Phase Contrast and Diffraction Enhanced Mammography at SYRMEP.

by F. Arfelli(1), V. Bonvicini(1), A. Bravin(2), G. Cantatore(1), E. Castelli(1), L. Dalla Palma(3), R. Longo(1), R. H. Menk(2), A. Olivo(1), S. Pani(1), P. Poropat(1), M. Prest(1), A. Rashevsky(1), L. Rigon(1), G. Tromba(2), A. Vacchi(1) and E. Vallazza(1)
(1)Dipartimento di Fisica - Università di Trieste and INFN - Sez. di Trieste, Italy
(2)Sincrotrone Trieste SCpA, Basovizza, Trieste, Italy
(3)Istituto di Radiologia dell'Università di Trieste, Ospedale di Cattinara, Trieste, Italy

( contact: tromba@elettra.trieste.it )

The high degree of coherence of third generation synchrotron sources such as Elettra offers the opportunity to investigate new x-ray imaging techniques that can be applied out also to medical radiology [1-4]. In conventional radiography the image contrast is based on absorption effects expressed by the amplitude properties of x-rays detected by the image receptor. Taking advantage of the beam coherence, additional information can be obtained about the phase perturbations of the incident wave: both spatial and contrast resolution can be significantly enhanced. This progress becomes particularly important for the cases of weak absorbing samples where small differences in the absorption coefficients result in poor contrast images. In the field of medical radiology this is a characteristic case of mammography where the recognition of low contrast nodules and/or small calcifications inside a soft tissue matrix is needed to perform accurate breast cancer diagnosis.
Two novel imaging techniques have been implemented at the SYRMEP beamline to produce images of test objects and of in vitro tissue specimens enhanced by the coherence of the beam: the PHase Contrast radiography (PHC) and the analyzer crystal Diffraction Enhanced Imaging (DEI). For these studies images have been taken using standard medical screen-film systems [4,5].
In the PHC radiography we are interested in the registration of the phase shifts produced on the incident wave by the sample and the details inside it. The interference pattern produced between the undiffracted and the diffracted (phase-shifted) wave occurs along the border of the objects inside a very narrow angular region. This signal can be detected if the distance between the sample and the detector matches with the detector spatial resolution. The implementation of this technique consists simply in selecting the sample-to-detector distance that maximize the interference signal and scanning the sample and the detector through the beam using two independent movement stages. The produced diffraction pattern appears superimposed to the conventional absorption image on the film.
In the DEI set-up, an analyzer Si(1,1,1) crystal, in addition to the channel-cut Si(1,1,1) crystal used as monochromator, is placed between the sample and the detector to select the angular emission of x-rays outgoing the sample. This crystal is supplied of appropriate movements to allow a fine tuning with the beam. Two ionization chambers are used to find out the alignment of the analyzer with respect to the monochromator, the latter is also used to evaluate the film exposure.
The rocking curve of the two crystals is the same. Thus, when the analyzer is aligned with the monochromator (i.e. the (1,1,1) planes of the crystals are parallel), the former acts as a scattering rejecter: most of the photons scattered by the sample at angles larger than the FWHM of the analyzer rocking curve are rejected by the crystal.
If a slight misalignment between monochromator and analyzer is introduced, the reflectivity of the analyzer is maximized for x-rays deviated of a given scattering angle (equal to the misalignment angle). In this way it is possible to convert the photon diffraction angles, and thus the phase shifts, into different reflectivity coefficients.
Since the scattering angle is - to first approximation - proportional to the gradient of the real part of the refractive index, the analyzer effectively converts the behavior of this real part of the sample into different reflectivity coefficients for the refracted wave, and consequently into intensity differences on the detector. The effectiveness of these techniques can be evaluated by comparing the images of different tissue specimens and test objects obtained at the beamline and at a clinical mammographic unit with the same screen-film system devices [4,5].
A comparison between a conventional radiography and a PHC image of a human breast tissue specimen containing a calcification and some calcium deposits can be done by observing Fig.1a and 1b.

Fig.1a - 1b: Images of a breast tissue specimen with patological calcifications obtained with the mammographic unit (a) and with the PHC technique (b).

The delivered mean glandular dose is the same (about 1.5 mGy) for the two images. In the PHC image (Fig. 1b) the visibility of the details is enhanced; furthermore, differently from the conventional image (Fig. 1a), one can realize that the calcification is surrounded by soft tissue having a varying thickness. The density of this tissue can be recognized from the opacity. At the right side of this image one can also clearly see an extended vascular calcification.
The images of another human breast tissue sample, obtained at the mammographic unit and with the DEI technique at 17 keV, are shown in Fig. 2a-2b.

Fig. 2a - 2b: Images of a breast tissue specimen obtained with the mammographic unit (a) and with the DEI technique (b).

The conventional image (Fig. 2a) has been taken with a mean glandular dose of 0.8 mGy. Compared to the previous one in the DEI image (Fig. 2b) the contrast as well as the resolution are enhanced, this results in an improved definition of the glandular component and of the visibility of calcifications. The mean glandular dose is 0.6 mGy.
Both the imaging techniques can allow significant improvements in mammography and do not require an increase in the delivered dose. Further investigations have to be carried out to complete the comprehension of the DEI technique. Certainly these novel imaging modalities will have a great impact in the near future of medical applications when applied to in-vivo investigations.

References
[1] A.Snigirev, I.Snigireva, V.Kohn, S.Kuznetsov, I.Schelokov, Rev. Sci. Instrum. 66 (12), (1995), p. 5486-5492.
[2] D. Chapman, W.Thomlinson, R.E. Johnston, D.Washburn, E.Pisano, N.Gmur, Z.Zhong, R.Menk, F.Arfelli, D.Sayers, Phys. Med. Biol. 42 (1997), p.2015-2025.
[3] M.Di Michiel, A.Olivo, G.Tromba, F.Arfelli, M.Assante, V.Bonvicini, A.Bravin, G.Cantatore, E.Castelli, L.Dalla Palma, R.Longo, S.Pani, D.Pontoni, P.Poropat, M.Prest, A.Rashevsky, A.Vacchi, E.Vallazza, Proceedings of Haga Conference 1997, Springer-Verlag ed., p.78-82.
[4] F.Arfelli, M.Assante, V.Bonvicini, A.Bravin, G.Cantatore, E.Castelli, L.Dalla Palma, M.Di Michiel, R.Longo, A.Olivo, S.Pani, D.Pontoni, P.Poropat, M.Prest, A.Rashevsky, G.Tromba, A.Vacchi, E.Vallazza, F.Zanconati, Phys. Med. Biol, Vol.43, n.10, oct. 1998, p.2845-2852.
[5] F.Arfelli, V.Bonvicini, A.Bravin, G.Cantatore, E.Castelli, L.Dalla Palma, M.Di Michiel, M.Fabrizioli, R.Longo, R.H.Menk, A.Olivo, S.Pani, D.Pontoni, P.Poropat, M.Prest, A.Rashevsky, M.Ratti, L.Rigon, G.Tromba, A.Vacchi, E.Vallazza, F.Zanconati, in press on Radiology.