Jennifer G. DyAssociate Professor
Department of Electrical and Computer Engineering
My research interests include machine learning, data mining, pattern recognition, medical image analysis and computer vision.
In particular, my research has focused on developing algorithms for clustering
high-dimensional data and on applied machine learning.
Clustering is an important data analysis tool. It helps
researchers discover relationships among samples, which would otherwise be
difficult with large volumes of data.
The need for clustering high-dimensional data
is extensive and includes applications to microarray
analysis, text document, web page clustering, and medical image analysis.
I have focused my applied research on biomedical image applications.
More specifically, some of the projects I am currently working on include
tumor tracking in fluoroscopy images of the lungs for radiotherapy and on
image segmentation of in vivo
confocal microscopy image stacks of the skin to detect cancer.
Beyond clustering, I have also developed algorithms for improving
support vector machines.