Yoshiyuki Asai

Optimum tube voltage for chest radiographs obtained by psychophysical analysis

Many kinds of x-ray films having various characteristic curves have been developed for chest radiographs. In general, a phototiming device for determination of a mAs value which gives a proper exposure has been used for a chest radiography. For each film, however, the x-ray tube voltage has been determined by the subjective evaluation of radiologists or radiological technologists. In this paper, we propose a new method for determining the optimum tube voltage for chest radiographs using psychophysical analysis. The optimum density and the optimum density range of a screen/film system are obtained from the gradient curve of film and the minimum perceptible contrast delta Dmin [Acta Radiol. Diagnos. 4, 463-476 (1966)]. The optimum tube voltage, by which the lowest density of a mediastinum and the highest density of a lung field just cover the optimum density range, is obtained using the x-ray photon spectrum and sensitivity spectrum of the screen. This objective method does not depend on personal subjective evaluation, therefore it is available for the determination of optimum tube voltage for chest radiographs to be observed by many doctors of various departments.

Investigation of optimal display size for detecting ground-glass opacity on high resolution computed tomography using a new digital contrast-detail phantom

The purpose of this study was to clarify the relationship between display sizes of high resolution computed tomography (HRCT) images for detecting ground-glass opacity (GGO) and observer performance using a digital contrast-detail (d-CD) phantom. A structure of the d-CD phantom was determined on the basis of the actual images of GGOs and background noises of 22 patients who were diagnosed as GGO by chest HRCT. The d-CD phantom has a 512×512 matrix in size and has total of 100 holes: the diameter of these holes increases stepwise from 2 to 20 pixels with 2 pixels interval in a vertical direction and the CT value varies stepwise from 2 to 200 HU in a horizontal direction. The observer performance study was carried out for three different display sizes (30 cm×30 cm as an enlarged size, 13 cm×13 cm as an original size, and 7 cm×7 cm as a reduced size) using a 2-megapixels LCD monitor, and it was analyzed using Friedman and Wilcoxon statistical tests. As a result, the observer performance for the original display and the reduced display sizes was superior to that for the enlarged size (P=0.006 and 0.037 for the original display and the reduced display sizes, respectively), whereas there was no significant difference between the original display and reduced display sizes (P=0.77). The d-CD phantom enables a short-term evaluation of observer performance and is useful in analyzing relationship between display size and observer performance.

Equivalent spatial frequency and optimum film densities for the perceptibility of radiographic contrast of step-edge images

AbstractPrevious experimental results for minimum perceptible contrasts of step-edge images and sinusoidal periodic patterns were transformed to minimum perceptible luminous-exitance differences. By comparing the differences between the two kinds of patterns, the equivalent spatial frequency for the perceptibility of step edges was determined as 1.6-1.8 mm 1. From the minimum perceptible contrasts of step-edge images, the optimum film densities for step edges were obtained as 1.1 - 1.3 for Kodak XRP and OC medical films and 1.7 -2.0 for Fuji Industrial 100 and FR industrial films.

Evaluation of grey-scale standard display function as a calibration tool for diagnostic liquid crystal display monitors using psychophysical analysis

The digital imaging and communications in medicine (DICOM) standard proposed the grey-scale standard display function (GSDF) as a calibration tool for making the gradation characteristic of a radiographic output image consistent. This is designed in such a manner that the contrast visually recognised by observers (called psychophysical contrast) becomes a constant for all digital driving levels. The DICOM standard calls such an ideal characteristic perceptual linearisation. The psychophysical gradient that can express the psychophysical contrast was introduced for the evaluation of the GSDF using a liquid crystal display monitor. Investigations regarding its ability to yield a constant psychophysical contrast, independent of the digital driving level change and under an actual observation environment, such as for clinical radiographic diagnosis in hospital, were carried out. The psychophysical gradients of the GSDF were obtained for two kinds of observation environment: one was a restricted environment such as in a dark room, under steady-state adaptation, using the sinusoidal grading pattern corresponding to the peak frequency of the human eye response. The other was an actual environment reflecting that encountered during clinical diagnosis in a hospital. As a result, the psychophysical gradient under the restricted environment became almost constant and independent of the change in digital driving level, i.e. perceptual linearisation could be satisfied. Furthermore, under the actual observation environment, the psychophysical gradient decreased gradually with the increase in digital driving level, i.e. the perceptual linearisation could not be satisfied. The percentage decrease in the value of the psychophysical gradient at the maximum luminance area was approximately 60% compared with that at the minimum luminance area. Accordingly, the GSDF is unsuitable as a calibration tool for the liquid crystal display monitor, which will be used in actual clinical diagnosis, as it cannot achieve ‘perceptual linearisation’ under the actual environment. For the purpose of clinical diagnosis, it is necessary to enlarge the physical gradient of GSDF further in the high digital driving level range (which relates to a high luminance area) to give an approximation that is as close to the idealised from as possible.

Psychophysical RMS Granularity for the Evaluation of Radiographic Mottle

AbstractThe psychophysical RMS (root mean square) granularity ap is proposed for the evaluation of radiographic mottle, i.e., the granularity appearing in a radiograph made with an X-ray screen-film system. The value of is calculated from the Wiener spectrum of the radiographic mottle and the frequency dependence of the minimum perceptible contrast, which expresses a psychophysical property of human vision. Examples are presen ted for three typical X-ray films. The density dependence of a,, is in close agreement with the results of subjective rating that (V the most perceptible density is 1.0 and (2) the perceptibility at density 2.5 is much worse than that at 0.6. Whereas the Selwyn granularity coefficient, which relates to the physical RMS granularity, does not correlate with the results of subjective rating.

