WO2022197074A1 - Appareil de traitement d'image médicale, méthode d'apprentissage d'image médicale et méthode de traitement d'image médicale - Google Patents

Appareil de traitement d'image médicale, méthode d'apprentissage d'image médicale et méthode de traitement d'image médicale Download PDF

Info

Publication number
WO2022197074A1
WO2022197074A1 PCT/KR2022/003627 KR2022003627W WO2022197074A1 WO 2022197074 A1 WO2022197074 A1 WO 2022197074A1 KR 2022003627 W KR2022003627 W KR 2022003627W WO 2022197074 A1 WO2022197074 A1 WO 2022197074A1
Authority
WO
WIPO (PCT)
Prior art keywords
artificial
neural network
network model
medical image
type
Prior art date
Application number
PCT/KR2022/003627
Other languages
English (en)
Korean (ko)
Inventor
양동현
이준구
Original Assignee
재단법인 아산사회복지재단
울산대학교 산학협력단
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 재단법인 아산사회복지재단, 울산대학교 산학협력단 filed Critical 재단법인 아산사회복지재단
Publication of WO2022197074A1 publication Critical patent/WO2022197074A1/fr

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
    • A61B6/5217—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data extracting a diagnostic or physiological parameter from medical diagnostic data
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/12—Arrangements for detecting or locating foreign bodies
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/503—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of the heart
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/504—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of blood vessels, e.g. by angiography
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10116—X-ray image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30048—Heart; Cardiac
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
    • G06T2207/30104—Vascular flow; Blood flow; Perfusion

