Classification improvement by segmentation refinement: Application to contrast-enhanced MR-mammography

C Barillot (Editor), D R Haynor (Editor), P Hellier (Editor), M O Leach, D J Hawkes

Research output: Chapter in Book/Report/Conference proceedingConference paper

11 Citations (Scopus)

Abstract

In this study we investigated whether automatic refinement of manually segmented MR breast lesions improves the discrimination of benign and malignant breast lesions. A constrained maximum a-posteriori scheme was employed to extract the most probable lesion for a user-provided coarse manual segmentation. Standard shape, texture and contrast enhancement features were derived from both the manual and the refined segmentations for 10 benign and 16 malignant lesions and their discrimination ability was compared. The refined segmentations were more consistent than the manual segmentations from a radiologist and a non-expert. The automatic refinement was robust to inaccuracies of the manual segmentation. Classification accuracy improved on average from 69% to 82% after segmentation refinement.
Original languageEnglish
Title of host publicationLECT NOTE COMPUT SCI
Place of PublicationBERLIN
PublisherSpringer
Pages184 - 191
Number of pages8
ISBN (Print)3-540-22976-0
Publication statusPublished - 2004
Event7th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2004) - St Malo, France
Duration: 1 Jan 2004 → …

Publication series

NameLECTURE NOTES IN COMPUTER SCIENCE

Conference

Conference7th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2004)
Country/TerritoryFrance
CitySt Malo
Period1/01/2004 → …

Fingerprint

Dive into the research topics of 'Classification improvement by segmentation refinement: Application to contrast-enhanced MR-mammography'. Together they form a unique fingerprint.

Cite this