Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Arfelli, F.
Right arrow Articles by Zanconati, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arfelli, F.
Right arrow Articles by Zanconati, F.
(Radiology. 2000;215:286-293.)
© RSNA, 2000

Technical Developments

Mammography with Synchrotron Radiation: Phase-Detection Techniques1

Fulvia Arfelli, PhD, Valter Bonvicini, PhD, Alberto Bravin, PhD, Giovanni Cantatore, PhD, Edoardo Castelli, PhD, Ludovico Dalla Palma, MD, Marco Di Michiel, PhD 2, Mauro Fabrizioli, MS, Renata Longo, PhD, Ralf Hendrik Menk, PhD, Alessandro Olivo, PhD, Silvia Pani, MS, Diego Pontoni, MS, Paolo Poropat, PhD, Michela Prest, PhD, Alexander Rashevsky, PhD, Marina Ratti, MD, Luigi Rigon, MS, Giuliana Tromba, PhD, Andrea Vacchi, PhD, Erik Vallazza, PhD and Fabrizio Zanconati, MD

1 From the Depart of Physics (F.A., V.B., A.B., G.C., E.C., M.D.M., R.L., A.O., S.P., P.P., M.P., A.R.) and Institutes of Radiology (L.D.P., M.R.) and Pathologic Anatomy (F.Z.), Università di Trieste, Italy; National Institute of Nuclear Physics, Area di Ricerca, Padriciano 99, 34012 Trieste, Italy (F.A., A.B., G.C., E.C., R.L., A.O., S.P., D.P., P.P., M.P., A.V., E.V.); and Sincrotrone Trieste Società Consortile per Azioni, Basovizza, Italy (M.F., R.H.M., D.P., L.R., G.T.). From 1998 RSNA scientific assembly. Received Oct 12, 1998; revision requested Dec 23; final revision received Jul 22, 1999; accepted Aug 2. R.H.M. supported in part by European Community contract ERBFMBICT961694. Address reprint requests to A.O. (e-mail: olivo@trieste.infn.it).


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The authors evaluated the effect on mammographic examinations of the use of synchrotron radiation to detect phase-perturbation effects, which are higher than absorption effects for soft tissue in the energy range of 15–25 keV. Detection of phase-perturbation effects was possible because of the high degree of coherence of synchrotron radiation sources. Synchrotron radiation images were obtained of a mammographic phantom and in vitro breast tissue specimens and compared with conventional mammographic studies. On the basis of grades assigned by three reviewers, image quality of the former was considerably higher, and the delivered dose was fully compatible.

Index terms: Breast radiography, technology, 00.119 • Phantoms • Test objects • Synchrotron


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Conventional radiologic studies are based on only absorption effects, and therefore image contrast is entirely due to differences in the absorption properties of details inside an object (1). This leads to several problems when soft biologic tissues (such as breast tissue) are imaged. Small x-ray absorption differences result in poor image contrast (2). The refractive index (n) of an object is given with the following equation (3):

where {delta} is the phase-shift term, i is the imaginary unit, and ß is the absorption term. For biologic soft tissues in the energy range of 15–25 keV, the absorption term is considerably smaller than the phase-shift term, and therefore the effects due to phase shift are considerably more relevant than are those due to absorption effects (4). Recently, new imaging modalities with the capability of investigating phase-shift effects have been developed, namely, phase-contrast and diffraction imaging (5–9).

In phase-contrast imaging, the phase shift in the x-ray wave field when it crosses a detail is of interest. Beyond the detail, the waves refracted (phase shifted) by the detail itself strongly interfere with the unrefracted waves. This interference effect takes place along the border of the detail inside a narrow angular region (about 10 µradians), and it results in strong interference patterns inside this region that could be detected (9) (Fig 1).



View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Schematic depicts the basic principle of phase-contrast imaging: interference pattern formation due to interference between diffracted and nondiffracted waves. {alpha} = angle within which the interference effect takes place, x and y refer to the axes of imaging, z is the propagation direction of x rays.

 
Since the interference angle is very small, it is impossible to detect the interference patterns immediately behind the sample, and therefore the sample-to-detector distance must be optimized to match the spatial resolution of the detector device (10). This leads to strong interference patterns in the intensity detected (ie, sharp black and white lines with use of film or screen-film systems) along the edges of the details in the imaged sample. Therefore, visibility of the details will be highly enhanced. In particular, the thin and small details that are usually invisible on absorption images will become detectable as a result of this edge-enhancement effect.

With diffraction imaging, two silicon (111) crystals are used. In addition to the monochromator crystal, which is always present between the source and the sample, a second (analyzer) crystal is placed between the sample and the detector. The reflectivity of the analyzer crystal is described with a narrow bell-shaped function of the incident angle at a fixed energy (11), and the reflectivity of the monochromator crystal is characterized by approximately the same rocking curve. Thus the analyzer crystal, when aligned with the monochromator crystal (ie, the [111] planes of the crystals are parallel), acts to reject scattering. Most of the photons scattered by the sample at angles larger than half the full width at half maximum (FWHM) of its rocking curve are rejected by the analyzer crystal. This allows removal of image blurring due to scattered radiation. Photons that cross the sample along the borders of the details are scattered at relevant angles as a result of the sharp gradient of the refractive index at the interface between materials. The diffraction angle, to the first approximation, is proportional to the gradient of the phase shift term (6). Therefore, these photons are almost completely rejected by the analyzer crystal and sharp white lines appear along the borders of all details in the image, which enhances their visibility.

