KR102357751B1 - Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image - Google Patents

Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image Download PDF

Info

Publication number
KR102357751B1
KR102357751B1 KR1020190170135A KR20190170135A KR102357751B1 KR 102357751 B1 KR102357751 B1 KR 102357751B1 KR 1020190170135 A KR1020190170135 A KR 1020190170135A KR 20190170135 A KR20190170135 A KR 20190170135A KR 102357751 B1 KR102357751 B1 KR 102357751B1
Authority
KR
South Korea
Prior art keywords
image
blood vessel
ultrasound
dimensional
blood vessels
Prior art date
Application number
KR1020190170135A
Other languages
Korean (ko)
Other versions
KR20210078273A (en
Inventor
김제
전기완
하태영
현윤경
Original Assignee
충남대학교병원
기초과학연구원
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 충남대학교병원, 기초과학연구원 filed Critical 충남대학교병원
Priority to KR1020190170135A priority Critical patent/KR102357751B1/en
Publication of KR20210078273A publication Critical patent/KR20210078273A/en
Application granted granted Critical
Publication of KR102357751B1 publication Critical patent/KR102357751B1/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
    • A61B8/5223—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for extracting a diagnostic or physiological parameter from medical diagnostic data
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
    • A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0891—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of blood vessels
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
    • A61B8/523—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for generating planar views from image data in a user selectable plane not corresponding to the acquisition plane
    • 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
    • G06T2207/10132—Ultrasound 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
    • G06T2207/30096—Tumor; Lesion
    • 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

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Surgery (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Vascular Medicine (AREA)
  • General Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Physiology (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

본 발명은 보편화된 통상의 초음파검사장치로부터 얻어진 2차원 초음파이미지 동영상으로부터 동맥경화 경화반 등의 3차원이미지를 획득하여 동맥경화의 진단과 치료에 적절히 활용하기 위한 혈관객체분석방법에 관한 것으로서, 보다 상세하게는 체외 초음파 검사장비로부터 획득된 길이방향 혈관단면 초음파이미지를 기초자료로 하여 의료 전문가에 의해 혈관외벽의 내측에 존재하는 혈관을 포함하는 혈관객체의 경계선이 확정된 경동맥 등 혈관단면 초음파이미지를 레이블이미지로 하여 심층인공신경망에서 기계학습을 수행하여 혈관단면 초음파이미지로부터 객체2차원이미지를 추출하는 기능을 하는 추출장치를 획득하는 장치획득단계; 체외 초음파 검사장비로부터 경동맥 등 혈관의 길이방향을 따라 복수개의 이미지프레임으로 이루어진 혈관단면 초음파 동이미지(cine image)을 획득하는 이미지획득단계; 상기 각 이미지프레임을 상기 추출장치에 제공하고, 상기 추출장치가 상기 각 이미지프레임으로부터 객체2차원이미지를 추출하는 추출단계; 및 상기 각 이미지프레임에 대응되는 연속된 복수개의 객체2차원이미지로부터 소정의 방법으로 객체3차원이미지를 획득하는 입체화단계;를 포함하는 혈관객체분석방법에 관한 것이다.The present invention relates to a vascular object analysis method for obtaining a three-dimensional image such as an arteriosclerotic plaque from a two-dimensional ultrasound image video obtained from a generalized conventional ultrasound examination device and appropriately utilizing it for the diagnosis and treatment of arteriosclerosis. In detail, a cross-sectional ultrasound image of blood vessels, such as the carotid artery, in which the boundary of a vascular object including blood vessels existing inside the outer wall of blood vessels is determined by a medical professional using the longitudinal blood vessel cross-section ultrasound image obtained from the in vitro ultrasound examination equipment as basic data. A device acquisition step of acquiring an extraction device having a function of extracting an object two-dimensional image from a blood vessel cross-section ultrasound image by performing machine learning in a deep artificial neural network as a label image; An image acquisition step of acquiring a blood vessel cross-sectional ultrasound cine image consisting of a plurality of image frames along the longitudinal direction of blood vessels, such as the carotid artery, from an in vitro ultrasound examination device; an extraction step of providing each of the image frames to the extraction device, and wherein the extraction device extracts a two-dimensional object image from each of the image frames; and a three-dimensionalization step of obtaining a three-dimensional object image by a predetermined method from a plurality of consecutive object two-dimensional images corresponding to the respective image frames.

Description

2차원 초음파이미지를 자료로 하는 혈관객체분석방법{Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image}Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image

본 발명은 보편화된 통상의 초음파검사장치로부터 얻어진 2차원 초음파이미지 동영상으로부터 동맥경화 경화반 등의 3차원이미지를 획득하여 동맥경화의 진단과 치료에 적절히 활용하기 위한 혈관객체분석방법에 관한 것이다.The present invention relates to a blood vessel object analysis method for obtaining a three-dimensional image such as an arteriosclerotic plaque from a two-dimensional ultrasound image video obtained from a generalized conventional ultrasound examination device and using it appropriately for diagnosis and treatment of arteriosclerosis.

