• Journal of the Korean Society of Radiology
• /
• v.18 no.2
• /
• pp.93-99
• /
• 2024

The digital medical imaging, especially, computed tomography (CT), should necessarily be considered in terms of noise distribution caused by converting to X-ray photon to digital imaging signal. Recently, the denoising technique based on deep learning architecture is increasingly used in the medical imaging field. Here, we evaluated noise reduction effect according to various noise types based on the U-net deep learning model in the lung CT images. The input data for deep learning was generated by applying Gaussian noise, Poisson noise, salt and pepper noise and speckle noise from the ground truth (GT) image. In particular, two types of Gaussian noise input data were applied with standard deviation values of 30 and 50. There are applied hyper-parameters, which were Adam as optimizer function, 100 as epochs, and 0.0001 as learning rate, respectively. To analyze the quantitative values, the mean square error (MSE), the peak signal to noise ratio (PSNR) and coefficient of variation (COV) were calculated. According to the results, it was confirmed that the U-net model was effective for noise reduction all of the set conditions in this study. Especially, it showed the best performance in Gaussian noise.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.10
• /
• pp.1709-1714
• /
• 2007

This paper proposes a compensating algorithm for the secondary current of the measurement current transformer (CT) that removes the effects of the hysteresis characteristics of the iron-core. The exciting current resulting from the hysteresis characteristics of the core causes an error between the primary current and the secondary current of the measurement CT. The exciting current can be decomposed into the magnetizing current and the core loss current. The core loss current is obtained from the measured secondary current and the core loss resistance. The core flux linkage is calculated by integrating the measured secondary current, and then inserted into the flux-magnetizing current curve to obtain the magnetizing current. The exciting current at every sampling interval is obtained by summing the core-loss and magnetizing currents and then added to the measured current to obtain the correct current. The performance of the proposed algorithm is validated under various conditions using EMTP generated data. The results indicate that the proposed algorithm can improve the accuracy of the measurement CT significantly, and thus reduce the size and the cost of the measurement CT.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2002.09a
• /
• pp.376-378
• /
• 2002

This work was carried out to evaluate the basic performances for 4D CT, which employed continuously rotating conebeam. The performances were evaluated with the same method as the conventional CT, because the standard method of evaluating 4D CT has not yet been established, and we think this result was helpful to establish it. 4D CT can give dynamic volume imaging data continuously and with high-speed. The results were isotropic except for the evaluation of distortion in which small distortions gradually appeared as coming off the center of phantom in longitudinal direction.

• The Korean Journal of Health Service Management
• /
• v.8 no.4
• /
• pp.25-34
• /
• 2014

This paper examined the effects of private health insurance (PHI) on the use of endoscopy, sonography, CT, MRI, and PET among Koreans. The data used was the four waves of the Korea Health Panel (2008-2011), and the number of subjects was 141,862 visits. We employed a pooled logistic regression mode which was corrected with cluster errors. The results showed that PHI was positively related to the use of endoscopy, but not related to the use of sonography, CT, MRI, and PET. It is, however, recommended that health policy makers pay attention to the effects of PHI on the use of sonography, CT, MRI, and PET.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.65-66
• /
• 2008

This paper proposes a compensating algorithm for the secondary current of the measurement current transformer (CT) that removes the effects of the hysteresis characteristics of the iron-core. The exciting current resulting from the hysteresis characteristics of the core causes an error between the primary current and the secondary current of the CT. The proposed algorithm decomposes the exciting current into the magnetizing current and the core loss current and each of them is estimated. The core loss current is calculated from the secondary voltage and the secondary voltage-core loss current curve. The core flux linkage is calculated and then inserted into the flux-current curve to estimate the magnetizing current. The exciting current at every sampling interval is obtained by summing the core-loss and magnetizing currents and then added to the measured current to compensate the secondary current. The performance of the proposed algorithm is validated under various conditions using EMTP generated data. The test results of the real CT were also included. The results indicate that the proposed algorithm can improve the accuracy of the measurement CT significantly, and thus reduce the size and the cost of the CT.