When a slight misalignment is adopted between the monochromator and analyzer crystals, the reflectivity of the latter is maximized for a scattering angle equal to the misalignment angle. Owing to the relationship between the scattering angle and the phase-shift term of the refractive index, the analyzer crystal allows the conversion of the behavior of the phase shift term inside the imaged sample into different reflectivity coefficients for the refracted wave, and therefore into intensity differences on the detector.

To detect phase effects, a high degree of spatial coherence must be provided by the x-ray source. With a conventional x-ray tube, this condition can be achieved only by greatly increasing the source-to-sample distance. This would lead to unacceptably long radiologic examinations of about 2 hours (6) as a result of the low flux emitted by a conventional tube. In contrast, a synchrotron radiation source is characterized by very high photon flux over a wide range of energies. Owing to this high flux, an image of a full breast sample can be acquired in a few seconds. Moreover, the source size is on the order of a few hundred micrometers, and the emitted radiation is highly collimated; therefore, a very high degree of spatial coherence is provided (12).

The purpose of this study was to evaluate the effect on mammographic examinations of the use of synchrotron radiation to detect phase-perturbation effects.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Syrmep Beam Line
Our experiments were performed at the Syrmep beam line (13) at the synchrotron radiation facility in Trieste, Italy. The radiation source results from a bending magnet of the storage ring. The cross-sectional dimension of the electron bunches circulating in the storage ring is approximately 1,100 x 140 µm2 (FWHM). The asymmetry of the source results in a higher degree of spatial coherence, and therefore higher image quality, in one direction compared with those in the other. A monolithic channel-cut Si (111) crystal is used to narrow the energy bandwidth of the incoming white beam. A monochromatic beam, with energy tuneable within the range of 10–35 keV with an energy resolution of about 0.2%, is thus available in the experimental area.

The experimental area, located approximately 22 m from the source, is equipped with a micrometric tungsten slit system, which determines the beam cross section (the typical value used for this experiment was 1 x 100 mm2). An ionization chamber readout with an amperometer is used to monitor the radiation flux in the experimental area and to evaluate the entrance dose and the mean glandular dose delivered to the samples.

All samples were first imaged at the Trieste Hospital with a conventional mammographic unit (Senographe 500T Senix HF; GE Medical Systems, Milwaukee, Wis) in combination with a mammographic screen-film system (Trimax T2M; Imation Enterprises, St Paul, Minn) and an antiscattering grid. The x-ray tube of the unit has a molybdenum anode with 30-µm molybdenum filtration, the focal spot diameter is 300 µm, and the source-to-detector distance is 65 cm. The samples are placed on the support that lies on the screen-film cassette and antiscattering grid combination, that is, where the breast of the patient is usually placed. On the basis of the x-ray tube parameters and sample thickness, the mean glandular dose was evaluated according to the method of Dance (14).

The same samples were then imaged with synchrotron radiation. The experimental set-up is shown in Figure 2.



View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Schematic depicts the conceptual beam-line layout. In the experimental set-up for absorption imaging, the sample-to-detector distance is equal to zero, and for phase-contrast imaging, the sample-to-detector distance is about 2 m. ELETTRA = the synchrotron radiation (SR) facility in Trieste, Italy.

 
The same screen-film systems were used as detector devices for both methods. Owing to the small vertical size of the beam (1 mm), two-dimensional images were obtained by scanning the sample and the film simultaneously through the beam by means of vertical movement stages driven by 1-µm-resolution stepping motors (Microcontrole model UE63PP; Newport, Irvine, Calif) controlled by a computer (Macintosh II FX; Apple Computers, Cupertino, Calif). The screen-film system speed was chosen to achieve the correct exposure on the film. On the basis of the ionization chamber current and the beam energy, beam size, scanning speed, and sample thickness, the mean glandular dose was calculated by assuming a 50% glandular and 50% adipose tissue breast model (15). This model included an outer 0.5-cm-thick adipose layer. Further details on the dose evaluation algorithm can be found in reference 16.

We first produced an absorption image to compare with the conventional screen-film mammogram by placing the sample in contact with the screen-film system on the vertical movement stage.

Phase-Contrast Imaging
Phase-contrast image quality was optimized by selecting the sample-to-detector distance that maximized the interference signal. The sample and detector were placed on two different vertical movement stages, each driven with an independent stepping motor, and simultaneously scanned through the beam with different speeds to obtain a two-dimensional image. The detector, or film, scanning speed was determined on the basis of exposure requirements, and the sample scanning speed was chosen to preserve the angular velocity with respect to the source, thus avoiding artifacts. The phase-contrast signal depends, among others things, on the structure of the imaged sample, on the radiation energy, and on the sample-to-detector distance. By means of simulations based on Fresnel-Kirchoff integrals, it was possible to maximize the signal as a function of the sample-to-detector distance for a certain sample and a certain energy (10). Images were acquired at 17 KeV to retain all the absorption signal together with the phase-contrast signal. By accepting a loss in the absorption signal, however, it would be possible to also preserve the phase-contrast signal at higher radiation energies and thus reduce the delivered dose. This is due to the different energy dependence of absorption effects and phase-shift effects. Since phase-shift effects decrease more slowly than do absorption effects for increasing energies, the former are detectable over a wider energy range (17).

