An effective classification procedure for diagnosis of prostate cancer in near infrared spectra

Seoung Bum Kim, Chivalai Temiyasathit, Karim Bensalah, Altug Tuncel, Jeffrey Cadeddu, Wareef Kabbani, Aditya V. Mathker, Hanli Liu

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

The main purpose of this study is to develop an effective classification procedure that discriminates between normal spectra and cancerous spectra in near infrared (NIR) spectroscopic data in which the classes are highly imbalanced and overlapped. Our proposed procedure consists of several steps. First, to ensure the comparability between spectra, normalization was done by dividing each spectral point by the area of the total intensity of the spectrum. Second, clustering analysis was performed with these normalized spectra to separate the spectra that represent the normal pattern from a mixed group that contains both normal and tumor spectra. Third, we conducted two-stage classification, the first being an effort to construct a classification model with the labels obtained from the preceding clustering analysis and the second being a classification to focus on the mixed group classified from the first classification model. To increase the accuracy, the second classification model was constructed based on the selected features that capture important characteristics of the spectral data. Our proposed procedure was evaluated by its classification ability in testing samples using a leave-one-out cross validation technique, yielding acceptable classification accuracy.

Original languageEnglish
Pages (from-to)3863-3869
Number of pages7
JournalExpert Systems with Applications
Volume37
Issue number5
DOIs
Publication statusPublished - 2010 May 1

Fingerprint

Infrared radiation
Labels
Tumors
Testing

Keywords

  • Class imbalance
  • Classification
  • Clustering
  • Near infrared
  • Prostate cancer

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Engineering(all)

Cite this

An effective classification procedure for diagnosis of prostate cancer in near infrared spectra. / Kim, Seoung Bum; Temiyasathit, Chivalai; Bensalah, Karim; Tuncel, Altug; Cadeddu, Jeffrey; Kabbani, Wareef; Mathker, Aditya V.; Liu, Hanli.

In: Expert Systems with Applications, Vol. 37, No. 5, 01.05.2010, p. 3863-3869.

Research output: Contribution to journalArticle

Kim, SB, Temiyasathit, C, Bensalah, K, Tuncel, A, Cadeddu, J, Kabbani, W, Mathker, AV & Liu, H 2010, 'An effective classification procedure for diagnosis of prostate cancer in near infrared spectra', Expert Systems with Applications, vol. 37, no. 5, pp. 3863-3869. https://doi.org/10.1016/j.eswa.2009.11.032
Kim, Seoung Bum ; Temiyasathit, Chivalai ; Bensalah, Karim ; Tuncel, Altug ; Cadeddu, Jeffrey ; Kabbani, Wareef ; Mathker, Aditya V. ; Liu, Hanli. / An effective classification procedure for diagnosis of prostate cancer in near infrared spectra. In: Expert Systems with Applications. 2010 ; Vol. 37, No. 5. pp. 3863-3869.
@article{b82cace397dc4454b17e27b44d591368,
title = "An effective classification procedure for diagnosis of prostate cancer in near infrared spectra",
abstract = "The main purpose of this study is to develop an effective classification procedure that discriminates between normal spectra and cancerous spectra in near infrared (NIR) spectroscopic data in which the classes are highly imbalanced and overlapped. Our proposed procedure consists of several steps. First, to ensure the comparability between spectra, normalization was done by dividing each spectral point by the area of the total intensity of the spectrum. Second, clustering analysis was performed with these normalized spectra to separate the spectra that represent the normal pattern from a mixed group that contains both normal and tumor spectra. Third, we conducted two-stage classification, the first being an effort to construct a classification model with the labels obtained from the preceding clustering analysis and the second being a classification to focus on the mixed group classified from the first classification model. To increase the accuracy, the second classification model was constructed based on the selected features that capture important characteristics of the spectral data. Our proposed procedure was evaluated by its classification ability in testing samples using a leave-one-out cross validation technique, yielding acceptable classification accuracy.",
keywords = "Class imbalance, Classification, Clustering, Near infrared, Prostate cancer",
author = "Kim, {Seoung Bum} and Chivalai Temiyasathit and Karim Bensalah and Altug Tuncel and Jeffrey Cadeddu and Wareef Kabbani and Mathker, {Aditya V.} and Hanli Liu",
year = "2010",
month = "5",
day = "1",
doi = "10.1016/j.eswa.2009.11.032",
language = "English",
volume = "37",
pages = "3863--3869",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",
number = "5",

}

TY - JOUR

T1 - An effective classification procedure for diagnosis of prostate cancer in near infrared spectra

AU - Kim, Seoung Bum

AU - Temiyasathit, Chivalai

AU - Bensalah, Karim

AU - Tuncel, Altug

AU - Cadeddu, Jeffrey

AU - Kabbani, Wareef

AU - Mathker, Aditya V.

AU - Liu, Hanli

PY - 2010/5/1

Y1 - 2010/5/1

N2 - The main purpose of this study is to develop an effective classification procedure that discriminates between normal spectra and cancerous spectra in near infrared (NIR) spectroscopic data in which the classes are highly imbalanced and overlapped. Our proposed procedure consists of several steps. First, to ensure the comparability between spectra, normalization was done by dividing each spectral point by the area of the total intensity of the spectrum. Second, clustering analysis was performed with these normalized spectra to separate the spectra that represent the normal pattern from a mixed group that contains both normal and tumor spectra. Third, we conducted two-stage classification, the first being an effort to construct a classification model with the labels obtained from the preceding clustering analysis and the second being a classification to focus on the mixed group classified from the first classification model. To increase the accuracy, the second classification model was constructed based on the selected features that capture important characteristics of the spectral data. Our proposed procedure was evaluated by its classification ability in testing samples using a leave-one-out cross validation technique, yielding acceptable classification accuracy.

AB - The main purpose of this study is to develop an effective classification procedure that discriminates between normal spectra and cancerous spectra in near infrared (NIR) spectroscopic data in which the classes are highly imbalanced and overlapped. Our proposed procedure consists of several steps. First, to ensure the comparability between spectra, normalization was done by dividing each spectral point by the area of the total intensity of the spectrum. Second, clustering analysis was performed with these normalized spectra to separate the spectra that represent the normal pattern from a mixed group that contains both normal and tumor spectra. Third, we conducted two-stage classification, the first being an effort to construct a classification model with the labels obtained from the preceding clustering analysis and the second being a classification to focus on the mixed group classified from the first classification model. To increase the accuracy, the second classification model was constructed based on the selected features that capture important characteristics of the spectral data. Our proposed procedure was evaluated by its classification ability in testing samples using a leave-one-out cross validation technique, yielding acceptable classification accuracy.

KW - Class imbalance

KW - Classification

KW - Clustering

KW - Near infrared

KW - Prostate cancer

UR - http://www.scopus.com/inward/record.url?scp=73749083191&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=73749083191&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2009.11.032

DO - 10.1016/j.eswa.2009.11.032

M3 - Article

AN - SCOPUS:73749083191

VL - 37

SP - 3863

EP - 3869

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 5

ER -