Definitions

  • the present invention relates to a method and apparatus for learning a medical image of a body, and a method and apparatus for processing a medical image of the body.
  • a medical image refers to an image obtained by using a medical imaging device to include information on the internal structure of the body of a subject to be diagnosed.
  • a medical imaging apparatus used to acquire such a medical image is a non-invasive examination apparatus, and displays structural details in the body, internal tissues and fluid flow, etc. by photographing and processing the image to the user.
  • a user such as a doctor may diagnose a patient's health condition and disease by using a medical image output from the medical imaging apparatus.
  • a medical imaging device an X-ray imaging device that irradiates X-rays to an object and detects X-rays passing through the object to image an image, and a magnetic resonance imaging device for providing a magnetic resonance image.
  • MRI magnetic resonance imaging
  • CT computed tomography
  • ultrasound Ultrasound
  • a prosthetic valve is implanted in a human or animal body and can be used, for example, as a passive one direct prosthetic valve in or near a blood vessel.
  • the prosthetic valve may be completely preformed and implanted as is, or it may be formed in situ using artificial and/or natural components necessary to form a functional prosthetic valve.
  • These artificial valves can be broadly divided into mechanical valves and tissue valves according to materials, and depending on the location, a mitral valve, an aortic valve, a tricuspid valve, and It can be divided into pulmonary valves, etc.
  • the artificial valves are also photographed in medical images taken of humans or animals with artificial valves, and medical practices, etc. For this, it is necessary to distinguish the types.
  • the types of artificial valves included in the medical images were classified depending on the doctor's visual observation of the medical images. Accordingly, there was a problem in that the accuracy of classifying the types of artificial valves included in the medical image was changed.
  • a medical image processing method and apparatus capable of accurately distinguishing types of artificial valves included in a medical image by using an artificial neural network model and a learning technique thereof.
  • the medical image processing method performed by the medical image processing apparatus includes inputting a body X-ray image of a heart of a subject to be diagnosed into a pre-trained first artificial neural network model, and the first artificial neural network model is estimating the type of the artificial valve in the heart of the body X-ray image as an output and obtaining shape information related to the artificial valve; and determining the type of the artificial valve as a result of re-estimating the type of the artificial valve based on the blood flow direction.
  • a medical image processing method includes inputting a body X-ray image of a heart of a subject to be diagnosed into a pre-trained first artificial neural network model, and the body X-ray image as an output of the first artificial neural network model. Estimating the type of the artificial valve in the heart and obtaining shape information related to the artificial valve, and learning the shape information related to the artificial valve and the type of the artificial valve provided by the first artificial neural network model inputting to a second artificial neural network model, the blood flow direction estimated based on the shape information related to the artificial valve by the second artificial neural network model, and the type of the artificial valve provided by the first artificial neural network model and determining the re-estimated result as the type of the artificial valve based on the .
  • instructions for causing the processor to perform the medical image processing method performed by the medical image processing apparatus include
  • a computer program stored in a computer-readable recording medium includes instructions for, when executed by a processor, to cause the processor to perform a medical image processing method performed by the medical image processing apparatus.
  • a method for learning a medical image of a medical image processing apparatus comprises the steps of: preparing body X-ray images for learning of a heart as an input constituting a first learning data set, and a label constituting the first learning data set Preparing the type information of the artificial valve and the shape information related to the artificial valve respectively corresponding to the learning body X-ray images as .
  • a method for learning a medical image of a medical image processing apparatus includes the steps of preparing body X-ray images for learning of a heart as an input constituting a first learning data set, and a label constituting the first learning data set preparing information on types of artificial valves and shape information related to the artificial valves respectively corresponding to the X-ray images of the body for learning, and a plurality of shape information and the plurality of shapes as inputs constituting a second learning data set Preparing cardiac blood flow direction information corresponding to the information, and preparing information on the type of artificial valve corresponding to the plurality of shape information and the cardiac blood flow direction information as a label constituting the second learning data set; and training a first layer of the artificial neural network model using the first training data set and training a second layer of the artificial neural network model using the second training data set.
  • the computer program of the computer-readable recording medium storing the computer program includes instructions for, when executed by a processor, to cause the processor to perform the method for learning a medical image of the medical image processing apparatus. .
  • a computer program stored in a computer-readable recording medium includes instructions for, when executed by a processor, to cause the processor to perform a method for learning a medical image of the medical image processing apparatus.
  • a medical image processing apparatus includes: a memory storing a medical image processing program; and a processor that loads the medical image processing program from the memory and executes the medical image processing program, wherein the processor receives a body X-ray image of a heart of a diagnosed subject, and receives the input of the received medical image processing program.
  • a first artificial neural network model estimates the type of artificial valve in the body X-ray image from the body X-ray image, obtains shape information related to the artificial valve, and calculates the estimated type of artificial valve and the obtained shape information. As a result of re-estimating the type of the artificial valve based on the estimated blood flow direction, the type of the artificial valve is determined, and information on the determined type of the artificial valve is output.
