PLS-ANN based classification model for oral submucous fibrosis and oral carcinogenesis

Chih Yu Wang, Tsuimin Tsai, Hsin Ming Chen, Chin Tin Chen, Chun Pin Chiang

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

Background and Objectives: For effective management of oral neoplasia, autofluorescence spectroscopy was conducted on patients with different characteristics of oral lesions in vivo. This study tested the possibility of using a multivariate statistical algorithm to differentiate human oral premalignant and malignant lesions from benign lesions or normal oral mucosa. Study Design/Materials and Methods: A fiber optics-based fluorospectrometer was used to measure the autofluorescence spectra from healthy volunteers (NOM) and patients with oral lesions of submucous fibrosis (OSF), epithelial hyperkeratosis (EH), epithelial dysplasia (ED), and squamous cell carcinoma (SCC). A partial least-squares and artificial neural network (PLS-ANN) classification algorithm was used to characterize these oral lesions to discriminate premalignant (ED) and malignant (SCC) tissues from "benign" (NOM, OSF, and EH) tissues. Results: The normalized and centerized spectra of the different kinds of samples showed similar but divergent patterns. Our PLS-ANN classification algorithm could differentiate "premalignant and malignant" tissues from "benign" tissues with a sensitivity of 81%, a specificity of 96%, and a positive predictive value of 88%. Conclusions: We conclude that the PLS-ANN classification algorithm based on autofluorescence spectroscopy at 330-nm excitation is useful for in vivo diagnosis of OSF as well as oral premalignant and malignant lesions.

Original languageEnglish
Pages (from-to)318-326
Number of pages9
JournalLasers in Surgery and Medicine
Volume32
Issue number4
DOIs
Publication statusPublished - 2003

Fingerprint

Oral Submucous Fibrosis
Least-Squares Analysis
Carcinogenesis
Squamous Cell Carcinoma
Spectrum Analysis
Fibrosis
Myelinated Nerve Fibers
Mouth Mucosa
Healthy Volunteers
Epithelium
Neoplasms

Keywords

  • Artificial neural networks
  • Autofluorescence spectroscopy
  • Oral cancer diagnosis
  • Partial-least squares

ASJC Scopus subject areas

  • Surgery

Cite this

PLS-ANN based classification model for oral submucous fibrosis and oral carcinogenesis. / Wang, Chih Yu; Tsai, Tsuimin; Chen, Hsin Ming; Chen, Chin Tin; Chiang, Chun Pin.

In: Lasers in Surgery and Medicine, Vol. 32, No. 4, 2003, p. 318-326.

Research output: Contribution to journalArticle

Wang, Chih Yu ; Tsai, Tsuimin ; Chen, Hsin Ming ; Chen, Chin Tin ; Chiang, Chun Pin. / PLS-ANN based classification model for oral submucous fibrosis and oral carcinogenesis. In: Lasers in Surgery and Medicine. 2003 ; Vol. 32, No. 4. pp. 318-326.
@article{96b9e2a34ed9447d9c9e40eca03a7fc9,
title = "PLS-ANN based classification model for oral submucous fibrosis and oral carcinogenesis",
abstract = "Background and Objectives: For effective management of oral neoplasia, autofluorescence spectroscopy was conducted on patients with different characteristics of oral lesions in vivo. This study tested the possibility of using a multivariate statistical algorithm to differentiate human oral premalignant and malignant lesions from benign lesions or normal oral mucosa. Study Design/Materials and Methods: A fiber optics-based fluorospectrometer was used to measure the autofluorescence spectra from healthy volunteers (NOM) and patients with oral lesions of submucous fibrosis (OSF), epithelial hyperkeratosis (EH), epithelial dysplasia (ED), and squamous cell carcinoma (SCC). A partial least-squares and artificial neural network (PLS-ANN) classification algorithm was used to characterize these oral lesions to discriminate premalignant (ED) and malignant (SCC) tissues from {"}benign{"} (NOM, OSF, and EH) tissues. Results: The normalized and centerized spectra of the different kinds of samples showed similar but divergent patterns. Our PLS-ANN classification algorithm could differentiate {"}premalignant and malignant{"} tissues from {"}benign{"} tissues with a sensitivity of 81{\%}, a specificity of 96{\%}, and a positive predictive value of 88{\%}. Conclusions: We conclude that the PLS-ANN classification algorithm based on autofluorescence spectroscopy at 330-nm excitation is useful for in vivo diagnosis of OSF as well as oral premalignant and malignant lesions.",
keywords = "Artificial neural networks, Autofluorescence spectroscopy, Oral cancer diagnosis, Partial-least squares",
author = "Wang, {Chih Yu} and Tsuimin Tsai and Chen, {Hsin Ming} and Chen, {Chin Tin} and Chiang, {Chun Pin}",
year = "2003",
doi = "10.1002/lsm.10153",
language = "English",
volume = "32",
pages = "318--326",
journal = "Lasers in Surgery and Medicine",
issn = "0196-8092",
publisher = "Wiley-Liss Inc.",
number = "4",

