Identification of cervical neoplasia using a simulation of human vision

Eileen D. Dickman, Theodore J. Doll, Chun Kit Chiu, Daron Gale Ferris

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Objective. To determine whether a computer simulation of human vision could detect and discriminate cervical cancer precursors and cancer from normal epithelium. Methods. The Georgia Tech Vision (TGV) system was trained to recognize normal and abnormal cervical features. GTV then assessed a new series of images to determine whether it could detect and diagnose cervical neoplasia without any a priori information about the images. Results. After training on one set and testing on another set of images. GTV demonstrated 100% sensitivity and 98% specificity to detect cervical intraepithelial neoplasia 3 (CIN3). GTV also had 88% sensitivity and 93% specificity to detect cervical cancer following training on one set and testing on another set of digitized cervical colposcopic images. Conclusion: This study demonstrated that GTV can detect CIN3 and cancer from digitized cervical colposcopic images. Furthermore, GTV could discriminate cervical cancer precursors and cancer from normal cervical epithelim, including immature metaplasia.

Original languageEnglish (US)
Pages (from-to)144-152
Number of pages9
JournalJournal of Lower Genital Tract Disease
Volume5
Issue number3
DOIs
StatePublished - Dec 27 2001

Fingerprint

Uterine Cervical Neoplasms
Cervical Intraepithelial Neoplasia
Neoplasms
Sensitivity and Specificity
Metaplasia
Computer Simulation
Epithelium

Keywords

  • Cervical neoplasia
  • Colposcopy
  • Computer-aided diagnosis
  • Image analysis

ASJC Scopus subject areas

  • Obstetrics and Gynecology

Cite this

Identification of cervical neoplasia using a simulation of human vision. / Dickman, Eileen D.; Doll, Theodore J.; Chiu, Chun Kit; Ferris, Daron Gale.

In: Journal of Lower Genital Tract Disease, Vol. 5, No. 3, 27.12.2001, p. 144-152.

Research output: Contribution to journalArticle

Dickman, Eileen D. ; Doll, Theodore J. ; Chiu, Chun Kit ; Ferris, Daron Gale. / Identification of cervical neoplasia using a simulation of human vision. In: Journal of Lower Genital Tract Disease. 2001 ; Vol. 5, No. 3. pp. 144-152.
@article{e0a355567bf940a6af3c6ab3e8cb08b7,
title = "Identification of cervical neoplasia using a simulation of human vision",
abstract = "Objective. To determine whether a computer simulation of human vision could detect and discriminate cervical cancer precursors and cancer from normal epithelium. Methods. The Georgia Tech Vision (TGV) system was trained to recognize normal and abnormal cervical features. GTV then assessed a new series of images to determine whether it could detect and diagnose cervical neoplasia without any a priori information about the images. Results. After training on one set and testing on another set of images. GTV demonstrated 100{\%} sensitivity and 98{\%} specificity to detect cervical intraepithelial neoplasia 3 (CIN3). GTV also had 88{\%} sensitivity and 93{\%} specificity to detect cervical cancer following training on one set and testing on another set of digitized cervical colposcopic images. Conclusion: This study demonstrated that GTV can detect CIN3 and cancer from digitized cervical colposcopic images. Furthermore, GTV could discriminate cervical cancer precursors and cancer from normal cervical epithelim, including immature metaplasia.",
keywords = "Cervical neoplasia, Colposcopy, Computer-aided diagnosis, Image analysis",
author = "Dickman, {Eileen D.} and Doll, {Theodore J.} and Chiu, {Chun Kit} and Ferris, {Daron Gale}",
year = "2001",
month = "12",
day = "27",
doi = "10.1046/j.1526-0976.2001.53006.x",
language = "English (US)",
volume = "5",
pages = "144--152",
journal = "Journal of Lower Genital Tract Disease",
issn = "1089-2591",
publisher = "Lippincott Williams and Wilkins",
number = "3",

}

TY - JOUR

T1 - Identification of cervical neoplasia using a simulation of human vision

AU - Dickman, Eileen D.

AU - Doll, Theodore J.

AU - Chiu, Chun Kit

AU - Ferris, Daron Gale

PY - 2001/12/27

Y1 - 2001/12/27

N2 - Objective. To determine whether a computer simulation of human vision could detect and discriminate cervical cancer precursors and cancer from normal epithelium. Methods. The Georgia Tech Vision (TGV) system was trained to recognize normal and abnormal cervical features. GTV then assessed a new series of images to determine whether it could detect and diagnose cervical neoplasia without any a priori information about the images. Results. After training on one set and testing on another set of images. GTV demonstrated 100% sensitivity and 98% specificity to detect cervical intraepithelial neoplasia 3 (CIN3). GTV also had 88% sensitivity and 93% specificity to detect cervical cancer following training on one set and testing on another set of digitized cervical colposcopic images. Conclusion: This study demonstrated that GTV can detect CIN3 and cancer from digitized cervical colposcopic images. Furthermore, GTV could discriminate cervical cancer precursors and cancer from normal cervical epithelim, including immature metaplasia.

AB - Objective. To determine whether a computer simulation of human vision could detect and discriminate cervical cancer precursors and cancer from normal epithelium. Methods. The Georgia Tech Vision (TGV) system was trained to recognize normal and abnormal cervical features. GTV then assessed a new series of images to determine whether it could detect and diagnose cervical neoplasia without any a priori information about the images. Results. After training on one set and testing on another set of images. GTV demonstrated 100% sensitivity and 98% specificity to detect cervical intraepithelial neoplasia 3 (CIN3). GTV also had 88% sensitivity and 93% specificity to detect cervical cancer following training on one set and testing on another set of digitized cervical colposcopic images. Conclusion: This study demonstrated that GTV can detect CIN3 and cancer from digitized cervical colposcopic images. Furthermore, GTV could discriminate cervical cancer precursors and cancer from normal cervical epithelim, including immature metaplasia.

KW - Cervical neoplasia

KW - Colposcopy

KW - Computer-aided diagnosis

KW - Image analysis

UR - http://www.scopus.com/inward/record.url?scp=0035213617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035213617&partnerID=8YFLogxK

U2 - 10.1046/j.1526-0976.2001.53006.x

DO - 10.1046/j.1526-0976.2001.53006.x

M3 - Article

VL - 5

SP - 144

EP - 152

JO - Journal of Lower Genital Tract Disease

JF - Journal of Lower Genital Tract Disease

SN - 1089-2591

IS - 3

ER -