Cervical cancer detection using SVM based feature screening

Jiayong Zhang, Yanxi Liu

Research output: Contribution to journalConference article

39 Citations (Scopus)

Abstract

We present a novel feature screening algorithm by deriving relevance measures from the decision boundary of Support Vector Machines. It alleviates the "independence" assumption of traditional screening methods, e.g. those based on Information Gain and Augmented Variance Ratio, without sacrificing computational efficiency. We applied the proposed method to a bottom-up approach for automatic cervical cancer detection in multispectral microscopic thin PAP smear images. An initial set of around 4,000 multispectral texture features is effectively reduced to a computationally manageable size. The experimental results show significant improvements in pixel-level classification accuracy compared to traditional screening methods.

Original languageEnglish (US)
Pages (from-to)873-880
Number of pages8
JournalLecture Notes in Computer Science
Volume3217
Issue number1 PART 2
StatePublished - Dec 1 2004
EventMedical Image Computing and Computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings - Saint-Malo, France
Duration: Sep 26 2004Sep 29 2004

Fingerprint

Screening
Cancer
Variance Ratio
Information Gain
Texture Feature
Bottom-up
Computational efficiency
Computational Efficiency
Support vector machines
Support Vector Machine
Textures
Pixel
Pixels
Experimental Results
Relevance
Independence

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Zhang, Jiayong ; Liu, Yanxi. / Cervical cancer detection using SVM based feature screening. In: Lecture Notes in Computer Science. 2004 ; Vol. 3217, No. 1 PART 2. pp. 873-880.
@article{ce304460cdf1498f8817012595907a7c,
title = "Cervical cancer detection using SVM based feature screening",
abstract = "We present a novel feature screening algorithm by deriving relevance measures from the decision boundary of Support Vector Machines. It alleviates the {"}independence{"} assumption of traditional screening methods, e.g. those based on Information Gain and Augmented Variance Ratio, without sacrificing computational efficiency. We applied the proposed method to a bottom-up approach for automatic cervical cancer detection in multispectral microscopic thin PAP smear images. An initial set of around 4,000 multispectral texture features is effectively reduced to a computationally manageable size. The experimental results show significant improvements in pixel-level classification accuracy compared to traditional screening methods.",
author = "Jiayong Zhang and Yanxi Liu",
year = "2004",
month = "12",
day = "1",
language = "English (US)",
volume = "3217",
pages = "873--880",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",
number = "1 PART 2",

}

Cervical cancer detection using SVM based feature screening. / Zhang, Jiayong; Liu, Yanxi.

In: Lecture Notes in Computer Science, Vol. 3217, No. 1 PART 2, 01.12.2004, p. 873-880.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Cervical cancer detection using SVM based feature screening

AU - Zhang, Jiayong

AU - Liu, Yanxi

PY - 2004/12/1

Y1 - 2004/12/1

N2 - We present a novel feature screening algorithm by deriving relevance measures from the decision boundary of Support Vector Machines. It alleviates the "independence" assumption of traditional screening methods, e.g. those based on Information Gain and Augmented Variance Ratio, without sacrificing computational efficiency. We applied the proposed method to a bottom-up approach for automatic cervical cancer detection in multispectral microscopic thin PAP smear images. An initial set of around 4,000 multispectral texture features is effectively reduced to a computationally manageable size. The experimental results show significant improvements in pixel-level classification accuracy compared to traditional screening methods.

AB - We present a novel feature screening algorithm by deriving relevance measures from the decision boundary of Support Vector Machines. It alleviates the "independence" assumption of traditional screening methods, e.g. those based on Information Gain and Augmented Variance Ratio, without sacrificing computational efficiency. We applied the proposed method to a bottom-up approach for automatic cervical cancer detection in multispectral microscopic thin PAP smear images. An initial set of around 4,000 multispectral texture features is effectively reduced to a computationally manageable size. The experimental results show significant improvements in pixel-level classification accuracy compared to traditional screening methods.

UR - http://www.scopus.com/inward/record.url?scp=20344379432&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=20344379432&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:20344379432

VL - 3217

SP - 873

EP - 880

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

IS - 1 PART 2

ER -