Our aim in this study was to evaluate the effect of geometry for measuring section thickness in tomosynthesis by using a metal bead device (bead method). Tomosynthesis images were obtained from two types of tomosynthesis equipment, Safire17 (ST, Shimadzu, Kyoto, Japan) and XR650 (GT, GE Healthcare, Milwaukee, WI). After tomosynthesis radiography with each device, the bead tomosynthesis images were obtained by image reconstruction. The digital profile was obtained from the digital value of the bead central coordinate in the perpendicular direction, and we acquired the slice sensitivity profile (SSP). The section thickness was defined with the full width at half maximum obtained from the SSP. We investigated the change in section thickness under different evaluation conditions: the angular range, the height of the bead position, the source-image receptor distance (SID), and image processing. The section thickness decreased when the angular range and height of the bead position increased. Also, the section thickness varied with a change in the SID. The section thickness differed according to the geometry for measuring the section thickness. Thus, the effect of the geometry used for measurement should be considered when the section thickness in tomosynthesis is measured by the bead method.
The half-value layer (HVL) is an important index of the image quality or radiation risk in mammography. Radiation risk of the breast tissue is evaluated with the average glandular dose. The HVL index is indispensable for the average glandular dose computations. We investigated the influence of multiple factors that affect HVL value, such as thickness or purity of the aluminum attenuator, detector material of dosimeter, fluctuation of X-ray output, detector location in X-ray field and so on, for accurate average glandular dose computations. We found some aluminum plates about 20% thicker than nominal thickness. The HVL values between seven filter sets were different in about 5% at the maximum. In addition, we reduced a fluctuation of X-ray output with dose monitoring. Then, the standard deviation of HVL value decreased from 1.114% to 0.105%. HVL value obtained from a solid-state detector was statistically thicker than that measured by ionization chamber. It has been reported that there was a difference in the half-value layer under the influence of a heel effect by location of the measurement. Accompanied with alternation of detector location, HVL value of PCM (Konica Minolta) had a significant difference, while Novation (Siemens) and Senographe 2000D (GE) had no change.
Liquid-crystal displays (LCDs) used for medical imaging, such as luminance characteristics and panel surface processing, have been developed as a medium substituting for film. There are various models of and specifications for display, but those compatible with high-resolution radiographic diagnostic images have been required with the recent progression of high-resolution modalities. Displays are necessary for faithful presentation of radiographic diagnostic images acquired by X-ray systems. In independent sub-pixel driving (ISD) technology, aiming at high-resolution display, three sub-pixels contained in one pixel of the LCD independently display images, which increases the threefold resolutions in direction of the sub-pixels, facilitating faithful image display with less curtailed pixels. This is a new display technology which may improve the diagnostic performance with regard to reading of medical images. We evaluated the characteristics of ISD technology and performed a visual evaluation of phantom images to investigate its usefulness. After confirming the physical properties of LCDs, we performed a visual evaluation of CDMAM phantom images employing the calculated image quality figure (IQF). The detectability of 15 mega-sub-pixel (15 MsP) significantly improved despite the specification being 5 mega-pixel (5 MP), and that of 9 MsP was higher than that of 5 MP despite the specification being 3 MP. The usefulness of ISD for 6 MsP was also confirmed. Therefore, ISD technology was useful for all LCDs. ISD technology markedly advanced the LCD display performance for medical use.
The measurement methods of contrast to noise ratio (CNR) and signal difference to noise ratio (SDNR) in digital mammography are different among several quality assurance (QA) guidelines, that is, the type of pixel value (PV), phantom shape, location of aluminum plate, and the size of region of interest (ROI) principally differ in data acquisition. We compared CNR (SDNR) obtained from three QA guidelines. They are the European Reference Organisation for Quality Assured Breast Screening and Diagnostic Services (EUREF), the International Electrotechnical Commission (IEC), and the International Atomic Energy Agency (IAEA). In EUREF and IEC, CNR was calculated using linearized pixel value (LPV). In IAEA, because the type of pixel value to use in SDNR was not specified, SDNR was calculated using PV and LPV, and CNR was calculated using LPV. Target/filter combinations are molybdenum/molybdenum (Mo/Mo) and molybdenum/rhodium (Mo/Rh). Applied various tube voltages are 25, 30, and 35 kV, and various phantom thicknesses are 20, 45, and 70 mm of polymethyl methacrylate (PMMA). The PV-SDNR of IAEA showed the largest value among the three methods, following LPV-CNR of IEC, LPV-CNR of EUREF at 20 mm PMMA thickness. In IAEA, SDNR changed by the kind of pixel value (PV or LPV). When CNR is calculated, every researcher should describe the type of guidelines, the kind of pixel value, and formula for calculation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.