Diffraction Imaging
In diffraction imaging, a second Si (111) crystal (analyzer) was placed between the sample and the detector (Fig 3). This crystal was mounted on a holder fixed to a rotational cradle (Huber, Rimsting, Germany) moved by a motor (Berger Lahr, Lahr, Germany) with 5 x 10-5-degree angular resolution, for fine tuning of the beam incidence angle on the crystal surface (Bragg angle, {theta} in Fig 3). This stage was then mounted on a second rotational cradle moved by another motor, which allowed alignment of the crystal lattice planes with the beam along the x direction ({phi} in Fig 3). The whole system could be moved vertically to align the incoming beam with the center of the crystal surface.



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Schematic depicts the experimental set-up for diffraction imaging measurements. z is the propagation direction of x rays. Beam size is equal to 10 cm along the x direction and 1 mm along the y direction, {phi} is the rotation angle around z, and {theta} is the rotation angle around x. The ionization chambers are used to monitor the flux and are then removed from the beam during imaging.

 
When the analyzer crystal was aligned with the monochromator crystal and no sample was placed between the two crystals, nearly all the photons were reflected by the second crystal, and there were almost no flux losses. By slightly tilting the analyzer crystal, its reflectivity became smaller, and a part of the photons was not reflected anymore. At a fixed energy, the function that describes the reflectivity of the crystal as a function of the incidence angle is called the rocking curve of the crystal (11), and it is shown in Figure 4. This curve for the analyzer crystal was measured at each energy by scanning the Bragg angle of the incident photons while two ionization chambers (Fig 3) were used to monitor the fluxes of the incident and the reflected beams. The second ionization chamber was also used to determine the film exposure.



View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Graph depicts the rocking curve of the Si (111) analyzer crystal at 17 keV. The dots represent the experimental data, and the solid line is the Gaussian fit. The SEMs of the data are about 2%. A.U. = arbitrary units.

 
The width of the rocking curve was approximately 20 µradians at 17 keV. The top of the curve corresponded to the perfect alignment angle between the analyzer and monochromator crystals. By selecting a misalignment angle between the two crystals, the reflectivity of the analyzer crystal was maximized for photons deflected at the same angle. Therefore, when a sample was placed between the two crystals, photons diffracted by the sample at that angle were totally reflected. This meant that by choosing the misalignment angle between the two crystals, it was possible to convert the photon diffraction angles into different reflectivity values for the analyzer crystal and, therefore, into intensity differences on the detector. Since the analyzer crystal consisted of a single crystal (Fig 3), images were spatially inverted and, therefore, were obtained by scanning the sample and the detector in opposite directions.

Test Object and Breast Tissue
The test object used in this experiment was an Ackermann mammographic phantom (RMI 160; Gammex, Middleton, Wis), which includes a box (Mammochip) with 9 x 11-cm2 cross section that contains various details (eg, microcalcifications, fibers, simulated tissue). Two uniform slabs of Lucite (polymerized methacrylate), 1 and 3 cm thick, respectively, are added to the box. In this way, the overall thickness of the phantom is equivalent to a Lucite thickness of 50 mm (18), which corresponds to the thickness of a standard breast of about 54 mm (15).

Five fixed and 10 unfixed fresh breast tissues were studied; all of these specimens were obtained at postmortem excision. To contain the specimens during imaging, the tissue samples were compressed into a Lucite holder that comprised two 1-mm-thick acrylic windows, one in front and one behind the sample.

All images were scored by two physicists (F.A., A.O.) and one radiologist (L.D.P.) on a scale of 1, undetectable, to 5, very sharp. The scores of the three readers for each image were rounded, and the mean data are presented in the Table as are the values of the mean glandular dose.


View this table:
[in this window]
[in a new window]

  		Visibility of Details Embedded in Ackermann Phantom on Various Images Obtained with Synchrotron Radiation Imaging
 

   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Ackermann Mammographic Phantom
In the first part of this study, the Ackermann phantom was imaged with both conventional mammographic equipment and synchrotron radiation. Synchrotron beam energy was equal to 17 kV to preserve all the absorption effects; in this way, they are added to the improvements due to phase shift effects. Figure 5a is a schematic of the phantom. Figure 5b shows the image quality provided until now with conventional mammography. Figure 5c shows the absorption image obtained with synchrotron radiation. The screen-film system was placed in contact with the phantom, and no phase-shift effects were detectable. The quality of the absorption image obtained with synchrotron radiation was higher than that of the conventional image.