동맥경화(atherosclerosis)란 혈관의 가장 안쪽 막(내피)에 콜레스테롤이 침착되고 혈관 내피세포의 증식이 일어나 경화반(plaque)이 생기면서 혈관이 협착(stenosis)되어(도 1 참조) 그 혈관이 말초로의 혈류 장애를 일으키는 질환이다. 동맥경화로 결국 말초로의 혈액순환에 장애가 생기면서 다양한 증상이 나타난다. 2018년 현재 한국에서는 1100 만명이 동맥경화증에 노출되어 있으며(비특허문헌1), 사망률 2~3위를 차지하는 심혈관질환 및 뇌혈관질환의 중요한 원인이 되는 것으로 알려져 있다. 따라서 동맥경화를 조기에 진단하고, 동맥경화 환자의 병증 진행을 주기적으로 추적관찰하는 것이 필요하다.Atherosclerosis refers to the deposition of cholesterol on the innermost membrane (endothelium) of blood vessels and proliferation of endothelial cells in the blood vessels to form plaques, causing the blood vessels to narrow (stenosis) (see FIG. 1), and the blood vessels become peripheral It is a disease that causes blood flow to the lungs. Atherosclerosis eventually leads to disturbance of blood circulation to the periphery, resulting in various symptoms. As of 2018, 11 million people in Korea are exposed to atherosclerosis (Non-Patent Document 1), and it is known to be an important cause of cardiovascular and cerebrovascular diseases, which rank second to third in mortality. Therefore, it is necessary to diagnose arteriosclerosis at an early stage and to periodically follow-up the disease progression of arteriosclerosis patients.

이러한 동맥경화증은 다음과 같은 몇 가지 방법으로 검사 및 진단되고 있다. Atherosclerosis is being tested and diagnosed by several methods as follows.

자기공명영상(MRI)이나 컴퓨터단층영상(CT)에 의한 3차원이미지를 이용하여 인체 뇌혈관에 발생한 동맥경화의 면적 및 체적을 측정할 수 있다. 그러나 이에 의하면 고가의 비용이 소요될 뿐 아니라 방사선 조사나 조영제 사용에 따라 인체에 좋지 않은 영향을 줄 우려가 있다. 그리고 병증의 진행을 추적관찰해야 하는 경우에는 고비용이나 방사선 피폭으로 인해 적지 않은 부담이 되고 있다. The area and volume of arteriosclerosis occurring in the human brain blood vessels can be measured using a three-dimensional image by magnetic resonance imaging (MRI) or computed tomography (CT). However, according to this, not only expensive costs are required, but also there is a risk of adverse effects on the human body according to irradiation or use of contrast agents. And when it is necessary to follow up the progression of the disease, it becomes a burden due to high cost and radiation exposure.

이에 방사선을 사용하지 않고, 실시간으로 동맥경화의 조직 상태 및 혈류속도의 획득이 가능하고, 검사 비용이 저렴한 경동맥초음파검사가 많이 이용되고 있다. 이에 의하면 혈관 협착의 면적, 협착으로 인한 혈류속도 및 경화반의 길이/면적을 측정할 수 있다. Therefore, carotid artery ultrasonography, which does not use radiation, can obtain the tissue state of arteriosclerosis and blood flow velocity in real time, and is inexpensive, is widely used. According to this, it is possible to measure the area of the vascular stenosis, the blood flow rate due to the stenosis, and the length/area of the plaque.

그러나 종래의 경동맥초음파검사에 의하면 협착면적은 전체 동맥경화 조직의 한 면에서의 면적을 재는 것이고, 가장 심한 협착 위치에서 측정한 혈류속도는 개인별/혈관별로 다를 뿐 아니라 동맥경화의 간접 정보이며, 경화반의 길이/면적은 장축 및 단면에 대한 정보일 뿐이다. 특히 한번에 관찰할 수 있는 시야가 좁다는 초음파이미지 고유의 단점 때문에 혈관 또는 동맥경화(경화반)의 전체적인 윤곽 및 체적을 알 수가 없었다. However, according to the conventional carotid ultrasound examination, the stenosis area measures the area on one side of the entire arteriosclerotic tissue, and the blood flow velocity measured at the most severe stenosis location differs for each individual/vessel as well as indirect information of arteriosclerosis, and hardening The length/area of a half is only information about the major axis and cross section. In particular, the overall outline and volume of blood vessels or arteriosclerosis (sclerotic plaques) could not be known because of the inherent disadvantage of ultrasound images in that the field of view that can be observed at once was narrow.

JP 2010-88795는 화상의 삼차원 볼륨 데이터로부터 혈관의 석회화, 혈관 내강, 혈관 외벽 및 혈관 플라그 정보를 추출하는 장치에 관한 것이다. 이에 의하면 혈관 플라그의 3차원 정보를 획득할 수 있지만, 특수한 장치 및 '삼차원 볼륨 데이터'를 필요로 한다.JP 2010-88795 relates to an apparatus for extracting blood vessel calcification, vessel lumen, vessel outer wall, and vessel plaque information from three-dimensional volume data of an image. According to this method, three-dimensional information of vascular plaque can be obtained, but a special device and 'three-dimensional volume data' are required.

