에 대한 분류 정보를 출력하는 방법 및 이를 위한 장치
Reddy GRAPPA reconstruction using quantum computing: a thesis in Computer Science

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: 20867297

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021549039

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2020867297

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2020867297

Country of ref document: EP

Effective date: 20220428