Detection of breast thermograms using ensemble classifiers

Dayakshini Sathish, Surekha Kamath

Research output: Contribution to journalArticle

Abstract

Mortality rate of breast cancer can be reduced bydetecting breast cancer in its early stage. Breast thermographyplays an important role in early detection of breast cancer, as itcan detect tumors when the physiological changes start in thebreast prior to structural changes. Computer Aided Detection(CAD) systems improve the diagnostic accuracy by providing adetailed analysis of images, which are not visible to the nakedeye. The performance of CAD systems depends on many factors.One of the important factors is the classifier used forclassification of breast thermograms. In this paper, we made acomparison of classifier performances using two ensembleclassifiers namely Ensemble Bagged Trees and AdaBoost.Spatial and spectral features are used for classification.Ensemble Bagged Trees classifier performed better thanAdaBoost in terms of accuracy of classification, but trainingtime required is higher than AdaBoost classifier. An accuracy of87%, sensitivity of 83% and specificity of 90.6% is obtainedusing Ensemble Bagged Trees classifier.

Original languageEnglish
Pages (from-to)35-39
Number of pages5
JournalJournal of Telecommunication, Electronic and Computer Engineering
Volume10
Issue number3-2
Publication statusPublished - 01-01-2018

Fingerprint

Classifiers
Adaptive boosting
Tumors

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

@article{221eaf045edf4bc89960f4d25f48a0b5,
title = "Detection of breast thermograms using ensemble classifiers",
abstract = "Mortality rate of breast cancer can be reduced bydetecting breast cancer in its early stage. Breast thermographyplays an important role in early detection of breast cancer, as itcan detect tumors when the physiological changes start in thebreast prior to structural changes. Computer Aided Detection(CAD) systems improve the diagnostic accuracy by providing adetailed analysis of images, which are not visible to the nakedeye. The performance of CAD systems depends on many factors.One of the important factors is the classifier used forclassification of breast thermograms. In this paper, we made acomparison of classifier performances using two ensembleclassifiers namely Ensemble Bagged Trees and AdaBoost.Spatial and spectral features are used for classification.Ensemble Bagged Trees classifier performed better thanAdaBoost in terms of accuracy of classification, but trainingtime required is higher than AdaBoost classifier. An accuracy of87{\%}, sensitivity of 83{\%} and specificity of 90.6{\%} is obtainedusing Ensemble Bagged Trees classifier.",
author = "Dayakshini Sathish and Surekha Kamath",
year = "2018",
month = "1",
day = "1",
language = "English",
volume = "10",
pages = "35--39",
journal = "Journal of Telecommunication, Electronic and Computer Engineering",
issn = "2180-1843",
publisher = "Universiti Teknikal Malaysia Melaka",
number = "3-2",

}

Detection of breast thermograms using ensemble classifiers. / Sathish, Dayakshini; Kamath, Surekha.

In: Journal of Telecommunication, Electronic and Computer Engineering, Vol. 10, No. 3-2, 01.01.2018, p. 35-39.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Detection of breast thermograms using ensemble classifiers

AU - Sathish, Dayakshini

AU - Kamath, Surekha

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Mortality rate of breast cancer can be reduced bydetecting breast cancer in its early stage. Breast thermographyplays an important role in early detection of breast cancer, as itcan detect tumors when the physiological changes start in thebreast prior to structural changes. Computer Aided Detection(CAD) systems improve the diagnostic accuracy by providing adetailed analysis of images, which are not visible to the nakedeye. The performance of CAD systems depends on many factors.One of the important factors is the classifier used forclassification of breast thermograms. In this paper, we made acomparison of classifier performances using two ensembleclassifiers namely Ensemble Bagged Trees and AdaBoost.Spatial and spectral features are used for classification.Ensemble Bagged Trees classifier performed better thanAdaBoost in terms of accuracy of classification, but trainingtime required is higher than AdaBoost classifier. An accuracy of87%, sensitivity of 83% and specificity of 90.6% is obtainedusing Ensemble Bagged Trees classifier.

AB - Mortality rate of breast cancer can be reduced bydetecting breast cancer in its early stage. Breast thermographyplays an important role in early detection of breast cancer, as itcan detect tumors when the physiological changes start in thebreast prior to structural changes. Computer Aided Detection(CAD) systems improve the diagnostic accuracy by providing adetailed analysis of images, which are not visible to the nakedeye. The performance of CAD systems depends on many factors.One of the important factors is the classifier used forclassification of breast thermograms. In this paper, we made acomparison of classifier performances using two ensembleclassifiers namely Ensemble Bagged Trees and AdaBoost.Spatial and spectral features are used for classification.Ensemble Bagged Trees classifier performed better thanAdaBoost in terms of accuracy of classification, but trainingtime required is higher than AdaBoost classifier. An accuracy of87%, sensitivity of 83% and specificity of 90.6% is obtainedusing Ensemble Bagged Trees classifier.

UR - http://www.scopus.com/inward/record.url?scp=85054522233&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054522233&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:85054522233

VL - 10

SP - 35

EP - 39

JO - Journal of Telecommunication, Electronic and Computer Engineering

JF - Journal of Telecommunication, Electronic and Computer Engineering

SN - 2180-1843

IS - 3-2

ER -