I-Corps: Commercialization of Multiplatform Gigapixel Imaging and Annotation Software

University Of Utah, Salt Lake City UT

Investigators

Abstract

The Analysis and Visualization of large scale data is a major challenge of modern science. As new data acquisition devices are built to generate more accurate, high resolution information, scientist struggle with the difficulty of gaining access and making practical use of the information generated. Doctors need to be able to make diagnoses with high resolution pathology images to make the best diagnoses for their patients. Researchers need to use high resolution microscopy data to gain a better understanding of human biology. Engineers need to explore high resolution sensing and simulated data to develop new processes and devices. For these reasons, large data analysis and visualization tools proposed in this project are becoming a critical component in the science discovery process. This new management and processing of Big Data allows translating raw data into faster discovery and understanding leading to actionable information, such as the diagnosis from a doctor. The proposed ViSUS software framework was designed with the primary philosophy that the visualization of massive data need not be tied to specialized hardware or infrastructure. In other words, a visualization environment for large data can be designed to be lightweight, highly scalable and run on a variety of platforms or hardware. The proposed software can visualize data from nearly any image capture source or simulation. Due to the large size of datasets, it is emphasized that the data samples cannot all be loaded into main memory. As a result, it is not feasible to use standard implementations of visualization and analysis algorithms on commodity hardware. This project aims to complete the first release with the minimum features needed to turn the proposed software into a commercial product. The team also intends to conduct a pilot test to gather concrete customer feedback in functionalities and user interactions while making the first release available.