| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Commentary |
In this issue we publish the first paper to appear in the Journal on a relatively new imaging technique—diffraction enhanced imaging (Lewis et al Br J Radiol 2003;76:301–308).
It is well known that differences in X-ray absorption between different soft tissues are very small, sometimes generating little or no contrast in the resulting images. This has led to a search for other properties of soft tissues that might be exploited in the imaging process.
One such property is refractive index, which results in a phase shift of the wave – notice that this technique is explained in terms of the X-ray beam acting as a wave, not as a stream of photons. Although the differences in refractive index between tissues, and hence phase changes, are extremely small, three extremely sensitive methods for analysing phase changes are available. These are: (a) interferometry, in which a beam that has passed through the object is made to interfere with a coherent beam that has not been modified; (b) Fresnel diffraction, which has been used for many years in the highly successful phase contrast microscope; (c) diffraction enhanced imaging (DEI). One potential advantage of phase-related techniques is that the interaction cross-section is much higher than for X-ray absorption so they are potentially much more sensitive.
DEI relies on the fact that strong Bragg reflection from the atomic layers in certain crystals is critically dependent on the angle of incidence of the beam. X-rays striking the crystal at slightly different angles, for example after being slightly refracted at the boundary between tissues with different refractive indices, are much less strongly reflected. In other words the angular resolution of the crystal analyser is very high.
By setting the crystal analyser at different angles, it is possible to obtain either absorption contrast images that are almost free from scatter, or refraction images. Lewis et al compare these images with conventional absorption + scatter images for excised mouse liver, lung, heart and leg. The extra image information provided by the new technique is impressive and we note that, as expected, the scatter-free absorption images and the refraction images show different features. Also predicted and confirmed is that the method is very good for highlighting tissue boundaries. Other authors have shown that phase imaging results in better structure detail than digital mammography for mammography phantoms and breast tissue samples.
We welcome this contribution to the Journal for a number of reasons. First, there is always a place for looking at what is scientifically feasible and radiologically interesting, without too much worry initially about whether the concept is technically feasible. This is the philosophy of "looking under stones just to see what is hidden below". Second, it is important to publicise novel imaging techniques. Imaging is increasingly becoming "multimodality" with a need to combine information from a variety of techniques and in different ways. By exploiting differences in refractive index between different tissues, DEI presents a completely new source of information. Finally, the collaboration that has resulted in this paper reflects the long-held "multidisciplinary" principles of the British Institute of Radiology with contributions from physical scientists, biological scientists and radiologists. It is also good to see the high energy laboratories at Daresbury and Trieste collaborating in research that may be transferrable to diagnostic medicine.
As with any new technique, there are a number of difficulties to overcome. First, the radiation source is an intense beam of highly monochromatic, well-collimated, low energy X-rays (about 17 keV). Such sources are not available in a compact form for use in hospitals at present. Even a heavily filtered beam from a conventional X-ray source contains much too great a spread of wavelengths (photon energies) for diffraction work, although previously published phase contrast enhancement of images of mammographic test objects using conventional X-ray sources has shown that the potential exploitation of phase effects is not restricted to monochromatic synchrotron radiation. Second, although the authors express the opinion that high quality images could be obtained at doses that are similar to, or even lower than those used for conventional imaging, more work needs to be done on the optimum relationship between phase shift, dose and resolution. Finally the work has been done with small samples of dissected tissues—some way removed from in vivo imaging.
Nevertheless, past experience has shown, especially in CT, MRI and PET, that when a new imaging technique is capable of providing greater insights into body anatomy and physiology, the technical problems can be, and are, overcome. Dedicated DEI systems will strive for a more monochromatic output without Bremsstrahlung; higher X-ray energy; virtually no scatter; better detectors. DEI and other forms of diffraction imaging may represent the next window on the human body.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
