Medical Imaging and Information Sciences
Online ISSN : 1880-4977
Print ISSN : 0910-1543
ISSN-L : 0910-1543
Volume 4, Issue 2
Displaying 1-4 of 4 articles from this issue
  • Hiroshi FUJITA, Kunio DOI
    1987 Volume 4 Issue 2 Pages 31-45
    Published: 1987
    Released on J-STAGE: August 27, 2012
    JOURNAL FREE ACCESS
    How to measure the "characteristic curve" of image intensifier (I. I)/TV digital imaging systems is illustrated based upon our recent research. The curve represents the relationship between the input in terms of relative x-ray intensity and the output in terms of the pixel value. The method used to determine the characteristic curve involves exposing an aluminum stepwedge with a beam that was narrowly collimated in order to reduce the contribution of veiling glare. The relative x-ray intensities transmitted through each step of the stepwedge were determined by using a 100-mm camera, which was attached to the I. I. output, as well as screen/film systems selected. The gradient curve of the system was was derived from the slope of the characteristic curve. Results obtained with a Siemens Digitron 2DSA system showed that its characteristic and gradient curves depended upon the matrix size used only at an initial stage, and did not change with the I. I. field size, iris size, and x-ray beam quality (different kVs). It was found, for accurate measurement of the curve, that it was necessary to collimate the x-ray beams, to narrow width, approximately 0.6cm×10cm, so that the effect of veiling glare from the thinner to the thicker parts of the stepwedge was eliminated. The validity of the characteristic curve was demonstrated by measurement of iodine attenuation curves obtained with the I. I/TV digital system at different exposure levels. The application of the characteristic curve (or the gradient) to image analysis was discussed. The characteristic curve is useful for quantitative analysi (linearization of the system response) and also for monitoring of the system performance for quality assurance.
    Download PDF (2716K)
  • Masa MATSUMOTO, Hitoshi KANAMORI
    1987 Volume 4 Issue 2 Pages 46-58
    Published: 1987
    Released on J-STAGE: August 27, 2012
    JOURNAL FREE ACCESS
    In x-ray diagnosis, the photon fluence rates on the recording system vary in a range of six to seven order of magnitudes, even if the tube voltage is fixed at 70 kV. The reasons are mainly due to the wide range of tube currents from less than 1 mA (fluoroscopy) to 1 A (radiography) and the wide range of attenuation rates in objects. These reasons result in the difficulty of measuring diagnostic x-ray spectra using semiconductor detector systems. We have solved the difficulties by the two following methods: (1) Low-capacity high-voltage cables were developed, in order to reduce photon fluence rates by decreasing tube currents and keeping radiographic tube-voltage waveforms, (2) Collimators having various diameters from 0.13 to 7mm (the effective diameter of a Ge detector) were prepared.
    We carried out the measurement of photon spectra for various tube-voltage ripple-rates (which correspond to various type of generators and tube currents) at 70 kV using objects of various thicknesses made of aluminium and acrylic resin. Photon spectra were transformed to energy spectra and exposure spectra, and then exposure attenuation curves were obtained. The effects of tube-voltage ripple-rates on the spectra and the attenuation curves were quantitatively clarified.
    Download PDF (2267K)
  • Masao MATSUMOTO, Hitoshi KANAMORI
    1987 Volume 4 Issue 2 Pages 59-65
    Published: 1987
    Released on J-STAGE: August 27, 2012
    JOURNAL FREE ACCESS
  • Shinichiro KUWABARA, Takashi SUZUKI, Takahiro SAITO, Hiroshi HARASHIMA ...
    1987 Volume 4 Issue 2 Pages 66-73
    Published: 1987
    Released on J-STAGE: August 27, 2012
    JOURNAL FREE ACCESS
    Several type of CT, such as X-RAY CT, MRI, and positron CT, are used for diagnosis of brain diseases. These CTs are complements each of the other, and complementary data analysis of various CTs will make a large contribution to the establishment of a new diagnostic system for brain diseases. As the first step of this, we previously proposed an image processing algorithm for automatic brainstructureiden tification by using a series of MRI images. However, this algorithm cannot be applied to a quite metamorphosed brain. Thus this paper develops an interactive structure-analysis-system for identification and volume estimation of brain anatomies by incorporating interactive interfaces.
    Download PDF (2408K)
feedback
Top