  • the processor may compare the acquired shape information with preset shape information, and estimate the blood flow direction of the heart based on a result of the comparison.
  • a medical image processing apparatus includes: a memory storing a medical image processing program; and a processor that loads the medical image processing program from the memory and executes the medical image processing program, wherein the processor receives a body X-ray image of a heart of a diagnosed subject, and receives the input of the received medical image processing program.
  • a first artificial neural network model estimates the type of artificial valve in the body X-ray image from the body X-ray image, obtains shape information related to the artificial valve, and the type of the artificial valve output by the first artificial neural network model and input the shape information to the second artificial neural network model learned in advance, the blood flow direction estimated based on the shape information by the second artificial neural network model, and the type of the artificial valve output by the first artificial neural network model Based on the re-estimated result, it is determined and output as the type of the artificial valve.
  • a medical image learning apparatus includes: a memory storing a medical image learning program; and a processor that loads the medical image processing program from the memory and executes the medical image learning program, wherein the processor is an input constituting a first learning data set for learning body X-rays for the heart.
  • Prepare images prepare information on types of artificial valves corresponding to each of the body X-ray images for training and shape information related to the artificial valves as labels constituting the first training data set, and prepare the first training data set can be used to train an artificial neural network model.
  • a medical image learning apparatus includes: a memory storing a medical image learning program; and a processor that loads the medical image processing program from the memory and executes the medical image learning program, wherein the processor is an input constituting a first learning data set for learning body X-rays for the heart.
  • a plurality of shape information and cardiac blood flow direction information corresponding to the plurality of shape information are prepared as a constituting input, and a label constituting the second learning data set corresponds to the plurality of shape information and the cardiac blood flow direction information.
  • Prepare information on the type of artificial valve, train the first layer of the artificial neural network model using the first training data set, and train the second layer of the artificial neural network model using the second training data set can
  • types of artificial valves included in a medical image may be accurately distinguished by using an artificial neural network model and a learning technique thereof.
  • an artificial neural network model By providing the medical personnel with information on the types of artificial valves classified using the artificial neural network model, it has the effect of supporting the medical personnel to accurately determine the end of the artificial valve included in the body X-ray image.
  • FIG. 1 is a block diagram of a medical image processing apparatus according to an exemplary embodiment.
  • FIG. 2 is a flowchart illustrating a process of learning an artificial neural network model of an artificial neural network model unit of a medical image processing apparatus according to a first embodiment of the present invention.
  • FIG. 3 is a flowchart illustrating a process in which the medical image processing apparatus including the artificial neural network model learned according to the first embodiment of the present invention determines the type of the artificial valve in the heart of a diagnosis subject.
  • FIG. 4 is a flowchart illustrating a process of learning the artificial neural network model of the artificial neural network model unit of the medical image processing apparatus according to the second embodiment of the present invention.
  • 5 is a view showing the shape of a shadow for each type of artificial valve.
  • 'unit' used in the specification means software or a hardware component such as an FPGA or ASIC, and 'unit' performs certain roles.
  • 'part' is not limited to software or hardware.
  • the 'unit' may be configured to reside on an addressable storage medium or it may be configured to refresh one or more processors.
  • 'part' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. Functions provided within components and 'units' may be combined into a smaller number of components and 'units' or further divided into additional components and 'units'.
  • a subject or patient to be diagnosed may include a human or an animal, or a part of a human or animal.
  • 'image' may mean multi-dimensional data composed of discrete image elements (eg, pixels in a two-dimensional image and voxels in a three-dimensional image). have.
  • the medical image processing apparatus, the medical image learning method and the medical image processing method of the present invention can be implemented in various embodiments, and in the following description, the first embodiment and the second embodiment will be representatively described.
  • 1 is a block diagram of a medical image processing apparatus 100 according to an embodiment of the present invention, and the illustrated medical image processing apparatus 100 performs a medical image learning method and a medical image processing method implemented in various forms.
  • the description of the medical image processing apparatus 100 will be divided into a first embodiment and a second embodiment.
  • FIG. 1 shows a configuration of a medical image processing apparatus 100 according to an exemplary embodiment, but what is shown in FIG. 1 is merely exemplary.
  • the medical image processing apparatus 100 may be implemented in a PC or a server or may include the same.
  • the medical image processing apparatus 100 includes an input unit 110 , an artificial neutral network model unit 120 , a processing unit 130 , and an output unit 140 .
  • the input unit 110 receives body X-ray images for heart training as an input constituting the training data set.
  • the input unit 110 receives type information of the artificial valve in the heart of the body X-ray image and shape information related to the artificial valve as a label constituting the learning data set.
  • the input unit 110 receives a body X-ray image of the heart of the subject to be diagnosed.
  • the training data set is to be used when training the artificial neural network model unit 120 , and the body X-ray image of the heart of the diagnosis subject is input into the artificial neural network model unit 120 that has been previously learned and the type of artificial valve. is to determine
