Hongki Yoo
Korea Advanced Institute of Science and Technology · 工学
研究室紹介
Professor Hongki Yoo's research lab specializes in advanced optical imaging and photonics, with a strong focus on developing innovative, eco-friendly manufacturing techniques for planar diffractive optics and next-generation intravascular imaging systems. The lab integrates spectroscopic OCT with weakly supervised deep learning to enable automated, composition-aware tissue characterization—particularly for lipid detection in coronary arteries—without hardware modifications. A key research direction involves enhancing image quality and diagnostic accuracy through novel optical component design, such as index-matched cladding collimators for multimodal OCT systems. The lab also emphasizes green manufacturing processes, particularly direct laser writing, to support sustainable development in biomedical optics.
Research Overview
Research Output Trend
Figures are computed from collected data and may differ slightly.
Selected Papers
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
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
Supplementary Materials
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