• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.178-185
• /
• 2021

Three-dimensional high-resolution magnetic resonance imaging (MRI) provides fine-level anatomical information for disease diagnosis. However, there is a limitation in obtaining high resolution due to the long scan time for wide spatial coverage. Therefore, in order to obtain a clear high-resolution(HR) image in a wide spatial coverage, a super-resolution technology that converts a low-resolution(LR) MRI image into a high-resolution is required. In this paper, we propose a super-resolution technique through filter learning based on information on the surrounding gradient information in 3D space from 3D MRI images. In the learning step, the gradient features of each voxel are computed through eigen-decomposition from 3D patch. Based on these features, we get the learned filters that minimize the difference of intensity between pairs of LR and HR images for similar features. In test step, the gradient feature of the patch is obtained for each voxel, and the filter is applied by selecting a filter corresponding to the feature closest to it. As a result of learning 100 T1 brain MRI images of HCP which is publicly opened, we showed that the performance improved by up to about 11% compared to the traditional interpolation method.

• Journal of Gastric Cancer
• /
• v.3 no.3
• /
• pp.151-157
• /
• 2003

Purpose: In this study, we attempted to ascertain the proton magnetic resonance spectroscopy (${1}^H$ MRS) peak characteristics of human gastric tissue layers and finally to use the metabolic peaks of MRS to distinguish between normal and abnormal gastric specimens. Materials and Methods: Ex-vivo ${1}^H$ MRS examinations of thirty-five gastric specimens were performed to distinguish abnormal gastric tissues invaded by carcinoma cells from normal stomach-wall tissues. High-resolution 400-MHz (9.4-T) ${1}^H$ nuclear magnetic resonance (NMR) spectra of two gastric layers, a proper muscle layer, and a composite mucosasubmucosa layer were compared with those of clinical 64- MHz (1.5-T) MR spectra. Three-dimensional spoiled gradient recalled (SPGR) images were used to determine the size and the position of a voxel for MRS data collection. Results: For normal gastric tissue layers, the metabolite peaks of 400-MHz ${1}^H$ MRS were primarily found to be as follows: lipids at 0.9 ppm and 1.3 ppm; alanine at 1.58 ppm; N-acetyl neuraminic acid (sialic acid) at 2.03 ppm; and glutathione at 2.25 ppm in common. The broad and featureless featureless spectral peaks of the 64-MHz MRS were bunched near 0.9, 1.3, and 2.0, and 2.2 ppm in human specimens without respect to layers. In a specimen (Borrmmann type III) with a tubular adenocarcinoma, the resonance peaks were measured at 1.26, 1.36 and 3.22 ppm. All the peak intensities of the spectrum of the normal gastric tissue were reduced, but for gastric tumor tissue layers, the lactate peak split into 1.26 and 1.39 ppm, and the peak intensity of choline at 3.21 ppm was increased. Conclusion: We found that decreasing lipids, an increasing lactate peak that split into two peaks, 1.26 ppm and 1.36 ppm, and an increasing choline peak at 3.22 ppm were markers of tumor invasion into the gastric tissue layers. This study implies that MR spectroscopy can be a useful diagnostic tool for gastric cancer.

• Journal of the Korean Society of Radiology
• /
• v.10 no.5
• /
• pp.367-373
• /
• 2016

This study conducted an analysis to compare the differences in the properties of the magnetic field and the generation of artifacts because of the difference in the magnetic field between 1.5 T equipment and 3.0 T equipment, centering around four types of pulse sequences, mainly applied to the abdominal Magnetic Resonance Imaging (MRI). With data on 500 persons transmitted to the PACS, this study analyzed the SNR value, quantitatively and carried out a qualitative evaluation, dividing MSA, CSA, and DA into three steps. As a result of the quantitative evaluation, the SNR value was significantly higher in the 1.5 T equipment; however, there was a factor deteriorating the image quality, too, as artifacts were generated in the images. The 1.5 T equipment generated fewer artifacts than the 3.0 T equipment did, so it could compensate the image quality for 3.0 T. In conclusion, based on these findings, this study could understand the differences in the properties of the magnetic field and the generation of artifacts occurring because of the difference in the magnetic field and could provide a measure for them. This study would be guidelines for MRI users who directly examine the patients in abdominal MRI using the two types of equipment in the clinical setting in the future.

