CN112932535B - Medical image segmentation and detection method - Google Patents
Medical image segmentation and detection method Download PDFInfo
- Publication number
- CN112932535B CN112932535B CN202110137243.XA CN202110137243A CN112932535B CN 112932535 B CN112932535 B CN 112932535B CN 202110137243 A CN202110137243 A CN 202110137243A CN 112932535 B CN112932535 B CN 112932535B
- Authority
- CN
- China
- Prior art keywords
- myocardium
- image
- detection
- image segmentation
- target area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 46
- 238000003709 image segmentation Methods 0.000 title claims abstract description 23
- 210000004165 myocardium Anatomy 0.000 claims abstract description 60
- 238000000034 method Methods 0.000 claims abstract description 17
- 230000000302 ischemic effect Effects 0.000 claims abstract description 16
- 238000002604 ultrasonography Methods 0.000 claims abstract description 5
- 230000002107 myocardial effect Effects 0.000 claims description 20
- 208000028867 ischemia Diseases 0.000 claims description 11
- 230000002861 ventricular Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 2
- 230000007547 defect Effects 0.000 abstract description 6
- 230000011218 segmentation Effects 0.000 abstract description 5
- 230000000747 cardiac effect Effects 0.000 abstract description 3
- 230000008859 change Effects 0.000 abstract description 2
- 210000005240 left ventricle Anatomy 0.000 description 7
- 238000003745 diagnosis Methods 0.000 description 6
- 208000031225 myocardial ischemia Diseases 0.000 description 6
- 208000024172 Cardiovascular disease Diseases 0.000 description 3
- 206010000891 acute myocardial infarction Diseases 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 230000000877 morphologic effect Effects 0.000 description 3
- 210000001519 tissue Anatomy 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000002592 echocardiography Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 208000010125 myocardial infarction Diseases 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 241000270295 Serpentes Species 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000005242 cardiac chamber Anatomy 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 210000002808 connective tissue Anatomy 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 210000001174 endocardium Anatomy 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004217 heart function Effects 0.000 description 1
- 230000007654 ischemic lesion Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007659 motor function Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
-
- 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
- A61B8/0883—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of the heart
-
- 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
- A61B8/5207—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of raw data to produce diagnostic data, e.g. for generating an image
-
- 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
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
Abstract
The invention discloses a medical image segmentation and detection method, which relates to the technical field of ultrasonic medical engineering, and takes the motion characteristics of cardiac muscle into consideration, uses a sequence cardiac ultrasound short-axis image to describe the motion change of the cardiac muscle on time and space, and also uses the space position relationship between a ventricle and the cardiac muscle to determine the accurate position of the cardiac muscle in the image, thereby removing the influence of noise on the segmentation result and keeping the accuracy and the integrity of the shape of the cardiac muscle. The invention uses the time series polar coordinate deep convolution network to classify the ischemic myocardium of each segment of the myocardium zone, thereby not only solving the defects of the traditional deep convolution network on the description of the characteristics of the annular myocardium, but also solving the defects of the traditional method on the description of the dynamic myocardium, thereby better extracting the texture and the motion characteristics of the myocardium and accurately measuring the ischemic condition of each segment of the myocardium.
Description
Technical Field
The invention relates to the technical field of ultrasonic medical engineering, in particular to a medical image segmentation and detection method.
Background
Cardiovascular disease is a leading cause of death in most developed countries and in many developing countries. Among them, the incidence and mortality of Acute Myocardial Infarction (AMI) continue to rise, AMI has caused huge consumption of medical resources in our country, and the average cost of the in-hospital treatment is in the leading position of the in-hospital reason. Therefore, the rapid and accurate discovery and diagnosis of the early ischemic lesions are of great significance for constructing a harmonious medical environment. The cardiac ultrasonic examination has the specific advantages of no wound, real time, low price and the like, is a preferred examination method for patients with cardiovascular diseases, and compared with other detection methods, the ultrasonic examination can quickly identify various cardiovascular diseases in real time and can quickly detect the cardiac function examination. However, when identifying the abnormal motion of the left ventricular wall, the sonographer relies on his own experience to diagnose the disease, and most of the young physicians cannot effectively find the abnormal motion cardiac muscle or ischemic cardiac muscle, and have a very subjective basis and no objective quantitative or qualitative basis. Moreover, the left ventricle is the most important part of the heart, and the abnormality of the left ventricle is also an important basis for judging the heart lesion clinically, which is expressed in the form and motion state. Therefore, how to accurately analyze the myocardial ischemia quantitatively and qualitatively by using echocardiography examination is a hot spot of medical research.
