In the 1990s, Australian physicist Professor Veronica James and colleagues were studying the structure of hair, and particularly its chief protein component, keratin, using x-ray diffraction. In 1997, she published a study on the unusual change she saw in the structure of the hair from monkeys with diabetes.

In 1999, she and other Australian researchers published a paper in Nature which showed discernable changes in the diffraction pattern of hair from women with breast cancer. A ring was noticeable in the pattern in the breast cancer samples that was absent in healthy women. This was followed by a 2005 paper in the International Journal of Cancer confirming the result.

The science was so intriguing, that when Sydney investment banker Leon Carr and cell biologist Dr Peter French came across the published hair studies, they immediately began investigating their scientific and commercial potential. ,. The science was particularly interesting to French, whose PhD is in the use of monoclonal antibodies in determining the structure of keratin in wool

Carr decided to set up a company to commercialise the potential of the technique, with French and Gary Corino, one of the authors on the 2005 paper, as scientific advisors and businessman David Young as managing director.

Their new company, Fermiscan, was listed in October 2006 and is about to finalise a validation study of the technology in 2000 women in Australia. The idea is to have a diagnostic test for breast cancer that is far less invasive than current screening techniques like mammography but which has a comparable or better outcome.

The science behind the test is reasonably straightforward even if the actual reason these changes can be seen in hair is still not fully understood. Samples of hair are collected from close to the scalp, thus avoiding the problems that are caused by chemical treatments such as dyeing, perming and straightening, which can penetrate the hair and break down its proteins.

The samples are loaded onto specially designed holders and shipped to Chicago, where Fermiscan has a group based at that city's synchrotron, and placed into the centre of an x-ray beam. "The hair is exposed for less than 30 seconds - we are doing multiple hairs per person - and the time to run a sample is about two to three minutes," French says. "The result is then transferred to a PC where image processing occurs to enhance the features we are looking for."

The x-rays penetrate the hair and diffract off its molecular structure, he says. "There is a standard, well-known alpha keratin pattern of hair that has been looked at by x-ray diffraction for more than 50 years - it's not a new technique. But with the synchrotron there is much greater flux than can be achieved by standard x-ray sources and as a result very subtle, fine features can be seen. It is one of these that is associated with the presence of breast cancer."

What the data shows is a peculiar ring at a specific position or spacing in the diffraction pattern, he says. Normally the pattern shows a series of black arcs on a white background, but in samples from women with breast cancer, a circular feature is frequently observable.

"What causes that ring is not yet known. We have a few theories and what we think happens is the breast cancer secretes a range of molecules like growth factors and cytokines into the bloodstream. Some of those can interact with the hair follicle.

"It is possible that what we are seeing is a slight disturbance of the normal filament arrangement that causes the arcs to become rings. Whether this distubance is due to something that the cancer secretes that is taken up by the follicle and is put into the fibre, or whether the factor itself alters the way the hair follicle functions so that it lays down filaments slightly differently, is still open to debate. We have a research program in place to look at those sorts of issues but it's at a fairly early stage."