비특허문헌 1(Med Phys. 2002 Oct;29(10):2319-27.)은 3차원초음파이미지장치를 활용하여 경동맥 플라그 부피의 정량화를 시도하고 있으나, 고가이며 사용이 복잡한 3차원 초음파이미지장치가 요구된다.Non-Patent Document 1 (Med Phys. 2002 Oct;29(10):2319-27.) attempts to quantify carotid plaque volume using a 3D ultrasound imaging device, but a 3D ultrasound imaging device that is expensive and complicated to use. is required

한편, 관련 하드웨어와 소프트웨어의 발달에 따라 탐촉자(프로브)로 검사 범위를 천천히 움직이면서 받은 이미지를 고속으로 처리하여 연속프레임(cine image)을 통해 넓은 부위의 이미지를 실시간으로 관찰할 수 있게 되어 혈관 등을 검사할 때 유용하게 이용되고 있다. 나아가 2D이미지를 편집하고 재구성하여 3D입체이미지로 전환하는 프로그램을 활용하여 혈관의 3D입체이미지를 획득하는 기술도 알려져 있다.On the other hand, with the development of related hardware and software, it is possible to observe images of a wide area in real time through continuous frames (cine images) by slowly moving the inspection range with a probe (probe) and processing the received image in real time. It is useful for inspection. Furthermore, a technique for acquiring a 3D stereoscopic image of a blood vessel by using a program that edits and reconstructs a 2D image and converts it into a 3D stereoscopic image is also known.

그러나 의료분야에서 이들 3D입체이미지는 전술하였듯이 고가이면서 인체에 피해를 줄 수 있는 MRI나 CT 기반의 2D이미지가 요구된다는 단점이 있으며, 초음파이미지로부터 3D입체이미지를 얻는 것에 대하여 알려진 바 없다. However, as described above, these 3D stereoscopic images in the medical field have the disadvantage of requiring an MRI or CT-based 2D image that is expensive and can damage the human body, and there is no known about obtaining a 3D stereoscopic image from an ultrasound image.

한편, 동맥경화와 관련해서 혈관의 구조 뿐만 아니라 협착부의 크기와 형태, 생성위치, 협착부 내의 구성물의 종류(석회질 또는 콜레스테롤)와 이들의 크기, 형태 등 객체에 대한 정보가 동맥경화의 진행과 치료과정을 진단하고 관찰하는데 매우 중요하다. 그러나 보편화되어 있고 저렴하며 인체에 무해한 초음파이미지장치에 의해 얻어진 이미지로부터 이러한 객체 정보를 얻는 방법은 알려져 있지 않다.On the other hand, in relation to arteriosclerosis, not only the structure of blood vessels, but also the size and shape of the stenosis, the location of its formation, the type of components (calcareous or cholesterol) in the stenosis, and their size and shape, etc. It is very important for diagnosing and observing However, a method for obtaining such object information from an image obtained by a universal, inexpensive, and harmless ultrasound imaging device is not known.

JP 2010-88795JP 2010-88795

Med Phys. 2002 Oct;29(10):2319-27.Med Phys. 2002 Oct;29(10):2319-27.

본 발명은 동맥 2차원 초음파이미지를 이용하여 동맥경화 경화반의 면적, 부피 및 형태 뿐만 아니라 경화반조직 내부의 성분종류 및 이의 크기와 형태 등 혈관객체정보를 간편하게 분석하는 방법을 제공하는 것을 목적으로 한다.An object of the present invention is to provide a method for conveniently analyzing vascular object information such as the area, volume, and shape of atherosclerotic plaque, as well as the type of component inside the plaque tissue and its size and shape, using a two-dimensional ultrasound image of an artery. .

또한 본 발명은 일반화된 통상적인 초음파 검사장비를 활용하여 손쉽고 저렴한 안전한 동맥경화 검사방법을 제시함으로써 환자의 신체적, 경제적 부담 및 의료기관의 장비에 대한 부담을 최소화하는 것을 목적으로 한다.Another object of the present invention is to minimize the physical and economic burden of a patient and the burden on equipment of a medical institution by presenting an easy, inexpensive and safe arteriosclerosis examination method using generalized conventional ultrasound examination equipment.

전술한 목적을 달성하기 위한 본 발명은 The present invention for achieving the above object

체외 초음파 검사장비로부터 획득된 길이방향 혈관단면 초음파이미지를 기초자료로 하여 의료 전문가에 의해 혈관외벽의 내측에 존재하는 혈관을 포함하는 혈관객체의 경계선이 확정된 경동맥 등 혈관단면 초음파이미지를 레이블이미지로 하여 심층인공신경망에서 기계학습을 수행하여 혈관단면 초음파이미지로부터 객체2차원이미지를 추출하는 기능을 하는 추출장치를 획득하는 장치획득단계; 체외 초음파 검사장비로부터 경동맥 등 혈관의 길이방향을 따라 복수개의 이미지프레임으로 이루어진 혈관단면 초음파 동이미지(cine image)을 획득하는 이미지획득단계; 상기 각 이미지프레임을 상기 추출장치에 제공하고, 상기 추출장치가 상기 각 이미지프레임으로부터 객체2차원이미지를 추출하는 추출단계; 및 상기 각 이미지프레임에 대응되는 연속된 복수개의 객체2차원이미지로부터 소정의 방법으로 객체3차원이미지를 획득하는 입체화단계;를 포함하는 혈관객체분석방법인 것을 특징으로 한다. Using the longitudinal blood vessel cross-section ultrasound image obtained from the in vitro ultrasound examination equipment as the basic data, the blood vessel cross-section ultrasound image, such as the carotid artery, in which the boundary line of the vessel object including the blood vessel existing inside the outer vessel wall is determined by a medical professional as the label image a device acquisition step of acquiring an extraction device having a function of extracting an object two-dimensional image from a blood vessel cross-section ultrasound image by performing machine learning in a deep artificial neural network; An image acquisition step of acquiring a blood vessel cross-sectional ultrasound cine image consisting of a plurality of image frames along the longitudinal direction of blood vessels, such as the carotid artery, from an in vitro ultrasound examination device; an extraction step of providing each of the image frames to the extraction device, and wherein the extraction device extracts a two-dimensional object image from each of the image frames; and a three-dimensionalization step of acquiring a three-dimensional object image by a predetermined method from a plurality of consecutive two-dimensional object images corresponding to the respective image frames.