View larger version (46K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. (a-c) Schematic and images of the Ackermann mammographic phantom. (a) Column 1, Al2O3 specks with diameters from 0.083 (1A) to 0.32 (1E) mm. Column 2, plastic wires with diameters from 0.3 (2A) to 0.8 (2E) mm. Column 3, carbohydrate spheres with diameters from 0.35 (3A) to 1.2 (3E) mm. Column 4, plastic disks that contain 0.6-mm-diameter holes that form single digits to indicate the depth of the holes in 10ths of a millimeter: {alpha}, simulated tissue samples; ß, simulated lymph node with embedded steel needle and copper wire; {gamma}, measuring point for film density. (b) Radiograph obtained with conventional mammographic equipment (27 kVp, 109 mAs). (c) Absorption image obtained with synchrotron radiation. The beam energy is 17 KeV. (d-f) Images of the Ackermann mammographic phantom obtained with synchrotron radiation at 17 KeV and phase detection techniques. (d) Phase-contrast image. (e) Diffraction image with the analyzer and monochromator crystals perfectly aligned. (f) Diffraction image obtained when the misalignment angle between the monochromator and analyzer crystals was nearly equal to half the FWHM of the rocking curve of the latter. Arrows 1 and 2 indicate findings in d-f that are almost invisible in b and c: 1, an opaque ring in the middle of a simulated tissue sample; 2, internal structure in a second sample.

 


View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. (a-c) Schematic and images of the Ackermann mammographic phantom. (a) Column 1, Al2O3 specks with diameters from 0.083 (1A) to 0.32 (1E) mm. Column 2, plastic wires with diameters from 0.3 (2A) to 0.8 (2E) mm. Column 3, carbohydrate spheres with diameters from 0.35 (3A) to 1.2 (3E) mm. Column 4, plastic disks that contain 0.6-mm-diameter holes that form single digits to indicate the depth of the holes in 10ths of a millimeter: {alpha}, simulated tissue samples; ß, simulated lymph node with embedded steel needle and copper wire; {gamma}, measuring point for film density. (b) Radiograph obtained with conventional mammographic equipment (27 kVp, 109 mAs). (c) Absorption image obtained with synchrotron radiation. The beam energy is 17 KeV. (d-f) Images of the Ackermann mammographic phantom obtained with synchrotron radiation at 17 KeV and phase detection techniques. (d) Phase-contrast image. (e) Diffraction image with the analyzer and monochromator crystals perfectly aligned. (f) Diffraction image obtained when the misalignment angle between the monochromator and analyzer crystals was nearly equal to half the FWHM of the rocking curve of the latter. Arrows 1 and 2 indicate findings in d-f that are almost invisible in b and c: 1, an opaque ring in the middle of a simulated tissue sample; 2, internal structure in a second sample.

 


View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c. (a-c) Schematic and images of the Ackermann mammographic phantom. (a) Column 1, Al2O3 specks with diameters from 0.083 (1A) to 0.32 (1E) mm. Column 2, plastic wires with diameters from 0.3 (2A) to 0.8 (2E) mm. Column 3, carbohydrate spheres with diameters from 0.35 (3A) to 1.2 (3E) mm. Column 4, plastic disks that contain 0.6-mm-diameter holes that form single digits to indicate the depth of the holes in 10ths of a millimeter: {alpha}, simulated tissue samples; ß, simulated lymph node with embedded steel needle and copper wire; {gamma}, measuring point for film density. (b) Radiograph obtained with conventional mammographic equipment (27 kVp, 109 mAs). (c) Absorption image obtained with synchrotron radiation. The beam energy is 17 KeV. (d-f) Images of the Ackermann mammographic phantom obtained with synchrotron radiation at 17 KeV and phase detection techniques. (d) Phase-contrast image. (e) Diffraction image with the analyzer and monochromator crystals perfectly aligned. (f) Diffraction image obtained when the misalignment angle between the monochromator and analyzer crystals was nearly equal to half the FWHM of the rocking curve of the latter. Arrows 1 and 2 indicate findings in d-f that are almost invisible in b and c: 1, an opaque ring in the middle of a simulated tissue sample; 2, internal structure in a second sample.

 


View larger version (171K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d. (a-c) Schematic and images of the Ackermann mammographic phantom. (a) Column 1, Al2O3 specks with diameters from 0.083 (1A) to 0.32 (1E) mm. Column 2, plastic wires with diameters from 0.3 (2A) to 0.8 (2E) mm. Column 3, carbohydrate spheres with diameters from 0.35 (3A) to 1.2 (3E) mm. Column 4, plastic disks that contain 0.6-mm-diameter holes that form single digits to indicate the depth of the holes in 10ths of a millimeter: {alpha}, simulated tissue samples; ß, simulated lymph node with embedded steel needle and copper wire; {gamma}, measuring point for film density. (b) Radiograph obtained with conventional mammographic equipment (27 kVp, 109 mAs). (c) Absorption image obtained with synchrotron radiation. The beam energy is 17 KeV. (d-f) Images of the Ackermann mammographic phantom obtained with synchrotron radiation at 17 KeV and phase detection techniques. (d) Phase-contrast image. (e) Diffraction image with the analyzer and monochromator crystals perfectly aligned. (f) Diffraction image obtained when the misalignment angle between the monochromator and analyzer crystals was nearly equal to half the FWHM of the rocking curve of the latter. Arrows 1 and 2 indicate findings in d-f that are almost invisible in b and c: 1, an opaque ring in the middle of a simulated tissue sample; 2, internal structure in a second sample.

