The amount of information obtained from the non-visible information source was calculated by means of the entropy method. This method was applied to the radiographic imaging system of a clinical chest examination. The relative efficiency of the transmission transmitted from anatomically fixed positions of the human body to the photographic densities on the chest radiographs through the system was as follows. They were 43.8% for the male's chest, 41.6% for the female's chest and 51.9% for the chest phantom at one facility, and 50.4% for the male's chest at another facility. Accordingly, it is considered and made sure that the entropy method can be used as one of the calculation methods for information quantities derived from the non-visible information source.

View full abstract

Image characteristics inherent to Xeroradiography were investigated in combination with exposure and surface voltage of the selenium plate, which allowed us an easy way to predict the surface voltage merely with the result of the visibility of glass beads, which therefore, did not requires an actually measurement. The results obtained by this study are summarized as follows : 1) It was comfirmed that the surface voltage was reduced according to the exponential low, when the selenium plate was exposed to the radiation with uniform intensity, as follows in which, V=V_<0e>^<-R/R_0> where, V_0 was the initial uniform potential, V was the surface potential vs. exposure R. R_0 was the constant required to reduce the voltage down to 1/e of its initial voltage. 2) When sensitivity S of the selenium plate was reffered to as S=R^<-1>, its maximum value was obtained at approximately 40 keV. 3) Fatouros reported that the developed image density could be approximated by the next formula : D=D_0 (D_∞-D_0)e^<-R/R_0> D was the optical density, D_0 and D_∞ were the densities under a given set of development conditions for zero and for infinite exposure. The formula will enable us to predict image density at any exposure only using the experimentally determined values of D_0 and D_∞. Practical density, however, was higher than D, about 0〜0.04,for the positive mode, and 0〜0.06,for the negative mode. This density needs to be compensated by the above values. 4) The developed image density depends to a large degree on the exposed field, if the field is smaller than an area of 1/4 the selenium plate, and also the edge effect decreases at both development modes. 5) The edge effect was maximized with the exposure around R_0. 6) The maximum perceptible diameter of minute glass beads was investigated. The visual assessment revealed that optical exposures corresponding to minute glass beads detected was as follows : Ropt (posi)=1.3R_0(Back Bias 1,600 V) Ropt (nega)=0.8R_0 (Back Bias -3,500 V) This study will make it possible to predict the values, by just knowing the R_0 s in advance, about erectrostatic sensitometry, such as surface potential, potential steps (ΔV), sensitivity and image density as well, without doing any other experiments.

View full abstract

A square wave chart has been used to measure the MTF of a screen-film system. The ploblem is that the scattered X-ray from the chart may give rise to meansurement errors. In this paper, the authors proposed two methods to reduce the scattered X-ray : the first method is the use of a Pb mask and second is to provide for an air gap between the chart and the screen-film system. In these methods, the scattered X-ray from the chart was reduced. MTFs were measured by both of the new methods and the conventional method, and MTF values of the new methods were in good agreement while that of the conventional method was not. It was concluded that these new methods are able to reduce errors in the measurment of MTF.

View full abstract

Renogram curves of two nuclides were obtained by simultaneous using a mixture solution of ^<131>I-Hippuran and ^<111>In-DTPA. In this report, a simple quantitative analysis method for obtaining RPF and GFR from these renogram curves was investigated. These results closely approximate the RPF and GFR values which were obtained from conventional renal clearance test.

View full abstract

It is important to check periodically the quality and quantity of the X-rays emitted, in order to obtain the best possible performance from your diagnostic X-ray apparatus. The best way of checking the exact quality of the X-ray is to measure the spectrum of the X-ray, but it is not an easy task to carry out. The second way is to plot the attenuation curve of the X-rays. We have developed a method to plot the attenuation curve by a single exposure, utilizing J. R. Greening's empirical formula. The output of the three cavity ionization chambers, one with 7 mmAl filter, another with a 3 mmAl and the third without any filter, exposed to the same X-ray, were put into a microcomputer. The programming was arranged to display the attenuation curve of the X-rays, effective energy of the X-rays, the 1st HVL and the 2nd HVL on the CRT. The attenuation curves of the X-rays, emitted at a tube voltage at between 60 and 140 kV obtained by this method, agreed with the experimental results with an error of ±4%. The effective energy obtained by this method agreed with the experimental data with an error of ±1%.