이상과 같이 본 발명에 의하면 널리 보급된 일반적인 초음파 검사장비를 활용하여 동맥경화의 정도, 협착의 범위, 경화반의 크기와 구조, 경화반 내부의 특성 등 객체정보를 획득하여 분석할 수 있게 된다.As described above, according to the present invention, it is possible to obtain and analyze object information such as the degree of arteriosclerosis, the range of stenosis, the size and structure of the plaque, and the characteristics of the inside of the plaque by using the widely spread general ultrasound examination equipment as described above.

또한 본 발명에 의하면 널리 보급된 통상의 초음파 검사장비로 동맥경화를 검사할 수 있으므로 안전하면서도 저비용이고 신속하게 동맥경화의 진단 및 환자의 경시적 추적관찰이 용이하게 된다.In addition, according to the present invention, since arteriosclerosis can be inspected with a widely spread conventional ultrasound examination device, it is easy to diagnose arteriosclerosis quickly, at low cost and safely, and to follow-up the patient over time.

도 1은 혈관에 동맥경화반이 생긴 예를 보여주는 개념도.
도 2는 혈관단면 초음파이미지 및 그에 대응되는 레이블이미지 세트의 예시적 사진.
도 3은 본 발명에 적용된 인공신경망의 구조를 보여주는 개념도.
도 4는 통상의 2차원 초음파검사장치에서 혈관단면 초음파이미지를 얻는 방식을 보여주는 일예의 사진.
도 5는 추출단계에서 좌측의 초음파이미지로부터 추출된 우측의 혈관객체 추출도를 보여주는 개념도.
도 6은 본 발명에 의한 방법에 따라 분석(추출)된 혈관 협착조직의 예시적 입체도.1 is a conceptual diagram showing an example in which an atherosclerotic plaque is formed in a blood vessel.
2 is an exemplary photograph of a blood vessel cross-section ultrasound image and a label image set corresponding thereto.
3 is a conceptual diagram showing the structure of an artificial neural network applied to the present invention.
4 is a photograph showing an example of a method of obtaining a blood vessel cross-section ultrasound image in a conventional two-dimensional ultrasound examination apparatus.
5 is a conceptual diagram showing a blood vessel object extraction diagram on the right side extracted from the ultrasound image on the left side in the extraction step.
6 is an exemplary stereoscopic view of a vascular stenosis tissue analyzed (extracted) according to the method according to the present invention.

이하 첨부된 도면을 참조하여 본 발명을 보다 상세히 설명한다. 그러나 첨부된 도면은 본 발명의 기술적 사상의 내용과 범위를 쉽게 설명하기 위한 예시일 뿐, 이에 의해 본 발명의 기술적 범위가 한정되거나 변경되는 것은 아니다. 이러한 예시에 기초하여 본 발명의 기술적 사상의 범위 안에서 다양한 변형과 변경이 가능함은 당업자에게는 당연할 것이다. Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings. However, the accompanying drawings are only examples for easily explaining the content and scope of the technical idea of the present invention, and thereby the technical scope of the present invention is not limited or changed. It will be natural for those skilled in the art that various modifications and changes are possible within the scope of the technical spirit of the present invention based on these examples.

본 발명에서 '혈관객체' 또는 '객체'란 동맥경화와 관련하여 3차원이미지 정보를 얻고자 하는 대상을 지칭하는 것으로서, 혈관, 혈관내 혈액, 혈관내 협착조직, 협착조직 내부의 일정 성분으로 된 덩어리 등을 의미한다. In the present invention, 'vascular object' or 'object' refers to an object for which three-dimensional image information is to be obtained in relation to arteriosclerosis, lumps, etc.

본 발명에서 '혈관'이란 경동맥을 포함하는 개념이며, 건강한 혈관 뿐만 아니라 경화반이 생성된 혈관을 포함한다.In the present invention, 'vascular' is a concept including carotid arteries, and includes healthy blood vessels as well as blood vessels in which hard plaques are generated.

본 발명에서 '2차원 초음파이미지를 자료로 한다'는 것은 2차원 초음파이미지를 혈관객체분석의 기초자료(source)로 한다는 것이다.In the present invention, 'using a two-dimensional ultrasound image as data' means that a two-dimensional ultrasound image is used as a basic data (source) for blood vessel object analysis.

전술한 바와 같이 본 발명은 장치획득단계, 이미지획득단계, 추출단계 및 입체화단계를 포함하는 과정을 거치는, 2차원 초음파이미지를 자료로 하는 혈관객체분석방법에 관한 것이다. As described above, the present invention relates to a blood vessel object analysis method using a two-dimensional ultrasound image as data, which is subjected to a process including a device acquisition step , an image acquisition step , an extraction step , and a three-dimensionalization step .