  • the artificial neural network model unit 120 learns the training data set received through the input unit 110 . And, the artificial neural network model unit 120 learned through the input unit 110, the artificial neural network model from the body X-ray image of the subject to be diagnosed, estimating the type of artificial valve in the body X-ray image, the form related to the artificial valve get information
  • the artificial neural network model unit 120 may obtain information about the shadow shape of the artificial valve as shape information related to the artificial valve.
  • the artificial neural network model may be a Mask R-CNN model.
  • the Mask R-CNN model can use body X-ray images for heart training as training images of the Mask R-CNN model, and information on the type of artificial valve in the heart of the body X-ray image corresponding to each of the training body X-ray images. And shape information related to the artificial valve can be used as a mask of the Mask R-CNN model.
  • the artificial neural network model unit 120 may include a memory for storing instructions programmed to perform a function as an artificial neural network model, and a microprocessor for executing these instructions.
  • the processing unit 130 estimates the blood flow direction of the heart based on the shape information related to the artificial valve obtained by the artificial neural network model unit 120 , and the type and estimation of the artificial valve estimated by the artificial neural network model unit 120 .
  • the type of artificial valve is determined as a result of re-estimating the type of artificial valve based on the direction of blood flow.
  • the processor 130 may estimate the blood flow direction of the heart based on a result of comparing the obtained shape information with preset shape information.
  • the preset shape information may be a plurality of crown shapes, and when estimating the blood flow direction, the processor 130 may estimate the blood flow direction for any one of the tissue valves respectively corresponding to the plurality of crown shapes.
  • the processing unit 130 performs a function of determining the type of artificial valve as a result of re-estimating the type of artificial valve based on the type of artificial valve estimated by the artificial neural network model unit 120 and the estimated blood flow direction.
  • memory for storing instructions programmed to do so and a microprocessor for executing such instructions.
  • the output unit 140 outputs information on the type of artificial valve determined by the processing unit 130 .
  • the output unit 140 may output information about the type of artificial valve determined by the processing unit 130 so that it can be recognized from the outside.
  • the output unit 140 may include a port for outputting information on the type of artificial valve, a wired communication module, or a wireless communication module.
  • the output unit 140 may include an image display device capable of outputting information on the type of the artificial valve in the form of an image.
  • FIG. 2 is a flowchart illustrating a process of learning an artificial neural network model of the artificial neural network model unit 120 of the medical image processing apparatus 100 according to a first embodiment of the present invention
  • FIG. 3 is a first embodiment of the present invention. It is a flowchart illustrating a process in which the medical image processing apparatus 100 including the artificial neural network model learned according to the method determines the type of the artificial valve in the heart of the patient to be diagnosed.
  • the operator of the medical image processing apparatus 100 may prepare body X-ray images for learning of the heart as an input constituting the learning data set (S210), and the body X-ray image as a label constituting the learning data set information on the type of artificial valve in the heart and information on the shape related to the artificial valve may be prepared (S220).
  • the shape information related to the artificial valve included in the training data set as the label may be information about the shadow shape of the artificial valve.
  • the artificial neural network model of the artificial neural network model unit 120 learns the training data set ( S230 ).
  • the number of times the artificial neural network model unit 120 learns the training data set is not particularly limited, but as one embodiment, learning may be performed within an appropriate number of times for the best learning effect.
  • an operator of the medical image processing apparatus 100 After learning of the artificial neural network model unit 120 of the medical image processing apparatus 100 is performed, an operator of the medical image processing apparatus 100 prepares a body X-ray image of a subject to be diagnosed and inputs it through the input unit 110 . can do.
  • the artificial neural network model unit 120 learned through steps S210 to S230 estimates the type of artificial valve in the body X-ray image of the artificial neural network model from the body X-ray image of the diagnosis subject received through the input unit 110 . and acquires shape information related to the artificial valve.
  • the artificial neural network model unit 120 may acquire information about the shadow shape of the artificial valve as shape information related to the artificial valve ( S310 ).
  • the type of artificial valve estimated by the artificial neural network model unit 120 through step S310 may or may not coincide with the actual type of the artificial valve existing in the body of the subject to be diagnosed.
  • Several artificial valves existing in the body of the diagnosis subject may be included in the X-ray image in a similar position, or in the form of overlapping a plurality of artificial valves in a certain direction due to an enlarged heart or a change in body shape, in this case, in step S310
  • the processing unit 130 of the medical image processing apparatus 100 estimates the blood flow direction of the heart based on the shape information related to the artificial valve obtained in step S310.
  • the reason that the processing unit 130 estimates the blood flow direction of the heart is to more accurately classify the types of artificial valves by reflecting the estimated blood flow direction of the heart. is likely to decrease.
  • the processing unit 130 compares the shape information related to the artificial valve acquired by the artificial neural network model unit 120 with preset shape information in step S310 to check whether the acquired shape information is included in the preset shape information.
  • the preset shape information may be a plurality of crown shapes, that is, various crown shapes.
  • 5 is a view showing the shape of a shadow for each type of artificial valve. 5, it can be seen that among the tissue valves, the aortic valve (Aortic Bio), the pulmonary valve (Pulmonary Bio), the mitral valve (Mitral Bio), and the tricuspid valve (Tricuspid Bio) have a crown-shaped shadow.
