Voxel-wise displacement as independent features in classification of multiple sclerosis

Min Chen, Aaron Carass, Daniel S. Reich, Peter Calabresi, Dzung Pham, Jerry Ladd Prince

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

Abstract

We present a method that utilizes registration displacement fields to perform accurate classification of magnetic resonance images (MRI) of the brain acquired from healthy individuals and patients diagnosed with multiple sclerosis (MS). Contrary to standard approaches, each voxel in the displacement field is treated as an independent feature that is classified individually. Results show that when used with a simple linear discriminant and majority voting, the approach is superior to using the displacement field with a single classifier, even when compared against more sophisticated classification methods such as adaptive boosting, random forests, and support vector machines. Leave-one-out cross-validation was used to evaluate this method for classifying images by disease, MS subtype (Acc: 77%{88%), and age (Acc: 96%{100%).

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume8669
DOIs
StatePublished - 2013
EventMedical Imaging 2013: Image Processing - Lake Buena Vista, FL, United States
Duration: Feb 10 2013Feb 12 2013

Other

OtherMedical Imaging 2013: Image Processing
CountryUnited States
CityLake Buena Vista, FL
Period2/10/132/12/13

Fingerprint

Multiple Sclerosis
Adaptive boosting
Magnetic resonance
voting
Support vector machines
Brain
Classifiers
Politics
classifiers
classifying
brain
magnetic resonance
Magnetic Resonance Spectroscopy
Support Vector Machine
Forests

Keywords

  • Classification
  • Image registration
  • Magnetic resonance imaging
  • Multiple sclerosis

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Chen, M., Carass, A., Reich, D. S., Calabresi, P., Pham, D., & Prince, J. L. (2013). Voxel-wise displacement as independent features in classification of multiple sclerosis. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 8669). [86690K] https://doi.org/10.1117/12.2007150

Voxel-wise displacement as independent features in classification of multiple sclerosis. / Chen, Min; Carass, Aaron; Reich, Daniel S.; Calabresi, Peter; Pham, Dzung; Prince, Jerry Ladd.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8669 2013. 86690K.

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

Chen, M, Carass, A, Reich, DS, Calabresi, P, Pham, D & Prince, JL 2013, Voxel-wise displacement as independent features in classification of multiple sclerosis. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 8669, 86690K, Medical Imaging 2013: Image Processing, Lake Buena Vista, FL, United States, 2/10/13. https://doi.org/10.1117/12.2007150
Chen M, Carass A, Reich DS, Calabresi P, Pham D, Prince JL. Voxel-wise displacement as independent features in classification of multiple sclerosis. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8669. 2013. 86690K https://doi.org/10.1117/12.2007150
Chen, Min ; Carass, Aaron ; Reich, Daniel S. ; Calabresi, Peter ; Pham, Dzung ; Prince, Jerry Ladd. / Voxel-wise displacement as independent features in classification of multiple sclerosis. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8669 2013.
@inproceedings{b86106ac84b54d9b81d5b6f9f4c689de,
title = "Voxel-wise displacement as independent features in classification of multiple sclerosis",
abstract = "We present a method that utilizes registration displacement fields to perform accurate classification of magnetic resonance images (MRI) of the brain acquired from healthy individuals and patients diagnosed with multiple sclerosis (MS). Contrary to standard approaches, each voxel in the displacement field is treated as an independent feature that is classified individually. Results show that when used with a simple linear discriminant and majority voting, the approach is superior to using the displacement field with a single classifier, even when compared against more sophisticated classification methods such as adaptive boosting, random forests, and support vector machines. Leave-one-out cross-validation was used to evaluate this method for classifying images by disease, MS subtype (Acc: 77{\%}{88{\%}), and age (Acc: 96{\%}{100{\%}).",
keywords = "Classification, Image registration, Magnetic resonance imaging, Multiple sclerosis",
author = "Min Chen and Aaron Carass and Reich, {Daniel S.} and Peter Calabresi and Dzung Pham and Prince, {Jerry Ladd}",
year = "2013",
doi = "10.1117/12.2007150",
language = "English (US)",
isbn = "9780819494436",
volume = "8669",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Voxel-wise displacement as independent features in classification of multiple sclerosis

AU - Chen, Min

AU - Carass, Aaron

AU - Reich, Daniel S.

AU - Calabresi, Peter

AU - Pham, Dzung

AU - Prince, Jerry Ladd

PY - 2013

Y1 - 2013

N2 - We present a method that utilizes registration displacement fields to perform accurate classification of magnetic resonance images (MRI) of the brain acquired from healthy individuals and patients diagnosed with multiple sclerosis (MS). Contrary to standard approaches, each voxel in the displacement field is treated as an independent feature that is classified individually. Results show that when used with a simple linear discriminant and majority voting, the approach is superior to using the displacement field with a single classifier, even when compared against more sophisticated classification methods such as adaptive boosting, random forests, and support vector machines. Leave-one-out cross-validation was used to evaluate this method for classifying images by disease, MS subtype (Acc: 77%{88%), and age (Acc: 96%{100%).

AB - We present a method that utilizes registration displacement fields to perform accurate classification of magnetic resonance images (MRI) of the brain acquired from healthy individuals and patients diagnosed with multiple sclerosis (MS). Contrary to standard approaches, each voxel in the displacement field is treated as an independent feature that is classified individually. Results show that when used with a simple linear discriminant and majority voting, the approach is superior to using the displacement field with a single classifier, even when compared against more sophisticated classification methods such as adaptive boosting, random forests, and support vector machines. Leave-one-out cross-validation was used to evaluate this method for classifying images by disease, MS subtype (Acc: 77%{88%), and age (Acc: 96%{100%).

KW - Classification

KW - Image registration

KW - Magnetic resonance imaging

KW - Multiple sclerosis

UR - http://www.scopus.com/inward/record.url?scp=84878309669&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878309669&partnerID=8YFLogxK

U2 - 10.1117/12.2007150

DO - 10.1117/12.2007150

M3 - Conference contribution

C2 - 24236227

AN - SCOPUS:84878309669

SN - 9780819494436

VL - 8669

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

ER -