본문 바로가기 주메뉴 바로가기
22대 대한민국 국회 국회정보길라잡이 국회의원검색
회원가입 로그인
HOME

정보검색

소장정보 검색

국회도서관 홈으로 정보검색 소장정보 검색
결과 내 검색 동의어 포함
결과 내 검색 동의어 포함

-

목차보기

Title Page

Contents

Abstract 9

Ⅰ. Introduction 10

Ⅱ. Methods and Materials 14

A. Reverberation artifact in medical ultrasound B-mode imaging 14

B. Experimental setup 16

1) Reverberation artifact phantom 16

2) Data acquisition 18

C. DCL for Reverberation artifact reduction (DCL-Reverb) 19

D. Image evaluation 23

Ⅲ. Results and Discussion 24

A. Results on ex-vivo images 24

B. Results on in-vivo images 28

Ⅳ. Conclusion 31

References 32

List of Tables

Table.1. Results of CNR (dB) and gCNR measured in B-mode images for ex-vivo cyst-shape phantom in Fig.8-(a). The orange and green boxes indicate of the... 27

Table.2. Results of CNR (dB) and gCNR measured in B-mode images of transverse section for ex-vivo carotid-shape phantom in Fig.8-(b). The orange and green boxes... 27

Table.3. Results of CNR (dB) and gCNR measured in B-mode images of longitudinal section for ex-vivo carotid-shape phantom in Fig.8-(c). The orange and green boxes... 27

Table.4. Results of CNR (dB) and gCNR measured in B-mode images for in-vivo carotid artery in Fig.10. The orange and green boxes indicate of the anechoic and... 30

List of Figures

Fig.1. A graphic illustration showing (a) the process by which reverberation artifact occur and (b) the process by which attenuation occurs between organizations. 15

Fig.2. (a) Reverberation artifact seen at evenly spaced echogenic bands beyond a copper intrauterine device. (b) Reverberation artifact seen in the anterior portion of... 15

Fig.3. (a) A cylindrical mold with a diameter of 6 mm and a length of 75 mm was used to produce the carotid-shaped reverberation artifact model. (b) A spherical... 16

Fig.4. A pork phantom with (a) carotid-shape, and (b) cyst-shape for reverberation artifact. (c) The completed reverberation artifact phantom for RF data acquisition. 17

Fig.5. Plane wave images with reverberation artifacts using (a) carotid-shape and (b) cyst-shape phantoms. 18

Fig.6. Schematic representation of the DCL-Reverb framework about (a) training phase and (b) inference phase. 20

Fig.7. CNN based on the U-Net architecture for DCL-Reverb training. 22

Fig.8. Reconstructed B-mode images by DAS, 3DCNN, DCL, and DCL-Reverb from the results scanned in ex-vivo (a) cyst shape, (b) carotid-shape with transverse... 25

Fig.9. Normalized intensity (dB) of axial profiles for four beamforming methods in the yellow arrow scanline from Fig.8-(a), (b), and (c). The red arrow point in the... 26

Fig.10. Reconstructed B-mode images by (a) DAS, (b) 3DCNN, (c) DCL, and (d) DCL-Reverb from the results scanned in in-vivo carotid artery. CNR and gCNR... 29

Fig.11. Normalized intensity (dB) of axial profiles for four beamforming methods in the yellow arrow scanline from Fig.10-(a). The red arrow point in the graph is the... 30

초록보기

 Reverberation artifacts in medical ultrasound B-mode images, caused by multiple repetitive reflections of the echo signal, significantly degrade image quality and hinder accurate diagnoses. Various studies, particularly those focusing on deep learning, have proposed techniques to mitigate these artifacts. Deep learning-based methods face a primary challenge in training strategy, which is categorized into supervised, semi-supervised, and unsupervised approaches. While supervised learning is simple and effective when input and ground-truth data are available, it is often impractical for artifact reduction due to the difficulty of obtaining suitable data. Conversely, unsupervised learning presents a promising alternative to overcome these data acquisition challenges. A recent advancement is deep coherence learning (DCL), an unsupervised technique specifically for enhancing ultrasound imaging quality. In this thesis, a custom phantom is designed to leverage DCL for suppressing reverberation artifacts. The effectiveness of the developed deep coherence learning with reverberation dataset (DCL-Reverb) was assessed using real-world experimental data. Quantitatively, DCL-Reverb demonstrated higher contrast-to-noise ratio (CNR) and generalized contrast-to-noise ratio (gCNR) compared to conventional methods. Qualitatively, it also achieved clearer B-mode images and superior artifact suppression in axial profiles.

추천서가 (다양한 추천 자료를 만나보세요)

권호기사보기

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 기사목차