• Imaging Science in Dentistry
• /
• 제54권1호
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Applied Microscopy
• /
• 제39권4호
• /
• pp.349-354
• /
• 2009

TFT-LCD의 구조불량이 발생한 박막 트랜지스터에 대해서 집속이온빔 가공장치(Dual-beam FIB/SEM)를 이용하여 연속절편법(Serial sectioning)과 일련의 연속적인 2차원 주사전자현미경 이미지를 얻었고, IMOD 소프트웨어를 통해서 3차원 구조구현(3D reconstruction) 연구를 하였다. 3차원 구조구현 결과, Gate막과 Data막이 접합되어 있는 불량이 관찰되었다. 두 막이 접합되어서 ON/OFF 역할을 하는 Gate의 기능이 상실되었고, Data신호는 Drain을 통해서 투명전극에 전류를 공급하여 계속 빛나는 선 불량(line defect)이 발생한 것으로 판단된다. 이 논문의 결과인 집속이온빔 가공장치(Dual-Beam FIB/SEM)를 이용한 3차원 구조구현 연구와 연속절편법, 주사전자현미경 이미지작업, 이미지 프로세싱에 대한 결과는 향후 연구의 기초자료로 활용될 수 있을 것으로 판단된다.

• 한국수산과학회지
• /
• 제42권2호
• /
• pp.171-178
• /
• 2009

The purpose of this study was the development of a digital water color measurement system using a CCD optical device. Photographs of the standard medium of Forel and Ule water color scales in the laboratory, and one of sea surface above a Secchi disc (Z=SD/2) immersed in seawater were taken. The colors of these pictures were estimated with the value of CIE $L^*a^*b^*$. Water color assessment was conducted with a digital photo-image. In the case of the Forel scale (No. 1-11), the $b^*$ value was so high that the water color number was large. In the Ule scale (No.11-21), the $a^*$ value became high, which is why the number on the water color scale was large. The color of these pictures showed that the $a^*$ value in the Forel scale and the $b^*$ value in the Ule scale increase with the increase of the F value. The $a^*$ value of seawater color was always lower than the one on the Forel and Ule water color scales. This indicates that the color of the scales differs from actual seawater color. It was concluded that water color number can more effectively be assessed by estimating the ${\Delta}Eab^*$ color difference between the water scales and actual seawater color.

• 제어로봇시스템학회논문지
• /
• 제21권12호
• /
• pp.1100-1108
• /
• 2015

A camera-based real-time hand posture and gesture recognition system is proposed for controlling various devices inside automobiles. It uses an imaging system composed of a camera with a proper filter and an infrared lighting device to acquire images of hand-motion sequences. Several steps of pre-processing algorithms are applied, followed by a background normalization process before segmenting the hand from the background. The hand posture is determined by first separating the fingers from the main body of the hand and then by finding the relative position of the fingers from the center of the hand. The beginning and ending of the hand motion from the sequence of the acquired images are detected using pre-defined motion rules to start the hand gesture recognition. A set of carefully designed features is computed and extracted from the raw sequence and is fed into a decision tree-like decision rule for determining the hand gesture. Many experiments are performed to verify the system. In this paper, we show the performance results from tests on the 550 sequences of hand motion images collected from five different individuals to cover the variations among many users of the system in a real-time environment. Among them, 539 sequences are correctly recognized, showing a recognition rate of 98%.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제1권4호
• /
• pp.239-247
• /
• 2001

The Fluorescence-Activated Cell Sorter (FACS) is a well-established instrument used for identifying, enumerating, classifying and sorting cells by their physical and optical characteristics. For a miniaturized FACS device, a disposable plastic microchip has been developed which has a hydrodynamic focusing chamber using soft lithography. As the characteristics of the spatially confined sample stream have an effect on sample throughput, detection efficiency, and the accuracy of cell sorting, systematic fluid dynamic studies are required. Flow visualization is conducted with a laser scanning confocal microscopy (LSCM), and three-dimensional flow structure of the focused sample stream is reconstructed from 2D slices acquired at $1\mutextrm{m}$ intervals in depth. It was observed that the flow structure in the focusing chamber is skewed by unsymmetrical velocity profile arising from trapezoidal cross section of the microchannel. For a quantitative analysis of a microscopic flow structure, Confocal Micro-PIV system has been developed to evaluate the accelerated flow field in the focusing chamber. This study proposes a method which defines the depth of the measurement volume using a detection pinhole. The trajectories of red blood cells (RBCs) and their interactions with surrounding flow field in the squeezed sample stream are evaluated to find optimal shape of the focusing chamber and fluid manipulation scheme for stable cell transporting, efficient detection, and sorting

• 한국CDE학회논문집
• /
• 제20권3호
• /
• pp.246-253
• /
• 2015

In this paper, we propose calibration methods that can be applied to the markerless surgical robotic system for Intracerebral Hematoma (ICH) Surgery. This surgical robotic system does not require additional process of patient imaging but only uses CT images that are initially taken for a diagnosis purpose. Furthermore, the system applies markerless registration method other than using stereotactic frames. Thus, in overall, our system has many advantages when compared to other conventional ICH surgeries in that they are non-invasive, much less exposed to radiation exposure, and most importantly reduces a total operation time. In the paper, we specifically focus on the application of calibration methods and their verification which is one of the most critical factors that determine the accuracy of the system. We implemented three applications of calibration methods between the coordinates of robot's end-effector and the coordinates of 3D facial surface scanner, based on the hand-eye calibration method. Phantom tests were conducted to validate the feasibility and accuracy of our proposed calibration methods and the surgical robotic system.

