A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography

Salih Burak Göktürk, Carlo Tomasi, Burak Acar, Christopher F. Beaulieu, David S. Paik, R. Brooke Jeffrey, Judy Yee, Sandy Napel

Research output: Contribution to journalArticle

130 Citations (Scopus)

Abstract

Adenomatous polyps in the colon are believed to be the precursor to colorectal carcinoma, the second leading cause of cancer deaths in United States. In this paper, we propose a new method for computer-aided detection of polyps in computed tomography (CT) colonography (virtual colonoscopy), a technique in which polyps are imaged along the wall of the air-inflated, cleansed colon with X-ray CT. Initial work with computer aided detection has shown high sensitivity, but at a cost of too many false positives. We present a statistical approach that uses support vector machines to distinguish the differentiating characteristics of polyps and healthy tissue, and uses this information for the classification of the new cases. One of the main contributions of the paper is the new three-dimensional pattern processing approach, called random orthogonal shape sections method, which combines the information from many random images to generate reliable signatures of shape. The input to the proposed system is a collection of volume data from candidate polyps obtained by a high-sensitivity, low-specificity system that we developed previously. The results of our tenfold cross-validation experiments show that, on the average, the system increases the specificity from 0.19 (0.35) to 0.69 (0.74) at a sensitivity level of 1.0 (0.95).

Original languageEnglish (US)
Pages (from-to)1251-1260
Number of pages10
JournalIEEE Transactions on Medical Imaging
Volume20
Issue number12
DOIs
StatePublished - Dec 1 2001
Externally publishedYes

Fingerprint

Polyps
Tomography
Information use
Processing
Support vector machines
Colon
Computed Tomographic Colonography
Tissue
Adenomatous Polyps
X rays
X Ray Computed Tomography
Air
Costs
Cause of Death
Colorectal Neoplasms
Experiments
Costs and Cost Analysis
Sensitivity and Specificity
Neoplasms

Keywords

  • Computer aided diagnosis
  • CT colonography
  • Pattern recognition
  • Random orthogonal shape section (ROSS) method
  • Support vector machines classification

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Göktürk, S. B., Tomasi, C., Acar, B., Beaulieu, C. F., Paik, D. S., Jeffrey, R. B., ... Napel, S. (2001). A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography. IEEE Transactions on Medical Imaging, 20(12), 1251-1260. https://doi.org/10.1109/42.974920

A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography. / Göktürk, Salih Burak; Tomasi, Carlo; Acar, Burak; Beaulieu, Christopher F.; Paik, David S.; Jeffrey, R. Brooke; Yee, Judy; Napel, Sandy.

In: IEEE Transactions on Medical Imaging, Vol. 20, No. 12, 01.12.2001, p. 1251-1260.

Research output: Contribution to journalArticle

Göktürk, SB, Tomasi, C, Acar, B, Beaulieu, CF, Paik, DS, Jeffrey, RB, Yee, J & Napel, S 2001, 'A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography', IEEE Transactions on Medical Imaging, vol. 20, no. 12, pp. 1251-1260. https://doi.org/10.1109/42.974920
Göktürk, Salih Burak ; Tomasi, Carlo ; Acar, Burak ; Beaulieu, Christopher F. ; Paik, David S. ; Jeffrey, R. Brooke ; Yee, Judy ; Napel, Sandy. / A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography. In: IEEE Transactions on Medical Imaging. 2001 ; Vol. 20, No. 12. pp. 1251-1260.
@article{67900fc9f82345a5a1445740e7b9dd32,
title = "A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography",
abstract = "Adenomatous polyps in the colon are believed to be the precursor to colorectal carcinoma, the second leading cause of cancer deaths in United States. In this paper, we propose a new method for computer-aided detection of polyps in computed tomography (CT) colonography (virtual colonoscopy), a technique in which polyps are imaged along the wall of the air-inflated, cleansed colon with X-ray CT. Initial work with computer aided detection has shown high sensitivity, but at a cost of too many false positives. We present a statistical approach that uses support vector machines to distinguish the differentiating characteristics of polyps and healthy tissue, and uses this information for the classification of the new cases. One of the main contributions of the paper is the new three-dimensional pattern processing approach, called random orthogonal shape sections method, which combines the information from many random images to generate reliable signatures of shape. The input to the proposed system is a collection of volume data from candidate polyps obtained by a high-sensitivity, low-specificity system that we developed previously. The results of our tenfold cross-validation experiments show that, on the average, the system increases the specificity from 0.19 (0.35) to 0.69 (0.74) at a sensitivity level of 1.0 (0.95).",
keywords = "Computer aided diagnosis, CT colonography, Pattern recognition, Random orthogonal shape section (ROSS) method, Support vector machines classification",
author = "G{\"o}kt{\"u}rk, {Salih Burak} and Carlo Tomasi and Burak Acar and Beaulieu, {Christopher F.} and Paik, {David S.} and Jeffrey, {R. Brooke} and Judy Yee and Sandy Napel",
year = "2001",
month = "12",
day = "1",
doi = "10.1109/42.974920",
language = "English (US)",
volume = "20",
pages = "1251--1260",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

