WVU spin-off company IstoVisio, Inc. awarded funding for virtual reality research

The main menu of syGlass, a scientific data visualization and annotation system, created by engineers at West Virginia University. (Photo submitted)

Engineers at West Virginia University are aiding neuroscientists in data collection through a software that enables them to visualize and explore data in 3D for a more accurate image.

Story by Adrianne Uphold, Graduate Assistant  

IstoVisio, Inc., a WVU spin-off company that created syGlass, a software that analyzes data collected from large multidimensional volumetric data — stacks of 2D images where the pixels have multiple values — that use virtual reality head-mounted displays, has been awarded a National Institutes of Health Small Business Innovation Research grant totaling $1.6 million. WVU will receive $380,000 as part of the grant.

Gianfranco Doretto, associate professor in the Lane Department of Computer Science and Electrical Engineering in the Statler College of Engineering and Mineral Resources and principal investigator of the grant, explained the software was created to address issues experienced by neuroscientists while collecting image volumes of biological tissues.

“Neuroscientists collect image volumes of biological tissues and then look at them to formulate research hypotheses triggered by what they observe in those image volumes, or to draw conclusions about physiological or morphological characteristics based on those observations,” Doretto said. “Often, neuroscientists also need to further process the image volumes to analyze what they see.”

This is where the software syGlass comes into play. SyGlass enables neuroscientists to visualize and explore their data in 3D by using virtual reality head-mounted displays. Before IstoVisio, Inc. developed syGlass, neuroscientists would only see image slices of the data, which Doretto said is not always accurate.

“SyGlass enabled neuroscientists to visualize and explore their data in 3D by enabling virtual reality technology for image volumes, as opposed to just 3D models made of 2D surfaces,” Doretto said.

SyGlass is now used in over 100 labs across the world to help aid in the visualization of data. 

“While we developed syGlass to satisfy the needs of neuroscientists, we continue to find other application domains in need of exploration and analysis of volumetric data,” Doretto said. “In general, structural biologists collect and analyze this type of data. In clinical applications, CT scans and MRIs are better observed in virtual reality for different applications, like educational or surgical planning. At the Department of Energy, geologists have been using syGlass to analyze the internal structure of rocks through their CT scans.”

The grant will help address the challenges large image volume collections have for image acquisition, analysis and communication of a raw data set. The acquisition runs can be lengthy and expensive, and often errors are not identified until after the completion of scanning. To solve these issues, Doretto said syGlass will integrate virtual reality into the microscope controls for tuning the microscope and then efficiently inspecting images in 3D as they are acquired.

Novel transfer learning techniques will be introduced to scale up 3D image quantification capabilities for current acquisition sizes, by coupling them with user-optimized experiences that do not require machine learning expertise, and yet provide automated and accurate results. Tools to efficiently generate narrated scientific presentations in virtual reality for use in the lab setting, as manuscript publications and for production of educational materials.

SyGlass was developed with the contribution of two graduate students and a team of nearly a dozen skilled undergraduate students. WVU licensed the syGlass technology to IstoVisio, Inc. in spring 2017 and is a company stakeholder. 

The grant will also support the supervision of multiple graduate and undergraduate students who will be engaged in developing novel image analytics tools based on deep learning. The same tools will then be the basis for their engineered version that will be transitioned into syGlass.

The contribution from WVU will enable syGlass to embed advanced features based on machine learning technology that are much needed by its users, and that no other product currently provides. This will lead to widening the customer base of syGlass and will significantly strengthen the growth of IstoVisio, Inc.

---