• Biomedical Science Letters
• /
• v.25 no.4
• /
• pp.398-406
• /
• 2019

Cancer immunotherapy using gene-modified tumor cells is safe and customized cancer treatment method. In this study, we made gene-modified tumor cells by transferring costimulatory molecules, 4-1BBL and OX40L, into tumor cells using lentivirus vector, and identified anti-cancer effect of gene-modified tumor cells in CT26 mouse colorectal tumor model. We construct pLVX-puro-4-1BBL, -OX40L vector for lentivirus production and optimized the transfection efficiency and transduction efficiency. The transfection efficiency is maximal at DNA:cationic polymer = 1:0.5 and DNA 2 ㎍ for lentivirus production. Then, the lentiviral including 4-1BBL and OX40L was used to deliver CT26 mouse tumor cells to establish optimal delivery conditions according to the amount of virus. The transduction efficiency is maximal at 500 μL volume of lentiviral stock without change in cell shape or growth rate. CT26-4-1BBL, CT26-OX40L significantly inhibited the tumor growth compare with CT26-WT or CT26-β-gal cell line. These data showed the possibility the use of genetically modified tumor cells with costimulatory molecule as cancer immunotherapy agent.

• Journal of the Korean Society of Radiology
• /
• v.7 no.4
• /
• pp.301-305
• /
• 2013

This thesis intends to analyze tooth distortion by mandibular arch form by reproducing existing panorama image and reconstructed panorama image of Cone Beam CT data with the three-dimensional computer program. The diameter of tooth measured in Cone Beam CT's cross-section image and reconstructed panorama was synchronized without any big change from incisors to posteriors. But, panorama showed serious distortion as going to posteriors after showing a little distortion in incisors. The panorama reconstructed for patients' individual arch showed reduced distortion than panoramas used generally. In addition, panorama showed serious distortion from incisors to posteriors and it means that distortion is reduced in reconstructed panorama.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.252-255
• /
• 2005

Mandibular first premolars obtained from the middle-aged men about the ages of 50 were scanned using a Micro-CT. A Jig was made for a Micro-CT measurement to get reliable data from irregular teeth shape. Data were measured from the scanned 2-D images by way of measurement software. the methodology fer measurement of the mandibular first premolar was presented and according to this, the standardized mandibular first premolars of middle aged Korean males and females were made by using a rapid prototyping system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2359-2376
• /
• 2022

With the exponential growth of medical image big data represented by high-resolution CT images(CTI), the high-resolution CTI data is of great importance for clinical research and diagnosis. The paper takes lung CTI as an example to study. Retrieving answer CTIs similar to the input one from the large-scale lung CTI database can effectively assist physicians to diagnose. Compared with the conventional content-based image retrieval(CBIR) methods, the CBIR for lung CTIs demands higher retrieval accuracy in both the contour shape and the internal details of the organ. In traditional supervised deep learning networks, the learning of the network relies on the labeling of CTIs which is a very time-consuming task. To address this issue, the paper proposes a Weakly Supervised Similarity Evaluation Network (WSSENet) for efficiently support similarity analysis of lung CTIs. We conducted extensive experiments to verify the effectiveness of the WSSENet based on which the CBIR is performed.

• Korean Journal of Radiology
• /
• v.22 no.11
• /
• pp.1894-1908
• /
• 2021

Double outlet right ventricle (DORV) is a relatively common congenital heart disease in which both great arteries are connected completely or predominantly to the morphologic RV. Unlike other congenital heart diseases, DORV demonstrates various anatomic and hemodynamic subtypes, mimicking ventricular septal defect, tetralogy of Fallot, transposition of the great arteries, and functional single ventricle. Because different surgical strategies are applied to different subtypes of DORV with ventricular septal defects, a detailed assessment of intracardiac anatomy should be performed preoperatively. Due to high spatial and contrast resolutions, cardiac CT can provide an accurate characterization of various intracardiac morphologic features of DORV. In this pictorial essay, major anatomic factors affecting surgical decision-making in DORV with ventricular septal defects were comprehensively reviewed using three-dimensional cardiac CT data. In addition, the surgical procedures available for these patients and major postoperative complications are described.