DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Ye, Jongchul | - |
dc.contributor.advisor | 예종철 | - |
dc.contributor.author | Yang, Serin | - |
dc.date.accessioned | 2022-04-21T19:30:56Z | - |
dc.date.available | 2022-04-21T19:30:56Z | - |
dc.date.issued | 2021 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=948585&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/295271 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 바이오및뇌공학과, 2021.2,[iv, 34 p. :] | - |
dc.description.abstract | In X-ray computed tomography (CT) reconstruction, different filter kernels are used for different structures being emphasized. Since the raw sinogram data is usually removed after reconstruction, in case there are additional requirements for reconstructed images with other types of kernels that were not previously generated, the patient may need to be scanned again. Accordingly, there exists increasing demand for post-hoc image domain conversion from one kernel to another without sacrificing the image content. In this paper, we propose a novel unsupervised kernel conversion method using cycle-consistent generative adversarial network (cycleGAN) with adaptive instance normalization (AdaIN). In contrast to the existing deep learning approaches for kernel conversion, our method does not require paired dataset for training. In addition, our network can not only translate the images between two different kernels but also generate images on every interpolating path along an optimal transport between the two kernel image domains, enabling synergestic combination of the two filter kernels. Experimental results confirm the advantages of the proposed algorithm. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Computed tomography | - |
dc.subject | reconstruction kernels | - |
dc.subject | cycle-consistent adversarial networks | - |
dc.subject | style transfer | - |
dc.subject | adaptive instance normalization (AdaIN) | - |
dc.subject | 컴퓨터 단층촬영 | - |
dc.subject | 복원 커널 | - |
dc.subject | 순환적 생성 신경망 | - |
dc.subject | 스타일 변환 | - |
dc.subject | 적응적 인스턴스 정규화 | - |
dc.title | Continuous conversion of CT kernel using switchable CycleGAN with AdaIN | - |
dc.title.alternative | 적응적 인스턴스 정규화가 적용된 전환가능 CycleGAN을 이용한 비지도학습 기반의 연속적인 CT 커널 생성에 관한 연구 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :바이오및뇌공학과, | - |
dc.contributor.alternativeauthor | 양세린 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.