유홍기 교수
Hongki Yoo
KAIST 기계공학과 · 공학
연구실 소개
유홍기 교수의 연구실은 광학 및 생체의학 영상 기술 분야에서 혁신적인 연구를 수행하고 있습니다. 주요 연구 방향은 스펙트럼 정보를 통합한 내시적 광간섭단층촬영(S-OCT) 기반의 자동화된 조직 특성 분석과, 약물 치료에 영향을 미칠 수 있는 립프드 분포의 정밀 진단입니다. 특히, 깊이 학습 기반의 약한 지도 학습 프레임워크를 활용해 레이저 조명 및 나노제조 공정을 통한 환경 친화적 광학 소자의 개발도 함께 진행하고 있습니다.
연구 현황
연구 성과 추이
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
주요 논문
14Evolving demands for compact, light-weight, and versatile optical systems across various industries require the facile integration of planar diffractive optics. For the manufacturing of diffractive optics, green manufacturing becomes the prerequisite with timely considerations of Environmental, Social, and Governance (ESG). Conventional manufacturing processes such as semiconductor lithography or nano /micro imprinting utilize a large amount of harmful chemicals. Meanwhile, direct laser writing
Recent advances in label-free optical imaging are transforming cardiovascular diagnostics by combining structural and biochemical insights. Intravascular optical coherence tomography (OCT) and fluorescence lifetime imaging (FLIm) provide complementary information for assessing atherosclerotic disease. This talk will highlight combined intravascular OCT/FLIm for clinical studies, deep learning–based reconstruction of saturation-distorted lifetime signals, chromatic OCT for extended depth of focus
Reliable identification of lipid distribution is critical for assessing coronary vulnerability, yet conventional optical coherence tomography (OCT) lacks compositional specificity. Expanding OCT into the spectral information through spectroscopic OCT (S-OCT), in combination with deep learning, enables automated, composition-aware tissue characterization without requiring hardware modification. This study aims to develop a weakly supervised deep learning framework for lipid detection and localiza
Reliable identification of lipid distribution is critical for assessing coronary vulnerability, yet conventional optical coherence tomography (OCT) lacks compositional specificity. Expanding OCT into the spectral information through spectroscopic OCT (S-OCT), in combination with deep learning, enables automated, composition-aware tissue characterization without requiring hardware modification. This study aims to develop a weakly supervised deep learning framework for lipid detection and localiza
Supplementary Materials
Reliable identification of lipid distribution is critical for assessing coronary vulnerability, yet conventional optical coherence tomography (OCT) lacks compositional specificity. Expanding OCT into the spectral information through spectroscopic OCT (S-OCT), in combination with deep learning, enables automated, composition-aware tissue characterization without requiring hardware modification. This study aims to develop a weakly supervised deep learning framework for lipid detection and localiza
Supplementary Materials
Supplementary Materials
Reliable identification of lipid distribution is critical for assessing coronary vulnerability, yet conventional optical coherence tomography (OCT) lacks compositional specificity. Expanding OCT into the spectral information through spectroscopic OCT (S-OCT), in combination with deep learning, enables automated, composition-aware tissue characterization without requiring hardware modification. This study aims to develop a weakly supervised deep learning framework for lipid detection and localiza
We developed a novel collimator with an expanded, index-matched cladding for double-clad fiber-based multimodal intravascular optical coherence tomography (OCT) systems. The proposed design mitigates multipath artifacts by reducing cladding-to-core coupling at fiber discontinuities, resulting in an improvement in signal-to-noise ratio. This enhancement effectively suppresses image artifacts and improves overall OCT image quality, thereby enabling more accurate diagnostic assessment in multimodal
Reliable identification of lipid distribution is critical for assessing coronary vulnerability, yet conventional optical coherence tomography (OCT) lacks compositional specificity. Expanding OCT into the spectral information through spectroscopic OCT (S-OCT), in combination with deep learning, enables automated, composition-aware tissue characterization without requiring hardware modification. This study aims to develop a weakly supervised deep learning framework for lipid detection and localiza
Reliable identification of lipid distribution is critical for assessing coronary vulnerability, yet conventional optical coherence tomography (OCT) lacks compositional specificity. Expanding OCT into the spectral information through spectroscopic OCT (S-OCT), in combination with deep learning, enables automated, composition-aware tissue characterization without requiring hardware modification. This study aims to develop a weakly supervised deep learning framework for lipid detection and localiza
대표 연구 분야
유홍기 교수의 연구를 Nubint에서 더 깊이 살펴보세요
이 연구실의 논문을 앱에서 열어 AI와 함께 읽고, 핵심을 요약하고, 내 글에 인용하세요.