Practical evaluation of image quality in computed radiographic (CR) imaging systems

Two main spatial frequency and quantitative calculations and measurements am, used to study the imaging system of three industrial digital radiography (IDR) modules for non-destructive testing (NDT), namely complementary metal oxide semiconductor (CMOS) flat-panel digital detector with 50 mu m pixel pitch, computed radiography (CR) with phosphor imaging plate with 25 mu m pixel pitch and laser-type film digitiser with 50 mu m pixel pitch. The modulation transfer function (MTF) and noise power spectrum (NPS) measurement and calculations were adapted on order to evaluate the image quality of IDR images. The MTF has been used to characterise the resolution properties of analogue and digital X-ray imaging systems and measure image deterioration due to optical factors, such as diffusion of image forming radiation. The NPS is used in determining noise in images produced and the choices of normalisation of the NPS vary slightly from researcher to researcher, but the underlying methodology is consistent. In order to easily compare these two measurements and calculations, the integral of up to the system Nyquist frequency was used as the final image quality quantitative evaluation. For the MTF measurement and calculation, four regions of interest (ROI) were, defined to portray the randomness in performing real testing. The four ROI are then compared with each other. For the NPS measurement and calculation, ten ROI were defined, scattered and averaged to have overall noise estimation. From the measurements and calculations which were done, the averaged MTF at 20% modulation for the laser-type film digitiser system is 6.55 cycles/mm, CR modulated 4.48 and CMOS modulated 2.83 cycles/mm. For NPS measurement and calculation, the result obtained in decreasing order is laser-type film digitiser, CMOS and CR. The laser-type film digitiser system modulates the best transfer function at 20% but does not have the lowest NPS. The lowest and stable NPS is the CR system but it has the lowest modulating capability at 20%. The study shows that in order to perform NDT by using the evaluated modules, the user must know the true capability, of the system and how, it is designed for specific application and discontinuity detection.

Until recently, advanced image quality metrics such as the frequency-dependent detective quantum efficiency, DQE (f), have been employed by a relatively small cadre of expert imaging scientists to compare the performance of digital radiography systems, often using differing methods. To address the potential for differences in measured DQE(f) arising from differences in test methodology, in 2003 the International Electro-technical Commission (IEC) published the first in a series of performance standards to define techniques for measuring the DQE(f) and its constitutent metrics, the modulation transfer function, MTF(f), and noise power spectrum, NPS(f). More recently, the increasing prevalence of digital radiographic systems in the clinical environment and availability of both commercial and public-domain resources have made DQE(f) evaluations both more accessible and increasingly relevant to the clinical medical physicist. This course will review the measurement and analysis methods for DQE evaluations of both general digital radiography and digital mammography systems, using the relevant IEC standard method as a guide. Learning Objectives: 1. Review the definition of the DQE(f) and its constitutent metrics the MTF(f) and NPS(f). 2. Gain an understanding of the IEC standard method for measuring detective quantum efficiency, for both general digital radiography and digital mammography imaging systems. 3. Learn which commercial and public-domain resources are available for measuring the DQE(f) of digital radiographic imaging systems.