• Investigative Magnetic Resonance Imaging
• /
• v.22 no.1
• /
• pp.1-9
• /
• 2018

Purpose: This study was designed to optimize the flip angle (FA) and scan timing of the hepatobiliary phase (HBP) using the 3D T1-weighted, gradient-echo (GRE) imaging with controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) technique on gadoxetic acid-enhanced 3T liver MR imaging. Materials and Methods: Sixty-two patients who underwent gadoxetic acid-enhanced 3T liver MR imaging were included in this study. Four 3D T1-weighted GRE imaging studies using the CAIPIRINHA technique and FAs of $9^{\circ}$ and $13^{\circ}$ were acquired during HBP at 15 and 20 min after intravenous injection of gadoxetic acid. Two abdominal radiologists, who were blinded to the FA and the timing of image acquisition, assessed the sharpness of liver edge, hepatic vessel clarity, lesion conspicuity, artifact severity, and overall image quality using a five-point scale. Quantitative analysis was performed by another radiologist to estimate the relative liver enhancement (RLE) and the signal-to-noise ratio (SNR). Statistical analyses were performed using the Wilcoxon signed rank test and one-way analysis of variance. Results: The scores of the HBP with an FA of $13^{\circ}$ during the same delayed time were significantly higher than those of the HBP with an FA of $9^{\circ}$ in all the assessment items (P < 0.01). In terms of the delay time, images at the same FA obtained with a 20-min-HBP showed better quality than those obtained with a 15-min-HBP. There was no significant difference in qualitative scores between the 20-min-HBP and the 15-min-HBP images in the non-liver cirrhosis (LC) group except for the hepatic vessel clarity score with $9^{\circ}$ FA. In the quantitative analysis, a statistically significant difference was found in the degree of RLE in the four HBP images (P = 0.012). However, in the subgroup analysis, no significant difference in RLE was found in the four HBP images in either the LC or the non-LC groups. The SNR did not differ significantly in the four HBP images. In the subgroup analysis, 20-min-HBP imaging with a $13^{\circ}$ FA showed the highest SNR value in the LC-group, whereas 15-min-HBP imaging with a $13^{\circ}$ FA showed the best value of SNR in the non-LC group. Conclusion: The use of a moderately high FA improves the image quality and lesion conspicuity on 3D, T1-weighted GRE imaging using the CAIPIRINHA technique on gadoxetic acid, 3T liver MR imaging. In patients with normal liver function, the 15-min-HBP with a $13^{\circ}$ FA represents a feasible option without a significant decrease in image quality.

• Investigative Magnetic Resonance Imaging
• /
• v.6 no.2
• /
• pp.152-157
• /
• 2002

Purpose : The differential diagnosis between Modic type I degenerative spine and infectious spondylitis sometimes is difficult, because the affected bone marrows in both disease show similar signal intensity on conventional MR imaging. We evaluate the usefulness of diffusion-wighted MR imaging for differential diagnosis between Modic type I degenerative spine and infectious spondylitis. Materials and methods : The spin-echo and diffusion-weighted MR images of eight patients with Modic type I degenerative spines and 14 patients with infectious spondylitis diagnosed by clinical findings or CT-guided biopsies we re analyzed. The diffusion-weighted imaging sequence was based on reversed fast imaging with steady-state precession (PSIF). Signal intensity changes of the vertebral bone marrow on conventional spin-echo and diffusion-weighted MR imaging were compared between degenerative spine and infectious spondylitis. Results : On T1-weighte d images, the affeted bone marrow in both disease showed hypointense signals. On T 2-weighted images, all of type I degenerative spine and 11 of infectious spondylitis showed hyperintensity, and three of infectious spondylitis showed heterogeneo us mixed signal intensity. On diffusion-weighted MR images, all of type I degenerative spine were hypointense with peripheral high signal intensity to normal vertebral body, but infectious spondylitis was hyperintense (n = 11) and hypointense (n=3). Conclusion : Diffusion-weighted MR imaging is useful to differentiate Modic type I degenerative spine from infectious spondylitis. On diffusion-weighted images, the high singal intensity of bone marrow suggests infectious spondylitis, whereas the low signal intensity of bone marrow with peripheral focal high signal intensity suggests type I degenerative spine.

• Investigative Magnetic Resonance Imaging
• /
• v.4 no.2
• /
• pp.107-112
• /
• 2000

Purpose : To evaluate the usefulness of comparison of the signal intensity of uterine septum in the differential diagnosis of bicornuate and septate uterus on magnetic resonance (MR) imaging. Materials and methods : Preoperative MR imaging findings of surgically proven 5 bicornuate and 6 septate uteri were retrospectively analyzed. Because preoperative differential diagnosis of both was possible in all cases in terms of the intercornual distance, external contour of uterine fungus, and divergent angle of two uterine cavities, these criteria were excluded in this study. The signal intensity of uterine septum in patients with bicornuate and septate uterus was analyzed on T1-weighted and fast spin echo T2-weighted images obtained in the axial and coronal planes, using a 1.5-T MR scanner. The signal intensity of uterine septum especially on T2-weighted images was compared with that of myometrium or junctional zone. Results : The signal intensity of uterine septum in patients with bicornuate uterus (n=5) and septate uterus (n=6) was similar to that of myometrium in all cases on T1-weighted images. The septum of bicornuate uterus (n=5) on fast spin echo T2-weighted images was isointense with myometrium in three and hypointense in two cases. The uterine septum of septate uterus (n=6) on T2-weighted images was isointense with myometrium in two, hypointense in two, and isointense with or more hypointense than junctional zone in two cases. No patient showed different signal intensity between upper and lower uterine septum. Conclusion : Because the MR signal intensity of the uterine septum in bicornuate or septate uterus is variable, it should not be used alone in the differential diagnosis of them. In these clinically important differentiation, therefore, comprehensive analysis of MR findings in terms of the external contour of uterine fundus, intercornual distance, divergent angle of two uterine cavities, in addition to the signal intensity of the uterine septum, should be considered.