Identification of elemental weight fraction and mass density of humanoid tissue-equivalent materials using dual energy computed tomography

The purpose of this study was to obtain the fraction by weight of the elemental composition and mass density of a humanoid tissue phantom that includes lung tissue, soft tissue, and bone tissue, by using dual energy computed tomography (DECT). The fraction by weight and the mass density for tissue-equivalent materials were calculated by means of a least-squares method with a linear attenuation coefficient, using monochromatic photon energies of 10-140keV, as obtained from DECT. The accuracy of calculated values for the fractions by weight of H (hydrogen), C (carbon), N (nitrogen), and O (oxygen) as verified by comparing the values with those that were analyzed using the combustion technique. The fraction by weight for other elements was confirmed by comparing with the analyzed values by means of energy dispersive photon spectroscopy. The calculated mass densities for each tissue were compared with those that were obtained by dividing the weight by volume. The calculated values of the fraction by weight that were obtained by means of DECT had differences of 1.9%, 9.2%, 6.6%, 7.8%, 0.8%, and 0.2% at a maximum for H, C, N, O, P (phosphorus), and Ca (calcium), respectively, from the reference values analyzed by the combustion technique and energy dispersive photon spectroscopy. The difference in the mass density for tissue was 0.011g/cm3 at a maximum. This study demonstrated the fraction by weight and the mass density of the humanoid tissue-equivalent materials that were calculated by means of DECT were expected high accuracy.

Reduction of patient dose in medical radiography by utilizing scattered X-rays: relation between permissible limit of scatter fraction, viewer brightness, and perceptibility of vision.

This paper proposes a new technique for reducing the patient dose when employing medical radiographs prepared by using screen-film systems. In this technique the patient dose can be reduced by employing scattered X-rays in order to obtain the same film density as that realized without the use of scattered X-rays. The minimum perceptible thickness difference ΔX(min), which can be recognized by liminal vision, was psychophysically calculated by considering the energy spectrum of incident X-ray, sensitivity spectrum of the screen layer, and the perception capability of human vision. From the calculated ΔX(mins) in various conditions, the permissible upper limit of scatter fraction for obtaining the same ΔX(min) for three kinds of luminances, and the fraction of reduction in the primary X-rays were determined. As an example of the results, when the object size required for perception is 1.3 mm, a scatter fraction up to 42% can be permitted at a density D of 1.0 for a luminance of 2548 cd m(-2). When we increase the luminance of the viewer from 478 cd m(-2) to 2548 cd m(-2), the upper limit of the permitted scatter fraction varies from 30% to 42% at a D of 1.0, i.e., the patient dose can be reduced by 17% under the same perceptibility of ΔX(min) by utilizing scattered X-rays. This reduction can be successfully achieved by changing the lead content of the grid from 0.45 to 0.38 g cm(-2).

Edge enhancement effect of vision on X-ray radiographs made using screen—film systems

Abstract Theoretical and experimental approaches for the edge enhancement effect of vision on X-ray radiographic images were carried out. Psychophysically enhanced portions in the density unit at both sides of a step edge image were derived from modulation transfer functions (MTFs) of vision and radiographic systems. Experimental values were obtained using a new method, in which a step edge image was compared with a standard film having continuous density change. Experimental values distributed around theoretical values and their qualitative behaviour were similar to those of theoretical values. Discrepancy between theoretical and experimental results was due to the non-linearity of the human visual system.

Evaluation of radiographic mottle using a single frequency value of the psychophysical Wiener spectrum

AbstractIn radiographs, the Wiener spectral values al low frequency have been used for the evaluation of radiographic mottle. The validity of the psychophysical RMS granularity a,, as a new measure for the evaluation of radiographic mottle was confirmed in our previous paper IJ. Photogr. Sci., 43, 103 (1995)1.We calculate the density dependence at various frequencies of the Wiener spectrum W(v) and the psychophysical Wiener spectrum Wp(v), which is psycliophysically corrected, and compare with that of σp2. The density dependence of Wr(v) at a frequency of 2.0 mm -1 is in good agreement with that of σp2. It is therefore possible to evaluate the radiographic mottle using the W/v) value, at a frequency of 2.0 mm -1 instead of ap in this example. The time to obtain the density dependence of ap, which requires the measurement of ΔD,mm, values at many frequencies and integration with respect to frequency, can be reduced using this method.

Prediction of glandularity and breast radiation dose from mammography results in Japanese women

AbstractGlandularity has a marked impact on the incidence of breast cancer and the missed lesion rate of mammography. The aim of this study was to develop a novel model for predicting glandularity and patient radiation dose using physical factors that are easily determined prior to mammography. Data regarding glandularity and mean glandular dose were obtained from 331 mammograms. A stepwise multiple regression analysis model was developed to predict glandularity using age, compressed breast thickness and body mass index (BMI), while a model to predict mean glandular dose was created using quantified glandularity, age, compressed breast thickness, height and body weight. The most significant factor for predicting glandularity was age, the influence of which was 1.8 times that of BMI. The most significant factor for predicting mean glandular dose was compressed breast thickness, the influence of which was 1.4 times that of glandularity, 3.5 times that of age and 6.1 times that of height. Both models were statistically significant (both p < 0.0001). Easily determined physical factors were able to explain 42.8% of the total variance in glandularity and 62.4% of the variance in mean glandular dose. Graphical abstractValidation results of the above prediction model made using physical factors in Japanese women. The plotted points of actual vs. prediction glandularity shown in a are distributed in the vicinity of the diagonal line, and the residual plot for predicted glandularity shows an almost random distribution as shown in b. These distributions indicate the appropriateness of the prediction model.