Medical images are acquired for the purpose of gathering information regarding the disease state of a patient. The quality of the image should be a measure of how a specified task may be performed given the resulting image. Image quality is task dependent; one imaging system may provide images that allow better task performance than another system's for one task, but worse for another task. Determining the quality of an imaging system involves specifying the task of interest, the range of objects that will be imaged, and the observer (human or machine reader) who will perform the task. In radiography it has become common to use Fourier measures to describe image quality through a combination of Fourier‐based measures of system resolution and noise. The Line Spread Function, when Fourier transformed, yields the Modulation Transfer Function (MTF) as a measure of system resolution. Use of the MTF to describe a system's resolution is appropriate when the imaging system is shift‐invariant, that is, the line spread function is independent of position. Image noise is described in the Fourier domain by the Noise Power Spectrum (NPS). We can use the NPS to describe the noise in an imaging system whenever the noise properties are stationary, meaning that the variance and correlations between fluctuations in different locations are not dependent on their absolute location. These Fourier measures of system resolution and noise are commonly combined to give an overall figure of merit known as the Detective Quantum Efficiency (DQE). The DQE describes the signal‐to‐noise transfer characteristics of an imaging system as long as those assumptions of shift‐invariance and noise stationarity are valid. In digital systems the image of an object is affected by the location of the object with respect to the pixel grid. In other words, the detector pixels cause the system to be shift‐variant. In this case the MTF does not describe the resolution properties completely. This is particularly the case for tomographic systems, where the point‐response function is quite position‐dependent. Moreover, in real imaging systems the noise properties will depend on position, most notably because real patients always have structure. So, the Fourier‐domain noise concept of Noise Power Spectrum does not capture all the properties of noise for real imaging systems and tasks. In this talk we will consider a general treatment of the imaging system as a mapping of information from the patient to the data, without the assumptions inherent in Fourier analysis. We will discuss approaches to the characterization of system resolution and noise in terms of position‐dependent parameters. Once such position‐dependent imaging properties of a system are determined, we can use them to calculate measures of image quality that summarize the usefulness of the images for the performance of various visual tasks.

Amorphous silicon (a-Si:H) flat-panel (FP) imaging systems have recently become commercially available for both chest and mammographic imaging applications. It has been shown that this new detector technology offers better image quality and various operational advantages over the computed radiography (CR) which to date has been the most widely implemented and used digital radiography technique. However, most image quality measurements reported on flat-panel systems have been performed on prototype systems in laboratories while those for CR systems were typically independently performed and reported on in separate studies. To directly compare the two technologies, we have measured the image properties for a commercial amorphous silicon/cesium iodide [a-Si:H/CsI(Tl)] flat-panel based digital chest system and a commercial CR system under clinical imaging conditions. In this paper, measurements of image quality metrics, including the modulation transfer functions (MTFs), noise power spectra (NPSs), and detective quantum efficiencies (DQEs), for the FP and CR systems are presented and compared. Methods and issues related to these measurements are discussed. The results show that the flat-panel system has slightly lower MTF but significantly higher DQEs than the CR system. The DQEs of the flat-panel system were found to increase with the exposure while those of the CR system decrease slightly with the exposure.

Over the past several years, digital mammography systems have been installed clinically across North America in small but growing numbers. A photostimulable phosphor-based full-field digital mammography image was evaluated in this investigation. Commonly known as computed radiography (CR), its use closely mimics the screen-film mammography paradigm. System performance using modulation transfer function (MTF) and detective quantum efficiency (DQE) metrics show MTF(2.5 mm(-1)) = 0.5, DQE(2.5 mm(-1)) = 0.3, and MTF(5.0 mm(-1)) = 0.2, DQE(5.0 mm(-1)) = 0.05, for a 26 kVp beam, 0.03 mm molybdenum tube filtration, 4.5 cm tissue attenuation, and 15 mR incident exposure to the detector. Slightly higher DQE values were measured at 32 kVp with 0.025 mm rhodium tube filtration. CR mammography advantages include the ability to use existing mammography machines, where multiple rooms can be converted to "digital" operation, which allows overall cost savings compared to integrated digital mammography systems. Chief disadvantages include the labor-intensive handling of the cassettes prior to and after the imaging exam, lack of a direct interface to the x-ray system for recording technique parameters, and relatively slow processing time. Clinical experience in an IRB-approved research trial has suggested that digital mammography with photostimulable storage phosphors and a dedicated CR reader is a viable alternative to conventional screen-film mammography.

