Paper: | FR-PM-PS4.11 |
Session: | Cellular and Molecular Image Analysis |
Time: | Friday, April 7, 13:30 - 14:50 |
Presentation: |
Poster
|
Title: |
A Multiresolution Enhancement to Generic Classifiers of Subcellular Protein Location Images |
Authors: |
Thomas Merryman; Carnegie Mellon University | | |
| Keridon Williams; University of Virgin Islands | | |
| Gowri Srinivasa; Carnegie Mellon University | | |
| Amina Chebira; Carnegie Mellon University | | |
| Jelena Kovacevic; Carnegie Mellon University | | |
Abstract: |
We propose an algorithm for classification of fluorescence microscopy images depicting protein subcellular locations. We aim to improve upon previous systems by adding the true power of multiresolution---adaptivity. In the process, we build a system able to work with any feature sets and any classifiers, which we denote as a Generic Classification System. Our system consists of multiresolution decomposition in the front, followed by feature computation and classification in each subband, yielding local decisions. This is followed by the crucial step of combining all those local decisions into a global one, while at the same time ensuring that the resulting system does no worse than a no-decomposition one. We obtain a high accuracy of 89.8%, effectively proving that the space-frequency localized information in the subbands adds to the discriminative power of the system. |