Improving the classification accuracy for IR spectroscopic diagnosis of stomach and colon malignancy using non-linear spectral feature extraction methods

Sanguk Lee, Kyoungok Kim, Hyeseon Lee, Chi Hyuck Jun, Hoeil Chung, Jong Jae Park

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Non-linear feature extraction methods, neighborhood preserving embedding (NPE) and supervised NPE (SNPE), were employed to effectively represent the IR spectral features of stomach and colon biopsy tissues for classification, and improve the classification accuracy for diagnosis of malignancy. The motivation was to utilize the NPE and SNPE's capability of capturing non-linear spectral behaviors by simultaneously preserving local relationships in order that minute spectral differences among classes would be effectively recognized. NPE and SNPE derive an optimal embedding feature such that the local neighborhood structure can be preserved in reduced spaces (variables). The IR spectra collected from stomach and colon tissues were represented by several new variables through NPE and SNPE, and also by using the principal component analysis (PCA). Then, the feature-extracted variables were subsequently classified into normal, adenoma and cancer tissues by using both fc-nearest neighbor (k- NN) and support vector machine (SVM), and the resulting accuracies were compared with each other. In both cases, the combination of SNPE-SVM provided the best classification performance, and the accuracy was substantially improved compared to when PCA-SVM was used. Overall results demonstrate that NPE and SNPE could be potential feature-representation strategies useful in biomedical diagnosis based on vibrational spectroscopy where effective recognition of minute spectral differences is critical.

Original languageEnglish
Pages (from-to)4076-4082
Number of pages7
JournalAnalyst
Volume138
Issue number14
DOIs
Publication statusPublished - 2013 Jul 21

Fingerprint

extraction method
Support vector machines
Feature extraction
Stomach
Colon
Tissue
Principal component analysis
Vibrational spectroscopy
Neoplasms
Biopsy
Principal Component Analysis
principal component analysis
Adenoma
cancer
Spectrum Analysis
spectroscopy
Support Vector Machine
support vector machine
tissue

ASJC Scopus subject areas

  • Analytical Chemistry
  • Environmental Chemistry
  • Biochemistry
  • Spectroscopy
  • Electrochemistry

Cite this

Improving the classification accuracy for IR spectroscopic diagnosis of stomach and colon malignancy using non-linear spectral feature extraction methods. / Lee, Sanguk; Kim, Kyoungok; Lee, Hyeseon; Jun, Chi Hyuck; Chung, Hoeil; Park, Jong Jae.

In: Analyst, Vol. 138, No. 14, 21.07.2013, p. 4076-4082.

Research output: Contribution to journalArticle

Lee, Sanguk ; Kim, Kyoungok ; Lee, Hyeseon ; Jun, Chi Hyuck ; Chung, Hoeil ; Park, Jong Jae. / Improving the classification accuracy for IR spectroscopic diagnosis of stomach and colon malignancy using non-linear spectral feature extraction methods. In: Analyst. 2013 ; Vol. 138, No. 14. pp. 4076-4082.
@article{0c0bcacde8054b02bc3fa2fe7dcb38b6,
title = "Improving the classification accuracy for IR spectroscopic diagnosis of stomach and colon malignancy using non-linear spectral feature extraction methods",
abstract = "Non-linear feature extraction methods, neighborhood preserving embedding (NPE) and supervised NPE (SNPE), were employed to effectively represent the IR spectral features of stomach and colon biopsy tissues for classification, and improve the classification accuracy for diagnosis of malignancy. The motivation was to utilize the NPE and SNPE's capability of capturing non-linear spectral behaviors by simultaneously preserving local relationships in order that minute spectral differences among classes would be effectively recognized. NPE and SNPE derive an optimal embedding feature such that the local neighborhood structure can be preserved in reduced spaces (variables). The IR spectra collected from stomach and colon tissues were represented by several new variables through NPE and SNPE, and also by using the principal component analysis (PCA). Then, the feature-extracted variables were subsequently classified into normal, adenoma and cancer tissues by using both fc-nearest neighbor (k- NN) and support vector machine (SVM), and the resulting accuracies were compared with each other. In both cases, the combination of SNPE-SVM provided the best classification performance, and the accuracy was substantially improved compared to when PCA-SVM was used. Overall results demonstrate that NPE and SNPE could be potential feature-representation strategies useful in biomedical diagnosis based on vibrational spectroscopy where effective recognition of minute spectral differences is critical.",
author = "Sanguk Lee and Kyoungok Kim and Hyeseon Lee and Jun, {Chi Hyuck} and Hoeil Chung and Park, {Jong Jae}",
year = "2013",
month = "7",
day = "21",
doi = "10.1039/c3an00256j",
language = "English",
volume = "138",
pages = "4076--4082",
journal = "The Analyst",
issn = "0003-2654",
publisher = "Royal Society of Chemistry",
number = "14",