  • the processing unit 130 estimates the blood flow direction of the heart based on a result of comparing the shape information related to the artificial valve acquired by the artificial neural network model unit 120 with preset shape information in step S310 . For example, whether the blood flows from the lower side to the upper right side of the heart or the blood flow from the right side to the left side of the heart may be estimated according to which crown shape among the plurality of crown shapes ( S320 ).
  • the processing unit 130 re-estimates the type of the artificial valve in the heart of the body X-ray image input in step S310 based on the type of artificial valve estimated in step S310 and the blood flow direction estimated in step S320. to determine the type of artificial valve. If it is confirmed in step S320 that the shape information related to the artificial valve is not included in the preset shape information, the result estimated in step S310 is finally determined as the type of the artificial valve as it is. However, when it is confirmed in step S320 that the shape information related to the artificial valve is included in the preset shape information, the type of the artificial valve can be corrected by reflecting the estimated blood flow direction instead of using the result estimated in step S310 as it is. .
  • the type of the artificial valve may be finally determined as the aortic valve regardless of the estimation result of step S310, and the blood flow estimated as the blood flow from the right side to the left side of the heart In this case, the type of the artificial valve may be finally determined as the tricuspid valve regardless of the estimation result of step S310 (S330).
  • the output unit 140 outputs information on the type of the artificial valve determined by the processing unit 130 .
  • the output unit 140 outputs information on the type of artificial valve determined by the processing unit 130 in the form of an externally recognizable image, or as a separate electronic device through a wired communication module or a wireless communication module. can send
  • the second embodiment can be said to be a case in which the artificial neural network model unit 120 performs until the processing unit 130 re-estimates the type of the artificial valve in the heart in the first embodiment.
  • FIG. 1 shows a configuration of a medical image processing apparatus 100 according to an exemplary embodiment, but what is shown in FIG. 1 is merely exemplary.
  • the medical image processing apparatus 100 may be implemented in a PC or a server or may include the same.
  • the medical image processing apparatus 100 includes an input unit 110 , an artificial neural network model unit 120 , a processing unit 130 , and an output unit 140 .
  • the input unit 110 receives body X-ray images for heart training as an input constituting the first training data set.
  • the input unit 110 receives type information of the artificial valve in the heart of the body X-ray image and shape information related to the artificial valve as a label constituting the first learning data set.
  • the input unit 110 receives a plurality of shape information and cardiac blood flow direction information corresponding to the plurality of shape information as inputs constituting the second learning data set.
  • the input unit 110 receives information on types of artificial valves corresponding to a plurality of shape information and cardiac blood flow direction information as labels constituting the second learning data set.
  • the shape information related to the artificial valve included as a label of the first training data set may be information about the shadow shape of the artificial valve.
  • the plurality of shape information may be a plurality of crown shapes
  • the heart blood flow direction information may be a blood flow direction for any one of the tissue valves respectively corresponding to the plurality of crown shapes.
  • the input unit 110 receives a body X-ray image of the heart of the subject to be diagnosed.
  • the first training data set and the second training data set are to be used when training the artificial neural network model unit 120
  • the body X-ray image of the heart of the subject to be diagnosed is the previously learned artificial neural network model unit 120 . ) to determine the type of artificial valve.
  • the first layer of the artificial neural network model learns the first training data set input through the input unit 110 , and uses the second learning data set input through the input unit 110 to model the artificial neural network model.
  • the second layer of And, the artificial neural network model unit 120 including the learned artificial neural network model is the first layer of the artificial neural network model from the body X-ray image of the subject to be diagnosed, input through the input unit 110, of the artificial valve in the body X-ray image. Estimate the type and obtain information about the shape related to the artificial valve. Then, the second layer of the artificial neural network model estimates the blood flow direction of the heart based on the obtained shape information related to the artificial valve, and the first layer estimates the type of artificial valve and the estimated blood flow direction of the artificial valve.
  • the type of artificial valve is determined.
  • the artificial neural network model may be a Mask R-CNN model.
  • the artificial neural network model unit 120 may include a memory for storing instructions programmed to perform a function as an artificial neural network model, and a microprocessor for executing these instructions.
  • the processing unit 130 controls the output unit 140 to output the type of artificial valve determined by the artificial neural network model unit 120 .
  • the processing unit 130 may include a microprocessor that executes various instructions.
  • the output unit 140 outputs information on the type of the artificial valve under the control of the processing unit 130 .
  • the output unit 140 may output information about the type of artificial valve determined by the processing unit 130 so that it can be recognized from the outside.
  • the output unit 140 may include a port for outputting information on the type of artificial valve, a wired communication module, or a wireless communication module.
  • the output unit 140 may include an image display device capable of outputting information on the type of the artificial valve in the form of an image.
  • FIG. 4 is a flowchart illustrating a process of learning the artificial neural network model of the artificial neural network model unit 120 of the medical image processing apparatus 100 according to the second embodiment of the present invention.
