DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Ye, Jong Chul | - |
dc.contributor.advisor | 예종철 | - |
dc.contributor.author | Kwon, Tae-Sung | - |
dc.date.accessioned | 2023-06-22T19:32:17Z | - |
dc.date.available | 2023-06-22T19:32:17Z | - |
dc.date.issued | 2022 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=997758&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/308380 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 바이오및뇌공학과, 2022.2,[iv, 36 p. :] | - |
dc.description.abstract | We address the application of the invertible neural network to the unsupervised denoising of low dose CT image data. Recently, CycleGAN was shown to provide high-performance, ultra-fast denoising for low dose X-ray computed tomography (CT) without the need for a paired training dataset. Although this was possible thanks to cycle consistency, CycleGAN requires two generators and two discriminators to enforce cycle consistency, demanding significant GPU resources and technical skills for training. To solve this problem, here we present a novel cycle-free CycleGAN architecture, which consists of a single generator and a discriminator but still guarantees the cycle consistency. The main innovation comes from the observation that the use of an invertible generator automatically fulfills the cycle consistency condition and eliminates the additional discriminator in the CycleGAN formulation. Extensive experiments using various levels of low dose CT images confirm that our method can improve denoising performance using only 24\% of learnable parameters, twice faster training time, and showed better trade-offs between the complexity versus reconstruction quality, compared to the existing methods. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.title | Unsupervised low dose CT image denoising by using invertible neural network | - |
dc.title.alternative | 가역적 뉴럴 네트워크를 활용한 저선량 전산화단층촬영 영상 잡음 제거에 관한 연구 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :바이오및뇌공학과, | - |
dc.contributor.alternativeauthor | 권태성 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.