Spectrozonal X ray detectors allow X ray image to be obtained at different energies and information value of projection X ray to be increased without increasing dose load on the patient [1,3].Subtraction algorithms of spectrozonal X ray imag ing can restore separate images of bone and soft tissues [5]. The resulting images are equivalent to images obtained using linear tomography of skeleton hidden pathology. Multienergy methods of radiography would provide an opportunity of not only density imaging, but also imaging in grades of atomic number or electron den sity [4,6], thereby providing noninvasive analysis of the nature of a lesion [7].Let us consider a method for calibration of an X ray diagnostic system equipped with a spectrozonal X ray detector.Let the detector have spectrometric properties and be able to distinguish X ray photons of 3 energy groups: low (l), medium (m), and high (h).A tissue identification criterion was suggested in [2]:where µ ob,h , µ ob,m , µ ob,l are X ray mass extinction coeffi cients of inclusion tissues at high, medium, and low ener gies, respectively. Practical implementation of this method includes recasting of Eq. (1) as:where I f и I ob are radiation intensities at background and inclusion regions, respectively.In case of calibration, Eq.(2) allows the effective atomic number of tissue to be determined using feedback filtration of background component of the image itself.Given the fact that X ray radiation extinction coeffi cient is a function of quantum energy:where summand τ(E)Z 4 determines photon absorption; σ(E)Z is Compton scattering; and χ(E)Z 2 is pair forma tion effect for energies used in medical radiography:Effective atomic number can be calculated from angle ϕ. Dependence of τ(E) and σ(E) is universal for all substances: Φ = tanϕ = µ ob,m /µ ob,l ; the expression for ϕ is:It follows from Eq. (2) that Φ = ln(I f /I ob ) m /ln(I f /I ob ) l .If coefficient Φ is experimentally found, effective atomic number of inclusion tissue can be derived from Eq. (4) as:Coefficients τ(E m ), τ(E l ), and σ(E m ), σ(E l ) are determined by calibration using reference substances with standard parameters, by radiation intensity ratio in direct and attenuated beams at different energies. A basis of two materials A and B (with standard parameters ρ A , t A , Z A and unknown parameters ρ B , t B , Z B meeting the condition Z A ≠ Z B , respectively) is a necessary con dition of calibration. Values Z A and Z B should fall

The energy spectrum of X ray radiation passed through the human body contains information about the physicochemical properties of tissues. For example, the structure of tissue can be studied from spectral data obtained in different spectral ranges [2,4,5].The goal of this work was to demonstrate that multi energy (spectrozonal) roentgenography could be imple mented in digital fluorographs.Spectrozonal fluorography can be used to analyze calcinates in diagnosis of thoracic cavity diseases. Analyses of calcinates in digital images is more difficult than in film images because the image borders in the lat ter case are significantly sharper than in the former case, like the vessel image borders in ortho projection. It was shown that organic inclusions visible in X ray photo graphs made in different spectral ranges could be differ entiated by the effective atomic number Z eff and electron ic density of tissue. However, monochromatic or quasi monochromatic X ray radiation should be used in quan titative analysis [1,4,5]. A special detection method should be developed to use the polychromatic radiation of the common X ray tube.Water and calcium oxalate can be regarded as mod els of blood vessels and calcinates, respectively: Z eff = 7.9 and Z eff = 18.4. Significant difference between effective atomic numbers makes it possible to distinguish between blood vessels and calcinates even against the background of substantial error in Z eff .Conventional methods of spectrozonal X ray detec tion are based on the use of X raying at two peak values of anode voltage (kV) or spectrozonal X ray detectors.The practical implementation of the method is based on characteristic angles:(1)where I ob , I f are intensities of X ray radiation at the X ray photograph plane at inclusion and background sites, respectively, at high (h), medium (m), and low (l) anode voltage.Local radiation intensity at image sites and charac teristic angles calculated from Eq. (1) allow the resulting image to be synthesized in terms of θ or ϕ and tissues to be differentiated regardless of their structure and density. Calibration of the system using samples with standard Z eff allows the method of subtractive synthesis of spectrozon al images, which improves the signal/noise ratio [3], to be supplemented with physical characteristics of inclu sion.Experiments were performed using a scanning digital X ray apparatus with a semiconductor linear detector.The 23 × 14.5 cm X ray phantom contained filter 19 mm Al and a container with samples filled with polyurethane foam. The main components of the samples are listed in Table 1.The phantom was in the detector plane, and the fil ter was oriented toward the detector. A series of three photographs was taken at anode voltage U l = 55 kV, U m = 85 kV, and U h = 125 kV.Characteristic angles were calculated as follows.

No abstract