Imaging

Perhaps better than any other method, images provide us with an intuitive understanding of the subject. It is therefore surprising that the exploitation of synchrotrons for imaging arrived rather later than their use for diffraction and spectroscopy. Nevertheless, the three largest synchrotrons now have substantial numbers of people and several beamlines dedicated to advanced X-ray imaging techniques, and imaging is one of the the most rapidly expanding areas of synchrotron science. The use of synchrotrons for imaging has led to the development of new approaches that provide unprecedented resolution and contrast of nature's smallest details. Australian scientists, based mainly in Melbourne, have helped pioneer the development of many of these techniques and are regarded as among the world leaders in the field. One aspect of this work relates to the development of phase-contrast based imaging techniques that transcend the conventional reliance on absorption to to produce contrast. A second relates to the development of theory to make these imaging techniques quantitative. These developments provide the basis for major advances in X-ray imaging science that are not only relevant for synchrotron-based imaging but also to radiography with conventional sources.

Biomedical imaging

Despite being by far the most popular medical imaging modality, lack of soft tissue contrast is a significant problem in both medical and biomedical X-ray imaging. The relatively small variations in density and composition of soft tissues mean that their X-ray attenuation characteristics are very similar. Conventional radiaography produces images through the differential absorption of X-rays, and so provides very little soft tissue contrast unless high doses are employed, as in computed tomography. Synchrotron-based imaging techniques produce high resolution images using differences in the refraction and scatter of X-rays as they pass through tissue. Genuine soft tissue contrast with micrometre-scale resolution is possible.

Furthermore, the collimation and monochromaticity of an imaging beamline allows high resolution images to be recorded at far lower doses than required by conventional equipment. This capability permits longitudinal studies (series imaging) to be performed for investigations where the dose required by conventional imaging would confound the experiment.

The power of these imaging techniques is particularly suited to the study of living processes. It will be possible to exploit the proximity of the Australian Synchrotron to Monash University, The University of Melboure, CSIRO and Monash Medical Centre to bring together the expertise and facilities that will make the imaging beamline (beamline 10) one of only two beamlines in the world capable of work on live animals. The studies of live animals for medical research is an area that is impractical under overseas access programs, and so relates to an essentially new and numerous Australian user class that has not been served in the past.

One of the problems at present is that animals are often sacrified in order to obtain anatomical information at high resolution. The proposed imaging beamline will allow in vivo imaging of small animals and so provide the major advantage of allowing longitudinal studies to be conducted. This has the significant advantage of following the same animal through the process and also dramatically reducing the number of animals sacrificed in a study.

Mammography

Two major areas where soft tissue contrast is vital are breast and lung imaging. Both breast and lung cancer are major killers and better methods of imaging these diseases would have a major impact on health care.

Screening for breast cancer, which is the biggest killer of women in the 35 to 55 year old age group, is based entirely on soft tissue X-ray absorption contrast. As a result, mammography, while having been proven to reduce mortality, suffers from some major deficiencies. In particular, it is non-specific, resulting in a large number of unnecessary biopsies, and it does not work well in women below the age of 50. The potential benefits of phase contrast imaging to improving the success of mammography in detecting cancer are enormous. Work by others has shown that the contrast increases by as much as 25 times by employing phase contrast. The beamline to be constructed would be used as a 'gold standard' facility to develop improved techniques for breast imaging.

Lung imaging

Anyone who has had a chest X-ray knows that the lungs are largely invisible to all but the highly-trained eye of a radiologist. However, the air-tissue interfaces in the lung appear with startling clarity using phase contrast imaging. This is particularly applicable to human babies, and an Australian project at a SPring-8 beamline is planned to investigate the potential for developing technology to image lung clearance at birth.

Phase contrast techniques offer enormous opportunities for the study of lung function and disease in both humans and animals. Examples include:

• The detailed study of the development of respiratory function in marsupials that are born in an embryonic state and yet can still breathe
• Longitudinal studies of the effect of anti-cancer therapies on mice and other animal models.

In addition to the dramatic improvements offered by phase contrast, workers at the ESRF have demonstrated xenon contrast respiration-gated synchrotron radiation computed tomography (SRCT) with a spatial resolution at the level of the respiratory lobule (terminal bronchiole and alveoli). This technique allows direct quantification of xenon as an inhaled contrast agent based on K-edge subtraction imaging and hence the dynamics of xenon wash-in can be used to calculate regionally specific quantitative maps of lung ventilation. Examples of the use of this technique include:

• Identifying local variations in lung function cased by diseases such as asthma and chronic obstructive pulmonary disease.
• Testing the efficacy of pharmaceuticals on respiratory dysfunction.

Imaging of advanced materials and manufactured products

The high contrast and microtomography capabilities can be exploited to great effect in the areas of materials science, nondestructive testing and mineralogy. Examples include:

• studies of precipitation and voids in industrially important light metal alloys
• the study of membranes for use in advanced fossil fuel cells
• studies of fracture in ceramics
• investigation of micro/nano structured devices by micro-CT in, for example, automotive applications
• the use of high-resolution computed tomography for the study of porosity in oil-bearing rocks. By tuning to different energies it will also be possible to image the amount of residual oil left in the rock following extraction.
• the study of advanced materials following and during various stresses, both mechanical and environmental. Many advanced materials, for example those in aerospace applications, are composed of materials that cannot be omaged with conventional X-ray techniques due to a lack of contrast. An example is aluminium embedded with graphite fibres: the fibrous features would be invisible in conventional X-ray absorption contrast radiaography.

Imaging of plants

The contrast mechanisms employed to visualise soft tissues can also render visible many of the structures inside plants. An enormous range of studies is envisaged, but of particular interest is the study of drought- and salt-tolerant species, with a view to developing more efficient crops for Australia. Phase contrast computed tomography techniques will be employed to study the development of root structures without removing the plant from the soil, while the K-edge imaging will be used to study protein hormone flow dynamics.

Image processing

Imaging may be regarded as any process by which spatial structural information about an object is acquired. A paradigm shift occurred with the Nobel Prize winning work of Gabor, where he demonstrated that the information in a coherent wave could be captured on film and decoded afterwards.

Australian researchers have already made major contributions to the fundamental understanding of this leading to the algorithms for decoding the information and thus the ability to extract readily interpretable information about a sample from the images that have been described above. However, there is still much to be done, and access to the Australian Synchrotron will be important for advancing the field.

Some fundamental topics that will be pursued are the non-crystallographic phase problem and complete wavefield recovery, leading to the detailed structural analysis of a sample.

This work will be of great value in obtaining structural information from proteins that are impossible to crystallize, such as membrane proteins. It will also enhance the ability to obtain very high resolution three-dimensional images of objects with poor contrast such as biological cells and tissue.

Progress in this field is currently limited by the lack of detailed theoretical understanding of the problems of coherence and phase recovery, the need for custom optics, such as cylindrical lenses, by the sophisticated software required to recover the multi-dimensional image information, and by access to a suitable imaging detector. The proposed Beamline 10 at the Australian Synchrotron will be designed for flexibility to support the development of the optics systems and detectors that are needed.