Japanese Journal of Radiological Technology
Online ISSN : 1881-4883
Print ISSN : 0369-4305
ISSN-L : 0369-4305
Volume 57, Issue 6
Displaying 1-41 of 41 articles from this issue
  • Article type: Cover
    2001 Volume 57 Issue 6 Pages Cover22-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (569K)
  • Article type: Index
    2001 Volume 57 Issue 6 Pages Toc17-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (232K)
  • Article type: Index
    2001 Volume 57 Issue 6 Pages Toc18-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (208K)
  • [in Japanese]
    Article type: Article
    2001 Volume 57 Issue 6 Pages I-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (500K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages II-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (74K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages III-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (95K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages IV-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (37K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages V-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (127K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages VI-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (117K)
  • [in Japanese], [in Japanese], [in Japanese], [in Japanese]
    Article type: Article
    2001 Volume 57 Issue 6 Pages 603-604
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (1389K)
  • MASAHIRO HASHIDA, SYUICHI YAMAUCHI, JING-LONG WU, HIROAKI MIZUHARA, KI ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 605-610
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Using functional magnetic resonance imaging (fMRI), we investigated the activated areas of the human brain related with calculation processing as an advanced function of the human brain. Furthermore, we investigated differences in activation between visual and auditory calculation processing. The eight subjects (all healthy men)were examined on a clinical MR unit (1.5 tesla)with a gradient echo-type EPI sequence. SPM99 software was used for data processing. Arithmetic problems were used for the visual stimulus (visual image)as well as for the auditory stimulus (audible voice). The stimuli were presented to the subjects as follows: no stimulation, presentation of random figures, and presentation of arithmetic problems. Activated areas of the human brain related with calculation processing were the inferior parietal lobule, middle frontal gyrus, and inferior frontal gyrus. Comparing the arithmetic problems with the presentation of random figures, we found that the activated areas of the human brain were not differently affected by visual and auditory systems. The areas activated in the visual and auditory experiments were observed at nearly the same place in the brain. It is possible to study advanced functions of the human brain such as calculation processing in a general clinical hospital when adequate tasks and methods of presentation are used.
    Download PDF (1708K)
  • HISAHARU TAKASHIMA, MITSUHIRO EJIMA, MAMORU YAKEYAMA, MASAMI YAMAGUCHI ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 611-615
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    This hospital has performed fMRI of language-competent areas of the brain to identify the language-dominant hemisphere and obtain the configuration of the focus in the language-dominant side of the brain. Until now, signals have been detected in only two of fifteen patients who were diagnosed by language tests of a lastsyllable word chain. In the present experiment, we tried to have subjects select the type of test. The result was that changes in signals were detected in eight of ten patients. Although the set-up of tests for fMRI is said to hold significant value, clear-cut studies to back this up have rarely been seen. Because clinical medicine treats patients who have difficulty in communication or suffer from aphasia, it is important to take into consideration individual variations and to set up a test suitable for, or achievable by, these individuals. The present method enabled us to avoid failure in examination caused by unsuccessful tests.
    Download PDF (1520K)
  • YASUYUKI TAKAHASHI, AKIRA MASUHARA, HISASHI SHINOHARA, ISAO KAWAMATA, ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 616-620
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    We have developed a new software program that makes it possible to conduct the Diamox one-day method (modified early method)using the microsphere model both quickly and easily, for the measurement of cerebral blood flow using ^<123>I-IMP. This software program obtains measurements before and after acetazolamide loading (hereafter referred to as Diamox), and allows the use of five imaging methods consisting of combinations of static, dynamic, and continuous and repeated SPECT, to meet the requirements of the singledetector-type camera as well as those of the multi-detector-type gamma camera. In addition, four methods of analysis are available, including continuous arterial blood collection, one-point arterial blood collection (Inoue, Fujioka, et al.), and one-point venous blood collection (Fujioka, et al). Furthermore, the program offers a variety of information such as at-rest SPECT view, response mapping, increase mapping, and subtraction mapping. The correlation between the quantitative value (n=30)obtained by continuous arterial blood collection and each of the above parameters was good: y=0.61X+6.32, r=0.82; y=1.24X+8.29, r=0.95; y=1.13x+5.17, r=0.90 (p<0.001).
    Download PDF (962K)
  • MEGUMI SAWAMOTO, JUNKO TAKABA, TOSHIO KUSHIMA, HIROYUKI KABASAWA
    Article type: Article
    2001 Volume 57 Issue 6 Pages 621-627
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    It is important to understand the motion probing gradient direction and nerve fiber tract structure to assess diffusion-weighted imaging, which is affected by the position of subjects. However, diffusion tensor imaging may be able to reproduce exact anisotropy regardless of the position of the subject. Therefore, we examined the effect of the position of subjects on diffusion tensor imaging. The value of fractional anisotropy and eigenvalue 1 in standard position were equal to the values of other documented records. Change in head position caused no significant difference in eigenvector imaging, the value of fractional anisotropy, or eigenvalue 1.