The basis of quantitative and qualitative analysis of cardiac left ventricular myocardial changes is to determine the ischemia condition of the left ventricular myocardium, currently, in clinical diagnosis, the positioning of the myocardium in an ultrasonic image is basically completed manually, which wastes time and labor and has large workload, the existing ultrasonic instrument can only complete some simple detection works generally, and does not achieve the expected effect every time, multiple attempts are often needed, and great difficulty still exists in accurate positioning of the ventricle. The localization of the ischemic myocardium is subjective, and the degree of ischemia is not a quantitative measure. Therefore, the left ventricle ultrasonic image needs to be automatically analyzed by means of a computer artificial intelligence technology, the left ventricle myocardial area is detected from background tissues, the endocardium and the epicardium of the suspicious myocardium are accurately drawn, characteristics are extracted to measure quantitative parameters such as myocardial motion and ischemia degree, and the ischemia degree of the myocardium is diagnosed by means of a classifier.
The existing myocardial positioning technology is influenced by noise, and the boundary of myocardial detection is not continuous and smooth; there are regions of the myocardium divided into several discrete regions; some of them remove the influence of noise, but the segmentation result is not accurate enough, and include regions other than the myocardium. In an ultrasonic image, the gray scale of the epicardium of the left ventricle is not uniform, the edge information is weak, the contour is discontinuous, and the like, so that great difficulty is caused for the automatic segmentation of the epicardium. In myocardial ischemia diagnosis, whether ischemia exists in a segment is determined by observing the myocardial motor function and gray texture information with the naked eye of a doctor. The diagnosis method has the defects of high subjectivity, low precision of diagnosis process and result, strong subjectivity and the like, and is easily influenced by subjective factors such as doctor experience, knowledge and the like.
In the automatic analysis of the ultrasonic image, the existing algorithm rarely considers the motion state of the myocardium along with the time, and for the doctor, the human eye has the deficiency in quantitatively measuring the motion of the myocardium. In addition, the deep convolution network extracts the features of the object to be classified by using a rectangular convolution kernel, and can only extract the features of the object in the horizontal and vertical directions, so that the convolution network has disadvantages in describing objects (such as myocardium) in a strip shape, and the extracted features cannot well describe the gray scale and texture features of the myocardium, thereby affecting the detection accuracy of the ischemic myocardium.
The application provides a medical image segmentation and detection method, which is based on a time series deep learning network, positions cardiac muscle in an ultrasonic image, and carries out quantitative analysis on tissue morphological characteristics and motion functional characteristics of the cardiac muscle, thereby solving the analysis and detection problems of clinical myocardial ischemia ultrasonic images, and completing the classification of each segment of the cardiac muscle and the automatic detection of ischemic cardiac muscle segments.
Disclosure of Invention
The invention aims to provide a medical image segmentation and detection method, which is based on a time series deep learning network, is used for positioning the myocardium in an ultrasonic image and quantitatively analyzing the tissue morphological characteristics and the motion functional characteristics of the myocardium, so that the analysis and detection problems of the myocardial ischemia ultrasonic image in clinic are solved, and the classification of each segment of the myocardium and the automatic detection of the ischemic myocardium segment are completed.
The invention provides a medical image segmentation and detection method, which comprises the following steps:
s1, establishing an image segmentation and detection system;
s2, inputting an ultrasonic two-dimensional time sequence gray level image SeqBUS to an image segmentation and detection system;
s3, manually marking the SeqBUS, and detecting a target area by using a SeqBUS training time sequence target detection depth network SeqObjDetNet;
s4, detecting the SeqBUS which is not manually identified by adopting the trained SeqObjDetNet, converting Hough of the outer boundary of the detected target area into candidate points, and performing elliptic curve fitting on the candidate points;
and S5, outputting an elliptic curve fitting result to obtain the position of the target area.
