Wednesday, August 19, 2009

No Bones About It: New Imaging Technology Lets Scientists See Right Through You

By Celia Smith, Cornell University

The image on the projector screen is eerie and a bit shocking. The ghostly form of a fish swims into view, its body completely transparent except for the pale bones of its skeleton: a moving, breathing x-ray. The graceful interlocking of the carp’s bones pushes it toward some scraps of food. Suddenly, its snout springs forward until the bones in its mouth form a hollow tube. Like a vacuum cleaner attachment, the carp’s extended mouthparts powerfully suck up the scraps, which then float lazily back into its shadowy stomach.

What looks like a marine take on a zombie movie is called XROMM, which stands for X-Ray Reconstruction of Moving Morphology. That’s just a fancy way of describing the creation of three-dimensional x-ray videos to study the symphony of muscle, tendon, and bone that is animal movement.

Until now, there have been many ways scientists have studied how body parts work together to produce all the movements we see in the animal kingdom. Previous research has included examining animal anatomy, and building models. XROMM is unique because it allows scientists to see the individual movements of bones while they are still inside a living, breathing animal.

With this ability to observe natural motion in such detail, scientists have re-created the workings of animal bodies with greater precision than ever before. The carp x-ray video is an example of XROMM’s uncanny ability to capture and display animal movements. It was taken by Nicholas Gidmark, a PhD student in the department of Ecology and Evolutionary Biology at Brown University, who is using XROMM to study the movements of fish jaws.

Jaws evolved in the animal kingdom after an evolutionary split left behind the jawless fish, such as lampreys and hagfish, around 500 million years ago. Since then, most other fish as well as humans have evolved strong jaws with wide ranges of motion. But despite their long history of development, human jaws are built quite simply. "Basically, there are two elements to chewing: the skull and the lower jaw," Gidmark says, "and at the end of the day, there are only so many ways the jaw can move."

Gidmark is interested in a group of animals with much more complex and evolved jaws than ours: the ray-finned fishes. This group of fish, which includes sturgeon, salmon, goldfish, and the carp from the x-ray video, is equipped with heavy-duty mouth machinery that can grasp, bite, and suck with astonishing power. Their many jaw bones and range of motion make them much more difficult to study than humans, and scientists have often had to rely on guesswork to determine which bones do what. Gidmark is now using XROMM technology to look at the movements of individual bones in these fish during the catching and chewing of food.

To create an XROMM image, researchers first take an x-ray ‘video’ of a subject from two different angles. The videos are similar to the x-rays you get on either side of your head at the dentist, only these are recorded as two sets of moving images at the same time. The x-rays capture the images of small metal markers, which have been implanted into specific areas in the bones.

Once the x-rays have been taken, a computer program combines information about the animal’s movements with a three-dimensional scan of its bones. The software uses the metal markers from the video as reference points to match precisely each bone-scan image with its ghostly x-ray outline. The end result is a lifelike image of a real bone, moving around in three-dimensional space.

Scientists can then use computer animation to make the images move around, either to examine how they behave in nature, or to experiment with other possible motions. This ability to play with the images on the computer is important for the scientists, who often find it difficult to get live animals to move the way they want in order to study them. "We use the same computer program Pixar used for Toy Story," says Gidmark proudly. He explains how the program also allows him to view the bone images from any angle, just as though he were a cameraman filming a scene.

To demonstrate, Gidmark shows a clip of a finished XROMM representation. In addition to the x-ray motion shown by the carp, this image displays the bones’ precise shape and pitted texture from the bone scan. The result looks just like a bunch of disembodied bones hanging in space, calmly opening and closing in an unmistakable chewing motion.

Gidmark isn’t the only one who sees how useful three-dimensional x-ray images can be for the study of animals. Henry Astley, also a Brown University researcher, is focusing on frogs. Any casual observer can see that frogs can jump very far for their size; up to six feet for an American bullfrog. Scientists always assumed that the frogs’ leg muscles were responsible for sending it over such impressive distances. But that didn’t always add up, since some frog jumps were estimated to generate up to eight times the amount of power one would expect from muscles alone.

Using XROMM, Astley has been able to show that, in fact, tendons should get the credit for these lengthy leaps. "Think of it like a bow and arrow," says Astley. "If you just throw an arrow, it doesn't go very far. But if you use a bow, you can slowly load energy into the bow and release it all at once. Like the bow, the tendon stores the energy produced by the muscle, then suddenly releases it all at once."

In addition to fish and frogs, XROMM can be used to explore the motion of animals we have never seen move before…like dinosaurs. Even though there are none alive today, artists and scientists can collaborate to re-create their movements, using what we know about their anatomy.

"How does a T. rex run? We can make some good guesses," says Gidmark, who adds that there is high demand from movie makers for this type of imaging. For example, the makers of Jurassic Park and other film producers have developed characters using "motion capture;" a cousin of XROMM technology in which markers are placed on the surface of a subject (such as a model or actor) for tracking movements.

Since motion capture markers are not actually implanted within muscle or bone, copying movements this way is less accurate than using XROMM. However, as entertainment technology becomes more and more sophisticated, XROMM may be introduced to create more lifelike characters.

The precision of XROMM may find its greatest use in human health and medicine. "There are applications for medical and clinical science no one ever though of until about twenty years ago," says Gidmark. The ability to look at exactly how bones and other body parts are working together could be a tremendous help to doctors in diagnosing their patients. For example, x-ray videos of someone walking on a treadmill combined with a bone scan of their leg could help pinpoint the cause of knee pain.

Patients in need of a diagnosis would not be able to wait for metal markers to be drilled into their bones; but without the markers, it is very difficult for scientists to properly match the three-dimensional bone scan images with the x-ray videos. Computer software may become sophisticated enough to efficiently match the images in a "markerless" version of XROMM, but seeing this technology in a doctor’s office is probably still a few years down the road. However, Gidmark believes it has the potential to drastically improve health and medicine, removing a lot of the guesswork from the process of figuring out where it hurts.

Other human uses for XROMM include creating what Gidmark calls a "gold standard" in medicine, meaning that the program could be used to show scientists exactly how a body part should ideally work. For example, if doctors have a standard model for the proper anatomy and motion of human limbs, building comfortable, functional prosthetics could become much easier.
"When we try to build something, XROMM can say: ‘yes, that’s how it moves,’ or ‘no, that’s not how it moves,’" explains Gidmark. In the case of prosthetics, XROMM can help create a standardized computer image of normal human arm movement; flexing, extending, making a fist, and so on. That image can then be used as the template for building a man-made arm that will work exactly like a natural one.

"What I like about this type of science is, it’s more concrete than, say, chemical analysis," says Gidmark. "There’s a bone, and I can see it, but I don’t know what it does." With a bit of luck, and some cutting-edge three-dimensional imaging technology, he soon will.

[Image of carp skeleton courtesy Nick Gidmark]

No comments:

Post a Comment