  • the operator of the medical image processing apparatus 100 may prepare body X-ray images for heart training as an input constituting the first learning data set (S410), and as a label constituting the first learning data set Information on the type of the artificial valve in the heart of the body X-ray image and information on the shape related to the artificial valve may be prepared ( S420 ).
  • the shape information related to the artificial valve included in the training data set as the label may be information about the shadow shape of the artificial valve.
  • the operator of the medical image processing apparatus 100 may prepare a plurality of shape information and cardiac blood flow direction information corresponding to the plurality of shape information as an input constituting the second learning data set (S430), and the second As a label constituting the learning data set, information on types of artificial valves corresponding to a plurality of shape information and cardiac blood flow direction information may be prepared ( S440 ).
  • the first layer of the artificial neural network model learns the first training data set
  • the second training data set of the artificial neural network model The second layer learns.
  • the number of times the artificial neural network model unit 120 learns the training data set is not particularly limited, but as one embodiment, learning may be performed within an appropriate number of times for the best learning effect (S450).
  • an operator of the medical image processing apparatus 100 After learning of the artificial neural network model unit 120 of the medical image processing apparatus 100 is performed, an operator of the medical image processing apparatus 100 prepares a body X-ray image of a subject to be diagnosed and inputs it through the input unit 110 . can do.
  • the artificial neural network model unit 120 learned through steps S410 to S450 determines the type of artificial valve in the body X-ray image by the artificial neural network model from the body X-ray image of the diagnosis subject received through the input unit 110 . do.
  • the process of the artificial neural network model unit 120 determining the type of artificial valve in the body X-ray image is that steps S310 to S330 of the first embodiment are sequentially performed by the first layer and the second layer of the artificial neural network model. Since it can be done and can be inferred from the above descriptions, a detailed description thereof will be omitted.
  • the output unit 140 outputs information on the type of artificial valve determined by the artificial neural network model unit 120 under the control of the processing unit 130 .
  • the output unit 140 outputs information on the type of artificial valve determined by the processing unit 130 in the form of an externally recognizable image, or as a separate electronic device through a wired communication module or a wireless communication module. can send
  • each step included in the medical image learning method and the medical image processing method according to the first and second embodiments described above is performed in a computer-readable recording medium for recording a computer program programmed to perform these steps. can be implemented.
  • each step included in the medical image learning method and the medical image processing method according to the above-described first and second embodiments is a computer program stored in a computer-readable recording medium programmed to perform these steps. It can be implemented in the form
  • the above-described input unit 110 , artificial neural network model unit 120 , processing unit 130 , and output unit 140 may be implemented by a hardware device such as a processor.
  • embodiments of the present invention may be implemented by hardware, firmware, software, or a combination thereof.
  • the method according to embodiments of the present invention may include one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), FPGAs (Field Programmable Gate Arrays), may be implemented by a processor, a controller, a microcontroller, a microprocessor, and the like.
  • ASICs Application Specific Integrated Circuits
  • DSPs Digital Signal Processors
  • DSPDs Digital Signal Processing Devices
  • PLDs Programmable Logic Devices
  • FPGAs Field Programmable Gate Arrays
  • the method according to the embodiments of the present invention may be implemented in the form of a module, procedure, or function that performs the functions or operations described above.
  • the software code may be stored in the memory unit and driven by the processor.
  • the memory unit may be located inside or outside the processor, and may transmit and receive data to and from the processor by various known means.
  • Combinations of each step in each flowchart attached to the present invention may be performed by computer program instructions.
  • These computer program instructions may be embodied in a processor of a general purpose computer, special purpose computer, or other programmable data processing equipment, such that the instructions performed by the processor of the computer or other programmable data processing equipment provide the functions described in each step of the flowchart. It creates a means to do these things.
  • These computer program instructions may also be stored in a computer-usable or computer-readable medium that may direct a computer or other programmable data processing equipment to implement a function in a particular manner, and thus the computer-usable or computer-readable medium.
  • the instructions stored in the recording medium are also capable of producing an article of manufacture including instruction means for performing the functions described in each step of the flowchart.
  • the computer program instructions may also be mounted on a computer or other programmable data processing equipment, such that a series of operational steps are performed on the computer or other programmable data processing equipment to create a computer-executed process to create a computer or other programmable data processing equipment. It is also possible that instructions for performing the processing equipment provide steps for performing the functions described in each step of the flowchart.
  • each step may represent a module, segment, or portion of code comprising one or more executable instructions for executing the specified logical function(s). It should also be noted that in some alternative embodiments it is also possible for the functions recited in the steps to occur out of order. For example, it is possible that two steps shown one after another may in fact be performed substantially simultaneously, or that the steps may sometimes be performed in the reverse order depending on the function in question.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Pathology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Veterinary Medicine (AREA)
  • Optics & Photonics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Cardiology (AREA)
  • Vascular Medicine (AREA)
  • Physiology (AREA)
  • Quality & Reliability (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