    Download PDF (1857K)
  • JUNICHIROU KODAMA, KOJI SUZUKI, YUJI KAGA, HIROYUKI KABASAWA
    Article type: Article
    2001 Volume 57 Issue 6 Pages 628-633
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Conventional magnetic resonance axonography based on anisotropic diffusion-weighted imaging often suffers from motion artifacts.We developed a new diffusion-weighted imaging method by radial scan with back projiction which is almost free from motion artifacts. Like x-ray CT scanning,rotational data acquisition in radial-scan diffusion proceeds according to a spin-echo method that varies the gradient magnetic field along the vertical and horizontal axes. Data sampled in the radial direction for k-space are analyzed through the back-projection method. A view number of 192 and b-factor ranging from 700 to 800 are the most suitable parameters for radial-scan diffusion-weighted imaging to provide fine imaging with fat suppression. This method is almost free from phase artifact and motion artifact. Therefore, it eliminates the need for cardiac gating and is valid for patients with involuntary movements. It causes no deformation and is suitable for imaging the posterior cranial fossa and for coronal imaging. Its disadvantages are the long scanning time resuiting from the spin-echo method and radial artifact due to the back-projiction method.
    Download PDF (2281K)
  • HITOSHI KUBO, MASAYUKI MAEDA, AKINOBU ARAKI
    Article type: Article
    2001 Volume 57 Issue 6 Pages 634-638
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    We evaluated the accuracy of calculating apparent diffusion coefficients (ADCs)using high-B-value diffusion images. Echo planar diffusion-weighted MR images were obtained at 1.5 tesla in five standard locations in six subjects using gradient strengths corresponding to B values from 0 to 3000 s/mm^2. Estimation of ADCs was made using two methods: a nonlinear regression model using measurements from a full set of B values (multi-point method) and linear estimation using B values of 0 and max only (two-point method). A high correlation between the two methods was noted (r=0.99), and the mean percentage differences were-0.53 % and 0.53 % in phantom and human brain, respectively. These results suggest there is little error in estimating ADCs calculated by the two-point technique using high-B-value diffusion MR images.
    Download PDF (524K)
  • MASATO UCHIKOSHI, TAKASHI UEDA, YASUSI KAJI, AKIO TSUJI, AKIHIKO WADA, ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 639-644
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Studies have shown that diffusion MR imaging is a reliable method for the diagnosis of central nervous system diseases, especially acute cerebral infarction. Although echo planar imaging (EPI) is a promising tool for that purpose, it is vulnerable to susceptibility artifacts that are responsible for image distortion or signal loss. Our purpose in this study was to evaluate the usefulness of diffusion MR imaging with PSIF (reversed fast imaging SSFP) and split acquisition of fast-spin-echo signals for diffusion imaging (SPLICE) in the central nervous system (CNS). First, PSIF and SPLICE were applied to the phantoms. Each phantom, including acetone, acetic acid, and water, was analyzed for ADC based on SPLICE and for diffusion related coefficient (DRC) based on PSIF. The ADCs based on SPLICE were 4.36±0.89 * 10^<-3> mm^2/sec, 1.25±0.04 * 10^<-3> mm^2/sec, and 2.35±0.04 * 10^<-3> mm^2/sec, and the DRCs based on PSIF were 0.353±0.25, 0.178±0.07, and 0.273±0.018 for acetone, acetic acid, and water, respectively. These calculated ADCs based on SPLICE were well correlated with known diffusion coefficients, showing a correlation coefficient of 0.995. Second, PSIF and SPLICE were applied to the CNS. The advantage of PSIF and SPLICE was that susceptibility artifacts were reduced in the images of spinal cord and brain stem. PSIF was especially useful for diffusion MR imaging in the spinal cord. The disadvantage of SPLICE was the decreased SN ratio. We conclude that PSIF or SPLICE may be helpful when EPI diffusion MR imaging is insufficient.
