서브메뉴
검색
Developing Computational Optical Imaging Systems With Artificial Intelligence
Developing Computational Optical Imaging Systems With Artificial Intelligence
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211153124
- ISBN
- 9798346851356
- DDC
- 610
- 저자명
- Du, Xiaoxi.
- 서명/저자
- Developing Computational Optical Imaging Systems With Artificial Intelligence
- 발행사항
- [Sl] : University of California, Los Angeles, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 123 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-06, Section: B.
- 주기사항
- Advisor: Gao, Liang.
- 학위논문주기
- Thesis (Ph.D.)--University of California, Los Angeles, 2024.
- 초록/해제
- 요약Alzheimer's disease (AD) is a major risk for the aging population. The pathological hallmarks of AD-an abnormal deposition of amyloid β-protein (Aβ) and phosphorylated tau (pTau)-have been demonstrated in the retinas of AD patients, including in prodromal patients with mild cognitive impairment (MCI). Aβ pathology, especially the accumulation of the amyloidogenic 42-residue long alloform (Aβ42), is considered an early and specific sign of AD, and together with tauopathy, confirms AD diagnosis. To visualize retinal Aβ and pTau, state-of-the-art methods use fluorescence. However, administering contrast agents complicates the imaging procedure. To address this problem from fundamentals, this dissertation performed ex vivo studies to develop a label-free hyperspectral imaging method to detect the spectral signatures of Aβ42 and pS396-Tau. A deep learning framework was developed to predict their abundance in retinal cross-sections and transform a label-free HSI image to either a DAB or an immunofluorescent stained image in high accuracy. For the first time, we reported the spectral signature of pTau and provided a direct validation through immunostaining.For small incision in vivo imaging, optical endoscopes are mostly limited by two-dimensional views or very small number of three-dimensional (3D) views of pathological sites, and are intrinsically low in resolution caused by the limited fiber cores. The dissertation demonstrated a flexible light field endoscopy (Flex- LFE) imaging system capable of capturing depth information in a single shot. To address the resolution challenges inherent in endoscopic imaging, an AI-powered super-resolution pipeline is developed to enhance the quality of reconstructed images.Additionally, this work explored the utilization of tunable image-mapping optical coherence tomography (TIM-OCT) to further advance in imaging of the retina. While most current OCT devices require extensive scanning, by combining phase-only spatial light modulators with spectral domain OCT, TIM-OCT achieves tailored imaging performance and enables "eye motion freeze" snapshot imaging. Computational spectroscopic analysis is discussed to extract spectral signatures.This dissertation demonstrates the potential of AI-driven optical imaging systems in addressing critical challenges in biomedical imaging, with a particular focus on the early detection of AD. The findings are expected to lay the groundwork for label-free detection of AD.
- 일반주제명
- Bioengineering
- 일반주제명
- Optics
- 일반주제명
- Biomedical engineering
- 일반주제명
- Medical imaging
- 키워드
- Image prediction
- 키워드
- Super resolution
- 기타저자
- University of California, Los Angeles Bioengineering 0288
- 기본자료저록
- Dissertations Abstracts International. 86-06B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017165105
■00520250211153124
■006m o d
■007cr#unu||||||||
■020 ▼a9798346851356
■035 ▼a(MiAaPQ)AAI31764485
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a610
■1001 ▼aDu, Xiaoxi.
■24510▼aDeveloping Computational Optical Imaging Systems With Artificial Intelligence
■260 ▼a[Sl]▼bUniversity of California, Los Angeles▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a123 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-06, Section: B.
■500 ▼aAdvisor: Gao, Liang.
■5021 ▼aThesis (Ph.D.)--University of California, Los Angeles, 2024.
■520 ▼aAlzheimer's disease (AD) is a major risk for the aging population. The pathological hallmarks of AD-an abnormal deposition of amyloid β-protein (Aβ) and phosphorylated tau (pTau)-have been demonstrated in the retinas of AD patients, including in prodromal patients with mild cognitive impairment (MCI). Aβ pathology, especially the accumulation of the amyloidogenic 42-residue long alloform (Aβ42), is considered an early and specific sign of AD, and together with tauopathy, confirms AD diagnosis. To visualize retinal Aβ and pTau, state-of-the-art methods use fluorescence. However, administering contrast agents complicates the imaging procedure. To address this problem from fundamentals, this dissertation performed ex vivo studies to develop a label-free hyperspectral imaging method to detect the spectral signatures of Aβ42 and pS396-Tau. A deep learning framework was developed to predict their abundance in retinal cross-sections and transform a label-free HSI image to either a DAB or an immunofluorescent stained image in high accuracy. For the first time, we reported the spectral signature of pTau and provided a direct validation through immunostaining.For small incision in vivo imaging, optical endoscopes are mostly limited by two-dimensional views or very small number of three-dimensional (3D) views of pathological sites, and are intrinsically low in resolution caused by the limited fiber cores. The dissertation demonstrated a flexible light field endoscopy (Flex- LFE) imaging system capable of capturing depth information in a single shot. To address the resolution challenges inherent in endoscopic imaging, an AI-powered super-resolution pipeline is developed to enhance the quality of reconstructed images.Additionally, this work explored the utilization of tunable image-mapping optical coherence tomography (TIM-OCT) to further advance in imaging of the retina. While most current OCT devices require extensive scanning, by combining phase-only spatial light modulators with spectral domain OCT, TIM-OCT achieves tailored imaging performance and enables "eye motion freeze" snapshot imaging. Computational spectroscopic analysis is discussed to extract spectral signatures.This dissertation demonstrates the potential of AI-driven optical imaging systems in addressing critical challenges in biomedical imaging, with a particular focus on the early detection of AD. The findings are expected to lay the groundwork for label-free detection of AD.
■590 ▼aSchool code: 0031.
■650 4▼aBioengineering
■650 4▼aOptics
■650 4▼aBiomedical engineering
■650 4▼aMedical imaging
■653 ▼aAlzheimer's disease
■653 ▼aHyperspectral imaging
■653 ▼aImage prediction
■653 ▼aLight field imaging
■653 ▼aSuper resolution
■690 ▼a0202
■690 ▼a0752
■690 ▼a0800
■690 ▼a0574
■690 ▼a0541
■71020▼aUniversity of California, Los Angeles▼bBioengineering 0288.
■7730 ▼tDissertations Abstracts International▼g86-06B.
■790 ▼a0031
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17165105▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
