An Automated Method for Segmenting White Matter Lesions through Multi-Level Morphometric Feature Classification with Application to Lupus

Mark Scully; Blake Anderson; Terran Lane; Charles Gasparovic; Vince Magnotta; Wilmer Sibbitt; Carlos Roldan; Ron Kikinis; Henry J Bockholt

doi:10.3389/fnhum.2010.00027

Back

An Automated Method for Segmenting White Matter Lesions through Multi-Level Morphometric Feature Classification with Application to Lupus

Journal article

Open access

Peer reviewed

An Automated Method for Segmenting White Matter Lesions through Multi-Level Morphometric Feature Classification with Application to Lupus

Mark Scully, Blake Anderson, Terran Lane, Charles Gasparovic, Vince Magnotta, Wilmer Sibbitt, Carlos Roldan, Ron Kikinis and Henry J Bockholt

Frontiers in human neuroscience, Vol.4, pp.27-27

2010

DOI: 10.3389/fnhum.2010.00027

PMCID: PMC2859868

PMID: 20428508

Files and links (1)

url

https://doi.org/10.3389/fnhum.2010.00027View

Published (Version of record) Open Access

Abstract

We demonstrate an automated, multi-level method to segment white matter brain lesions and apply it to lupus. The method makes use of local morphometric features based on multiple MR sequences, including T1-weighted, T2-weighted, and fluid attenuated inversion recovery. After preprocessing, including co-registration, brain extraction, bias correction, and intensity standardization, 49 features are calculated for each brain voxel based on local morphometry. At each level of segmentation a supervised classifier takes advantage of a different subset of the features to conservatively segment lesion voxels, passing on more difficult voxels to the next classifier. This multi-level approach allows for a fast lesion classification method with tunable trade-offs between sensitivity and specificity producing accuracy comparable to a human rater.

Neuroscience

support vector machine

method

segmentation

machine learning

classification

lupus

lesion

Details

Title: Subtitle: An Automated Method for Segmenting White Matter Lesions through Multi-Level Morphometric Feature Classification with Application to Lupus
Creators: Mark Scully - The Mind Research Network
Blake Anderson - Department of Computer Science, The University of New Mexico
Terran Lane - Department of Computer Science, The University of New Mexico
Charles Gasparovic - The Mind Research Network
Vince Magnotta - Radiology Department, Carver School of Medicine, The University of Iowa
Wilmer Sibbitt - Rheumatology, Department of Internal Medicine, School of Medicine, The University of New Mexico
Carlos Roldan - Cardiology, Department of Internal Medicine, School of Medicine, The University of New Mexico
Ron Kikinis - Surgical Planning Laboratory, Brigham and Women's Hospital, Harvard School of Medicine
Henry J Bockholt - The Mind Research Network
Resource Type: Journal article
Publication Details: Frontiers in human neuroscience, Vol.4, pp.27-27
DOI: 10.3389/fnhum.2010.00027
PMID: 20428508
PMCID: PMC2859868
NLM abbreviation: Front Hum Neurosci
ISSN: 1662-5161
eISSN: 1662-5161
Publisher: Frontiers Research Foundation
Language: English
Date published: 2010
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Radiology; Psychiatry; Iowa Neuroscience Institute
Record Identifier: 9984051780202771

Metrics

21 Record Views

20 Times Cited - Web of Science