Zayd Ghoggali

Algeria | Fulbright S&T grantee 2010 | University of Pittsburgh | Mathematics

profilepicghoggalizaydZayd Ghoggali is a Ph.D. candidate in Mathematics the University of Pittsburgh. With the support of the International Fulbright Science & Technology Award, he is investigating why multiresolution classification methods used in biomedical imaging outperform the non-multiresolution ones. His research focuses on using multiresolution methods in classification. The work performed in his current group over the last few years has focused on problems in biomedical imaging involving classification, that is, assigning labels to different classes of images (for example, cancerous versus noncancerous). The group has proposed a new classification system that uses a multiresolution block in front of a generic classifier. This allows a given image to be decomposed into multiresolution subspaces and classification to occur in each subspace, followed by voting. The system has been tested on a wide range of biomedical imaging data sets, and the results have been consistent: multiresolution systems always outperformed the non-multiresolution system. To that end, Zayd’s role is to continue experiments on biomedical data sets (such as histopathology data and middle-ear infection data, both in collaboration with the University of Pittsburgh Medical School) as well as to study the underlying models to develop a measure theoretic theory of multiresolution classification in order to set a mathematical framework within which to pose and investigate real-world classification problems. The goal of his Ph.D. thesis will be to understand for which classes of data this system can be used as well as to develop it into a tool that can be widely used by biologists and medical doctors.

A native of Algeria, Zayd received his State Engineer’s degree from the University of Batna. Over the past year, he devised novel mathematical models to study the scaling limits of the DG “Double-Gate” and GAA “Gate-All-Around” Metal Oxide Semiconductor Field Effect Transistors MOSFETs. These mathematical models allow a more accurate description of the “DG” and “GAA” behaviours with a simple set of mathematical equations.

When he is not in the lab, Zayd enjoys hiking and playing soccer.