Pushing for improvements in imaging

13 January 2010

A soldier is shot in Afghanistan and seriously injured. A scan is taken of his injuries, and a decision has to be made - but nobody with the necessary medical experience is available. What can be done?

Patrick Brennan, Professor of Diagnostic Imaging in the Faculty of Health Sciences, is working on solutions to this and other problems that confront medical specialists who deal with images - whether from X-rays, CT, ultrasound or MRI scans.

His area of research covers image optimisation and perception, improving both the quality of the image and our ability to interpret the information it contains.

"Can we image people better than we are currently doing?" he asks. "Are there better ways of doing things? We believe there are and we think we are doing patients a disservice by not trying to produce better images at lower risk to the patient."

In the case of the wounded soldier, Professor Brennan has recently completed a study which suggests that touch screen phones can successfully be used to display images to specialists around the world.

"So if you get an injury like a brain bleed in a war zone and you're scanned, the image can be sent to a world expert at the Mayo Clinic for an opinion. They can move it around and magnify it and the image quality is similar to that on a computer screen," he said, "We were the first people in the world to demonstrate this using large numbers of expert observers, and it's an important step forward. It enhances the immediacy of an expert opinion."

Professor Brennan comes from the small town of Monaghan in the north of the Republic of Ireland, and gained a PhD at Queens University Belfast in radiobiology and anatomy. He has been in Sydney for less than a year, moving from University College Dublin where he built up an international reputation as an imaging scientist.

"It's a small gamble to leave behind a comfortable position where grants are available and collaborations established, but the names of people working here at Sydney are well-known around the world and the University has an excellent reputation for postgraduate research and attracting funding."

His work involves a wide range of interdisciplinary collaboration, "not because it's trendy but because it's essential," he points out. "There's no way we can do what we do without working with other academics and clinicians - radiologists, anatomists, rheumatologists, obstetricians, neurologists, computer scientists, physiotherapists, nurses."

On top of that there are international collaborations with fellow researchers in America, the UK and Ireland.

A large slice of his work is focused on image perception, or interpreting what can be seen on the screen.

"Sixty per cent of radiological errors are made after the image is produced," he explains. "The system isn't foolproof. You can have the most fantastic images but still make mistakes."

As well as looking at ways of improving our ability to see disease in images, he is also looking at factors such as whether fatigue affects radiologists, and the optimum number of cases a year they should deal with.

In this area, there is a wide difference of opinion around the world. A radiologist specialising in breast imaging in Australia is expected to deal with 2000 cases a year, whereas in Europe they would handle 5000. But in America the minimum figure is only 480.

"This year we have carried out a study with Breast Screen NSW which suggests you need to look at between 1000 and 2000 images a year if you want to be up there with the median level of performance of experts," says Professor Brennan. "If you are down at 480 we would argue that you are possibly not seeing enough cases and the chances of diagnosis are reduced."