Me, Matt White, Ian Pengelly and Sarah Woolridge
at the Queensland X-ray's CT scanner
with Banjo's humerus (upper arm bone)

We want to bring Australian dinosaurs back to life as properly as the science, technology, and contemporary thinking allows. The aim is to show what our long-lost animals really looked like in the flesh, then make them walk, run and interact as they once did on the big screen! Crucial to the restoration process is the capture of every last bit of evidence the fossils can provide. They are, after all, the most important and the richest source of physical evidence we have.

In a previous post (see First 3D Digital Prints) I described how we kick-started the Digital Dinosaur reconstruction project when our palaeoartist Travis Tischler used Zbrush software to sculpt and restore all the bones of the right arm of our pin-up boy Banjo (Australovenator wintonensis) — the largest predatory animal ever discovered in Australia. The reconstruction and 3D prints of Banjo's arm that we made are impressive, but the bones were all interpretations sculpted by hand. The next step in our master plan was to more objectively and accurately capture the surface contours of each of Banjo's fossil bones directly to computer. But to do that required access to three dimensional (3D) digital scanners; expensive equipment we just didn't have. What to do?

Our palaeontologist Matt White came to the rescue. As part of his PhD project to study the biomechanics of Banjo's arms and claws, he approached Queensland X-ray at the Mater Hospital in Mackay, near where he lives. They had one of the best CT scanners in the country — the ultimate in 3D digital scanning equipment!

So what is a CT scanner? "CT" stands for Computed Tomography. You may have heard of the older term ‘CAT’ scan (Computer Assisted Tomography). It's a medical imaging procedure that uses x-rays and digital sampling to create detailed 2D and 3D images for computer analysis. Unlike other types of medical imaging, CT scans can record every type of body structure at once including bone, blood vessels and soft tissue. The equipment consists of a large square machine with a circular hole. Inside the machine is a rotating gantry that carries the x-ray source and sensitive x-ray detectors. An automated table moves the reclining subject (in our case a dinosaur bone, not a person) through the circular hole. Multiple x-rays are taken as thin cross-sections along the bone like slices of a loaf of bread. Detectors collect the x-ray information from each 2D cross-section and send them to a computer where they are combined into a single 3D image. The result is detailed 2D cross-section images and 3D scans of the internal and external structure of each bone. This is cutting edge science right now but I've no doubt it will one day be routine if not essential for palaeontological research.

Ian Pengelly (radiographer in charge) and Sarah Woolridge (radiographer) at Queensland X-ray were fantastic collaborators and generously donated several weekends of their time to the scanning project. The biggest hurdle was the time required to transport fossils to the CT scanner in Mackay by car. With fragile and priceless fossil bones meticulously packed on the back seat, each trip took me about about 30 hours driving time alone! With every trip to the CT scanner I aimed for minimum turnaround time as the fossils needed to be scanned and back on display in Winton as quickly as possible. A pity Winton hospital doesn't have a CT scanner but as they cost about a million dollars each I may be waiting a while yet!

The other intriguing thing about CT scans is they can potentially see right through jackets of plaster, aluminium foil and soil matrix to the fossil bones hidden inside. With batch processing of several jackets at once we could possibly pre-screen our entire backlog of smaller jackets and see what we have before even opening them! We haven't tested this idea properly yet but initial tests look promising.

So how did we go? Well, we scanned almost all of Banjo's bones (except for a smattering of small fragmentary material) and the results are amazing. We now have a "digital specimen" collection stored on CD right alongside our fossil specimens in our museum collection that not only act as a back-up virtual replica of each original fossil but are the only record we have of the delicate and sometimes surprising internal structures of each fossil — bone marrow hollows and bone density variations—spectacular!

As to limitations or drawbacks, well apart from the scarcity and cost of the the equipment, CT data doesn't record color information, so images are monotone greyscale only. Super fine details at the micro level are not captured either (resolution is limited to the sub-milimeter range; that's really good but could be better). However, a big advantage is that every dimension of each fossil is very accurately recorded. This ensures that each scan can be directly compared at the same relative size without error and you can measure each and every feature on the fossil, both internal and external, directly on the computer down to fractions of a millimeter, which sure beats using a tape measure!




2D CT scan cross-section of Banjo's large hand claw.
The spongy marrow and empty hollow in the middle
of the bone is evident, as are all the cracks in the fossil.





3D CT scan of Banjo's dentary (front half of lower jaw).
Three different rotated views show the outside, inside
and top view. Note all the empty tooth sockets.
In cross section internal views (not shown) we could
clearly see un-erupted teeth deep inside every socket!





3D print of the scan of Banjo's dentary bone. Here
we made a mirror-image of the scan to create the
matching bone from the other side of the jaw
(which so far remains undiscovered).




We made the local newspaper
(the Mackay Daily Mercury) with this cartoon!