Digital imaging systems require offset and gain calibration to normalize the behavior of individual pixels. This normalization corrects for imperfections in the system and also external variables that have effects on uniformity. Imaging metrics like Detective Quantum Efficiency (DQE) and Modulation Transfer Function (MTF) define how sensitivity and resolution are transferred through the system. Gain calibration can result in a loss of DQE due to the noise associated with its application. The typical technique to minimize this noise is to average several gain calibration procedures so that the introduced noise is minimized. This paper discuses the effects of gain calibration on DQE. It measures DQE as a function of the number of gain calibration procedures averaged and contrasts it with a novel technique that uses a single filtered gain calibration. It demonstrates that noise filter techniques, applied to a single gain calibration, regains the loss in DQE without any degradation in resolution. This paper also compares imaging performance of a system using a filtered gain map against a system that has many gain calibrations averaged. The technique is demonstrated using a Thin-Film-Transistor (TFT)-based large area medical imaging system.

As a continuation of our digital radiographic sensor R&D, we have developed a digital gamma imaging system based upon the cadmium-telluride (CdTe) photoconductor for the applications of industrial gamma-imaging. The imaging system consists of a commercially-available CMOS pixel array of a 100 × 100 μm2 pixel size and a 5.4 × 151.0 mm2 active area, coupled with a 750-μm-thick CdTe photoconductor, and a collimated selenium (75Se) radioisotope of an about 2.3 × 1012 Bq activity and a physical size of 3.0 mm in diameter. In this study, we, for the first time, succeeded in obtaining useful gamma images of several test phantoms with the 75Se radioisotope from the imaging system and evaluated its imaging performance in terms of the line spread function (LSF), the modulation transfer function (MTF), the noise power spectrum (NPS), and the detective quantum efficiency (DQE). For comparison, we also tested its X-ray imaging performance with a microfocus X-ray tube of an about 5 μm focal spot size at an operation condition of 90 kVp and 100 μA. According to our test results, the designed imaging system seems to be promising for our ongoing application of nondestructive testings, provided that the effective source size of the 75Se radioisotope can be further limited.

Photon counting detectors (PCDs) are an emerging technology with applications in spectral and low-dose radiographic and tomographic imaging. This paper develops an analytical model of PCD imaging performance, including the system gain, modulation transfer function (MTF), noise-power spectrum (NPS), and detective quantum efficiency (DQE). A cascaded systems analysis model describing the propagation of quanta through the imaging chain was developed. The model was validated in comparison to the physical performance of a silicon-strip PCD implemented on an experimental imaging bench. The signal response, MTF, and NPS were measured and compared to theory as a function of exposure conditions (70 kVp, 1-7 mA), detector threshold, and readout mode (i.e., the option for coincidence detection). The model sheds new light on the dependence of spatial resolution, charge sharing, and additive noise effects on threshold selection and was used to investigate the factors governing PCD performance, including the fundamental advantages and limitations of PCDs in comparison to energy-integrating detectors (EIDs) in the linear regime for which pulse pileup can be ignored. The detector exhibited highly linear mean signal response across the system operating range and agreed well with theoretical prediction, as did the system MTF and NPS. The DQE analyzed as a function of kilovolt (peak), exposure, detector threshold, and readout mode revealed important considerations for system optimization. The model also demonstrated the important implications of false counts from both additive electronic noise and charge sharing and highlighted the system design and operational parameters that most affect detector performance in the presence of such factors: for example, increasing the detector threshold from 0 to 100 (arbitrary units of pulse height threshold roughly equivalent to 0.5 and 6 keV energy threshold, respectively), increased the f50 (spatial-frequency at which the MTF falls to a value of 0.50) by ∼30% with corresponding improvement in DQE. The range in exposure and additive noise for which PCDs yield intrinsically higher DQE was quantified, showing performance advantages under conditions of very low-dose, high additive noise, and high fidelity rejection of coincident photons. The model for PCD signal and noise performance agreed with measurements of detector signal, MTF, and NPS and provided a useful basis for understanding complex dependencies in PCD imaging performance and the potential advantages (and disadvantages) in comparison to EIDs as well as an important guide to task-based optimization in developing new PCD imaging systems.

