https://scholar.dgist.ac.kr/handle/20.500.11750/191 2026-08-03T09:49:59Z https://scholar.dgist.ac.kr/handle/20.500.11750/60471 Title: Precision enhancement of epidural force-sensing needle with machine learning Author(s): Cho, Gichan; Na, Jongyeol; Lee, Myung Ho; Kwon, Hyun-Jung; Song, Cheol Abstract: Epidural injection is used in pain intervention, requiring precise needle placement within the epidural space. Traditional techniques, such as loss of resistance and fluoroscopy-guided procedures, have limitations, including reliance on subjective assessment and radiation exposure. We proposed an optical force-sensing probe with an offset criterion of the needle tip-distal end to enhance the precision of puncture detection. The offset between the needle tip and the force-sensing probe is adjusted using a piezoelectric motor-based system with feedback position control. A Long Short-Term Memory model is also trained to detect the puncture. Insertion test on silicone phantom and ex-vivo specimens demonstrates that the system's offset range for enhancing precision of puncture detection is between 0.6 mm and 1 mm. Compared to the offset in the previous study, the AUC score of puncture detection increased from 0.61 to 0.86. This approach secures the improvement of puncture detection reliability in robot-assisted epidural injection. 2026-11-30T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/59945 Title: Compact forward-viewing multimodal fluorescent and optical coherence tomography endomicroscopic probe Author(s): Im, Jintaek; Cho, Gichan; Song, Cheol Abstract: We present a compact multimodal endomicroscope that enables simultaneous fluorescence (FL) and optical coherence tomography (OCT) imaging. While current endoscopy techniques are effective for wide-area and rapid inspection, there is a growing demand for real-time precise diagnostics, including detailed tissue morphology and tumor invasion depth. Histological analysis through biopsy remains the diagnostic standard but involves a time-consuming process that can delay treatment decisions. Our approach integrates two complementary imaging modalities-FL for visualizing tissue morphology and OCT for cross-sectional imaging-within a single probe compatible with standard gastrointestinal endoscopic channels. The system employs a Lissajous scanning mechanism to achieve forward-viewing, uniform illumination, and high-speed imaging. A compact imaging probe is fabricated by assembling a composite fiber, piezoelectric tube actuator, and asymmetrically attached polymer stiffener in parallel, enabling combined fluorescence and optical coherence imaging with complementary performance characteristics. Real-time image reconstruction is implemented using parallel computing to support high-throughput data processing. Imaging experiments on phantom targets and ex-vivo animal tissues confirm the system's capability to produce detailed, co-registered images of tissue morphology and structure. This technology offers a promising platform for enhancing diagnostic accuracy and enabling real-time decision-making in gastrointestinal endoscopy. 2025-11-30T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/59116 Title: Compact Fiber-Optic Sensor for Simultaneous Force Measurement and Depth Profiling Author(s): Im, Jintaek; Cho, Gichan; Song, Cheol Abstract: This study presents a compact sensor for simultaneous force measurement and depth profiling. We have developed a fiber-optic sensor structure capable of integrating common-path optical coherence tomography (CP-OCT) and Fabry–Pérot Interferometry (FPI) within a single unit. The proposed FPI-OCT sensor, with an outer diameter of 0.25 mm, is encased in a 0.41 mm hypodermic tube. Analytical models were developed for efficient beam path optimization and signal interpretation, where multiple common-path references yielded distinct interferometric signals. The sensor has a sufficiently high sampling rate of 780 Hz for both CP-OCT and FPI signal processing using parallel computing. The fabricated sensor achieved a measurable force range of approximately 5 N with a resolution of 0.119 mN, and a depth sensing range of 3.6 mm with an axial resolution of 6.0 μm. Furthermore, injection experiments with multilayered phantoms and ex vivo porcine eyes demonstrated that the sensor could simultaneously measure force and depth profiles, providing enriched information during the tasks. 2026-01-31T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/57317 Title: Machine-learning-based diabetes classification method using blood flow oscillations and Pearson correlation analysis of feature importance Author(s): Jung, Hanbeen; Yeo, Chaebeom; Jang, Eunsil; Chang, Yeonhee; Song, Cheol Abstract: Diabetes is a global health issue affecting millions of people and is related to high morbidity and mortality rates. Current diagnostic methods are primarily invasive, involving blood sampling, which can lead to infection and increased patient stress. As a result, there is a growing need for noninvasive diabetes diagnostic methods that are both accurate and fast. High measurement accuracy and fast measurement time are essential for effective noninvasive diabetes diagnosis; these can be achieved using diffuse speckle contrast analysis (DSCA) systems and artificial intelligence algorithms. In this study, we use a machine learning algorithm to analyze rat blood flow signals measured using a DSCA system with simple operation, easy fabrication, and fast measurement for helping diagnose diabetes. The results confirmed that the machine learning algorithm for analyzing blood flow oscillation data shows good potential for diabetes classification. Furthermore, analyzing the blood flow reactivity test revealed that blood flow signals can be quickly measured for diabetes classification. Finally, we evaluated the influence of each blood flow oscillation data on diabetes classification through feature importance and Pearson correlation analysis. The results of this study should provide a basis for the future development of hemodynamic-based disease diagnostic methods. © 2024 The Author(s). Published by IOP Publishing Ltd. 2024-11-30T15:00:00Z