• Journal of the Korean Society of Radiology
• /
• v.83 no.1
• /
• pp.162-167
• /
• 2022

Renal myxomas are very rare benign tumors. To date, a few cases have been reported in English literature, mostly in pathology and urology journals. Thus, there are few reports on the radiological findings associated with renal myxomas. We report on the imaging findings in a case of renal myxoma in a 62-year-old male. MRI demonstrated a well-defined mass in the left renal sinus, with intermediate high signal intensity on T2-weighted images and low signal intensity on T1-weighted images. The tumor showed gradual enhancement on contrast-enhanced T1-weighted images.

• The Korean Journal of Nuclear Medicine
• /
• v.38 no.3
• /
• pp.218-224
• /
• 2004

We compared rest perfusion PET with redistribution perfusion SPECT to investigate the concordant rate between PET and SPECT images and analyze the discordant pattern. Materials and Methods: Rest N-13 ammonia and F-18 FDG PET were performed on 18 patients with old myocardial infarction and left ventricular dysfunction whose dipyridamole - 4hr redistribution TI-201 SPECT showed one or more severe fixed defects. Regional perfusion and metabolism were evaluated visually and quantitatively with 5-segment myocardial model. Results: There were high concordant rate in uptake pattern (80/90 segments, 88.9%) and high correlation coefficient on quantitative analysis (R=0.81, p<0.001) between redistribution TI-201 SPECT and N-13 ammonia PET images. Nine of 18 patients had SPECT-PET concordant pattern (Group I). Ten segments (9 in inferior wall, 1 in apex) from the remaining 9 patients showed SPECT-PET discordant pattern with abnormal TI-201 defect and near normal N-13 ammonia uptake (Group II). The diastolic and systolic left ventricular dimensions were significantly increased in Group II compared to those of Group I. When attenuation uncorrected N-13 ammonia PET images were reconstructed in Group II, it resulted in PET images with severe inferior wall defects nearly identical to those seen in redistribution TI-201 SPECT images. Conclusion: Redistribution TI-201 SPECT images showed high concordant rate and correlation with rest N-13 ammonia PET images. Most of discordant segments had fixed thallium defects in inferior wall with nearly normal N-13 ammonia uptake, which may result from severe left ventricular dilatation and attenuation by the left hemidiaphragm and cardiac blood pool.

• Journal of Korean Neurosurgical Society
• /
• v.63 no.3
• /
• pp.386-396
• /
• 2020

Objective : To generate synthetic spine magnetic resonance (MR) images from spine computed tomography (CT) using generative adversarial networks (GANs), as well as to determine the similarities between synthesized and real MR images. Methods : GANs were trained to transform spine CT image slices into spine magnetic resonance T2 weighted (MRT2) axial image slices by combining adversarial loss and voxel-wise loss. Experiments were performed using 280 pairs of lumbar spine CT scans and MRT2 images. The MRT2 images were then synthesized from 15 other spine CT scans. To evaluate whether the synthetic MR images were realistic, two radiologists, two spine surgeons, and two residents blindly classified the real and synthetic MRT2 images. Two experienced radiologists then evaluated the similarities between subdivisions of the real and synthetic MRT2 images. Quantitative analysis of the synthetic MRT2 images was performed using the mean absolute error (MAE) and peak signal-to-noise ratio (PSNR). Results : The mean overall similarity of the synthetic MRT2 images evaluated by radiologists was 80.2%. In the blind classification of the real MRT2 images, the failure rate ranged from 0% to 40%. The MAE value of each image ranged from 13.75 to 34.24 pixels (mean, 21.19 pixels), and the PSNR of each image ranged from 61.96 to 68.16 dB (mean, 64.92 dB). Conclusion : This was the first study to apply GANs to synthesize spine MR images from CT images. Despite the small dataset of 280 pairs, the synthetic MR images were relatively well implemented. Synthesis of medical images using GANs is a new paradigm of artificial intelligence application in medical imaging. We expect that synthesis of MR images from spine CT images using GANs will improve the diagnostic usefulness of CT. To better inform the clinical applications of this technique, further studies are needed involving a large dataset, a variety of pathologies, and other MR sequence of the lumbar spine.