Further, after the position of the target myocardial area is obtained, feature extraction is performed on the target area, and the ischemic myocardial segment is classified and detected, which specifically comprises the following steps:
s101, inputting an ultrasonic two-dimensional time sequence gray image with a target area and a myocardial area black-and-white binary image segmentation and detection system;
s102, acquiring a gray level image SeqBUSMy of the characteristic area;
s103, training a time sequence polar coordinate deep convolution network SeqPolCNN according to the artificial labeling condition of the ischemic myocardium;
s104, performing feature detection and classification on a target area without manual labeling features by adopting a trained time series polar coordinate deep convolution network;
and S105, outputting a feature detection result of the myocardium in the medical image.
Further, the SeqBUS in step S2 is an ultrasound two-dimensional time series gray scale image of the ventricular region, and the target region is the position of the left ventricular myocardium.
Further, the feature detection result of the medical image comprises an ischemic condition result and an ischemic segment feature measurement result of each segment of the myocardium.
Compared with the prior art, the invention has the following remarkable advantages:
the invention provides a medical image segmentation and detection method, which considers the motion characteristic of the cardiac muscle, describes the motion change of the cardiac muscle on the space-time by using a sequence image, and determines the accurate position of the cardiac muscle by using the spatial position relation of the cardiac chamber and the cardiac muscle, thereby removing the influence of noise on the segmentation result and maintaining the accuracy and the integrity of the shape of the cardiac muscle. The invention classifies each segment of the myocardial zone by utilizing the time sequence polar coordinate deep convolution network, not only solves the defects of the traditional deep convolution network on the description of the characteristics of the annular myocardium, but also solves the defects of the traditional method on the description of the dynamic myocardium, thereby better extracting the texture and the motion characteristics of the myocardium and accurately measuring the ischemia condition of each segment of the myocardium.
Drawings
FIG. 1 is a flow chart of medical image segmentation and detection provided by the present invention;
fig. 2 is a flowchart of the feature extraction of the target region according to the present invention.
Detailed Description
The technical solutions of the embodiments of the present invention are clearly and completely described below with reference to the drawings in the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
At present, the research on the segmentation and positioning of the left ventricle and the myocardium of an ultrasound image is mainly focused on a frame based on curve evolution, and can be divided into two types according to a basic expression mode of a curve: the method comprises a parametric active contour model and a geometric active contour model, wherein the former representation method is a Snake model, and the latter representation method is a level set method.
The conventional level set method has two defects: (1) The fuzzy edge detection contour is difficult, even impossible to detect; (2) The level set function needs to be continuously reinitialized into the symbol distance function, and the calculation amount is huge. To solve the problem of fuzzy edge detection, machine Learning (ML) models have been applied to echocardiography to detect myocardial infarction and ischemic myocardium. The existing methods rely on describing new characteristics of the myocardium, and most of the characteristics need manual design and multiple tests, have certain experience and subjectivity, and reduce the performance of the automatic diagnosis method for myocardial infarction and ischemia, thereby limiting the universality of the automatic diagnosis method.
For easy understanding and explanation, as shown in fig. 1-2, the present invention provides a medical image segmentation and detection method, comprising the following steps:
s1, establishing an image segmentation and detection system;
s2, inputting an ultrasonic two-dimensional time sequence gray level image SeqBUS to an image segmentation and detection system;
s3, manually marking the SeqBUS, and training a time sequence target detection depth network SeqObjDetNet by using the SeqBUS to detect a target area so that the target area has the capacity of detecting the position of the left ventricle;
s4, detecting the SeqBUS which is not manually identified by adopting the trained SeqObjDetNet, carrying out Hough transformation on the outer boundary of the detected target area into candidate points, and carrying out elliptic curve fitting on the candidate points;
and S5, outputting an elliptic curve fitting result to obtain the position of the target area. The fitting result of the elliptic curve is the myocardial intima, and the myocardial adventitia is obtained by expanding the myocardial intima according to the morphological attribute of the myocardium. The region of the epicardium was assigned white and the remaining regions were assigned black, and this image was taken as an output image.