La présente invention concerne une méthode de traitement d'image médicale réalisée par un appareil de traitement d'image médicale qui, selon un mode de réalisation, comprend les étapes consistant à : entrer un cliché radiologique corporel du cœur d'un patient à diagnostiquer dans un premier modèle de réseau de neurones artificiels pré-entraîné, prédire le type de valvule artificielle dans le cœur du cliché radiologique corporel par une sortie du premier modèle de réseau de neurones artificiels et obtenir des informations de forme associées à la valvule artificielle ; et déterminer le type de la valvule artificielle sur la base du résultat de la nouvelle prédiction du type de la valvule artificielle sur la base du type prédit de la valvule artificielle et du sens de la circulation sanguine prédit sur la base des informations de forme obtenues.
PCT/KR2022/003627 2021-03-17 2022-03-15 Appareil de traitement d'image médicale, méthode d'apprentissage d'image médicale et méthode de traitement d'image médicale WO2022197074A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR1020210034596A KR102481564B1 (ko) 2021-03-17 2021-03-17 의료영상 처리 장치와 그 의료영상 학습 방법 및 의료영상 처리 방법
KR10-2021-0034596 2021-03-17

Publications (1)

Publication Number Publication Date
WO2022197074A1 true WO2022197074A1 (fr) 2022-09-22

Family

ID=83321148

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2022/003627 WO2022197074A1 (fr) 2021-03-17 2022-03-15 Appareil de traitement d'image médicale, méthode d'apprentissage d'image médicale et méthode de traitement d'image médicale

Country Status (2)

Country Link
KR (1) KR102481564B1 (fr)
WO (1) WO2022197074A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20070110837A (ko) * 2004-12-24 2007-11-20 가부시키가이샤 와이디 심장판막 데이터 계측 방법 및 장치
KR20140002998A (ko) * 2012-06-28 2014-01-09 삼성메디슨 주식회사 진단 영상 장치 및 그 동작 방법
KR20170016461A (ko) * 2014-06-09 2017-02-13 지멘스 메디컬 솔루션즈 유에스에이, 인크. 메디컬 이미징에서 공간 및 시간 제약들을 이용하는 랜드마크 검출
JP2017164077A (ja) * 2016-03-15 2017-09-21 株式会社日立製作所 超音波診断装置
KR20210021663A (ko) * 2019-08-19 2021-03-02 재단법인 아산사회복지재단 의료 영상 비식별화를 위한 신체 부위 위치 특정 방법, 프로그램 및 컴퓨팅 장치

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014182505A1 (fr) * 2013-05-10 2014-11-13 Stenomics, Inc. Systeme de modelisation et de simulation pour optimisation de traitement de valvule cardiaque prothetique
US9092743B2 (en) * 2013-10-23 2015-07-28 Stenomics, Inc. Machine learning system for assessing heart valves and surrounding cardiovascular tracts
KR101611475B1 (ko) * 2014-11-25 2016-04-12 재단법인 아산사회복지재단 경피적 인공심장판막치환을 위한 임상의사결정지원 방법 및 시스템
CN110268478B (zh) * 2016-12-15 2023-07-28 斯特凡Tto有限公司 提供用于心血管疾病的决策支持和诊断的受试者特异性计算模型的方法和过程
JP7334036B2 (ja) * 2018-11-22 2023-08-28 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

