Publications
Detailed Information
Effective Algorithm for Left Ventricle Segmentation in MRI : 심장 MRI에서 좌심실 세분화를 위한 효과적인 알고리즘
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 문병로 | - |
dc.contributor.author | 이일규 | - |
dc.date.accessioned | 2018-05-29T03:32:08Z | - |
dc.date.available | 2018-05-29T03:32:08Z | - |
dc.date.issued | 2018-02 | - |
dc.identifier.other | 000000150514 | - |
dc.identifier.uri | https://hdl.handle.net/10371/141551 | - |
dc.description | 학위논문 (석사)-- 서울대학교 대학원 : 공과대학 컴퓨터공학부, 2018. 2. 문병로. | - |
dc.description.abstract | The short-axis left ventricle segmentation of Cine MRI is a representative imaging analysis used as a medical care. It is difficult to obtain the same segmentation results when performing image analysis with a human hand. Time and effort are being consumed due to different segmentation results. The algorithm consists of two deep learning models proposed in this paper, it provides saving time and effort and also obtain same segmentation results always. The first model uses the selective search to detect the region of interest from irrespective size of image and obtain the center point of left ventricle by deep learning. The second model consists of applying
coordinate transformation to the image and finding the boundary of the endocardium and epicardium by deep learning. The number of patients used was 24, and totally 194 slices of Cine MRI were used. Among them, 19 were used for training and 5 were used for testing. deep learning for short-axis left ventricle segmentation and other algorithms were used to solve the problem, based on the analysis of the experimental results, we identify the problems and show the possibility. | - |
dc.description.tableofcontents | Chapter 1. Introduction 1
Chapter 2. Preliminaries 3 2.1 Selective Search 3 2.2 Artificial Neural Networks 6 Chapter 3. The Proposed System 10 3.1 Formulation of the Problem 10 3.2 ROI Detection Model 10 3.3 Boundary Detection Model 12 Chapter 4. Experiment 16 4.1 Data of Subjects 16 4.2 Experimental Setup 18 4.3 Comparision of Results 19 4.4 Generated Boundary 24 Chapter 5. Conclusion 25 Bibliography 26 요약 28 | - |
dc.format | application/pdf | - |
dc.format.extent | 850871 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject | Image processing | - |
dc.subject | left ventricle segmentation | - |
dc.subject | artificial neural network | - |
dc.subject | cardiac magnetic resonance imaging | - |
dc.subject.ddc | 621.39 | - |
dc.title | Effective Algorithm for Left Ventricle Segmentation in MRI | - |
dc.title.alternative | 심장 MRI에서 좌심실 세분화를 위한 효과적인 알고리즘 | - |
dc.type | Thesis | - |
dc.contributor.AlternativeAuthor | ILKYU LEE | - |
dc.description.degree | Master | - |
dc.contributor.affiliation | 공과대학 컴퓨터공학부 | - |
dc.date.awarded | 2018-02 | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.