Brainstorming with a difference in CAVE2
Remember the dot-to-dot and spot the difference puzzles you did as a kid? Turns out they hold the key to identifying new knowledge concealed in the masses of data generated by today’s scientific research – by letting people use their own eyes and the unmatched power of their own brains to identify patterns.
There’s a place at Monash Uni called CAVE2, where researchers can literally immerse themselves in their data as it plays across the curved wall of 3D screens surrounding them. By looking straight into the heart of their images – whether they be brain scans, radio telescope pictures of galaxies or architectural reconstructions of buildings – researchers can zoom in, zoom out, spot the difference or find the pattern to uncover the secrets hidden in their data.
‘CAVE’ stands for ‘cave automatic virtual environment’. CAVE2 combines the latest in high-end computing with 80 high-quality 3D TV screens – providing a space for people to bounce ideas off each other as they truly experience their data in every possible direction.
“You could think of it as a hypothesis generation tool,” says Associate Professor David Barnes, Director of the Monash Immersive Visualisation Platform (MIVP), which is home to CAVE2. “It’s a way to make big data small again – to let you look at a larger data set than would otherwise be possible.”
Part of the Monash Technology Research Platforms, MIVP provides the equipment and expertise needed to help researchers get the most out of their data in the CAVE2. Associate Professor Barnes and his growing team of scientific and graphic programmers at MIVP develop what researchers need, and operate it for them, leaving them free to explore and discover – effectively to connect the dots in their own data.
A case in point is the study of the neurodegenerative illness known as Huntington’s disease. Instead of the traditional method of comparing just two MRI brain scans side by side, CAVE2 lets people view up to 80 images at once. So, they could look at MRI scans from a group of say 30 people who don’t have the disease, 30 people who do, and 20 people who are ‘pre-symptomatic’ –those who have inherited the disease, but aren’t suffering symptoms yet.
The images in view are controlled by one person who wears camera-tracked 3D glasses. Using an iPad connected wirelessly to the computing system, they can interactively modify the images on the spot. For instance, they can decide what nerve fibres to show; select any region of the brain; sort by age, or subject type; and synchronise the rotation of the images in any direction.
“Because the system is so computationally powerful, we can filter data in real time,” says Associate Professor Barnes. “People can start asking questions, and the computer can respond. We can even bring untrained non-neuroscientists in, literally turning a highly complex dataset of brain images into a child’s game of spot the difference.”
Apart from CAVE2’s top-end computing power, Associate Professor Barnes says its main advantage is that several people can fit inside at the same time.
“Most big discoveries these days are accomplished by teams, not by individuals. That’s honestly the biggest drawcard of CAVE2 – its collaborative capability.”