The SeqBUS in step S2 in this application is an ultrasound two-dimensional time series grayscale image of a ventricular region, and the target region is the position of the left ventricular myocardium.
The image segmentation and detection method comprises a target area positioning module and a feature detection module, wherein the target area positioning module detects and positions a target area through training teaching, and the feature detection module detects features of the target area through the training teaching.
Example 1
After the position of the target myocardial area is acquired, feature extraction is carried out on the target area, and ischemic myocardial segments are classified and detected, and the method specifically comprises the following steps:
s101, inputting an ultrasonic two-dimensional time sequence gray image with a target area and a myocardial area black-and-white binary image to an image segmentation and detection system;
s102, acquiring a gray level image SeqBUSMy of a characteristic region (namely a myocardium ring shape);
s103, training a time sequence polar coordinate depth convolution network SeqPolCNN according to the artificial labeling condition of the ischemic myocardium, so that the SeqPolCNN has the capability of detecting the ischemia condition of each segment of the myocardium;
s104, performing feature detection and classification on a target area without manual labeling features by adopting a trained time series polar coordinate deep convolution network;
and S105, outputting a feature detection result of the myocardium in the medical image.
The feature detection result of the medical image comprises an ischemia condition result and an ischemia segment feature measurement result of each segment of the myocardium.
Compared with the same type of algorithms and methods, the method has excellent results through experiments and simulation, not only can accurately detect the myocardial area without being influenced by speckle noise on an ultrasonic image, but also achieves higher accuracy on the detection of the myocardial ischemia segment.
The above-mentioned embodiments are only preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, and any simple modifications or equivalent substitutions of the technical solutions that can be obviously obtained by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.
Claims (3)
1. A medical image segmentation and detection method is characterized by comprising the following steps:
s1, establishing an image segmentation and detection system;
s2, inputting an ultrasonic two-dimensional time sequence gray level image SeqBUS to an image segmentation and detection system;
s3, manually marking the SeqBUS, and utilizing the SeqBUS to train a time sequence target detection deep network SeqObjDetNet to detect a target area;
s4, detecting the SeqBUS which is not manually identified by adopting the trained SeqObjDetNet, carrying out Hough transformation on the outer boundary of the detected target area into candidate points, and carrying out elliptic curve fitting on the candidate points;
s5, outputting an elliptic curve fitting result to obtain the position of the target area;
after the position of the target myocardial area is acquired, feature extraction is carried out on the target area, and ischemic myocardial segments are classified and detected, and the method specifically comprises the following steps:
s101, inputting an ultrasonic two-dimensional time sequence gray image with a target area and a myocardial area black-and-white binary image to an image segmentation and detection system;
s102, obtaining a gray level image SeqBUSMy of the characteristic area;
s103, training a time sequence polar coordinate deep convolution network SeqPolCNN according to the artificial labeling condition of the ischemic myocardium;
s104, performing feature detection and classification on a target area which is not subjected to manual feature labeling on the SeqBUSMy by adopting a trained time sequence polar coordinate deep convolution network;
and S105, outputting a feature detection result of the cardiac muscle in the medical image.
2. The medical image segmentation and detection method as claimed in claim 1, wherein the SeqBUS in the step S2 is an ultrasound two-dimensional time series gray scale image of a ventricular region, and the target region is a position of a left ventricular myocardium.
