Adaptive multiresolution techniques for subcellular protein location classification

Gowri Srinivasa, Thomas Merryman, Amina Chebira, Jelena Kovacevic, Alexia Mintos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose an adaptive multiresolution (MR) approach for classification of fluorescence microscopy images of subcellular protein locations, providing biologically relevant information. These images have highly localized features both in space and frequency which naturally leads us to MR tools. Moreover, as the goal of the classification system is to distinguish between various protein classes, we aim for features adapted to individual proteins. These two requirements further lead us to adaptive MR tools. We start with a simple classification system consisting of Haralick texture feature computation followed by a maximum-likelihood classifier, and demonstrate that, by adding an MR block in front, we are able to raise the average classification accuracy by roughly 10%. We conclude that selecting features in MR subspaces allows us to custom-build discriminative feature sets for fluorescence microscopy images of protein subcellular location images.

Original languageEnglish (US)
Title of host publication2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings
Volume5
StatePublished - Dec 1 2006
Event2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006 - Toulouse, France
Duration: May 14 2006May 19 2006

Other

Other2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006
CountryFrance
CityToulouse
Period5/14/065/19/06

Fingerprint

Proteins
Fluorescence microscopy
Maximum likelihood
Classifiers
Textures

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Srinivasa, G., Merryman, T., Chebira, A., Kovacevic, J., & Mintos, A. (2006). Adaptive multiresolution techniques for subcellular protein location classification. In 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings (Vol. 5). [1661491]

Adaptive multiresolution techniques for subcellular protein location classification. / Srinivasa, Gowri; Merryman, Thomas; Chebira, Amina; Kovacevic, Jelena; Mintos, Alexia.

2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. Vol. 5 2006. 1661491.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Srinivasa, G, Merryman, T, Chebira, A, Kovacevic, J & Mintos, A 2006, Adaptive multiresolution techniques for subcellular protein location classification. in 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. vol. 5, 1661491, 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006, Toulouse, France, 5/14/06.
Srinivasa G, Merryman T, Chebira A, Kovacevic J, Mintos A. Adaptive multiresolution techniques for subcellular protein location classification. In 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. Vol. 5. 2006. 1661491
Srinivasa, Gowri ; Merryman, Thomas ; Chebira, Amina ; Kovacevic, Jelena ; Mintos, Alexia. / Adaptive multiresolution techniques for subcellular protein location classification. 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. Vol. 5 2006.
@inproceedings{2d04317164424d1485a40a0a1cc31f5b,
title = "Adaptive multiresolution techniques for subcellular protein location classification",
abstract = "We propose an adaptive multiresolution (MR) approach for classification of fluorescence microscopy images of subcellular protein locations, providing biologically relevant information. These images have highly localized features both in space and frequency which naturally leads us to MR tools. Moreover, as the goal of the classification system is to distinguish between various protein classes, we aim for features adapted to individual proteins. These two requirements further lead us to adaptive MR tools. We start with a simple classification system consisting of Haralick texture feature computation followed by a maximum-likelihood classifier, and demonstrate that, by adding an MR block in front, we are able to raise the average classification accuracy by roughly 10{\%}. We conclude that selecting features in MR subspaces allows us to custom-build discriminative feature sets for fluorescence microscopy images of protein subcellular location images.",
author = "Gowri Srinivasa and Thomas Merryman and Amina Chebira and Jelena Kovacevic and Alexia Mintos",
year = "2006",
month = "12",
day = "1",
language = "English (US)",
isbn = "142440469X",
volume = "5",
booktitle = "2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings",

}

TY - GEN

T1 - Adaptive multiresolution techniques for subcellular protein location classification

AU - Srinivasa, Gowri

AU - Merryman, Thomas

AU - Chebira, Amina

AU - Kovacevic, Jelena

AU - Mintos, Alexia

PY - 2006/12/1

Y1 - 2006/12/1

N2 - We propose an adaptive multiresolution (MR) approach for classification of fluorescence microscopy images of subcellular protein locations, providing biologically relevant information. These images have highly localized features both in space and frequency which naturally leads us to MR tools. Moreover, as the goal of the classification system is to distinguish between various protein classes, we aim for features adapted to individual proteins. These two requirements further lead us to adaptive MR tools. We start with a simple classification system consisting of Haralick texture feature computation followed by a maximum-likelihood classifier, and demonstrate that, by adding an MR block in front, we are able to raise the average classification accuracy by roughly 10%. We conclude that selecting features in MR subspaces allows us to custom-build discriminative feature sets for fluorescence microscopy images of protein subcellular location images.

AB - We propose an adaptive multiresolution (MR) approach for classification of fluorescence microscopy images of subcellular protein locations, providing biologically relevant information. These images have highly localized features both in space and frequency which naturally leads us to MR tools. Moreover, as the goal of the classification system is to distinguish between various protein classes, we aim for features adapted to individual proteins. These two requirements further lead us to adaptive MR tools. We start with a simple classification system consisting of Haralick texture feature computation followed by a maximum-likelihood classifier, and demonstrate that, by adding an MR block in front, we are able to raise the average classification accuracy by roughly 10%. We conclude that selecting features in MR subspaces allows us to custom-build discriminative feature sets for fluorescence microscopy images of protein subcellular location images.

UR - http://www.scopus.com/inward/record.url?scp=33947684906&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947684906&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:33947684906

SN - 142440469X

SN - 9781424404698

VL - 5

BT - 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings

ER -