Patch-based sparse representation for bacterial detection

A. K. Eldaly, Yoann Altmann, Ahsan Akram, Antonios Perperidis, Kevin Dhaliwal, Stephen McLaughlin

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

Abstract

In this paper, we propose an unsupervised approach for bacterial de- tection in optical endomicroscopy images. This approach splits each image into a set of overlapping patches and assumes that observed intensities are linear combinations of the actual intensity values as- sociated with background image structures, corrupted by additive Gaussian noise and potentially by a sparse outlier term modelling anomalies (which are considered to be candidate bacteria). The ac- tual intensity term representing background structures is modelled as a linear combination of a few atoms drawn from a dictionary which is learned from bacteria-free data and then fixed while analyzing new images. The bacteria detection task is formulated as a minimization problem and an alternating direction method of multipliers (ADMM) is then used to estimate the unknown parameters. Simulations con- ducted using two ex vivo lung datasets show good detection and cor- relation performance between bacteria counts identified by a trained clinician and those of the proposed method.
Original languageEnglish
Title of host publication2019 IEEE International Symposium on Biomedical Imaging (IEEE)
PublisherIEEE
Publication statusAccepted/In press - 18 Dec 2018

Fingerprint

Bacteria
Glossaries
Atoms

Cite this

Eldaly, A. K., Altmann, Y., Akram, A., Perperidis, A., Dhaliwal, K., & McLaughlin, S. (Accepted/In press). Patch-based sparse representation for bacterial detection. In 2019 IEEE International Symposium on Biomedical Imaging (IEEE) IEEE.
Eldaly, A. K. ; Altmann, Yoann ; Akram, Ahsan ; Perperidis, Antonios ; Dhaliwal, Kevin ; McLaughlin, Stephen. / Patch-based sparse representation for bacterial detection. 2019 IEEE International Symposium on Biomedical Imaging (IEEE). IEEE, 2018.
@inproceedings{75257ea8a30048e6b14d85f9d5a436f3,
title = "Patch-based sparse representation for bacterial detection",
abstract = "In this paper, we propose an unsupervised approach for bacterial de- tection in optical endomicroscopy images. This approach splits each image into a set of overlapping patches and assumes that observed intensities are linear combinations of the actual intensity values as- sociated with background image structures, corrupted by additive Gaussian noise and potentially by a sparse outlier term modelling anomalies (which are considered to be candidate bacteria). The ac- tual intensity term representing background structures is modelled as a linear combination of a few atoms drawn from a dictionary which is learned from bacteria-free data and then fixed while analyzing new images. The bacteria detection task is formulated as a minimization problem and an alternating direction method of multipliers (ADMM) is then used to estimate the unknown parameters. Simulations con- ducted using two ex vivo lung datasets show good detection and cor- relation performance between bacteria counts identified by a trained clinician and those of the proposed method.",
author = "Eldaly, {A. K.} and Yoann Altmann and Ahsan Akram and Antonios Perperidis and Kevin Dhaliwal and Stephen McLaughlin",
year = "2018",
month = "12",
day = "18",
language = "English",
booktitle = "2019 IEEE International Symposium on Biomedical Imaging (IEEE)",
publisher = "IEEE",
address = "United States",

}

Eldaly, AK, Altmann, Y, Akram, A, Perperidis, A, Dhaliwal, K & McLaughlin, S 2018, Patch-based sparse representation for bacterial detection. in 2019 IEEE International Symposium on Biomedical Imaging (IEEE). IEEE.

Patch-based sparse representation for bacterial detection. / Eldaly, A. K.; Altmann, Yoann; Akram, Ahsan; Perperidis, Antonios; Dhaliwal, Kevin; McLaughlin, Stephen.

2019 IEEE International Symposium on Biomedical Imaging (IEEE). IEEE, 2018.

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

TY - GEN

T1 - Patch-based sparse representation for bacterial detection

AU - Eldaly, A. K.

AU - Altmann, Yoann

AU - Akram, Ahsan

AU - Perperidis, Antonios

AU - Dhaliwal, Kevin

AU - McLaughlin, Stephen

PY - 2018/12/18

Y1 - 2018/12/18

N2 - In this paper, we propose an unsupervised approach for bacterial de- tection in optical endomicroscopy images. This approach splits each image into a set of overlapping patches and assumes that observed intensities are linear combinations of the actual intensity values as- sociated with background image structures, corrupted by additive Gaussian noise and potentially by a sparse outlier term modelling anomalies (which are considered to be candidate bacteria). The ac- tual intensity term representing background structures is modelled as a linear combination of a few atoms drawn from a dictionary which is learned from bacteria-free data and then fixed while analyzing new images. The bacteria detection task is formulated as a minimization problem and an alternating direction method of multipliers (ADMM) is then used to estimate the unknown parameters. Simulations con- ducted using two ex vivo lung datasets show good detection and cor- relation performance between bacteria counts identified by a trained clinician and those of the proposed method.

AB - In this paper, we propose an unsupervised approach for bacterial de- tection in optical endomicroscopy images. This approach splits each image into a set of overlapping patches and assumes that observed intensities are linear combinations of the actual intensity values as- sociated with background image structures, corrupted by additive Gaussian noise and potentially by a sparse outlier term modelling anomalies (which are considered to be candidate bacteria). The ac- tual intensity term representing background structures is modelled as a linear combination of a few atoms drawn from a dictionary which is learned from bacteria-free data and then fixed while analyzing new images. The bacteria detection task is formulated as a minimization problem and an alternating direction method of multipliers (ADMM) is then used to estimate the unknown parameters. Simulations con- ducted using two ex vivo lung datasets show good detection and cor- relation performance between bacteria counts identified by a trained clinician and those of the proposed method.

M3 - Conference contribution

BT - 2019 IEEE International Symposium on Biomedical Imaging (IEEE)

PB - IEEE

ER -

Eldaly AK, Altmann Y, Akram A, Perperidis A, Dhaliwal K, McLaughlin S. Patch-based sparse representation for bacterial detection. In 2019 IEEE International Symposium on Biomedical Imaging (IEEE). IEEE. 2018