Evaluation of automated and semi-automated skull-stripping algorithms using similarity index and segmentation error

The skull-stripping in the MR brain image appears to be a key issue in neuroimage analysis. In this paper, we evaluated the accuracy and efficiency of both automated and semi-automated skull-stripping methods. The evaluation was performed on both simulated and real data with the ground truth in skul...

Full description

Saved in:
Bibliographic Details
Published in:Computers in biology and medicine Vol. 33; no. 6; pp. 495 - 507
Main Authors: Lee, Jong-Min, Yoon, Uicheul, Nam, Sang Hee, Kim, Jung-Hyun, Kim, In-Young, Kim, Sun I.
Format: Journal Article
Language:English
Published: Oxford Elsevier Ltd 01.11.2003
New York, NY Elsevier Science
Elsevier Limited
Subjects:
Algorithms
Biological and medical sciences
Brain - anatomy & histology
False negative rate
False positive rate
Humans
Image Processing, Computer-Assisted
Inter-rater reliability
Magnetic Resonance Imaging - methods
Medical sciences
Reproducibility of Results
Similarity index
Simulated T1-weighted MRI
Skull-stripping
False positive rate
Inter-rater reliability
Similarity index
Skull-stripping
False negative rate
Simulated T1-weighted MRI
ISSN:0010-4825, 1879-0534
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The skull-stripping in the MR brain image appears to be a key issue in neuroimage analysis. In this paper, we evaluated the accuracy and efficiency of both automated and semi-automated skull-stripping methods. The evaluation was performed on both simulated and real data with the ground truth in skull-stripping. Although automated method showed better efficient results, it should require additional intervention. In contrast to that, semi-automated method showed better accurate results, but it was time consuming and prone to operator bias. Therefore, it might be practical that the semi-automated method was used as the post-processing of the automated one.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-News-1
content type line 14
ObjectType-Feature-3
ObjectType-Undefined-1
content type line 23
ISSN:0010-4825
1879-0534
DOI:10.1016/S0010-4825(03)00022-2