Research at RIT: Focus on cognition
Oct. 13, 2011
by Will Dube
Follow RITNEWS on Twitter
The Fall/Winter issue of Research at RIT, coming out in November, features RIT’s efforts in cognitive processing, with articles on human-centered computing, the psychology of deafness, undergraduate research and RIT’s 30-year involvement with the LANDSAT satellite program.
“RIT’s research efforts are expanding into novel aspects of machine learning, psychology and biomedical sciences,” notes Ryne Raffaelle, vice president of research and executive editor of the magazine. “At the same time we continue to advance our traditional focus areas such as engineering, imaging science and deaf education to enhance the next generation of discovery and create a full academic experience for our students.”
The cover story profiles an interdisciplinary team led by Anne Haake, a professor in the Golisano College of Computing and Information Sciences, which is seeking to advance the development of database technologies used in biomedical imaging. The team is using visual perception techniques, computer modeling and computational linguistics to infuse human expertise into content-based image retrieval systems, known as CBIR. The work has the potential to vastly improve image use for prognosis and diagnosis of a host of diseases from skin conditions to cancer.
“Through the use of machine learning, visual perception and linguistics, we can better account for how people perceive and categorize images and incorporate this data into the computer algorithms used in CBIR systems,” notes Haake. “Human-centered image retrieval will ultimately produce more robust data that more accurately simulates human analysis.”
Research at RIT began in 2008 and previous issues have focused on nanophotonics, microfluidics and modeling and simulation. The magazine also highlights national and international awards garnered by RIT faculty and related research conducted by students and faculty throughout the campus.
To view a PDF or request print copies, go to www.rit.edu/research.