Reduction of breast biopsies with a modified self-organizing map

Yi Zheng, James F Greenleaf, John J. Gisvold

Research output: Contribution to journalArticle

44 Citations (Scopus)

Abstract

A modified self-organizing map with nonlinear weight adjustments has been applied to reduce the number of breast biopsies necessary for breast cancer diagnosis. Tissue features representing texture information from digital sonographic breast images were extracted from sonograms of benign and malignant breast tumors. The resulting hyperspace of data points was then used in a modified self-organizing map that objectively segments population distributions of lesions and accurately establishes benign and malignant regions. These methods were applied to a group of 102 problematic breast cases with sonographic images, including 34 with malignant lesions. All lesions were substantiated by excisional biopsy. The system can isolate clusters of purely benign lesions from other clusters containing both benign and malignant lesions. The hybrid neural network defined a region in which about 60% of the benign lesions were located exclusive of any malignant lesions. Using a hybrid approach and leave-one-out method of data evaluation, we estimate that the number of biopsies in this group of women could be decreased by 40-59% with high confidence and that no malignancies were included in the nonbiopsied group. The experimental results also suggest that the modified self-organizing map provides more accurate population distribution maps than conventional Kohonen maps.

Original languageEnglish (US)
Pages (from-to)1386-1396
Number of pages11
JournalIEEE Transactions on Neural Networks
Volume8
Issue number6
DOIs
StatePublished - 1997

Fingerprint

Biopsy
Self organizing maps
Self-organizing Map
Population distribution
Hyperspace
Hybrid Approach
Breast Cancer
Confidence
Texture
Tumor
Adjustment
Neural Networks
Tumors
Necessary
Textures
Evaluation
Experimental Results
Tissue
Neural networks
Estimate

Keywords

  • Breast biopsy
  • Computer vision radiologist assistant
  • Modified self-organized map
  • Texture analysis
  • Ultrasound image

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Electrical and Electronic Engineering
  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Hardware and Architecture

Cite this

Reduction of breast biopsies with a modified self-organizing map. / Zheng, Yi; Greenleaf, James F; Gisvold, John J.

In: IEEE Transactions on Neural Networks, Vol. 8, No. 6, 1997, p. 1386-1396.

Research output: Contribution to journalArticle

@article{de88df27a36f41c1b31f79eeb3257cab,
title = "Reduction of breast biopsies with a modified self-organizing map",
abstract = "A modified self-organizing map with nonlinear weight adjustments has been applied to reduce the number of breast biopsies necessary for breast cancer diagnosis. Tissue features representing texture information from digital sonographic breast images were extracted from sonograms of benign and malignant breast tumors. The resulting hyperspace of data points was then used in a modified self-organizing map that objectively segments population distributions of lesions and accurately establishes benign and malignant regions. These methods were applied to a group of 102 problematic breast cases with sonographic images, including 34 with malignant lesions. All lesions were substantiated by excisional biopsy. The system can isolate clusters of purely benign lesions from other clusters containing both benign and malignant lesions. The hybrid neural network defined a region in which about 60{\%} of the benign lesions were located exclusive of any malignant lesions. Using a hybrid approach and leave-one-out method of data evaluation, we estimate that the number of biopsies in this group of women could be decreased by 40-59{\%} with high confidence and that no malignancies were included in the nonbiopsied group. The experimental results also suggest that the modified self-organizing map provides more accurate population distribution maps than conventional Kohonen maps.",
keywords = "Breast biopsy, Computer vision radiologist assistant, Modified self-organized map, Texture analysis, Ultrasound image",
author = "Yi Zheng and Greenleaf, {James F} and Gisvold, {John J.}",
year = "1997",
doi = "10.1109/72.641462",
language = "English (US)",
volume = "8",
pages = "1386--1396",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "6",

}

TY - JOUR

T1 - Reduction of breast biopsies with a modified self-organizing map

AU - Zheng, Yi

AU - Greenleaf, James F

AU - Gisvold, John J.

PY - 1997

Y1 - 1997

N2 - A modified self-organizing map with nonlinear weight adjustments has been applied to reduce the number of breast biopsies necessary for breast cancer diagnosis. Tissue features representing texture information from digital sonographic breast images were extracted from sonograms of benign and malignant breast tumors. The resulting hyperspace of data points was then used in a modified self-organizing map that objectively segments population distributions of lesions and accurately establishes benign and malignant regions. These methods were applied to a group of 102 problematic breast cases with sonographic images, including 34 with malignant lesions. All lesions were substantiated by excisional biopsy. The system can isolate clusters of purely benign lesions from other clusters containing both benign and malignant lesions. The hybrid neural network defined a region in which about 60% of the benign lesions were located exclusive of any malignant lesions. Using a hybrid approach and leave-one-out method of data evaluation, we estimate that the number of biopsies in this group of women could be decreased by 40-59% with high confidence and that no malignancies were included in the nonbiopsied group. The experimental results also suggest that the modified self-organizing map provides more accurate population distribution maps than conventional Kohonen maps.

AB - A modified self-organizing map with nonlinear weight adjustments has been applied to reduce the number of breast biopsies necessary for breast cancer diagnosis. Tissue features representing texture information from digital sonographic breast images were extracted from sonograms of benign and malignant breast tumors. The resulting hyperspace of data points was then used in a modified self-organizing map that objectively segments population distributions of lesions and accurately establishes benign and malignant regions. These methods were applied to a group of 102 problematic breast cases with sonographic images, including 34 with malignant lesions. All lesions were substantiated by excisional biopsy. The system can isolate clusters of purely benign lesions from other clusters containing both benign and malignant lesions. The hybrid neural network defined a region in which about 60% of the benign lesions were located exclusive of any malignant lesions. Using a hybrid approach and leave-one-out method of data evaluation, we estimate that the number of biopsies in this group of women could be decreased by 40-59% with high confidence and that no malignancies were included in the nonbiopsied group. The experimental results also suggest that the modified self-organizing map provides more accurate population distribution maps than conventional Kohonen maps.

KW - Breast biopsy

KW - Computer vision radiologist assistant

KW - Modified self-organized map

KW - Texture analysis

KW - Ultrasound image

UR - http://www.scopus.com/inward/record.url?scp=0031274866&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031274866&partnerID=8YFLogxK

U2 - 10.1109/72.641462

DO - 10.1109/72.641462

M3 - Article

VL - 8

SP - 1386

EP - 1396

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 6

ER -