 


View larger version (182K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5e. (a-c) Schematic and images of the Ackermann mammographic phantom. (a) Column 1, Al2O3 specks with diameters from 0.083 (1A) to 0.32 (1E) mm. Column 2, plastic wires with diameters from 0.3 (2A) to 0.8 (2E) mm. Column 3, carbohydrate spheres with diameters from 0.35 (3A) to 1.2 (3E) mm. Column 4, plastic disks that contain 0.6-mm-diameter holes that form single digits to indicate the depth of the holes in 10ths of a millimeter: {alpha}, simulated tissue samples; ß, simulated lymph node with embedded steel needle and copper wire; {gamma}, measuring point for film density. (b) Radiograph obtained with conventional mammographic equipment (27 kVp, 109 mAs). (c) Absorption image obtained with synchrotron radiation. The beam energy is 17 KeV. (d-f) Images of the Ackermann mammographic phantom obtained with synchrotron radiation at 17 KeV and phase detection techniques. (d) Phase-contrast image. (e) Diffraction image with the analyzer and monochromator crystals perfectly aligned. (f) Diffraction image obtained when the misalignment angle between the monochromator and analyzer crystals was nearly equal to half the FWHM of the rocking curve of the latter. Arrows 1 and 2 indicate findings in d-f that are almost invisible in b and c: 1, an opaque ring in the middle of a simulated tissue sample; 2, internal structure in a second sample.

 


View larger version (182K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5f. (a-c) Schematic and images of the Ackermann mammographic phantom. (a) Column 1, Al2O3 specks with diameters from 0.083 (1A) to 0.32 (1E) mm. Column 2, plastic wires with diameters from 0.3 (2A) to 0.8 (2E) mm. Column 3, carbohydrate spheres with diameters from 0.35 (3A) to 1.2 (3E) mm. Column 4, plastic disks that contain 0.6-mm-diameter holes that form single digits to indicate the depth of the holes in 10ths of a millimeter: {alpha}, simulated tissue samples; ß, simulated lymph node with embedded steel needle and copper wire; {gamma}, measuring point for film density. (b) Radiograph obtained with conventional mammographic equipment (27 kVp, 109 mAs). (c) Absorption image obtained with synchrotron radiation. The beam energy is 17 KeV. (d-f) Images of the Ackermann mammographic phantom obtained with synchrotron radiation at 17 KeV and phase detection techniques. (d) Phase-contrast image. (e) Diffraction image with the analyzer and monochromator crystals perfectly aligned. (f) Diffraction image obtained when the misalignment angle between the monochromator and analyzer crystals was nearly equal to half the FWHM of the rocking curve of the latter. Arrows 1 and 2 indicate findings in d-f that are almost invisible in b and c: 1, an opaque ring in the middle of a simulated tissue sample; 2, internal structure in a second sample.

 
Figure 5d shows the phase-contrast image obtained with synchrotron radiation with a distance of 2.3 m between the phantom and the screen-film system. Contrast was enhanced and the details were depicted more sharply than those on the conventional or absorption images. Figure 5e and 5f was obtained with the analyzer crystal placed between the phantom and the screen-film system (diffraction imaging). Figure 5e was obtained with the analyzer crystal perfectly aligned with the monochromator crystal. The contrast was increased, and the edges of the details were well defined. Figure 5f was obtained with a misalignment angle between the monochromator and analyzer crystals of nearly half the FWHM of the rocking curve of the latter (10 µradians). The contrast was further enhanced, and sharp black and white lines improved visibility of the details. Imaging results for Figure 5b–5f are summarized in the Table.

Breast Tissue Samples
To test the effectiveness of phase-detection techniques for biologic tissue, 15 fresh full breast tissue specimens were imaged. Images of two of these specimens are presented in this section. The first tissue sample was 3.3 cm thick and was unfixed (Fig 6), and the second sample was 3.0 cm thick and was fixed in formalin (Fig 7). Images of both samples showed partial glandular involution and calcifications of different sizes and natures.



View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a. Unfixed full breast tissue sample (3.3 cm thick). (a) Radiograph produced with conventional mammographic equipment (26 kVp and 32 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, upper arrows indicate thin linear opacities and lower arrows indicate calcifications with enhanced visibility in c and d compared with that in a and b.

 


View larger version (75K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b. Unfixed full breast tissue sample (3.3 cm thick). (a) Radiograph produced with conventional mammographic equipment (26 kVp and 32 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, upper arrows indicate thin linear opacities and lower arrows indicate calcifications with enhanced visibility in c and d compared with that in a and b.

 


View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6c. Unfixed full breast tissue sample (3.3 cm thick). (a) Radiograph produced with conventional mammographic equipment (26 kVp and 32 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, upper arrows indicate thin linear opacities and lower arrows indicate calcifications with enhanced visibility in c and d compared with that in a and b.

