A team of geoscientists at Princeton University has developed a method to create three-dimensional digital images of deconstructed rock samples that can be viewed from any angle. Algorithms, developed by the team, allow the computer to segment the images without human bias.

Close to five years ago Adam Maloof, an associate professor of geoscience, collaborated with SITU Studio to develop The Princeton Grinding Imaging and Reconstruction Instrument, known as GIRI, which enables scientists to see what rocks look like on the inside.

Now, Maloof and Akshay Mehra, a doctoral student and co-author of the study, are using GIRI to dissect rocks and minerals and study fossils.

The full paper was recently published in Proceedings of the National Academy of Sciences.

The study

The team recently used GIRI to disprove the common belief that Cloudina, a thin-shelled creature that lived all over the world 545 million years ago, were reef builders. In fact, they proved that Cloudina fossils were transported from other areas and had little to do with reef building.

“I thought going in we would learn all sorts about this amazing first biomineralizer and first reef builder, but Cloudina turned out to be more like a reef dweller,” Maloof said in a statement.

Many fossils, including that of Cloudina, had previously been resistant to detailed studying because traditional X-ray and CT scan machines can’t pick up on the density contrast between the fossils and the surrounding mineral.

“The reasons X-rays work, or even MRIs work, is because our bones have a different density than our skin and blood vessels,” said Mehra.

“Since there is no difference between a fossil and the matrix in which it is in in some of these rocks, you can’t tell anything apart using an X-ray machine. It would just come back as a white return.”

Scientists have failed to figure out what fossils look like in 3D for a long time.

When they’re embedded in rock, it’s hard to get them out, according to Maloof.

“People did serial sections just like this way back then — but perhaps not at this scale — where they would grind away a little rock, draw it, grind a little more, draw it. … It can be incredibly time-consuming,” he said in a statement.

GIRI speeds up the process, eliminates human error, and lessens destruction of the rocks and fossils.

The powers of GIRI

GIRI can cut slices of rock smaller than a one percent of a millimeter. Each slice in a rock takes about 90 seconds to cut and image, so a typical inch-thick sample takes about a day and a half to grind and image.

The technology can create a 3D sample of any solid object no matter the density contrast, and because GIRI takes a high-resolution photograph of every slice, viewers always see the rock itself, not only the density contrasts.

“The beauty of having a physical image is that we are seeing a real response. We are seeing color and texture,” said Mehra.

Despite all of the advantages, the researchers recognize that their technique is still destructive.

“[T]hat’s the disadvantage,” Maloof said in a statement.

“But what’s so nice is that you get to see photographs and make direct observations: That’s what’s been so life-changing to me: I love that it’s not a model. You can just see it. On any given slice, if you find something great, you can just find the slice and say, ‘What did it look like?’ … We’re on a virtual tour inside, rather than looking at waveforms and trying to interpret them.”

The evolution of GIRI

Since GIRI was first developed years ago, the scientists have made multiple physical improvements to the machine.

They have redesigned and replaced the camera housing and the mechanism for cleaning and preparing the rocks for photographs, and they’ve installed monitors that record the temperature and humidity during the time each photograph is taken.

In addition to the physical improvements to the machine, there have been steps taken to improve the running and analyzing software used in GIRI.

Maloof credits Akshay with designing machine-learning solutions that enable the control computer to send and receive signals from the grinder, verify image capture, and trigger the shutter.

“From the ground up, Akshay has designed machine-learning solutions to make the process of image segmentation automated and reliable,” Maloof said in a statement.

“He has developed techniques that ultimately will be important for any tomographic applications, including X-ray CT. Akshay also has developed ways to make quantitative measurements in the reconstructed 3D volumes. You’d be surprised how much 3D modeling out there only leads to visualization and qualitative interpretation, whereas Akshay actually measures the size, shape and 3D orientation of these critters.”

Future implications

Mehra believes that GIRI could be a widespread tool used in paleontology and geology.

“In paleontology and geology, one thing that is missing is the application of machine learning or AI to identify features of interest,” said Mehra.

“We have noticed that paleontology students will go out and often times get an X-ray or CT scan of an object and spend a year or two hand-tracing out the layers and identifying different pieces.”

“There are two issues with that,” he continued.

“One is that it is time-consuming, and the other is that you’re relying on an individual to make a decision about whether this grey or this color represents one thing or the other.”

If there is a trained professional who is interested in testing a sample using GIRI, he or she can take a slice or two from the entire sample and highlight a few areas to depict what the bone, fossil, or rock materials may look like. That information is then fed into the network, which is designed to take in the image information and make decisions on what each color in a slice might look like.

“This allows the segmentation to be done by a machine, with some input from human beings, and that removes a degree of bias,” said Mehra.

GIRI technology has already caught the eye of scientists all over the world.

Paleontologists have reached out to Maloof and Mehra to ask for virtual tours through all kinds of specimen, including shelled creatures, land creatures, fish, and dinosaur bones.

Planetary scientists are interested in GIRI because dissecting tiny grains called chondrules could give insight into how planets are formed.

Battery makers and engineers are interested in testing reservoir rocks for carbon sequestration, and want to grind graphite batteries to evaluate 3D structures of the porosity in the carbon.

“There’s really no limit to the contributions GIRI can make,” Maloof said in a statement.

“This represents five years of work. It’s the only instrument in the world like it.”