Rights: The University of Waikato Published 21 July 2007 Download

Dr Richard Watts from the Department of Physics and Astronomy at the University of Canterbury talks about functional MRI, which looks at how the brain responds over time as patients carry out tasks.

Richard explains that functional MRI is less concerned with high resolution images and more about looking at changes in the brain over time. It’s more like creating a movie than taking a picture. Richard is working at the Van der Veer Institute for Parkinson’s & Brain Research looking at patients with injury or disease to see how their brains are functioning.


In a standard MRI scan what we’re trying to get is a very high-resolution structural image of the brain. We want to see all the little detail there. And we’re just taking a single snapshot. Whereas with functional MRI what we’re looking at are dynamic changes so things which are happening and different from one time to another. What we’re looking at is blood flow in the brain, and when we ask somebody to do a mental task, whether it’s understanding language, so for instances you’re listening to me, there’s part of your brain that’s understanding what I’m saying and that’s a very specific part of the brain.
‘Ok, this test is just going to be you listening to some language so you’re gonna hear a passage of text.’

When they do these tasks what we see is a very small change in the blood supply to the brain, which actually changes the oxygenation of the blood. So blood becomes oxyhaemoglobin, and when you lose the oxygen it become de-oxyhaemoglobn. What we’re actually sensitive to in the functional MRI scan is the ratio of those two chemicals. Is the blood oxygenated or is it not? And it turns out that oxyhaemoglobin and de-oxyhaemoglobin have different magnetic properties. An MRI is very sensitive to the magnetic properties of the blood. So as you perform a task, we get a signal change, and the trick with MRI is that you scan the brain for a period of time and you ask people to do particular tasks and then you look at the data and you say, well which parts of the brain have a signal that’s well correlated with the task that we ask them to perform. We can then identify parts of the brain associated with that task. If part of the brain is not involved with that task at all, then there’ll be no change in blood supply when they’re doing that task to when they’re not doing the task.

So that’s the kind of scan that we do, and we’ll typically scan for maybe three or four or five minutes. We’ll acquire an image of the brain maybe every two or three seconds throughout that time period, and then we go back and we say which parts of the brain are well correlated, and we usually show these as colour overlaid onto a structural image of the brain to say, this is the part of the brain involved. And we can look at people who have injuries or diseases to say are these parts of the brain working properly? Is it this part of the brain that’s involved or another part? So in functional MRI we’re much less concerned with getting a really high-resolution, sharp looking picture of the brain, as getting a lot of pictures of the brain over a period of time and figuring out what things are changing. So it’s much more of a dynamic. It’s like taking a movie as opposed to taking a single snapshot.