TY - JOUR

T1 - A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography

AU - Göktürk, Salih Burak

AU - Tomasi, Carlo

AU - Acar, Burak

AU - Beaulieu, Christopher F.

AU - Paik, David S.

AU - Jeffrey, R. Brooke

AU - Yee, Judy

AU - Napel, Sandy

PY - 2001/12/1

Y1 - 2001/12/1

N2 - Adenomatous polyps in the colon are believed to be the precursor to colorectal carcinoma, the second leading cause of cancer deaths in United States. In this paper, we propose a new method for computer-aided detection of polyps in computed tomography (CT) colonography (virtual colonoscopy), a technique in which polyps are imaged along the wall of the air-inflated, cleansed colon with X-ray CT. Initial work with computer aided detection has shown high sensitivity, but at a cost of too many false positives. We present a statistical approach that uses support vector machines to distinguish the differentiating characteristics of polyps and healthy tissue, and uses this information for the classification of the new cases. One of the main contributions of the paper is the new three-dimensional pattern processing approach, called random orthogonal shape sections method, which combines the information from many random images to generate reliable signatures of shape. The input to the proposed system is a collection of volume data from candidate polyps obtained by a high-sensitivity, low-specificity system that we developed previously. The results of our tenfold cross-validation experiments show that, on the average, the system increases the specificity from 0.19 (0.35) to 0.69 (0.74) at a sensitivity level of 1.0 (0.95).

AB - Adenomatous polyps in the colon are believed to be the precursor to colorectal carcinoma, the second leading cause of cancer deaths in United States. In this paper, we propose a new method for computer-aided detection of polyps in computed tomography (CT) colonography (virtual colonoscopy), a technique in which polyps are imaged along the wall of the air-inflated, cleansed colon with X-ray CT. Initial work with computer aided detection has shown high sensitivity, but at a cost of too many false positives. We present a statistical approach that uses support vector machines to distinguish the differentiating characteristics of polyps and healthy tissue, and uses this information for the classification of the new cases. One of the main contributions of the paper is the new three-dimensional pattern processing approach, called random orthogonal shape sections method, which combines the information from many random images to generate reliable signatures of shape. The input to the proposed system is a collection of volume data from candidate polyps obtained by a high-sensitivity, low-specificity system that we developed previously. The results of our tenfold cross-validation experiments show that, on the average, the system increases the specificity from 0.19 (0.35) to 0.69 (0.74) at a sensitivity level of 1.0 (0.95).

KW - Computer aided diagnosis

KW - CT colonography

KW - Pattern recognition

KW - Random orthogonal shape section (ROSS) method

KW - Support vector machines classification

UR - http://www.scopus.com/inward/record.url?scp=0035544611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035544611&partnerID=8YFLogxK

U2 - 10.1109/42.974920

DO - 10.1109/42.974920

M3 - Article

C2 - 11811825

AN - SCOPUS:0035544611

VL - 20

SP - 1251

EP - 1260

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 12

ER -