Automated classification of fMRI data employing trial-based imagery tasks

Jong-Hwan Lee, Matthew Marzelli, Ferenc A. Jolesz, Seung Schik Yoo

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Automated interpretation and classification of functional MRI (fMRI) data is an emerging research field that enables the characterization of underlying cognitive processes with minimal human intervention. In this work, we present a method for the automated classification of human thoughts reflected on a trial-based paradigm using fMRI with a significantly shortened data acquisition time (less than one minute). Based on our preliminary experience with various cognitive imagery tasks, six characteristic thoughts were chosen as target tasks for the present work: right-hand motor imagery, left-hand motor imagery, right foot motor imagery, mental calculation, internal speech/word generation, and visual imagery. These six tasks were performed by five healthy volunteers and functional images were obtained using a T2*-weighted echo planar imaging (EPI) sequence. Feature vectors from activation maps, necessary for the classification of neural activity, were automatically extracted from the regions that were consistently and exclusively activated for a given task during the training process. Extracted feature vectors were classified using the support vector machine (SVM) algorithm. Parameter optimization, using a k-fold cross validation scheme, allowed the successful recognition of the six different categories of administered thought tasks with an accuracy of 74.5% (mean) ± 14.3% (standard deviation) across all five subjects. Our proposed study for the automated classification of fMRI data may be utilized in further investigations to monitor/identify human thought processes and their potential link to hardware/computer control.

Original languageEnglish
Pages (from-to)392-404
Number of pages13
JournalMedical Image Analysis
Volume13
Issue number3
DOIs
Publication statusPublished - 2009 Jun 1
Externally publishedYes

Fingerprint

Imagery (Psychotherapy)
Magnetic Resonance Imaging
Hand
Computer control
Echo-Planar Imaging
Support vector machines
Data acquisition
Chemical activation
Foot
Healthy Volunteers
Hardware
Imaging techniques
Research

Keywords

  • Brain-computer interface (BCI)
  • Functional MRI
  • Neuroimaging
  • Pattern recognition
  • Support vector machine (SVM)

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

Automated classification of fMRI data employing trial-based imagery tasks. / Lee, Jong-Hwan; Marzelli, Matthew; Jolesz, Ferenc A.; Yoo, Seung Schik.

In: Medical Image Analysis, Vol. 13, No. 3, 01.06.2009, p. 392-404.

Research output: Contribution to journalArticle

Lee, Jong-Hwan ; Marzelli, Matthew ; Jolesz, Ferenc A. ; Yoo, Seung Schik. / Automated classification of fMRI data employing trial-based imagery tasks. In: Medical Image Analysis. 2009 ; Vol. 13, No. 3. pp. 392-404.
@article{7652e6a8fd174eccabd727b037825eef,
title = "Automated classification of fMRI data employing trial-based imagery tasks",
abstract = "Automated interpretation and classification of functional MRI (fMRI) data is an emerging research field that enables the characterization of underlying cognitive processes with minimal human intervention. In this work, we present a method for the automated classification of human thoughts reflected on a trial-based paradigm using fMRI with a significantly shortened data acquisition time (less than one minute). Based on our preliminary experience with various cognitive imagery tasks, six characteristic thoughts were chosen as target tasks for the present work: right-hand motor imagery, left-hand motor imagery, right foot motor imagery, mental calculation, internal speech/word generation, and visual imagery. These six tasks were performed by five healthy volunteers and functional images were obtained using a T2*-weighted echo planar imaging (EPI) sequence. Feature vectors from activation maps, necessary for the classification of neural activity, were automatically extracted from the regions that were consistently and exclusively activated for a given task during the training process. Extracted feature vectors were classified using the support vector machine (SVM) algorithm. Parameter optimization, using a k-fold cross validation scheme, allowed the successful recognition of the six different categories of administered thought tasks with an accuracy of 74.5{\%} (mean) ± 14.3{\%} (standard deviation) across all five subjects. Our proposed study for the automated classification of fMRI data may be utilized in further investigations to monitor/identify human thought processes and their potential link to hardware/computer control.",
keywords = "Brain-computer interface (BCI), Functional MRI, Neuroimaging, Pattern recognition, Support vector machine (SVM)",
author = "Jong-Hwan Lee and Matthew Marzelli and Jolesz, {Ferenc A.} and Yoo, {Seung Schik}",
year = "2009",
month = "6",
day = "1",
doi = "10.1016/j.media.2009.01.001",
language = "English",
volume = "13",
pages = "392--404",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Automated classification of fMRI data employing trial-based imagery tasks

AU - Lee, Jong-Hwan

AU - Marzelli, Matthew

AU - Jolesz, Ferenc A.

AU - Yoo, Seung Schik

PY - 2009/6/1

Y1 - 2009/6/1

N2 - Automated interpretation and classification of functional MRI (fMRI) data is an emerging research field that enables the characterization of underlying cognitive processes with minimal human intervention. In this work, we present a method for the automated classification of human thoughts reflected on a trial-based paradigm using fMRI with a significantly shortened data acquisition time (less than one minute). Based on our preliminary experience with various cognitive imagery tasks, six characteristic thoughts were chosen as target tasks for the present work: right-hand motor imagery, left-hand motor imagery, right foot motor imagery, mental calculation, internal speech/word generation, and visual imagery. These six tasks were performed by five healthy volunteers and functional images were obtained using a T2*-weighted echo planar imaging (EPI) sequence. Feature vectors from activation maps, necessary for the classification of neural activity, were automatically extracted from the regions that were consistently and exclusively activated for a given task during the training process. Extracted feature vectors were classified using the support vector machine (SVM) algorithm. Parameter optimization, using a k-fold cross validation scheme, allowed the successful recognition of the six different categories of administered thought tasks with an accuracy of 74.5% (mean) ± 14.3% (standard deviation) across all five subjects. Our proposed study for the automated classification of fMRI data may be utilized in further investigations to monitor/identify human thought processes and their potential link to hardware/computer control.

AB - Automated interpretation and classification of functional MRI (fMRI) data is an emerging research field that enables the characterization of underlying cognitive processes with minimal human intervention. In this work, we present a method for the automated classification of human thoughts reflected on a trial-based paradigm using fMRI with a significantly shortened data acquisition time (less than one minute). Based on our preliminary experience with various cognitive imagery tasks, six characteristic thoughts were chosen as target tasks for the present work: right-hand motor imagery, left-hand motor imagery, right foot motor imagery, mental calculation, internal speech/word generation, and visual imagery. These six tasks were performed by five healthy volunteers and functional images were obtained using a T2*-weighted echo planar imaging (EPI) sequence. Feature vectors from activation maps, necessary for the classification of neural activity, were automatically extracted from the regions that were consistently and exclusively activated for a given task during the training process. Extracted feature vectors were classified using the support vector machine (SVM) algorithm. Parameter optimization, using a k-fold cross validation scheme, allowed the successful recognition of the six different categories of administered thought tasks with an accuracy of 74.5% (mean) ± 14.3% (standard deviation) across all five subjects. Our proposed study for the automated classification of fMRI data may be utilized in further investigations to monitor/identify human thought processes and their potential link to hardware/computer control.

KW - Brain-computer interface (BCI)

KW - Functional MRI

KW - Neuroimaging

KW - Pattern recognition

KW - Support vector machine (SVM)

UR - http://www.scopus.com/inward/record.url?scp=67349187397&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67349187397&partnerID=8YFLogxK

U2 - 10.1016/j.media.2009.01.001

DO - 10.1016/j.media.2009.01.001

M3 - Article

VL - 13

SP - 392

EP - 404

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 3

ER -