본 발명에서 상기 장치획득단계는, 의료 전문가에 의해 혈관을 포함하는 혈관객체의 경계선이 확정된 혈관단면 초음파이미지를 레이블이미지(=학습자료)로 하여 심층인공신경망에서 기계학습을 수행하고, 그 결과 레이블이미지가 아닌 혈관단면 초음파이미지로부터 객체2차원이미지를 추출하는 기능을 하는 추출장치를 획득하는 단계이다. 따라서 상기 추출장치는 일종의 인공지능이라 할 수 있다. 도 2에 경동맥 단면 초음파이미지 및 이 초음파이미지에 대응되는, 의료 전문가에 의해 혈액, 협착조직 및 협착조직내 석회질(calcification) 등 혈관객체의 경계선이 결정된 레이블이미지 세트의 예를 도시하였다. 도면에서 적색, 황색 및 백색 영역은 각각 혈액, 협착조직 및 석회질을 나타낸다. 도시되지는 않았지만 필요에 따라서는 혈관 자체 또는 콜레스테롤 축적영역도 경계선 및 색상이 표시되도록 할 수도 있을 것이다.In the present invention, the device acquisition step performs machine learning in a deep artificial neural network using a blood vessel cross-section ultrasound image in which the boundary line of a blood vessel object including a blood vessel is determined by a medical professional as a label image (= learning data), and the result It is a step of acquiring an extraction device having a function of extracting an object two-dimensional image from a blood vessel cross-section ultrasound image rather than a label image. Therefore, the extraction device can be said to be a kind of artificial intelligence. Fig. 2 shows an example of a set of label images in which a carotid artery cross-section ultrasound image and a boundary line of a vascular object such as blood, stenotic tissue, and calcification in the stenotic tissue are determined by a medical professional corresponding to the ultrasound image. In the figure, the red, yellow and white areas represent blood, stenosis and calcification, respectively. Although not shown, if necessary, the blood vessel itself or the cholesterol accumulation region may also be marked with borders and colors.

본 발명에서 상기 심층인공신경망은 도 3에 예시된 것처럼 5층으로 구성된 deep convolutional framelets(참조 : https://doi.org/10.1137/17M1141771)을 이용할 수 있는데, 이러한 인공신경망에 관해서는 종래 널리 알려져 있고 본 발명이 인공신경망 자체에 관한 것이 아니기 때문에 이에 대한 상세한 설명을 생략한다. In the present invention, the deep artificial neural network can use deep convolutional framelets (refer to: https://doi.org/10.1137/17M1141771) composed of five layers as illustrated in FIG. and since the present invention does not relate to the artificial neural network itself, a detailed description thereof will be omitted.

이러한 장치획득단계에서, 예를 들면 수십~수백장의 혈관단면 초음파이미지 및 그에 대응되는 레이블이미지 세트를 학습자료로 하여 기계학습이 이루어지고 그 결과로 레이블이미지가 아닌 생(raw) 초음파이미지로부터 객체2차원이미지를 추출하는 기능을 하는 추출장치를 얻게 된다.In this device acquisition step, for example, machine learning is performed using tens to hundreds of blood vessel cross-section ultrasound images and corresponding label image sets as learning materials, and as a result, object 2 is obtained from raw ultrasound images rather than label images. An extraction device having a function of extracting a dimensional image is obtained.

이러한 장치획득단계에서 얻어진 추출장치는 기능의 수정/보완이 필요할 때를 제외하고는 완결성을 가지게 되며, 이후 단계들에 대해서 [1 : 복수]의 대응관계가 된다. 즉, 얻어진 추출장치를 활용하여 모든 분석대상혈관에 대해 혈관객체분석이 가능하게 된다.The extraction device obtained in this device acquisition step has completeness except when correction/supplementation of functions is required, and it becomes a correspondence of [1: multiple] for subsequent steps. That is, the blood vessel object analysis is possible for all the analysis target blood vessels by using the obtained extraction device.

이어서 상기 이미지획득단계는, 2차원프로브가 장착된 통상의 초음파검사장치를 이용하여 혈관의 길이방향을 따라 복수개의 혈관단면 초음파이미지로 이루어진 동영상(cine image)을 얻는 단계이다. 도 3에 오퍼레이터가 경동맥의 혈관단면 초음파이미지를 얻는 과정의 일예 사진을 도시하였다. '혈관단면'에 대한 이미지를 얻는 것이므로 프로브는 혈류 방향에 직각으로 배치된다. 오퍼레이터는 프로브를 환자의 한쪽 빗장뼈(clavicle)에 직각으로, 혈관의 혈류 방향에 직각으로 프로브를 댄 후에, 프로브를 턱뼈(mandible)와 직각이 되는 곳(도시된 화면에서 화살표 영역)까지 이동시키면서 수백장의 혈관단면 초음파이미지로 이루어진 동영상을 얻게 된다. Subsequently, the image acquisition step is a step of obtaining a cine image composed of a plurality of cross-sectional ultrasound images of blood vessels along the longitudinal direction of the blood vessels using a conventional ultrasound examination device equipped with a two-dimensional probe. 3 shows an example photograph of a process in which an operator obtains an ultrasound image of a blood vessel cross-section of the carotid artery. Since the image of the 'vascular cross section' is obtained, the probe is placed perpendicular to the blood flow direction. The operator places the probe at a right angle to the patient's one clavicle and at a right angle to the blood flow direction of the blood vessel, and then moves the probe to a place perpendicular to the mandible (arrow area in the illustrated screen). A moving picture consisting of hundreds of blood vessel cross-section ultrasound images is obtained.