3. The medical image segmentation and detection method as claimed in claim 1, wherein the feature detection results of the medical image include ischemia results of each segment of the myocardium and ischemia segment feature measurement results.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110137243.XA CN112932535B (en) | 2021-02-01 | 2021-02-01 | Medical image segmentation and detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110137243.XA CN112932535B (en) | 2021-02-01 | 2021-02-01 | Medical image segmentation and detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112932535A CN112932535A (en) | 2021-06-11 |
CN112932535B true CN112932535B (en) | 2022-10-18 |
Family
ID=76240801
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110137243.XA Active CN112932535B (en) | 2021-02-01 | 2021-02-01 | Medical image segmentation and detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112932535B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102855623A (en) * | 2012-07-19 | 2013-01-02 | 哈尔滨工业大学 | Method for measuring myocardium ultrasonic angiography image physiological parameters based on empirical mode decomposition (EMD) |
EP2738741A1 (en) * | 2011-07-19 | 2014-06-04 | Kabushiki Kaisha Toshiba | Apparatus and method for tracking contour of moving object, and apparatus of and method for analyzing myocardial motion |
CN104978730A (en) * | 2014-04-10 | 2015-10-14 | 上海联影医疗科技有限公司 | Division method and device of left ventricular myocardium |
CN108898606A (en) * | 2018-06-20 | 2018-11-27 | 中南民族大学 | Automatic division method, system, equipment and the storage medium of medical image |
CN110475505A (en) * | 2017-01-27 | 2019-11-19 | 阿特瑞斯公司 | Utilize the automatic segmentation of full convolutional network |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7813537B2 (en) * | 2006-05-15 | 2010-10-12 | Siemens Medical Solutions Usa, Inc. | Motion-guided segmentation for cine DENSE images |
US8583209B2 (en) * | 2007-10-03 | 2013-11-12 | Siemens Aktiengesellschaft | Method and system for monitoring cardiac function of a patient during a magnetic resonance imaging (MRI) procedure |
CN102871686B (en) * | 2012-03-05 | 2015-08-19 | 杭州弘恩医疗科技有限公司 | The apparatus and method of physiological parameter are measured based on 3D medical image |
CN103549949B (en) * | 2013-10-21 | 2015-04-29 | 华南理工大学 | Myocardial ischemia auxiliary detecting method based on deterministic learning theory |
US9875581B2 (en) * | 2014-10-31 | 2018-01-23 | The Regents Of The University Of California | Automated 3D reconstruction of the cardiac chambers from MRI or ultrasound |
CN104794706A (en) * | 2015-04-03 | 2015-07-22 | 哈尔滨医科大学 | Method for examining cardiac muscles and measuring features by aid of ultrasonic images |
CN108013904B (en) * | 2017-12-15 | 2020-12-25 | 无锡祥生医疗科技股份有限公司 | Heart ultrasonic imaging method |
CN109377470A (en) * | 2018-03-20 | 2019-02-22 | 任昊星 | A kind of heart disease risk forecasting system |
KR20210010920A (en) * | 2018-05-17 | 2021-01-28 | 더 유나이티드 스테이츠 오브 아메리카, 디파트먼트 오브 헬스 앤드 휴먼 서비시즈, 내셔널 인스티튜츠 오브 헬스 | A method and system for automatically generating and analyzing fully quantitative pixel-by-pixel myocardial blood flow and myocardial perfusion reserve maps for detecting ischemic heart disease using cardiac perfusion magnetic resonance imaging. |
CN109285157A (en) * | 2018-07-24 | 2019-01-29 | 深圳先进技术研究院 | Myocardium of left ventricle dividing method, device and computer readable storage medium |
CN109512423A (en) * | 2018-12-06 | 2019-03-26 | 杭州电子科技大学 | A kind of myocardial ischemia Risk Stratification Methods based on determining study and deep learning |
CN111329467A (en) * | 2018-12-18 | 2020-06-26 | 上海图灵医疗科技有限公司 | Heart disease auxiliary detection method based on artificial intelligence |
CN111012377B (en) * | 2019-12-06 | 2020-11-03 | 北京安德医智科技有限公司 | Echocardiogram heart parameter calculation and myocardial strain measurement method and device |
CN111739000B (en) * | 2020-06-16 | 2022-09-13 | 山东大学 | System and device for improving left ventricle segmentation accuracy of multiple cardiac views |
-
2021
- 2021-02-01 CN CN202110137243.XA patent/CN112932535B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2738741A1 (en) * | 2011-07-19 | 2014-06-04 | Kabushiki Kaisha Toshiba | Apparatus and method for tracking contour of moving object, and apparatus of and method for analyzing myocardial motion |
