MRI Scanning the Stars

22 February 2009

Interview with

Professor Alyssa Goodman, Harvard-Smithsonian Center for Astrophysics

Chris - That's exactly what Professor Alyssa Goodman from Harvard University is doing. She's taking some of the systems that have been geared up to do better body scans with MRI and applying them  to images of the night sky to enhance those pictures and  she's with us now. Hello Alyssa. Thank you for joining us. Tell us a bit more about this work.

The star-forming region IC 348 in 13CO as displayed in 3D Slicer.Alyssa - We have a challenge in astronomy that more and more often we're able to get a third dimension of data: something like distance, not always exactly distance. Instead of seeing just a flat sky we can see thing where we know what distance. We can see in my work, for instance, gas clouds that are busy forming new stars. We like to see what they look like the way that we could go around 3D clouds in the sky. Of course you can't do that in astronomy. We need a way of putting the images back together into something that looks more like a 3D picture that turns out that the computer software to do that we need is similar to the software that they use in medical imaging.

Chris - Why is it such a problem, compiling images into three-dimensions like that?

Alyssa - It turns out it's less of a problem in other fields and in astronomy people are just not used to having that kind of information. They were starting to try to build their own software and then we realised that a lot of other people have faced this ad done a good job of, for example, making animated movies. You may know that Pixar and companies like that have some of the most powerful computers around. It turns out that 3D animations, moving 3D pictures is rather computationally intensive to do high resolution. People in other fields, as I mentioned before - notably medical imaging and in film, movies and Hollywood - had figured that out quite well. We're trying to borrow on what they  already learned.

Chris - When you start doing this do your images literally come alive? Can you see things that, can you identify details that have previously been overlooked?

Alyssa - One of the things that we're interested in, in my own work which has to do with star formation, is what the impact of jets of material and expanding shells from stars have on the clouds that the stars are forming in. Imagine terrestrial clouds and you set off a bomb in it and you want to see if the expansion wave, some kind of sphere expanding from that bomb looks like. You'd really love  to be able to see a 3D image of that. When stars set off either supernova explosions of just powerful winds from stars the same kind of thing happens to these clouds that they're in. It's very important for us to understand what that looks like. In a lot off cases it's difficult to see that happening. It's important to view what happens to these clouds over millions of years as they evolve. This software has let us, among other things, see the outflows and shells that come from these stars in a 3D way that the human brain understands which was very hard to see when looking at just slices of the images before.

Chris - Can researchers begin to speculate that, in fact, our own solar system (in other words, the sun and our clutch of planets) actually get buffeted into existence by a big star nearby that was doing something similar to what you're describing? It was putting a jet of material out which pushed a cloud of gas to make it fall into itself, which then formed us?

The star-forming region B5 in 12CO (purple) and 13CO (pink) as displayed in 3D Slicer.Alyssa - Absolutely. There's a theory called triggered star formations and the idea is there are these gas clouds out there which are marginally what's called self-gravitating. They're sort of held together by their own gravity but not quite. They might blow apart, they might collapse but if you come by and push them a little bit - sort of trigger - then they're more likely  to collapse quickly and make something like our solar system. One way to do that is having a big blast wave either from one of these outflows or some sort of shell, possibly a supernova come by.

Chris - What are you looking at, at the moment? What's the prime focus of study?

Alyssa - Right now we have something we've been doing over the past five or six years called the COMPLETE Survey of Star-Forming Regions. That's a long, funny acronym you can look up online. What it does is it looks at some of the nearby star-forming regions using every technique we can use from the ground. Optical wavelength and radio wavelength. Radio is where we can make these three dimensional images. With the Spitzer Space telescope, which is the infrared part of the Hubble telescope, has also looked at these same regions. They're essentially targeted regions where we essentially want to understand the whole process of star formation. What we've been able  to do with this 3D imaging project that we call astronomical medicine is to be able to give people 3D views of what these very large regions of space look like, to be able to  put back together in our minds a picture of what's going on. From that picture we make hypotheses. Recently our work is about the details of the role of self gravity, how likely little bits of this gas are to collapse over time on themselves and to understand whether our theories are right. The best way to do that is to see a picture of what they mean. We've been able to convince people that we think we're on the right track.

Chris - And you have some spectacular pictures on your own website. If anyone listening on the radio wants to check it out, where's the web address so they can take a look at those pictures?

Alyssa - The best way to do it is to just type my name: Alyssa Goodman in Google and I think it's the first link that comes up.

Alyssa's website can be found
here, and there are some fantastic
virtual reality objects and even a
3d pdf available on the
Harvard IIC Websites.

Add a comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.