이어서 상기 추출단계는, 위 이미지획득단계에서 얻어진 각각의 초음파이미지를 장치획득단계에서 획득한 추출장치(인공지능)에 적용하여 자동으로 객체2차원이미지를 추출하는 단계이다. 이 단계를 도면으로 설명하자면, 도 5에서 좌측의 초음파이미지로부터 우측의 혈관객체 추출도를 얻는 것이다. 참고로, 설명의 편의를 위해 도 2 좌측사진은 도 5의 좌측사진과 동일한 것으로 하였지만, 도 2 우측사진이 의료 전문가에 의해 결정된 혈관객체 단면임에 반하여 도 5 우측사진은 본 발명에 의한 추출장치에 의해 자동으로 추출된 혈관객체 단면이다. Subsequently, the extraction step is a step of automatically extracting a two-dimensional object image by applying each ultrasound image obtained in the image acquisition step to the extraction device (artificial intelligence) obtained in the device acquisition step. This step will be described with drawings, in which a blood vessel object extraction diagram on the right is obtained from the ultrasound image on the left in FIG. 5 . For reference, the left photo of FIG. 2 is the same as the left photo of FIG. 5 for convenience of explanation, whereas the right photo of FIG. 2 is a cross-section of a vascular object determined by a medical professional, whereas the right photo of FIG. It is a cross section of a blood vessel object automatically extracted by

이때 추출장치가 혈관, 혈액, 협착조직, 협착조직내 물질 등 모든 혈관객체에 대하여 경계선 추출을 학습을 한 것이라 하더라도 필요에 따라 이 중의 어느 하나 또는 복수개의 혈관객체에 대한 추출만 시도하는 것도 가능할 것이다.At this time, even if the extraction device has learned boundary line extraction for all vascular objects such as blood vessels, blood, stenotic tissue, and substances in stenotic tissue, it will be possible to try to extract only one or a plurality of vascular objects if necessary. .

상기 입체화단계는, 추출단계에서 얻어진 연속된 복수개의 객체2차원이미지를 입체로 변환시켜 객체3차원이미지를 얻는 단계이다. 하나의 사물(본 발명에서는 혈관객체)의 연속된 단면사진(본 발명에서는 객체2차원이미지)으로부터 사물의 입체구조를 계산하는 다양한 방법이나 프로그램이 종래 알려져 있어 이에 대한 구체적인 설명은 생략한다.The three- dimensionalization step is a step of obtaining a three-dimensional object image by converting a plurality of consecutive object two-dimensional images obtained in the extraction step into a three-dimensional image. Various methods or programs for calculating the three-dimensional structure of an object (in the present invention, an object two-dimensional image) from a series of cross-sectional photographs (in the present invention, a blood vessel object) are known in the prior art, and detailed description thereof will be omitted.

이러한 과정을 거쳐 혈관의 협착조직만을 분석한 것(입체구조를 획득한 것)의 예를 도 6에 도시하였다. 도면에서 황색수직선(면)에 의한 단면도도 함께 도시하였다. 도면은 협착조직 만의 입체구조이기 때문에 혈관 및 혈액은 보이지 않는다. 따라서 협착조직이 없는 부분은 개구된 것처럼 보인다.An example in which only the vascular stenosis was analyzed (a three-dimensional structure was obtained) through this process is shown in FIG. 6 . A cross-sectional view along a yellow vertical line (plane) is also shown in the drawing. Since the drawing is a three-dimensional structure of only the stenotic tissue, blood vessels and blood are not visible. Therefore, the part without stenosis appears to be open.

이러한 본 발명에 의한 혈관객체분석방법에 의하면 혈관단면 초음파이미지로부터 원하는 혈관객체(혈관, 혈액, 혈관의 협착조직 등)에 대한 3차원이미지를 얻을 수 있게 된다.According to the vascular object analysis method according to the present invention, it is possible to obtain a three-dimensional image of a desired vascular object (blood vessel, blood, vascular stenosis, etc.) from the vascular cross-sectional ultrasound image.

한편, 초음파검사장치 이용시 오퍼레이터가 충분히 숙련되어 있다고 하더라도 전체 검사대상 혈관의 범위, 경동맥을 예로 든다면 [빗장뼈 바로 위~턱뼈 바로 아래] 범위에서 비교적 일정한 속도로 프로브를 이동시키려고 노력하지만 전체적으로 일정한 속도로 혈관단면 동영상을 얻기는 쉽지 않다(측정내 불균일성). 나아가 한 환자에서 일정기간이 지난 후에 추적검사를 할 때마다 오퍼레이터가 다르거나 또는 동일한 오퍼레이터라도 '과거와 동일한 속도'로 프로브를 이동시킨다는 보장도 할 수 없다(측정간 불균일성). On the other hand, even if the operator is sufficiently skilled when using the ultrasound examination device, if the range of the entire examination target blood vessel, taking the carotid artery as an example, tries to move the probe at a relatively constant speed in the range [just above the clavicle ~ just below the jawbone], but the overall speed is constant. Therefore, it is not easy to obtain a cross-sectional video of blood vessels (non-uniformity within measurements). Furthermore, it cannot be guaranteed that the probe moves at the same speed as in the past even if the operator is different or the same operator every time a follow-up examination is performed after a certain period of time in one patient (non-uniformity between measurements).