CN105105775A (en) * | 2011-07-19 | 2015-12-02 | 株式会社东芝 | Myocardial motion analysis device |
CN102855623A (en) * | 2012-07-19 | 2013-01-02 | 哈尔滨工业大学 | Method for measuring myocardium ultrasonic angiography image physiological parameters based on empirical mode decomposition (EMD) |
CN104978730A (en) * | 2014-04-10 | 2015-10-14 | 上海联影医疗科技有限公司 | Division method and device of left ventricular myocardium |
CN110475505A (en) * | 2017-01-27 | 2019-11-19 | 阿特瑞斯公司 | Utilize the automatic segmentation of full convolutional network |
CN108898606A (en) * | 2018-06-20 | 2018-11-27 | 中南民族大学 | Automatic division method, system, equipment and the storage medium of medical image |
Non-Patent Citations (2)
Title |
---|
Sermesant, A ; Delingette, H ; Ayache, N.An electromechanical model of the heart for image analysis and simulation.《IEEE TRANSACTIONS ON MEDICAL IMAGING》.2006, * |
基于深度学习的全自动心肌分割算法研究;吴圣杰;《中国优秀硕士学位论文全文数据库》;20180615;第E080-14页 * |
Also Published As
Publication number | Publication date |
---|---|
CN112932535A (en) | 2021-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Neto et al. | An unsupervised coarse-to-fine algorithm for blood vessel segmentation in fundus images | |
CN108464840B (en) | Automatic detection method and system for breast lumps | |
Menchón-Lara et al. | Early-stage atherosclerosis detection using deep learning over carotid ultrasound images | |
Mendonça et al. | 13 PH2 | |
Acharya et al. | Atherosclerotic risk stratification strategy for carotid arteries using texture-based features | |
CN110197713B (en) | Medical image processing method, device, equipment and medium | |
US20150023578A1 (en) | Device and method for determining border of target region of medical images | |
Nurmaini et al. | Accurate detection of septal defects with fetal ultrasonography images using deep learning-based multiclass instance segmentation | |
US20140233818A1 (en) | Methods and systems for segmentation in echocardiography | |
Ni Ni et al. | Anterior chamber angle shape analysis and classification of glaucoma in SS-OCT images | |
Binder et al. | Artificial neural networks and spatial temporal contour linking for automated endocardial contour detection on echocardiograms: A novel approach to determine left ventricular contractile function | |
JPWO2020027228A1 (en) | Diagnostic support system and diagnostic support method | |
Wijata et al. | Unbiased validation of the algorithms for automatic needle localization in ultrasound-guided breast biopsies | |
Zeng et al. | TUSPM-NET: A multi-task model for thyroid ultrasound standard plane recognition and detection of key anatomical structures of the thyroid | |
Deng et al. | Automated detection of fetal nuchal translucency based on hierarchical structural model | |
CN112932535B (en) | Medical image segmentation and detection method | |
Kanca et al. | Learning hand-crafted features for k-NN based skin disease classification | |
Ni et al. | Automatic cystocele severity grading in transperineal ultrasound by random forest regression | |
Yahav et al. | Strain curve classification using supervised machine learning algorithm with physiologic constraints | |
WO2023133929A1 (en) | Ultrasound-based human tissue symmetry detection and analysis method | |
CN114399493A (en) | Automatic detection and display method for ultrasonic brain abnormal area | |
Patel et al. | Arterial parameters and elasticity estimation in common carotid artery using deep learning approach | |
CN113298773A (en) | Heart view identification and left ventricle detection device and system based on deep learning | |
Vázquez et al. | Automatic arteriovenous ratio computation: Emulating the experts | |
Upendra et al. | Artificial neural network application in classifying the left ventricular function of the human heart using echocardiography |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20240410 Address after: No.107, Yanjiang West Road, Guangzhou, Guangdong 510000 Patentee after: SUN YAT-SEN MEMORIAL HOSPITAL, SUN YAT-SEN University Country or region after: China Address before: No.246, Xuefu Road, Nangang District, Harbin City, Heilongjiang Province Patentee before: Du Guoqing Country or region before: China Patentee before: Guo Yanhui |