Visualization

The visualization needs for scientific computing has increased tremendously in recent years.  Simple visualization tools such as plots or charts are no longer sufficient to peer through large data sets.  A slew of visualization tools and associated hardware have been developed to meet these demands.

MSI is well aware of the need of the University community and has equipped its labs with high-end workstations with stereo capabilities. One of the labs, LMVL, has an immersive visualization power-wall system of size 28’ x 7.5’ along with active 3D stereo and motion tracking. The power wall allows users to visualize and navigate information in multiple dimensions.

Users needing remote access to visualization can get it using NX. For users with visualization needs that cannot be met using NX, a specialized high-end analysis and visualization system is needed.  MSI has been working actively to acquire a new system with large memory, faster CPUs and a top of the line graphics card.  Two such systems are the 4 graphics nodes in Koronis and the Visualization system connected to Itasca. These high-end visualization systems are connected to storage by a high-speed network connection. Currently, VirtualGL is used as the software of choice for remote visualization. We have used the visualization systems to visualize data sets as large as 50GB in size using commercial visualization packages such as Avizo and Ensight.

MSI has staff with specialization in visualization and image processing to help users with the visualization hardware. These staff consultants work with various faculty, staff, and students.  Here are some of the projects completed by MSI visualization staff.

Automated segmentation and Visualization of a large micro-CT data of mice

High-resolution large datasets were acquired to improve the understanding of murine bone physiology.  The analyzed dataset is more than 50 GB in size with more than 6,000 2,048 × 2,048 slices. The study was performed to automatically measure the bone mineral density (BMD) of the entire skeleton. The manual segmentation process took 6,300 min for 6,300 slices while our method took only 150 min for segmentation.  Iso-surface rendering was performed on the data-set and 10–12 frames per second of rendering was obtained. with 2:1 binning.

Bone segmentationQuantification of bone density by segmentation using Analyze Arthritis in mice is studied by knocking the corresponding genes. The bone density (a measure of arthritis) is then studied by segmenting the CT image of the bone into various regions like trabeculae, cortical and marrow. The volume of the various regions is then determined to quantify bone erosion.

 

 

 

Block Copolymers Defect Formation and Evolution MSI researchers Shuxia Zhang and David Porter developed a framework for characterizing defects of block copolymers.  This framework produces a graphically and tempo-spatial characterization of the defects. 

 

Virtual Colosseum Reconstruction  Professors Dan Keefe and Richard Graff using the LMVL visualization laboratory to explore a 3D model of a reconstructed Roman colosseum.

 

 

heart visualization

Generating 3D models from CT Scans MSI helped the Medical Devices Center create 3D models from CT scans of human anatomy. In the example shown at the right, 300 CT scans were used to create a 3D model of a human heart. After creating the models, MSI brought the models into a virtual reality environment available in MSI's LMVL lab enabling users to "fly" through the heart. The models can be used for anatomical training or to assist in placement of medical devices.

 

 

Deforestation Data Visualization Using the LMVL visualization system Professor Vipin Kumar's research group can overlay satellite deforestation data onto Google Earth.  Using the large display system in the LMVL they are able to visualize a large set of data without losing detail or frame of reference. 

 

See consulting for general information about MSI consulting.