    Download PDF (1504K)
  • JUN OKAMOTO
    Article type: Article
    2001 Volume 57 Issue 6 Pages 645-650
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (2043K)
  • YOSHIHIRO KANAI, MIYUKI NAKAMURA, NORIKO KIKUMA, GO YASUI, KIYOKAZU TA ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 651-656
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    The chemical shift of water in proton-MRS is dependent on the sample temperature. Thus the target temperature can be calculated by measuring changes in the frequency difference between water and metabolites. There have been some clinical reports on measuring brain temperature using this method. However, various questions arise as to whether this method can be applied in a severely damaged brain(head injury, cerebrovascular disease)where in-vivo pH has shifted far from normal. We here report the results of our basic experiments using phantoms to clarify whether pH as well as temperature of the target may change the frequency difference between water and metabolites. Our data showed that the frequency difference between water and metabolites (choline, creatine, NAA)changed by 0.01 ppm as phantom temperature declined 1℃ (32-40℃),but was not affected by changes in phantom pH (3.2-9.6). Since the chemical shift of metabolites was shown to be independent of target pH, it is feasible to measure in-vivo brain temperature less invasively using proton-MRS even in patients with severe brain damage.
    Download PDF (1316K)
  • SHIN-ICHIRO KUMITA, [in Japanese]
    Article type: Article
    2001 Volume 57 Issue 6 Pages 657-663
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (2050K)
  • TSUYOSHI MATSUDA, [in Japanese], [in Japanese], [in Japanese], [in Jap ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 664-670
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (2253K)
  • TAMIO FURUYAMA, NAOYA KOGAMO, YOSHIHIRO MIKAMI, TAKAYUKI MUKAI, HIDEO ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 671-678
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    The k-space segmented fat-saturation fast low angle shot (3D-FLASH) technique was used to obtain MR images of coronary arteries using navigator-echo-based respiratory gating. In recent years, it has been possible to obtain these images with a single breath-hold and contrast enhancement by rapid intravenous injection. Now IR-Prep 3D-FLASH has been developed and is available. We have obtained MR images of model arteries that had stenosis using 3D-FLASH and IR-Prep 3D-FLASH, and evaluated them physically and visually. Our research through this basic experiment has led to improvement. We have also evaluated the signal-to- noise ratio (SNR) and contrast-to-noise ratio (CNR) using the images of patients we have examined. The results have been better than those obtained by the usual sequences. Volume-rendering images of coronary arteries were very similar to CAG images. These images depicted peripheral vessels continuously and stenosis was detected clearly. In particular, images with high contrast between coronary artery and myocardium were obtained by IR-Prep 3D-FLASH.
    Download PDF (2332K)
  • KYOICHI KATO, TOSHIYUKI TAKAHASHI, MIKA ITO, MASARU SONODA, HISAYA SAT ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 679-684
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    In this study we selected two-dimensional and volume-rendering three-dimensional MR images to evaluate the patency of CABG in comparison with angiography. Twenty-six patients were included in the study (19 men and 7 women, aged 49-85 years, mean age 70.4 years). The total number of bypassed coronary arteries was 62. The MR apparatus used was a Siemens MAGNETOM VISION(1.5 T); the pulse sequence employed was FISP 3D with TR 5 msec, TE 2 msec, FA 15 degrees, and fat saturation. Slice thickness was 4.27 mm, slab thickness 111 mm, and matrix 128×256. A body array coil was used, and coronal images were obtained during a 20-second breath-hold and ECG gating. Scanning was initiated 13 seconds after the injection of contrast medium (Gd-DTPA 10 ml + 20 ml saline, 3 ml/sec.). Data sets were post-processed to obtain MIP images and volume-rendering 3D images. The MR images were equal to those of angiography in the evaluation of patency of the CABG. Volume rendering had more liberal parameter settings and could be used to observe the vessels under ideal conditions. 3D images enabled observation from multiple angles and were useful in evaluating overlapping vessels.
    Download PDF (1984K)
  • TADAMITSU IDEGUCHI, MASAAKI YONEMURA, TORU MURANAKA, SHIGEHIKO MATSUSH ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 685-690
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    The purpose of this study was to determine the volume of contrast medium sufficient for three-dimensional (3D) time-of-flight (TOF) MR-venography (MRV) of the intracranial venous systemS. 3D-TOF MRV was performed with contrast medium(Gd-DTPA: 0 ml, 5 ml, 15 ml)on 20 subjects. With the 3D-TOF sequence, post-processing maximum intensity projection (MIP)images were constructed throughout the entire imaging volume. Visual evaluation of the five stages by five radiologists with at least three years of experience was performed in respect to the independence of contrast medium volume from the suppression effect of arteries and visualization of the deep cerebral veins. The deep cerebral veins and transverse sinus were not clearly visualized on 3D-TOF MRV without contrast medium. "Excellent" and "good" evaluations accounted for 72% in terms of the suppression effect of the artery in the group administered 5 ml. This rate was significantly higher than the 33% in the group administered 15 ml. "Excellent" and "good" evaluations accounted for 68 % in terms of visualization of the deep cerebral veins in the 5-ml group, a significantly a higher rate than the 44% in the 15-ml group. In the examination of contrast medium volume independence in an identical case, the artery suppression effect occurred at a rate of 100% in the comparison of 5 ml administration with 15 ml administration. In conclusion, when carried out with a small amount of contrast medium (about 5 ml), 3D-TOF MRV was useful for demonstrating the intracranial venous system.