}

TY - JOUR

T1 - PLS-ANN based classification model for oral submucous fibrosis and oral carcinogenesis

AU - Wang, Chih Yu

AU - Tsai, Tsuimin

AU - Chen, Hsin Ming

AU - Chen, Chin Tin

AU - Chiang, Chun Pin

PY - 2003

Y1 - 2003

N2 - Background and Objectives: For effective management of oral neoplasia, autofluorescence spectroscopy was conducted on patients with different characteristics of oral lesions in vivo. This study tested the possibility of using a multivariate statistical algorithm to differentiate human oral premalignant and malignant lesions from benign lesions or normal oral mucosa. Study Design/Materials and Methods: A fiber optics-based fluorospectrometer was used to measure the autofluorescence spectra from healthy volunteers (NOM) and patients with oral lesions of submucous fibrosis (OSF), epithelial hyperkeratosis (EH), epithelial dysplasia (ED), and squamous cell carcinoma (SCC). A partial least-squares and artificial neural network (PLS-ANN) classification algorithm was used to characterize these oral lesions to discriminate premalignant (ED) and malignant (SCC) tissues from "benign" (NOM, OSF, and EH) tissues. Results: The normalized and centerized spectra of the different kinds of samples showed similar but divergent patterns. Our PLS-ANN classification algorithm could differentiate "premalignant and malignant" tissues from "benign" tissues with a sensitivity of 81%, a specificity of 96%, and a positive predictive value of 88%. Conclusions: We conclude that the PLS-ANN classification algorithm based on autofluorescence spectroscopy at 330-nm excitation is useful for in vivo diagnosis of OSF as well as oral premalignant and malignant lesions.

AB - Background and Objectives: For effective management of oral neoplasia, autofluorescence spectroscopy was conducted on patients with different characteristics of oral lesions in vivo. This study tested the possibility of using a multivariate statistical algorithm to differentiate human oral premalignant and malignant lesions from benign lesions or normal oral mucosa. Study Design/Materials and Methods: A fiber optics-based fluorospectrometer was used to measure the autofluorescence spectra from healthy volunteers (NOM) and patients with oral lesions of submucous fibrosis (OSF), epithelial hyperkeratosis (EH), epithelial dysplasia (ED), and squamous cell carcinoma (SCC). A partial least-squares and artificial neural network (PLS-ANN) classification algorithm was used to characterize these oral lesions to discriminate premalignant (ED) and malignant (SCC) tissues from "benign" (NOM, OSF, and EH) tissues. Results: The normalized and centerized spectra of the different kinds of samples showed similar but divergent patterns. Our PLS-ANN classification algorithm could differentiate "premalignant and malignant" tissues from "benign" tissues with a sensitivity of 81%, a specificity of 96%, and a positive predictive value of 88%. Conclusions: We conclude that the PLS-ANN classification algorithm based on autofluorescence spectroscopy at 330-nm excitation is useful for in vivo diagnosis of OSF as well as oral premalignant and malignant lesions.

KW - Artificial neural networks

KW - Autofluorescence spectroscopy

KW - Oral cancer diagnosis

KW - Partial-least squares

UR - http://www.scopus.com/inward/record.url?scp=0037963251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037963251&partnerID=8YFLogxK

U2 - 10.1002/lsm.10153

DO - 10.1002/lsm.10153

M3 - Article

C2 - 12696101

AN - SCOPUS:0037963251

VL - 32

SP - 318

EP - 326

JO - Lasers in Surgery and Medicine

JF - Lasers in Surgery and Medicine

SN - 0196-8092

IS - 4

ER -