• Journal of KIISE:Software and Applications
• /
• v.27 no.6
• /
• pp.651-659
• /
• 2000

It is very important to extract the heart region in the medical images. In this paper, we present the automatic heart region extraction in the EBT (electron beam tomography) images. We use contrast thresholding, anatomic knowledge, and mathematical morphology to extract the heart region. Using these results, we applied the active contour models (snakes) to search the exact region. We analyzed the experimental results by comparing the results with the results made by medical experts.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.268-272
• /
• 2003

Large medical images in PACS are compressed for saving storage space and improving network speed. The integrated compression environment was designed and developed for uniting of various compression methods. Various compression algorithm-RLE compression, lossless JEPG, JPEG, was built into it, complying with DICOM. A image compression using DWT was also implemented in it. And a unified algorithm of lossless compression and lossy compression was designed to improve images quality and to make compression ratios high. And integrated compression environment was operating together with a database program for efficient and user-friendly management.

• Proceedings of the KOSOMBE Conference
• /
• v.1995 no.11
• /
• pp.197-200
• /
• 1995

In this paper, we proposed the multi-step multi-layer artificial neural network(MMANN) to classify the chromosome, Which is used as a chromosome pattern classifier after learning. We extracted three chromosome morphological feature parameters such as centromeric index, relative length ratio, and relative area ratio by means of preprocessing method from ten chromosome images. The feature parameters of five chromosome images were used to learn neural network and the rest of them were used to classify the chromosome images. The experiment results show that the chromosome classification error is reduced much more, comparing with less feature parameters than that of the other researchers.

• Smart Media Journal
• /
• v.9 no.3
• /
• pp.59-70
• /
• 2020

This paper presents a fully automatic tool to recognize the liver region from CT images based on a deep learning model, namely Multiple Filter U-net, MFUnet. The advantages of both U-net and Multiple Filters were utilized to construct an autoencoder model, called MFUnet for segmenting the liver region from computed tomograph. The MFUnet architecture includes the autoencoding model which is used for regenerating the liver region, the backbone model for extracting features which is trained on ImageNet, and the predicting model used for liver segmentation. The LiTS dataset and Chaos dataset were used for the evaluation of our research. This result shows that the integration of Multiple Filter to U-net improves the performance of liver segmentation and it opens up many research directions in medical imaging processing field.

• Journal of Industrial Technology
• /
• v.25 no.A
• /
• pp.87-93
• /
• 2005

Three-dimensional ultrasonic probes being applied to the medical imaging can be grouped into three depending on the scanning methods, which are a mechanical type system, a free-hand system, and 2D phased arrays system. A mechanical type scanner uses a mechanically driven transducer to acquire series of 2D plane images. By integrating these images, a 3-D medical image can be constructed. A motor driving mechanism is a conventional choice for mechanically driving a transducer assembly which picks the raw ultrasonic images up. In this paper we attempt to design a 3D ultrasonic probe which has a operating mechanism of s tilting 3-D scanning. The motion of a transducer assembly of the ultrasonic probe is analytically modelled. We propose a selection procedure for the diameter of a wire rope driving the transducer assembly and the size of torsional spring which gives an initial tension to wire ropes.

• Proceedings of the KSMRM Conference
• /
• 2002.11a
• /
• pp.114-114
• /
• 2002

Purpose： To know the fate of Gadophrin-2-enhanced areas in hearts with acute reperfused myocardial infarction. Method： The left anterior descending branches of coronary arteries were occluded for 90 min and reperfused for 60 min in 15 rabbits. Then, Gadophrin-2 (0.05 mmol/kg) was injected via ear veins. Short-axial T1-weighted spin echo images and fast cine images were obtained 24 hours after injection of Gadophrin-2. After four weeks, short-axial cine MRI was performed and T1-weighted spin echo Images were obtained 5 min and 15 min after injection of Gd-DTPA (0.2 mmol/kg). The animals were sacrificed and short-axial slices of the specimen were stained with 1.5% triphenyltetrazolium chloride (TTC) solution.

• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.05
• /
• pp.103-107
• /
• 1996

In this paper, We propose an algorithm for reconstitution of chromosome images to extract its morphological feature parameters. It is reconstituted from 460 chromosome images using the 32 direction line algorithm. We extract three morphological feature parameters such as centromeric index, relative length ratio, and relative area ratio. The experiment results show that our method is batter than that of other researchers comparing with the error of feature parameters.