X-ray or gamma ray systems or devices – Accessory – Testing or calibration
Reexamination Certificate
2000-03-14
2002-06-25
Dunn, Drew (Department: 2882)
X-ray or gamma ray systems or devices
Accessory
Testing or calibration
C378S204000
Reexamination Certificate
active
06409383
ABSTRACT:
FIELD OF THE INVENTION
This invention relates in general to an automated method for the quality assurance (QA) of x-ray digital radiography imaging systems. More particularly, it relates to such a method that can provide a simple and quick way to quantitatively measure the characteristics of storage phosphor-based computed radiography (CR) imaging systems and direct-digital flat-panel detector-based direct radiography (DR) imaging systems.
BACKGROUND OF THE INVENTION
There are a number of important parameters that together characterize the performance of x-ray imaging systems. For CR and DR, most of these parameters are common to both, yet some are unique to the particular type of system. Those common parameters include spatial resolution, noise, detective efficiency, exposure response, dark image signal level, and artifacts etc.
For CR and DR, the modulation transfer function (MTF) of the imaging system is often used to characterize the system spatial resolution. MTF is a 2D (two-dimensional) function of spatial frequency and is usually measured for both x and y directions of the acquired image. There have been developed three major types of techniques for MTF measurement: resolution target, angular slit and angular edge. The resolution target technique depends on imaging either a commercially available bar-target or a star-pattern target (J. Anthony Seibert, “Photostimulable Phosphor System Acceptance Testing,” AAPM Medical Physics Monograph No. 20: Specification, Acceptance Tesing and Quality Control of Diagnostic X-ray Imaging Equipment, 1991) or a custom-made bar target of varying resolution (J. Daniel Newman, Daniel K. McBridge, James C. Montoro, “Automatic Technique for Calibrating A Storage Phosphor Reader,” U.S. Pat. No. 5,420,441, 1995). In this technique, the MTF is estimated either using a human observer to identify the blur frequency point of the target or calculating the visibility modulation. The error of the estimation depends on the orientation/resolution of the target and the subjective criteria of the observer. It requires that he target be perfectly aligned in the x or y direction, and the severity of the error increases when the target resolution is close to the Nyquist frequency of the imaging system.
Another method that makes use of a narrow slit can be used to measure the line spread function, followed by Fourier transformation to obtain the MTF of the imaging system in the slit transverse direction (Hiroshi Fujita, Du-Yih Tsai, Takumi Itoh, Kunio Doi, Junji Morishita, Katsuhiko Ueda, and Akiyoshi Ohtsuka, “A Simple Method for Determining the Modulation Transfer Function in Digital Radiography,” IEEE Transactions on Medical Imaging, Vol. 11, No. 1, 1992). In this technique, to obtain the MTF in the x or y direction, the orientation of the slit must be placed at a slight angle with the y or x direction in order to achieve super sampling for aliasing reduction. The width of the slit is required to be much narrower than the sampling pitch (pixel size) of the imaging system, and the slit needs to be long enough to cover at least one pixel in the slit transverse direction. Although this is a very accurate method for MTF measurement, it relies on a delicately made, expensive slit target.
A third method makes use of a sharp and straight edge target to measure the edge spread function of the imaging system. The MTF in the edge transverse direction can be obtained from the edge spread function by taking the Fourier transform of it's derivative (Stephen E. Reichenbach, Stephen K. Park, and Ramkumar Narayanswamy, “Characterizing Digital Image Acquisition Devices,” Optical Engineering, Vol. 30, No. 2, 1991). Because sharp and straight edges are relatively easy to manufacture accurately, this method is preferred for MTF measurement for the quality assurance of digital radiography imaging systems.
The noise of the imaging system determines the system low-contrast resolution as well as the x-ray detective efficiency etc. The noise characteristics can be described by the noise power spectrum (NPS) of the imaging system, which is also a 2D function of spatial frequency. To obtain the NPS, a flat image region is usually taken for Fourier analysis. Because the system noise level is also x-ray exposure dependent, the NPS is often measured at a certain exposure level to facilitate comparisons among imaging systems.
Detective efficiency is a secondary parameter of the imaging system that can be readily calculated from the system MTF and NPS:
DQE
(
u
,
v
)
=
DQE
(
0
,
0
)
MTF
2
(
u
,
v
)
NPS
(
u
,
v
)
.
Exposure response describes the relationship between the output of the imaging system (image pixel values) and the incident x-ray exposure. Ideally, the exposure response or logarithmic exposure response should be linear and equal for all the pixels across the whole image. Response accuracy, linearity, and uniformity are three of the major parameters for exposure response characterization. Exposure accuracy and linearity describes how accurately and linearly the output of the imaging system can track the incident x-ray exposure. Response uniformity describes the inter-pixel response variation. All three parameters are usually measured using the same x-ray spectrum but different exposure levels.
The dark image signal level determines the baseline noise of the imaging system and it is independent of x-ray exposure. For a CR image, this corresponds to the signal level that would result from reading an erased phosphor screen, and for a DR image, this corresponds to the accumulated noise level before the x-ray exposure and during the readout process.
Artifacts may also be present in images. The nature of artifacts is that they are often unpredictable and may take the form of spots, lines and low-frequency modulations etc and can also be periodic or non-periodic. The white and dark spots are usually caused by foreign dust/dirt residing on the image receptor or caused by the bad pixels (DR). There are two major types of line artifacts, periodic (banding), and non-periodic (streaks). Either artifact will result in objectionable image quality if the magnitude is large enough.
There are several other important parameters for image QA that are unique to either CR or DR. For a CR imaging system, a laser beam is used for raster scanning and reading the signal from the storage phosphor screen. Because there are moving optical devices, the image pixel size and the pixel aspect ratio can be spatially variant. The other two important parameters for CR image quality assurance are scan linearity and scan accuracy which are measures of the geometric integrity of the image. Although there is no variable geometry related quality assurance issues for DR because all the imaging pixels are solid state elements manufactured on an evenly distributed grid, the locations of failed pixels and the individual pixel response correction are unique to DR and need to be characterized.
The large number of diversified parameters to be measured for the QA of digital x-ray imaging systems, presents a considerable challenge for designing a quick, accurate, easy to use, and fully automatic method/procedure to conduct the measurements. There have been considerable efforts so far in this area. However, most of the proposed methods rely either on visually reading image pixel values from a computer screen or on printing a test image on film and then using visual examination combined with film densitometer measurements. For example, the methods using film prints and densitometer measurements include (1) the work by J. Anthony Seibert (“Photostimulable Phosphor System Acceptance Testing,” AAPM Medical Physics Monograph No. 20: Specification, Acceptance Testing and Quality Control of Diagnostic X-ray Imaging Equipment, 1991), (2) the work by Walter Huda, Manuel Arreola and Xhenxue Jing (“Computed Radiography Acceptance Testing,” SPIE Symposium on Medical Imaging 1995, Vol. 2432), (3) the work by Hamid Jafroudi, Dot Steller, Matthew Freeman and Seong Ki Mun (“Quality Control on Stor
Foos David L.
Steklenski David J.
Vanmetter Richard L.
Wang Xiao-hui
Barber Therese
Dunn Drew
Eastman Kodak Company
Noval William F.
LandOfFree
Automated and quantitative method for quality assurance of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automated and quantitative method for quality assurance of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automated and quantitative method for quality assurance of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2922367