}

TY - JOUR

T1 - Improving the classification accuracy for IR spectroscopic diagnosis of stomach and colon malignancy using non-linear spectral feature extraction methods

AU - Lee, Sanguk

AU - Kim, Kyoungok

AU - Lee, Hyeseon

AU - Jun, Chi Hyuck

AU - Chung, Hoeil

AU - Park, Jong Jae

PY - 2013/7/21

Y1 - 2013/7/21

N2 - Non-linear feature extraction methods, neighborhood preserving embedding (NPE) and supervised NPE (SNPE), were employed to effectively represent the IR spectral features of stomach and colon biopsy tissues for classification, and improve the classification accuracy for diagnosis of malignancy. The motivation was to utilize the NPE and SNPE's capability of capturing non-linear spectral behaviors by simultaneously preserving local relationships in order that minute spectral differences among classes would be effectively recognized. NPE and SNPE derive an optimal embedding feature such that the local neighborhood structure can be preserved in reduced spaces (variables). The IR spectra collected from stomach and colon tissues were represented by several new variables through NPE and SNPE, and also by using the principal component analysis (PCA). Then, the feature-extracted variables were subsequently classified into normal, adenoma and cancer tissues by using both fc-nearest neighbor (k- NN) and support vector machine (SVM), and the resulting accuracies were compared with each other. In both cases, the combination of SNPE-SVM provided the best classification performance, and the accuracy was substantially improved compared to when PCA-SVM was used. Overall results demonstrate that NPE and SNPE could be potential feature-representation strategies useful in biomedical diagnosis based on vibrational spectroscopy where effective recognition of minute spectral differences is critical.

AB - Non-linear feature extraction methods, neighborhood preserving embedding (NPE) and supervised NPE (SNPE), were employed to effectively represent the IR spectral features of stomach and colon biopsy tissues for classification, and improve the classification accuracy for diagnosis of malignancy. The motivation was to utilize the NPE and SNPE's capability of capturing non-linear spectral behaviors by simultaneously preserving local relationships in order that minute spectral differences among classes would be effectively recognized. NPE and SNPE derive an optimal embedding feature such that the local neighborhood structure can be preserved in reduced spaces (variables). The IR spectra collected from stomach and colon tissues were represented by several new variables through NPE and SNPE, and also by using the principal component analysis (PCA). Then, the feature-extracted variables were subsequently classified into normal, adenoma and cancer tissues by using both fc-nearest neighbor (k- NN) and support vector machine (SVM), and the resulting accuracies were compared with each other. In both cases, the combination of SNPE-SVM provided the best classification performance, and the accuracy was substantially improved compared to when PCA-SVM was used. Overall results demonstrate that NPE and SNPE could be potential feature-representation strategies useful in biomedical diagnosis based on vibrational spectroscopy where effective recognition of minute spectral differences is critical.

UR - http://www.scopus.com/inward/record.url?scp=84891698694&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84891698694&partnerID=8YFLogxK

U2 - 10.1039/c3an00256j

DO - 10.1039/c3an00256j

M3 - Article

C2 - 23687649

AN - SCOPUS:84891698694

VL - 138

SP - 4076

EP - 4082

JO - The Analyst

JF - The Analyst

SN - 0003-2654

IS - 14

ER -