View full abstract

K. Doi, G. Holje, L. N. Loo, H. P. Chan, J. M. Sandrik, R. J. Jennings, and R. F. Wagner. MTF's and Wiener Specrta of Radiographic Screen-Film Systems. HHS Publication FDA 82-8187 (April 1982). This report is a presentation, in two parts, of the measurements of the fundamental imaging characteristics of radiographic screen film systems. Part 1 describes the interlaboratory comparison of measurements of sensitometry, the modulation transfer function (MTF), and the Wiener spectrum (WS) undertaken between the Kurt Rossmann Laboratories for Radiologic Image Research, Department of Radiology, University of Chicago (UC) and the Medical Physics Branch, Division of Electronic Products, Bureau of Radiological Health (BRH). Part 2 consists of tabular and graphical presentations of the results of measurements of sensitivity, sensitivity, sensitometry, MTF, and WS performed at UC on a selection of screen-film systems. Both calcium tungstate screens and those made from recently developed phosphors ("rare earths") were evaluated. The first part of this report will be of general interest to those seeking information on the theory and practice of fundamental imaging measurements of screen-film systems and of particular interest to those with a serious commitment to the measurement of these quantities. The second part will be of value to those seeking imaging data as input to research on diagnostic radiologic imaging systesm and to those attempting to develop quantitative, physical comparisons among imaging systems. Although almost all of the measurements in Part 2 represent the performance of each screen with only one of three film types, suggestions for estimating the characteristics of the same screens with other films are given. In addition, the measurement effort to acquire data on more systems is continuing. The image analysis techniques used at UC and BRH are similar in principle. Sensitometry was performed using the inverse-square, x-ray, intensity-scale method. MTF's were derived from measurements of the line spread function acquired by exposing the system to x rays transmitted by a narrow slit. Wiener spectra were calculated from microdensitometric measurements of films exposed to a uniform gross diffuse density of 1.0. The techniques did, however, differ in detail and descriptions of the equipment used and the data acquisition and analysis methods employed by each laboratory are included. The intercomparison study demonstrated that MTF's derived from the same film sample were reproducible between laboratories to a precision comparable to the measurement uncertainty at either laboratory-about 0.02. Variations of MTF's of different samples of a single screen-film type were about 2.5 times larger. These variations were ascribed to variations in manufacture of the screens and film and in film processing. The magnitudes of contributions to the variability between samples due to such sources have not been indepently investigated. In the 0.4 to 4 cycle/mm frequency range, the WS of a film sample measured at one laboratory agreed to within 10 to 20 percent of the WS measured at the other when corrected to express fluctuations of diffuse density by the methods routinely used at each laboratory. Transmission fluctuations measured at UC were corrected according to the logarithmic relationship between density and transmission. Instrument (specular) density fluctuations measured at BRH were corrected according to the slope of the microdensitometric characteristic curve. A third means of intercomparing noise measurements is to scale the fluctuations measured in an image to fluctuations in exposure and determine the number of noise equivalent quanta (NEQ). These methods are still being refined. The tables and graphs of Part 2 provide convenient and consistent data on the imaging characteristcs of 18 screen filim systems Also included are general and specific descri

(View PDF for the rest of the abstract.)

View full abstract

Minimum perceptible contrast ΔD_<min> was obtained from Penetrameter made of alminum plates in which 40 minute holes were perforated. An object will be visualized when its contrast exceeds the ΔD_<min>. Visual threshold contrast ratio defined here as ΔD_<obj>/ΔD_<min>=1 offers minimum perceptible diameter MPD which theoretically obtained from ΔD_<obj>, corresponding to the technique you applied. The MPDs were obtained in prior to experiments as a function of system speed, beam quality, exposure range and magnification, which resulted in a reasonable agreement with the experimental results. Visibility of minute dot-lide objects depends much upon granularity as its threshold is expected to be governed by the surrounding grainy noise. Our investigation, however, revealed that MTF was as important as granularity, and the detail visibility defined here as an inverse function of the MPD is correlated to the physical factors as follows ; MPD^<-1>∝γ (MTF/σ(D)^<1/2> Dot and line-lide objects were simulated by micro glass beads and Nylon fibrils, respectively in order to make ROC analysis on the different radiographic techniques with a CGR SENOGRAPHE 500T mammographic unit. The 1.5X magnification technique showed, on phantom work basis, the greatest advantage for detecting glass beads over the rest such as grid technique and Xeroradiography as well as conventional contact technique. This implies that magnification can reduce relative granularity to the object. On the other hand, Xero scored the remarkably highest true positive rate, ie., 100 percent for detecting Nylon fibrils. Clinical evaluation proved that neither magnification nor grid technique offered any distinct advantage over the contact technique. This implies that weight of physical factors affecting visibility for actual mammograms differs from that of phantom work, ie., phantom images with uniform back ground require finner grain system, actual mammograms, on the contrary, sharper images or greater acutance clearly recognized from the surroundings. The results from ROC analysis were extended to entropy analysis. Transmitted information T (X ; Y) eloquently speaks of the nature of the ROC curves, but redundancy C (z) correlates well to the clinical results as wide scattering of the judgement due to lack of acutance developses a tendancy to reduce the score. This study on the visibility leads us to a conclusion that the detail visbility in actual mammography depends much upon acutance of the image as Xeroradiography proved. Author warned possible risks that lead to cancer induction by a long-term annual screening by mammography. Breast cancer incidence and detection probability of clinically occult cancer are especially low in Japan, ie., approximately one hundredth those of the West according to Feig's evaluation. Risks, therefore, should be always taken into consideration when you evaluate visibility.

View full abstract