 


View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6d. Unfixed full breast tissue sample (3.3 cm thick). (a) Radiograph produced with conventional mammographic equipment (26 kVp and 32 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, upper arrows indicate thin linear opacities and lower arrows indicate calcifications with enhanced visibility in c and d compared with that in a and b.

 


View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a. Full breast tissue sample (3.0 cm thick) fixed in formalin. (a) Radiograph produced with conventional mammographic equipment (27 kVp and 28 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, the arrow indicates an opacity in which contrast is highly enhanced in c and d compared with that in a and b.

 


View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b. Full breast tissue sample (3.0 cm thick) fixed in formalin. (a) Radiograph produced with conventional mammographic equipment (27 kVp and 28 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, the arrow indicates an opacity in which contrast is highly enhanced in c and d compared with that in a and b.

 


View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7c. Full breast tissue sample (3.0 cm thick) fixed in formalin. (a) Radiograph produced with conventional mammographic equipment (27 kVp and 28 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, the arrow indicates an opacity in which contrast is highly enhanced in c and d compared with that in a and b.

 


View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7d. Full breast tissue sample (3.0 cm thick) fixed in formalin. (a) Radiograph produced with conventional mammographic equipment (27 kVp and 28 mAs). (b-d) Images obtained with synchrotron radiation at 17 KeV. (b) Absorption image. (c) Phase-contrast image. (d) Diffraction image obtained with the monochromator and analyzer crystals slightly misaligned. In a-d, the arrow indicates an opacity in which contrast is highly enhanced in c and d compared with that in a and b.

 
Figures 6a and 7a show the images obtained with the conventional mammographic equipment. Image quality was poorer than that in Figures 6b–6d and 7b–7d. Figures 6b and 7b show the absorption images obtained with synchrotron radiation of the same specimens obtained with beam energy of 17 keV and the screen-film system placed in contact with the sample. Compared with the conventional images, the synchrotron absorption images showed improved image contrast and spatial resolution. Figures 6c and 7c show the phase-contrast images obtained with synchrotron radiation with beam energy of 17 keV and sample to screen-film system distance of 2.3 m. Compared with the conventional and absorption images, the phase-contrast images showed improved image contrast and spatial resolution, with better definition of the glandular component and improved visibility of all calcium deposits. Compared with the conventional image, on which the edges were blurred, the phase-contrast images depict the edges more clearly.

Figures 6d and 7d show the diffraction images obtained with synchrotron radiation of the same samples. To detect phase-shift effects while delivering the minimum mean glandular dose in both cases, the misalignment angle between the monochromator and analyzer crystals was very small (a few microradians). Compared with the phase-contrast images, the visibility of details and overall level of image quality are comparable and the contrast is further enhanced as a result of the phase shift term gradient. The arrows in Figures 6 and 7 indicate relevant details for comparison.

Air bubbles seen in both samples were not clearly recognizable on the conventional mammographic and synchrotron absorption images, whereas the edge-enhancement effect clarified their presence on the phase-contrast and diffraction images.


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The improvement in image quality on synchrotron absorption images compared with conventional mammograms was investigated by Burattini et al (19). On images of the test object and breast tissue samples obtained in our study, further improvements were achieved with phase-shift detection techniques. The dose delivered to obtain a synchrotron absorption image was reduced by about 20% compared with that required for conventional mammography as a result of the monochromaticity of the synchrotron x-ray beam.

Synchrotron phase-contrast imaging further improved image quality with only slightly increased dose compared with those for synchrotron absorption imaging and with reduced dose compared with that for conventional mammography (Fig 5d).

At diffraction imaging of the phantom, with the analyzer crystal parallel to the monochromator crystal, nearly all the scattered radiation was removed, and the contrast of the details was consequently increased. Furthermore, the amount of scattered x rays was extremely high along the edges of the details within the sample. Since the analyzer crystal rejected most of the scattered radiation, an edge-enhancement effect was obtained on the images. The only side effect was a sort of granularity in the background of the images owing to the granularity of the inner structure of the phantom. This means that diffraction images depict the structure in which the details are embedded correctly, and therefore this granularity should not be interpreted as image noise. This kind of effect was not seen on images of biologic tissues (Fig 6). In addition, shadowing was caused by the increasing phase shift term gradient along the edges of the details, which resulted in a discernible three-dimensional effect.

On the phantom images obtained with a misalignment angle between the monochromator and analyzer crystals equal to nearly half the FWHM of the rocking curve of the latter, the reflectivity of the latter was maximized for photons diffracted with the sample at an angle of about 10 µradians. Owing to the steep slope of the rocking curve around its FWHM, small changes in the diffraction angle resulted in high reflectivity differences and, therefore, in high radiation-intensity differences on the screen-film system. Since the diffraction angle was approximately proportional to the gradient of the phase shift term, this choice for the misalignment angle maximized the sensitivity of the technique to sudden changes in the phase shift term inside the imaged object. Because the gradient of the phase shift term was very large at the edges of the details, one of the most evident effects of this technique was the formation of sharp black and white lines along the edges of all details embedded in the imaged object. Details invisible with conventional techniques because of low absorption differences with respect to the background became detectable with this technique. Moreover, gray-scale levels were modulated by the phase shift term value inside the details, which resulted in a further enhancement of the contrast. The visual three-dimensional effect was amplified with the misalignment between the two crystals. Compared with parallel alignment, use of misalignment required an increase in the dose of about a factor of two because nearly half the primary beam was rejected by the analyzer crystal.

