Virtual reality in research
by Guest Author on 24 Jan 2019
Computational scientist Stephen Taylor and his team at the MRC Weatherall Institute of Molecular Medicine (MRC WIMM) are helping scientists and surgeons explore biological structures up closer than ever before. He takes us on a tour of his virtual reality vision.
If you’re an engineer looking to fix a problem in the network of tunnels in the London Underground, you wouldn’t find the standard 2D London Tube map much use.
The same applies to scientists trying to understand how DNA works in the body. Each of your 37 trillion cells contains a nucleus that is approximately 1/10th width of a human hair and contains 2 metres of DNA, tightly folded into intricate patterns. The pattern of folding is critical to the function of the cell and has implications in many diseases from cancer to diabetes. Until recently, we’ve been using 2D genome maps to try and understand this complex structure and how genes are turned on and off.
With this problem in mind, I wanted to create software to help us explore DNA in more than two dimensions. In computational biology, we have fantastic algorithms for analysing biological data but often only a small minority of scientists know how to use the programs.
I’ve always been interested in how computer games have great user-friendly interfaces. So, we turned to the video gaming expertise of a group at Goldsmiths. We wanted a system that’s accessible for as many people as possible, not just scientists.
Stepping inside DNA
That’s why we started developing CSynth: an interactive 3D visualisation platform for biological molecules. It provides a way to go inside the genome in virtual reality (VR) using genome sequencing data.
By using CSynth’s immersive environment and a VR headset, scientists step into a space where they can look at and interact with elements of DNA molecules. This is helping scientists look at DNA ‘control switches’ – switches associated with specific diseases which can turn genes on or off. By seeing where the switches are in 3D, they are better understanding the role of their structures in disease, and how to fix them when they go wrong.
Our first release of this software is now freely available. Anyone can upload data to explore models using a web browser and in VR if they have a headset. It’s incredibly quick and the public interface means the software can be used as a free learning tool too.
Going beyond DNA
Next, I started thinking about how we could apply this technology beyond biological molecules to more translational aspects of medicine. VR has been applied in surgery simulation and teaching for several years. Now its use in other areas has been opened up with VR headsets becoming cheaper and more widely available.
For example, abdominal aortic aneurysm (AAA) is a significant cause of death in developed countries. AAA is the ballooning of the main artery that pumps blood from the heart to the rest of the body. Burst AAAs kill around 200,000 people in the world each year.
As an aneurysm grows, the risk of it bursting increases exponentially. But if detected early, it can be effectively treated by surgery to insert a supporting structure called a stent, which alleviates the risk of the artery rupturing. Currently, the process of designing stents is time-consuming as the surgeon must look at the 3D scan 2D segment by segment, measuring the width of the artery at different positions.
In collaboration with surgeons at the Nuffield Department of Surgical Sciences, we are developing our “Babel VR” software that allows the surgeon to do the measurements in VR, using the 3D scan. This is much quicker and more intuitive than the old process.
We are also collaborating with radiologists at the Oxford Centre for Magnetic Resonance to look at liver and heart magnetic resonance imaging (MRI) data and help plan cancer treatment.
Named after the Babel fish in the ‘Hitchhiker’s Guide to the Galaxy’, Babel VR can take data in any 3D image format and ‘translate it’ into a VR model. Looking at medical data in 3D could be useful for planning different types of operations. Our surgeons are excited about the possibilities. We are beginning to collect more data to show how Babel VR can speed up tasks compared with existing 2D methods.
It’s still early days for VR. I’m keen to investigate where there are real niches and to improve how we interact with images and data compared to traditional 2D methods. Consumer VR, driven by computer gaming, is making equipment more lightweight and affordable but it will take time before it becomes more mainstream.
VR has been overhyped as ’the next big thing’ a few times in its history but it’s finding niches in many areas now, including in design, architecture, robotics, training and healthcare. The next few years will be interesting to see what’s developed, both at the hardware and software level. I’m excited to be able to pioneer new techniques that solve real problems using this technology.
CSynth was created by Stephen Taylor and Jim Hughes from the MRC WIMM Centre for Computational Biology, with experts in real-time computer graphics and human-machine interaction at Goldsmiths, University of London.
The initial development of Babel VR has been supported by a University of Oxford IT Innovation Challenge 2018 award.