• Investigative Magnetic Resonance Imaging
• /
• v.18 no.3
• /
• pp.200-207
• /
• 2014

Purpose : We evaluated the diagnostic value of susceptibility-weighted imaging (SWI) for the detection of developmental venous anomaly (DVA). Materials and Methods: Retrospective review of 1068 brain MR examinations found 28 DVAs in 28 patients (2.6%) on contrast-enhanced T1-weighted images. SWI, T2, and FLAIR images of 28 patients with DVA and 28 sex- and age-matched control patients without DVA were analyzed by blinded readers on each type of sequences. All images were independently reviewed by two radiologists who were blinded to other MR imaging finding. In cases of discrepancy, two reviewers reached a consensus later. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of each MR sequence for the detection of DVA were determined. Statistical analysis was performed by using the Mcnemar test. The significance level was p < 0.05. Results: The sensitivity, specificity, PPV, and NPV of SWI for the detection of DVA were 85.7%, 92.9%, 92.3%, and 86.7%, respectively. T2 and FLAIR images showed sensitivity of 35.7% and 35.7%, specificity of 92.9% and 96.4%, PPV of 83.3% and 90.9%, and NPV of 59.1% and 60.0%, respectively. On SWI, the sensitivity and NPV for the detection of DVAs were significantly higher than those of T2 and FLAIR images (p < 0.05). Conclusion: SWI was sensitive and specific for the detection of DVA.

• Investigative Magnetic Resonance Imaging
• /
• v.12 no.1
• /
• pp.49-54
• /
• 2008

Purpose : The purpose of this study was to determine the characteristics of hypoxic-ischemic encephalopathy (HIE) on diffusion-weighted imaging (DWI) and the role of DWI for the diagnosis of HIE. Materials and Methods : Six patients with HIE underwent MRI including DWI. MR examinations were performed within 4 - 32 days (mean, 11.8 days) after hypoxic brain insult. We assessed the distribution of the lesions and compared the DWI and T2, FLAIR images for the subjective conspicuity of the lesions. Results : In all patients, symmetrical hyperintense lesions were demonstrated in the bilateral basal ganglia on T2, FLAIR, and DWI. On ADC map image, the lesions were hypointense in four of six patients and isointense in other two patients. Lesion conspicuity on DWI was higher than on T2 and FLAIR images in four of six patients and similar in other two patients. For the involvement of the cortex and subcortical white matter, in five of six patients, bilateral symmetric hyperintense lesions were seen on T2, FLAIR, and DWI. Lesion conspicuity on DWI was higher than on T2 and FLAIR images in three of them and similar in other two patients. On ADC map image, the lesions showed hypointensity in three of five patients and isointensity in other two patients. For the involvement of the deep cerebral white matter, T2, FLAIR, and DWI showed bilateral symmetric hyperintense lesions in four of six patients. Among them, Lesion conspicuity on DWI was higher than on T2 and FLAIR images in only one patient. Conclusion : HIE is characterized by symmetrical hyperintense lesions in the bilateral basal ganglia, cerebral cortex, and white matter on DWI and the lesions are more conspicuously demonstrated on DWI than on T2 and FLAIR images.

• Journal of Korea Multimedia Society
• /
• v.14 no.3
• /
• pp.348-355
• /
• 2011

In this paper, we propose an automatic prostate segmentation method from dynamic magnetic resonance (MR) images. Our method detects contrast-enhanced images among the dynamic MR images using an average intensity analysis. Then, the candidate regions of prostate are detected by the B-spline non-rigid registration and subtraction between the pre-contrast and contrast-enhanced MR images. Finally, the prostate is segmented by performing a dilation operation outward, and sequential shape propagation inward. Our method was validated by ten data sets and the results were compared with the manually segmented results. The average volumetric overlap error was 6.8%, and average absolute volumetric measurement error was 2.5%. Our method could be used for the computer-aided prostate diagnosis, which requires an accurate prostate segmentation.

• Applied Microscopy
• /
• v.42 no.2
• /
• pp.105-109
• /
• 2012

Electron tomography (ET) of biological specimens is performed from a series of images obtained over a range of tilt angles in a transmission electron microscope. When using the high voltage electron microscope (HVEM), various noises appear in EM images acquired from thick sections by high voltage electron beam. In order to obtain an adequate result in electron tomograms that allow visualization of rather complex and mega-cellular structure such as brain tissue, it is necessary to remove the noise in each original tilt images of thick section. Using band-pass filtering of original tilt images, the filtered images are obtained and used to assemble a reconstructed tomogram. The qualified 3D tomogram from filtered images results in a considerable reduction of the noises compared to conventional tomogram. In conclusion, this study suggests that band-pass filtering is effective to improve the brightness and intensity of HVEM produced tomograms acquired from micron-thick sections of biological specimens.