• 방송공학회논문지
• /
• 제5권2호
• /
• pp.281-289
• /
• 2000

영상을 획득하는 과정에 있어서 영상 획득 장치 또는 피사체의 흔들림은 영상에 손상을 가져온다. 이러한 손상을 움직임 열화(motion blur)라고 부르며, 영상의 선명도를 떨어뜨리는 주된 원인이 된다. 최근 연구에서 밝힌 극점자취 방법을 통해 주어진 열화영상에서 열화의 PSF(Point Spread Function) 특성을 구하는데 사용되는 중요한 파라메터를 추출 할 수 있다. 이러한 극점 자취방법으로, 노이즈에 의한 열화에 관계없이 적은 연산량으로 움직임 열화의 방향을 추출할 수 있다. 본 논문에서는 통계적 극점 자취 접근법을 새롭게 제안한다. 저주파 영역에서 움직임 열화방향의 추정오차를 줄이기 위해 ML(Maximum Likelihood)분류를 통해 오차를 유발하는 극점을 선택하여 가중치를 적용, 그 영향을 최소화한다. 선형 예측법을 사용하여, 불규칙적 자료가 극점으로 선택되는 것을 방지한다 제안된 MALM(Moving average least mean)방법은 두번째로 큰 극점의 검출을 위해 움직임의 정도를 판별하는데 사용된다. MALM방법은 자체적으로 노이즈 제거 과정을 내포하고 있으므로 노이즈가 많은 환경에서도 파라메터를 추출할 수가 있다. 실험에서 우리는 제안된 방법을 통해 얻어진 정보를 사용하여, 열화 된 이미지를 효율적으로 복구해 낼 수 있었다.

• Imaging Science in Dentistry
• /
• 제49권4호
• /
• pp.301-306
• /
• 2019

Purpose: This report presents a procedure for performing power Doppler ultrasound-guided sialography using the phenomenon of increased blood flow and illustrates its application to practical patient cases. Materials and Methods: The salivary gland was scanned using ultrasound equipment (GE LOGIQ5 Expert^® device; GE Medical Systems, Milwaukee, WI, USA) to identify pathological findings related to the patient's chief complaint. To identify the orifice of the main duct, it should be cannulated using a lacrimal dilator. After inserting the catheter into the cannulated main duct, the position of the catheter within the duct was confirmed by ultrasound. A contrast agent was injected until the patient felt fullness, and ultrasound (B-mode) was used to confirm whether the contrast agent filled the main canal and secondary and tertiary ducts. Then, power Doppler ultrasound was performed to determine whether the salivary gland had increased blood flow. Results: In 2 cases in this report, a power Doppler ultrasound scan showed a significant increase in blood flow after contrast medium injection, which was not observed on a preoperative scan. Conclusion: Power Doppler ultrasound was found to be a simple, safe, and effective tool for real-time sialography monitoring.

• 한국가시화정보학회지
• /
• 제14권3호
• /
• pp.38-45
• /
• 2016

Separation of particles based on different sizes, detection of pathogenic bacteria and isolation of leukocytes from whole blood are typical applications of spiral or helical microchannels. The present study focuses on developing a CD4+ T-cell counting device for monitoring HIV/AIDS patients with the aid of a helical minichannel used for a sample cartridge. For the experiment, $10{\mu}m$ sized microbeads were used for visualization with a fluorescence imaging system. Alignment of microbeads was investigated in a stationary and spinning sample cartridge filled with glycerol-water mixtures of different densities. The helical minichannel was spun using a DC motor controlled by an Arduino board with a Bluetooth shield. It was found that when the sample cartridge was made stationary, no bead alignment was achieved for a medium with density (0% and 20% glycerol) lower than that of the beads, but when it was spun at 2000-3000 rpm for 1-4 min, an alignment was obtained at the top of the channel facilitating optical detection and enumeration of those microbeads. Since an alignment of microbeads was achieved for a medium with density as that of blood plasma, the same approach can be applied for aligning and counting CD4+ T-lymphocytes in whole blood samples collected from patients.

• 대한방사선기술학회지:방사선기술과학
• /
• 제41권3호
• /
• pp.209-214
• /
• 2018

The results of CBCT was obtained using image guided radiation therapy for radiation therapy in 5 prostate cancer patients. Using these results, we compared and evaluated the dose changes according to the treatment plan depending on the volume and position of bladder, rectum, and prostate. The 28 images of CBCT were acquired using On-Board Imaging device before radiotherapy. After the outline of bladder, rectum, and PTV, pCT images and CBCT images for radiotherapy were treated respectively. The volume of the bladder was increased by 105.6% and decreased by 45.2%. The volume of the rectum was increased by 30.5% and decreased by 20.3%. Prostate volume was increased by 6.3% and decreased by 12.3%. The mean dose of the rectum was higher in the CBCT than in the pCT, and V40 (equivalent to 40 Gy) of the bladder showed a reduction in all treatment regimens in the CBCT than in the pCT. Conformity treatment and homogeneity index of PTV showed better results in all treatment regimens using pCT than CBCT. It was found that the dose distribution of the pelvic internal organs varied greatly according to the patient 's condition and pretreatment.