이렇듯 이미지획득단계에서 측정내 불균일성과 측정간 불균일성 때문에 동일한 환자가 시차를 두고 본 발명에 의한 방법에 따라 동맥경화조직의 형태와 체적에 대한 분석을 하였는데 프로브 이동속도가 달라지면(실무적으로 이럴 가능성이 높음) 동맥경화조직이 변하지 않았는데도 과거에 비해 크게 또는 작게 인식되어 병증의 진단과 대응에 중대한 착오를 일으킬 수 있다.As such, in the image acquisition stage, the shape and volume of atherosclerotic tissue were analyzed according to the method according to the present invention with a time lag due to non-uniformity within measurements and non-uniformity between measurements. ) Even though the atherosclerotic tissue has not changed, it is recognized as larger or smaller than in the past, which can cause serious errors in diagnosis and response.

이미지획득단계에서의 이러한 문제를 해결하기 위하여 본 발명은 프로브 진행속도의 변이를 보정하는 보정단계를 추가로 가지는 것이 바람직하다. 이러한 보정단계는, 상기 복수개의 혈관단면 초음파이미지의 최초 프레임과 최후 프레임 촬영위치 사이의 거리를 측정하는 거리측정소단계, 및 초음파이미지의 수와 무관하게 최초 초음파이미지와 최후 초음파이미지를 상기 측정된 거리의 양단에 배치하고 나머지 초음파이미지를 균일하게 배치하는 배치소단계를 포함한다. In order to solve this problem in the image acquisition step, it is preferable that the present invention additionally include a correction step for correcting the variation in the probe progress speed. The correction step includes a distance measuring station step of measuring the distance between the first frame and the last frame photographing position of the plurality of blood vessel cross-sectional ultrasound images, and the measured distance between the first ultrasound image and the last ultrasound image regardless of the number of ultrasound images. It includes a batch small step of disposing at both ends of the and uniformly disposing the remaining ultrasound images.

이에 의해 복수개의 초음파이미지들이 분석되는 환자의 혈관범위(길이)에 대응하여 획득된 초음파이미지들을 균일하게 배치함으로써 측정내 불균일성은 최소화하고 측정간 불균일성은 제거하는 효과를 얻게 된다. 이에 따라 환자의 혈관객체에 대한 비교적 정확한 추적정보를 얻을 수 있게 되므로 동맥경화로 인해 발생하는 뇌혈관질환, 심혈관질환의 치료 방침을 빠르게 결정할 수 있고, 치료 개시 후 그 효과를 추적관찰 하는데 유용하게 활용될 수 있다.Thereby, by uniformly disposing the obtained ultrasound images corresponding to the blood vessel range (length) of the patient in which the plurality of ultrasound images are analyzed, it is possible to minimize non-uniformity within measurement and remove non-uniformity between measurements. As a result, it is possible to obtain relatively accurate tracking information about the patient's vascular object, so it is possible to quickly determine the treatment policy for cerebrovascular disease and cardiovascular disease caused by arteriosclerosis, and it is useful for tracking and observing the effect after starting treatment can be

Claims (5)

체외 초음파 검사장비로부터 획득된 길이방향 혈관단면 초음파이미지를 기초자료로 하여 의료 전문가에 의해 혈관을 포함하는 혈관객체의 경계선이 확정된 혈관단면 초음파이미지를 레이블이미지로 하여 심층인공신경망에서 기계학습을 수행하여 혈관단면 초음파이미지로부터 객체2차원이미지를 추출하는 기능을 하는 추출장치를 획득하는 장치획득단계;
체외 초음파 검사장비로부터 혈관의 길이방향을 따라 복수개의 혈관단면 초음파이미지에 의한 동영상을 획득하는 이미지획득단계;
상기 추출장치가 상기 각 초음파이미지로부터 객체2차원이미지를 추출하는 추출단계; 및
상기 각 초음파이미지에 대응되는 복수개의 객체2차원이미지를 차례로 배치하고 소정의 방법으로 객체3차원이미지를 획득하는 입체화단계;를 포함하는 혈관객체분석방법에 있어서,
상기 복수개의 혈관단면 초음파이미지의 최초 프레임과 최후 프레임 촬영위치 사이의 거리를 측정하는 거리측정소단계, 및 초음파이미지의 수와 무관하게 최초 초음파이미지와 최후 초음파이미지를 상기 측정된 거리의 양단에 배치하고 나머지 초음파이미지를 균일하게 배치하는 배치소단계를 포함하는 프로브 진행속도의 변이를 보정하는 보정단계;가 추가되는 것을 특징으로 하는 혈관객체분석방법.
Machine learning is performed in a deep artificial neural network by using a longitudinal blood vessel cross-section ultrasound image obtained from an in vitro ultrasound examination device as a basic data, and a blood vessel cross-section ultrasound image in which the boundary line of a blood vessel object including blood vessels is determined by a medical professional as a label image a device acquisition step of acquiring an extraction device having a function of extracting an object two-dimensional image from a blood vessel cross-section ultrasound image;
an image acquisition step of acquiring a moving picture using a plurality of cross-sectional ultrasound images of blood vessels along the longitudinal direction of blood vessels from an in vitro ultrasound examination device;
an extraction step of extracting, by the extraction device, a two-dimensional object image from each ultrasound image; and
In the blood vessel object analysis method comprising a; a three-dimensional step of sequentially arranging a plurality of object two-dimensional images corresponding to the respective ultrasound images and obtaining a three-dimensional object image by a predetermined method,
A distance measuring station step of measuring the distance between the first frame and the last frame photographing position of the plurality of blood vessel cross-section ultrasound images, and irrespective of the number of ultrasound images, placing the first ultrasound image and the last ultrasound image at both ends of the measured distance, A vascular object analysis method, characterized in that it is added; a correction step of correcting the variation in the probe progress speed, which includes a placement step of uniformly disposing the remaining ultrasound images.
 청구항 1에 있어서,
상기 혈관객체는,
혈관이거나,
혈관 및 혈관의 협착조직인 것을 특징으로 하는 혈관객체분석방법.
The method according to claim 1,
The blood vessel object,
blood vessels, or
A vascular object analysis method, characterized in that it is a blood vessel and a stenotic tissue of the blood vessel.
 청구항 2에 있어서,
상기 혈관객체는,
협착조직내의 석회질 또는 콜레스테롤을 포함하는 것을 특징으로 하는 혈관객체분석방법.
3. The method according to claim 2,
The blood vessel object,
Blood vessel object analysis method, characterized in that it contains calcareous or cholesterol in the stenotic tissue.
 청구항 1 또는 3에 있어서,
객체3차원이미지에서 객체가 다르면 다른 색상으로 표시되는 것을 특징으로 하는 혈관객체분석방법.
4. The method of claim 1 or 3,
Object A blood vessel object analysis method, characterized in that if objects are different in a three-dimensional image, they are displayed in different colors.
 삭제delete
KR1020190170135A 2019-12-18 2019-12-18 Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image KR102357751B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020190170135A KR102357751B1 (en) 2019-12-18 2019-12-18 Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020190170135A KR102357751B1 (en) 2019-12-18 2019-12-18 Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image