Diffraction imaging of the phantom was characterized by a dose increase compared with that for phase-contrast or absorption imaging. This was probably due to the inner granular structure of the box embedded in the phantom, which scattered high amounts of radiation that was rejected by the analyzer crystal.

To evaluate this effect at 17 keV, two separate measurements of scattering were performed with the box in the phantom and with a uniform slab of Lucite of the same thickness. The box scattered twice the amount of radiation at angles larger than 10 µradians with respect to the Lucite slab. When biologic tissue was imaged, however, the dose delivered with diffraction imaging was comparable to that delivered with absorption and phase-contrast imaging.

As discussed by Burattini et al (19), improvements on synchrotron absorption images compared with conventional images of biologic tissue are mostly due to the high collimation of the synchrotron beam. The delivered mean glandular dose was reduced from 0.8 to 0.5 mGy because of the monochromaticity of the beam.

Phase-contrast images of both samples had increased contrast compared with that on the conventional or synchrotron absorption images, and the delivered dose was 0.6 mGy for both images. Thus, a 25% dose reduction was provided compared with that at conventional mammography, even if the same detector was used, as a result of the monoenergetic nature of the synchrotron beam.

Further enhancement of the contrast due to the phase shift term was obtained with diffraction imaging. The delivered mean glandular dose was 0.7 mGy, which is fully compatible with that for conventional mammography. Even if the dose delivered at diffraction imaging of the phantom was relatively high at soft-tissue imaging, since there was no scattering increase due to granular structures, the nearly perfect reflectivity of the analyzer crystal when aligned with the monochromator crystal (ie, at the top of the rocking curve) allowed a sample exposure almost identical to that required with phase-contrast imaging. Diffraction imaging can therefore be performed within dose limitations required for mammography.

The only limitation for phase detection techniques lies in the stringent requirements for the radiation source, namely coherence and high emitted photon flux. Up to now, these features are provided at only synchrotron radiation facilities, which prevents widespread clinical application of phase detection techniques. Efforts are now directed toward the development of compact sources (20,21) that could provide high-quality radiation beams while solving the basic problems of applicability of synchrotron radiation (eg, high costs, large dimensions of the facilities, dedicated technology, maintenance). These sources, however, are not yet commercially available.

Therefore, this study should be considered a preclinical feasibility study that demonstrates the potential of phase detection techniques in the imaging of phantoms and excised tissue samples. Thus, the next step in research should be to validate the method for in vivo measurements. This will necessitate development of a dedicated clinical beam line, as was done by other research institutes for intravenous coronary angiography (22,23). Also in our case, the examination procedure will necessitate the use of a dedicated stage for high-precision vertical scanning of the patient through the stationary beam. The very high flux provided by a synchrotron radiation source allows an examination time that is fully compatible with clinical requirements.

Compared with conventional mammography, phase-contrast and diffraction imaging result in strong enhancement of image contrast and increased visibility of thin and small details. Because of the improved image contrast and detail visibility, these phase-detection techniques may be applied effectively in the field of mammography for both solving ambiguous cases and allowing earlier diagnosis of malignant lesions.


   Acknowledgments
 
We thank Luigi Di Bonito, MD, from the Institute of Pathologic Anatomy for his help and Stefano Reia from the National Institute of Nuclear Physics for sample holder design and manufacturing.