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20070110837A (ko) * 2004-12-24 2007-11-20 가부시키가이샤 와이디 심장판막 데이터 계측 방법 및 장치
KR20140002998A (ko) * 2012-06-28 2014-01-09 삼성메디슨 주식회사 진단 영상 장치 및 그 동작 방법
KR20170016461A (ko) * 2014-06-09 2017-02-13 지멘스 메디컬 솔루션즈 유에스에이, 인크. 메디컬 이미징에서 공간 및 시간 제약들을 이용하는 랜드마크 검출
JP2017164077A (ja) * 2016-03-15 2017-09-21 株式会社日立製作所 超音波診断装置
KR20210021663A (ko) * 2019-08-19 2021-03-02 재단법인 아산사회복지재단 의료 영상 비식별화를 위한 신체 부위 위치 특정 방법, 프로그램 및 컴퓨팅 장치

Also Published As

Publication number Publication date
KR102481564B1 (ko) 2022-12-29
KR20220130288A (ko) 2022-09-27

Similar Documents

Publication Publication Date Title
JP6947759B2 (ja) 解剖学的対象物を自動的に検出、位置特定、及びセマンティックセグメンテーションするシステム及び方法
WO2020050635A1 (fr) Méthode et système de segmentation automatique de vaisseaux sanguins dans une image médicale au moyen d'un apprentissage automatique et d'un algorithme de traitement d'image
WO2021060700A1 (fr) Appareil et procédé de confirmation d'étude de déglutition vidéofluoroscopique
CN107847289A (zh) 现实增强的形态学手术
WO2018155765A1 (fr) Procédé et dispositif d'analyse de plaque à partir d'une image de tomographie par ordinateur
US20210059758A1 (en) System and Method for Identification, Labeling, and Tracking of a Medical Instrument
WO2020101264A1 (fr) Procédé et appareil de calcul d'un score de calcium de l'artère coronaire
WO2015147595A1 (fr) Procédé pour distinguer un artéfact et une région malade dans des images médicales
WO2022139068A1 (fr) Système d'aide au diagnostic d'une maladie pulmonaire basé sur un apprentissage profond et procédé d'aide au diagnostic d'une maladie pulmonaire basé sur un apprentissage profond
US11783477B2 (en) Medical image process apparatus, medical image learning method, and medical image process method
CN109674463A (zh) 判断心电导联错接的方法和装置
WO2022055271A1 (fr) Appareil de traitement d'image médicale, procédé d'apprentissage d'image médicale et procédé de traitement d'image médicale
Almeida et al. Lung ultrasound for point-of-care COVID-19 pneumonia stratification: computer-aided diagnostics in a smartphone. First experiences classifying semiology from public datasets
CN116246778A (zh) 一种肺功能检测智能诊断平台
WO2022197074A1 (fr) Appareil de traitement d'image médicale, méthode d'apprentissage d'image médicale et méthode de traitement d'image médicale
CN111951950B (zh) 基于深度学习的三维数据医疗分类系统
US20220331556A1 (en) System and method for visualizing placement of a medical tube or line
WO2020101265A1 (fr) Procédé et dispositif d'analyse d'image myocardique
WO2021125851A1 (fr) Procédé et programme de modélisation de prothèse mammaire personnalisée
JPH09147082A (ja) 画像診断支援装置
WO2021251777A1 (fr) Procédé et système de modélisation 3d de tomodensitométrie corps entier
WO2023085910A1 (fr) Procédé, dispositif, programme et support d'enregistrement d'apprentissage d'image à l'aide d'un réseau antagoniste génératif
WO2022158727A1 (fr) Dispositif et procédé de prise en charge de résultats/diagnostics d'images biométriques
WO2022025477A1 (fr) Procédé d'apprentissage d'un réseau de neurones artificiels pour prédire une réponse de traitement à une maladie, et appareil de prédiction de réponse de traitement
WO2022211505A1 (fr) Procédé et appareil pour fournir des informations de confiance sur le résultat d'un modèle d'intelligence artificielle

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: 22771740

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22771740

Country of ref document: EP

Kind code of ref document: A1