For x-ray detectors, Compton interactions deposit photon energies along the paths of recoil electrons, which are not isotropic about the primary interaction sites. Light from each interacting x-ray is only generated near the path of a recoil electron. In this study, Compton scatter is modeled as an input-labeled cascade of the amplification and scattering processes to describe the transfer relationship of signal and noise in frequency space. The output of the model is the spatial distribution of secondary quanta generated by Compton recoil electrons. We determine the spatial dependence and statistical correlation of secondaries from the initial energy of the recoil electron and its range, resulting in the 'Compton' modulation transfer function (MTF) and noise power spectrum (NPS), respectively. Then the 'Compton' MTF and NPS are used to calculate the 'Compton' detective quantum efficiency (DQE). The probability density function of scattering angle of Compton recoil electron is developed using the Klein-Nishina coefficients. Results are applied to the description of a portal imaging system at 6 approximately MV where non-Compton interactions can be ignored. The MTF results are compared with a Monte Carlo calculation. This is the first model of how Compton interactions in the metal-plate/phosphor combination degrade image quality in terms of signal and noise. It is shown that Compton MTF depends on energy of x-ray photon in a complex way, and Compton scatter imposes a fundamental limitation on both the MTF and DQE of x-ray imaging system.© (2002) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Current flat-panel detectors either directly convert x-ray energy to electronic charge or use indirect conversion with an intermediate optical process. The purpose of this work was to compare direct and indirect detectors in terms of their modulation transfer function (MTF), noise power spectrum (NPS), and detective quantum efficiency (DQE). Measurements were made on three flat-panel detectors, Philips Digital Diagnost, GE Revolution XQ/i, and Hologic Direct-Ray DR1000 using the IEC-defined RQA5 (approximately 75 kVp, 21 mm Al) and RQA9 (approximately 120 kVp, 40 mm Al) radiographic techniques. The presampled MTF of the systems was measured using an edge method (Samei et al., Med Phys 25:102, 1998). The NPS of the systems was determined for a range of exposure levels by 2D Fourier analysis of uniformly exposed radiographs (Flynn and Samei, Med Phys 26:1612, 1999). The ideal signal-to-noise ratio per exposure for each system was estimated using a semi-empirical x-ray model. The DQE, reported only at the RQA5 technique, was assessed from the measured MTF, NPS, exposure, and the ideal signal-to-noise ratio. For the direct system, the MTF was found to be significantly higher than that for the indirect systems and very close to an ideal function associated with the detector pixel size. The NPS for the direct system was found to be constant in relation to frequency. The DQE results reflected expected differences based on the absorption efficiency of the different detector materials. Using RQA5 and 0.3 mR exposure, the measured DQE values at spatial frequencies of 0.15 mm-1 and 2.5 mm-1 were 64% and 14% for the XQ/i system and 35% and 19% for DR-1000. Using RQA5 and the averages at all exposures, the corresponding values were 58% and 13% for the XQ/i system and 36% and 19% for DR-1000.

The purpose of this study was to critically assess and compare the intrinsic resolution, noise, and signal-to-noise transfer characteristics of two modern digital photostimulable phosphor (PSP) radiographic systems. Two commercial PSP systems were evaluated by identical methodologies. Measurements were made at three beam qualities using standard-resolution and high-resolution screens. The presampled modulation transfer functions (MTF) of the systems were measured using an edge method. The noise power spectra (NPS) were determined by 2D Fourier analysis of uniformly exposed radiographs. The frequency-dependent detective quantum efficiencies (DQE) were obtained from the MTF and NPS measures and the input signal-to-noise ratio, determined by a computational model for x-ray spectrum which was verified against exposure and half-value-layer measurements. The physical performance of the systems were very similar; a DQE of 0.24 and a spatial frequency of 2.5 cycles/mm at 0.2 MTF were estimated for both systems at 115 kVp using a pixel size of 0.1 mm and standard-resolution screens. The HD screen provided improved MTF with an adverse effect on noise as compared to the GP-25 screen using the same pixel size and beam quality.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Getting to grips with quantitative image quality measurements for QA of digital mammography (DM) systems