Publications (2)

Publication Number Publication Date
KR20210078273A KR20210078273A (en) 2021-06-28
KR102357751B1 true KR102357751B1 (en) 2022-02-03

Family

ID=76607958

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020190170135A KR102357751B1 (en) 2019-12-18 2019-12-18 Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image

Country Status (1)

Country Link
KR (1) KR102357751B1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101059824B1 (en) 2010-09-09 2011-08-29 한국표준과학연구원 Method measuring the ratio of intima to media thickness in carotid artery using ultrasound image
WO2016088758A1 (en) 2014-12-01 2016-06-09 国立研究開発法人産業技術総合研究所 Ultrasound examination system and ultrasound examination method

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5305821B2 (en) 2008-10-10 2013-10-02 株式会社東芝 Medical image processing apparatus and medical image diagnostic apparatus
KR101049915B1 (en) * 2009-09-14 2011-07-15 울산대학교 산학협력단 Apparatus and method for measuring atherosclerotic plaque component using endovascular ultrasound image
KR20120028154A (en) * 2010-09-14 2012-03-22 울산대학교 산학협력단 Diagnose method and apparatus for atherosclerotic lesions
KR102212499B1 (en) * 2018-01-03 2021-02-04 주식회사 메디웨일 Ivus image analysis method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101059824B1 (en) 2010-09-09 2011-08-29 한국표준과학연구원 Method measuring the ratio of intima to media thickness in carotid artery using ultrasound image
WO2016088758A1 (en) 2014-12-01 2016-06-09 国立研究開発法人産業技術総合研究所 Ultrasound examination system and ultrasound examination method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
이욱진, 정민건, 조성욱. (2016). 관상동맥내 경화반 타입 기반의 이상화된 이차원 경화반 모델에 대한 수치 해석적 연구. 한국CDE학회 학술발표회 논문집, (2016.1), 177-180.

Also Published As

Publication number Publication date
KR20210078273A (en) 2021-06-28

Similar Documents

Publication Publication Date Title
US20230282365A1 (en) Diagnostically useful results in real time
US20180177474A1 (en) Creating a vascular tree model
JP2022169579A (en) Diagnostically useful results in real time
JP4728627B2 (en) Method and apparatus for segmenting structures in CT angiography
EP2875781B1 (en) Apparatus and method for processing a medical image of a body lumen
US9058665B2 (en) Systems and methods for identifying bone marrow in medical images
CN104363833B (en) Video generation device and image generating method
US20160066795A1 (en) Stenosis therapy planning
JP2007167656A (en) Method for analysis of motion of subject, and tomographic device
JP2009261651A (en) Image processing apparatus and program
JP2006102353A (en) Apparatus, method and program for analyzing joint motion
Chifor et al. Three-dimensional periodontal investigations using a prototype handheld ultrasound scanner with spatial positioning reading sensor
CN107106059A (en) Perfusion imaging
JP2020512885A (en) Method and apparatus for determining physiological function parameters
KR102357751B1 (en) Method for Analysing Blood Vessel Object Using Two Dimensional Ultrasonic Image
Mamabolo et al. Modern imaging modalities in forensic anthropology and the potential of low-dose X-rays
US20180315185A1 (en) System and method to analyze various retinal layers
CN104000618A (en) Breathing movement gating correction technology implemented with ring true photon number gating method
EP3129955A1 (en) Method for the analysis of image data representing a three-dimensional volume of biological tissue
EP3995081A1 (en) Diagnosis assisting program
JPH04270983A (en) Spect tomographic image shooting device
Lee et al. Human airway measurement from CT images
Natalinova et al. Computer spatially oriented reconstruction of a 3d heart shape based on its tomographic imaging
Strauss et al. Computer assistance in the interpretation and quantification of lung scans
KR101113747B1 (en) Assessing methode of in-stent restenosis using micro computed tomography

Legal Events

Date Code Title Description
E902 Notification of reason for refusal
AMND Amendment
E601 Decision to refuse application
AMND Amendment
X701 Decision to grant (after re-examination)
GRNT Written decision to grant