    Download PDF (1738K)
  • TATSUHIKO OKADA, MASAYUKI YAMADA, MASATO NAKANE, YOSHIFUMI KUWAYAMA, H ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 691-698
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    The purpose of this study was to evaluate k-space order in three-dimensional magnetic resonance angiography (3D-CE MRA) with intravenous bolus injection of contrast medium. It is important to use effectual k-space ordering to represent only the arterial phase. In addition, it is necessary to set an appropriate acquisition time after bolus injection. In this study, we compared elliptical centric encoding and conventional centric encoding in a flow phantom that simulated the cervical vasculature. We also carried out a clinical study. We found no difference between the two encoding methods in regard to the time to begin representation of the simulated blood vessels in the flow phantom, and obtained an appropriate starting point of scanning before venous return in an enhancement curve. In this case, however, central k-space data were acquired during the changing gadolinium concentration of the arterial phase. For this reason, artifacts occurred in images acquired by conventional centric encoding. In contrast, there were no artifacts in elliptical centric encoding, and the difference between the two encoding methods was significant in this respect. Accordingly, we concluded that elliptical centric encoding was useful for 3D-CE MRA of the cervical vasculature, which has a rapid venous return.
    Download PDF (1491K)
  • MASARU TAKAHASHI, MASANORI YAMADA, TOMOKAZU OKU, SHIGEO MIYAZAKI, AKIR ...
    Article type: Article
    2001 Volume 57 Issue 6 Pages 699-705
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    We attempted to perform renal three-dimensional (3D)magnetic resonance angiography (MRA) using the navigator echo sequence for abdominal MRA and a noninvasive method. We evaluated the best parameters for the pulse sequence, and examined delineation of the maximum intensity projection (MIP) and multi-planar reconstruction (MPR) images. We carried out noninvasive renal MRA without breath-hold using the navigator echo sequence. The best repetition time of slice encoding group to obtain images of good contrast was 700 msec. MIP images were adequate for observation of the renal arteries, and the contrast of MPR images (coronal) was very high. This method is effective if the patient cannot hold the breath or has renal functional disorder.
    Download PDF (1658K)
  • AKIHIRO NOGI, TSUKASA DOI, ASUKA KASAI, YOSHIMASA MIZUNO
    Article type: Article
    2001 Volume 57 Issue 6 Pages 706-712
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Enhanced MRA using the three-dimensional Fourier transform (3DFT) sequence for the body or its extremities is commonly adopted. Through advanced technology in the areas of turbo-MRA and ZIP(zerofilled interpolation) high-quality imaging as well as progress in spatial resolution, the anatomy of blood vessels can be depicted with the same high quality as that provided by DSA (digital subtraction angiography). In spite of these technological advances, it is still difficult to obtain clear delineation of the arteriovenous phase and hemodynamics because of the 7-30 second acquisition time. In this study, we examined the possibility of shortening the time to provide effective temporal resolution in the clinical setting. MR-DSA is useful for clinical application because high-quality images with good spatial resolution can be obtained with the 2DFT sequence.
    Download PDF (2390K)
  • TSUKASA DOI, AKIHIRO NOGI, YOSHIMASA MIZUNO
    Article type: Article
    2001 Volume 57 Issue 6 Pages 713-720
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    We examined the appearance of turbo NRA using zero-filling in k-nspace, and verified that overlapping scanning created a decline in signal intensity by the partial volume effect. In clinical application, in a case of multiangle visualization after MIP of the CE-3DMRA method, we obtained the effective slice taickness for maintaining the time resolution and spatial resolution. When using an angle within 45 degrees, the effective slice thickness has to be set at less than the diameter of the signal material(objective aortic diameter). And when using a 90-degree angle (lateral direction), the effective slice has to be set at less than 80% of the signal material diameter.
    Download PDF (1698K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 721-723
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (322K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 724-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (96K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 724-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (96K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 725-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (85K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 726-729
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (509K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 730-736
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (968K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 737-738
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (255K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 739-740
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (255K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 741-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (65K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 742-743
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (174K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 744-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (160K)
  • Article type: Appendix
    2001 Volume 57 Issue 6 Pages 745-746
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (247K)
  • Article type: Cover
    2001 Volume 57 Issue 6 Pages Cover23-
    Published: June 20, 2001
    Released on J-STAGE: June 30, 2017
    JOURNAL FREE ACCESS
    Download PDF (1293K)
feedback
Top