Subtle lung nodules (SLNs) are lesions that are commonly missed in radiologic examinations of the chest. This thesis explores three factors that influence the detection of these lesions: contrast‐diameter characteristics of SLNs, physical performance of the imaging system, and structured noise in the chest radiographs. The contrast‐diameter characteristics of SLNs were determined from radiographs containing confirmed lesions. The nodules were found to have similar subject contrast profiles with diverse variability in peak contrast (0.02–0.13) and diameter (3–16 mm). The physical performance of a digital chest radiographic system was assessed by measuring its Modulation Transfer Function (MTF), Noise Power Spectrum (NPS), and Detective Quantum Efficiency (DQE) responses. The results showed that the system only minimally blurs the appearance of SLNs, and that for a typical chest exposure, the quantum mottle is low enough not to obscure the visibility of even the smallest SLNs. The relative influence of the structured and quantum noise was explored in a Receiver Operating Characteristic (ROC) study in which simulated tissue‐equivalent nodules were superimposed on images of either structured or quantum noise backgrounds. The detection threshold =0.8) was achieved for nodules 4.5 mm in diameter against structured backgrounds, and for nodules 1.0 mm in diameter against quantum noise backgrounds. The results indicate that anatomical structured noise is the main factor limiting the detection of SLNs in chest radiographs.

Cardiac x-ray detectors are used to acquire moving images in real-time for angiography and interventional procedures. Detective quantum efficiency (DQE) is not generally measured on these dynamic detectors; the required "for processing" image data and control of x-ray settings have not been accessible. By 2016, USA hospital physicists will have the ability to measure DQE and will likely utilize the International Electrotechnical Commission (IEC) standard for measuring DQE of dynamic x-ray imaging devices. The current IEC standard requires an image of a tilted tungsten edge test object to obtain modulation transfer function (MTF) for DQE calculation. It specifies the range of edge angles to use; however, it does not specify a preferred method to determine this angle for image analysis. The study aimed to answer the question "will my choice in method impact my results?" Four different established edge angle determination methods were compared to investigate the impact on DQE. Following the IEC standard, edge and flat field images were acquired on a cardiac flat-panel detector to calculate MTF and noise power spectrum, respectively, to determine DQE. Accuracy of the methods in determining the correct angle was ascertained using a simulated edge image with known angulations. Precision of the methods was ascertained using variability of MTF and DQE, calculated via bootstrapping. Three methods provided near equal angles and the same MTF while the fourth, with an angular difference of 6%, had a MTF lower by 3% at 1.5 mm(-1) spatial frequency and 8% at 2.5 mm(-1); corresponding DQE differences were 6% at 1.5 mm(-1) and 17% at 2.5 mm(-1); differences were greater than standard deviations in the measurements. DQE measurements may vary by a significant amount, depending on the method used to determine the edge angle when following the IEC standard methodology for a cardiac x-ray detector. The most accurate and precise methods are recommended for absolute assessments and reproducible measurements, respectively.

We have analyzed the influence of stochastic amplifying and scattering mechanisms on the transfer of signal and noise through multistage imaging systems in terms of multivariate moment-generating functions. Stochastic amplification of photon noise by one stage of an imaging system is shown to constitute an effective signal to the next, while the underlying photon-noise component is unaffected by a subsequent scattering process. In the case of stationary, photon-limited inputs, these considerations then lead to useful expressions for the noise power spectrum and detective quantum efficiency for multistage image systems. The application of these results to the analysis of radiographic screen-film imaging systems is discussed.