   Footnotes
 
2 Current address: Medical Imaging Group, Grenoble, France. Back

Abbreviation: FWHM = full width at half maximum

Author contributions: Guarantors of integrity of entire study, E.C., L.D.P.; study concepts, F.A., A.B., M.D.M., R.H.M., A.O.; study design, F.A., A.B., R.H.M., A.O.; definition of intellectual content, F.A., A.B., E.C., L.D.P., R.H.M., A.O.; literature research, F.A., A.B., R.H.M., A.O.; experimental studies, all authors; data acquisition, F.A., A.B., R.H.M., A.O., S.P.; data analysis, all authors; manuscript preparation, F.A., E.C., L.D.P., R.H.M., A.O.; manuscript editing, F.A., A.O., R.H.M.; manuscript review, F.A., A.B., G.C., E.C., L.D.P., R.L., R.H.M., A.O.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Dance DR. Diagnostic radiology with x-rays. In: Webb S, eds. The physics of medical imaging. Bristol, England: Institute of Physics Publishing, 1988; 20-26.
  2. Johns HE, Cunningam JR. The physics of radiology Springfield, Ill: Thomas, 1971; 645-646.
  3. Takeda T, Momose A, Itai Y, Wu J, Hirano K. Phase contrast imaging with synchrotron x-rays for detecting cancer lesions. Acad Radiol 1995; 2:799-803.[Medline]
  4. Raven C, Snigirev A, Snigireva I, et al. Phase-contrast microtomography with coherent high-energy synchrotron x-rays. Appl Phys Lett 1996; 69:1826-1828.
  5. Ingal VN, Beliaevskaya EA. X-ray plane wave topography observation of the phase contrast from a non crystalline object. J Phys Appl Phys 1995; 28:2314-2317.
  6. Wilkins SW, Gureyev TE, Gao D, Pogany A, Stevenson AW. Phase-contrast imaging using polychromatic hard x-rays. Nature 1996; 384:335-338.
  7. Chapman D, Thomlinson W, Johnston RE, et al. Diffraction enhanced x-ray imaging. Phys Med Biol 1997; 42:2015-2025.[Medline]
  8. Di Michiel M, Olivo A, Tromba G, et al. Phase contrast imaging in the mammography field In: Medical applications of synchrotron radiation. Tokyo, Japan: Springer-Verlag, 1998; 78-82.
  9. Snigirev A, Snigireva I, Kohn V, Kuznetsov S, Schelokov I. On the possibilities of x-ray phase contrast microimaging by coherent high-energy synchrotron radiation. Rev Sci Instrum 1995; 66:5486-5492.
  10. Arfelli F, Bonvicini V, Bravin A, et al. Low dose phase contrast x-ray medical imaging. Phys Med Biol 1998; 43:2845-2852.[Medline]
  11. Pinskers ZG. Dynamical scattering of x-rays in crystals Berlin, Germany: Springer-Verlag, 1978; 213-228.
  12. Margaritondo G. Introduction to synchrotron radiation New York, NY: Oxford University Press, 1988; 3-8.
  13. Arfelli F, Bravin A, Barbiellini G, et al. Digital mammography with synchrotron radiation. Rev Sci Instrum 1995; 66:1325-1328.
  14. Dance DR. Monte Carlo calculation of conversion factors for the estimation of mean glandular breast dose. Phys Med Biol 1990; 35:1211-1219.[Medline]
  15. National Council on Radiation Protection and Measurements. Mammography: a user's guide NCRP report no. 85. Washington, DC: Government Printing Office, 1986.
  16. Arfelli F, Bonvicini V, Bravin A, et al. Mammography of a phantom and breast tissue with synchrotron radiation and a linear-array silicon detector. Radiology 1998; 208:709-715.[Abstract/Free Full Text]
  17. Beckmann F, Bonse U, Busch F, Günnewig O. X-ray microtomography (µCT) using phase contrast for the investigation of organic matter. J Comput Assist Tomogr 1997; 21:539-553.[Medline]
  18. Law J. A new phantom for mammography. Br J Radiol 1991; 64:116-120.[Abstract]
  19. Burattini E, Cossu E, Di Maggio C, et al. Mammography with synchrotron radiation. Radiology 1996; 200:659-663.[Abstract/Free Full Text]
  20. Krol A, Ikhlef A, Kieffer JC, et al. Laser-based microfocused x-ray source for mammography: feasibility study. Med Phys 1997; 24:725-732.[Medline]
  21. Zhong Z, Chapman D, Menk R, et al. Monochromatic energy-subtraction radiography using a rotating anode source and a bent Laue monochromator. Phys Med Biol 1997; 42:1751-1762.[Medline]
  22. Dill T, Dix W, Hamm W, et al. Intravenous coronary angiography with synchrotron radiation. Eur J Phys 1998; 19:499-511.
  23. Thomlinson W. Current status and perspectives of synchrotron radiation in medicine. Phys Med 1997; 13(suppl 1):19-23.



This article has been cited by other articles:


Home page
 RadiologyHome page
J. Keyrilainen, M. Fernandez, M.-L. Karjalainen-Lindsberg, P. Virkkunen, M. Leidenius, K. von Smitten, P. Sipila, S. Fiedler, H. Suhonen, P. Suortti, et al.
Toward High-Contrast Breast CT at Low Radiation Dose
Radiology, October 1, 2008; 249(1): 321 - 327.
[Abstract] [Full Text] [PDF]

Home page
 J. Am. Podiatr. Med. Assoc.Home page
C. Muehleman, D. R. Sumner, and Z. Zhong
Refraction Effects of Diffraction-Enhanced Radiographic Imaging: A New Look at Bone
J Am Podiatr Med Assoc, September 1, 2004; 94(5): 453 - 455.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
R A Lewis, C J Hall, A P Hufton, S Evans, R H Menk, F Arfelli, L Rigon, G Tromba, D R Dance, I O Ellis, et al.
X-ray refraction effects: application to the imaging of biological tissues
Br. J. Radiol., May 1, 2003; 76(905): 301 - 308.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
F. E. Carroll
Tunable Monochromatic X Rays: A New Paradigm in Medicine
Am. J. Roentgenol., September 1, 2002; 179(3): 583 - 590.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Arfelli, F.
Right arrow Articles by Zanconati, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arfelli, F.
Right arrow Articles by Zanconati, F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH