Statistical analysis of functional MRI data in the wavelet domain

Urs E. Ruttimann, Michael Unser, Robert R. Rawlings, Daniel Rio, Nick F. Ramsey, Venkata Mattay, Daniel W. Hommer, Joseph A. Frank, Daniel Weinberger

Research output: Contribution to journalArticle

Abstract

The use of the wavelet transform is explored for the detection of differences between brain functional magnetic resonance images (fMRI's) acquired under two different experimental conditions. The method benefits from the fact that a smooth and spatially localized signal can be represented by a small set of localized wavelet coefficients, while the power of white noise is uniformly spread throughout the wavelet space. Hence, a statistical procedure is developed that uses the imposed decomposition orthogonality to locate wavelet-space partitions with large signal-to-noise ratio (SNR), and subsequently restricts the testing for significant wavelet coefficients to these partitions. This results in a higher SNR and a smaller number of statistical tests, yielding a lower detection threshold compared to spatial-domain testing and, thus, a higher detection sensitivity without increasing type I errors. The multiresolution approach of the wavelet method is particularly suited to applications where the signal bandwidth and/or the characteristics of an imaging modality cannot be well specified. The proposed method was applied to compare two different fMRI acquisition modalities. Differences of the respective useful signal bandwidths could be clearly demonstrated; the estimated signal, due to the smoothness of the wavelet representation, yielded more compact regions of neuroactivity than standard spatial-domain testing.

Original languageEnglish (US)
Pages (from-to)142-154
Number of pages13
JournalIEEE Transactions on Medical Imaging
Volume17
Issue number2
StatePublished - 1998
Externally publishedYes

Fingerprint

Statistical methods
Magnetic Resonance Imaging
Magnetic resonance
Signal to noise ratio
Testing
Signal-To-Noise Ratio
Bandwidth
Magnetic Resonance Spectroscopy
Statistical tests
Image acquisition
White noise
Wavelet Analysis
Wavelet transforms
Brain
Decomposition
Imaging techniques

Keywords

  • Functional magnetic resonance imaging
  • Multiresolution analysis
  • Statistical models
  • Wavelet transform

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Ruttimann, U. E., Unser, M., Rawlings, R. R., Rio, D., Ramsey, N. F., Mattay, V., ... Weinberger, D. (1998). Statistical analysis of functional MRI data in the wavelet domain. IEEE Transactions on Medical Imaging, 17(2), 142-154.

Statistical analysis of functional MRI data in the wavelet domain. / Ruttimann, Urs E.; Unser, Michael; Rawlings, Robert R.; Rio, Daniel; Ramsey, Nick F.; Mattay, Venkata; Hommer, Daniel W.; Frank, Joseph A.; Weinberger, Daniel.

In: IEEE Transactions on Medical Imaging, Vol. 17, No. 2, 1998, p. 142-154.

Research output: Contribution to journalArticle

Ruttimann, UE, Unser, M, Rawlings, RR, Rio, D, Ramsey, NF, Mattay, V, Hommer, DW, Frank, JA & Weinberger, D 1998, 'Statistical analysis of functional MRI data in the wavelet domain', IEEE Transactions on Medical Imaging, vol. 17, no. 2, pp. 142-154.
Ruttimann UE, Unser M, Rawlings RR, Rio D, Ramsey NF, Mattay V et al. Statistical analysis of functional MRI data in the wavelet domain. IEEE Transactions on Medical Imaging. 1998;17(2):142-154.
Ruttimann, Urs E. ; Unser, Michael ; Rawlings, Robert R. ; Rio, Daniel ; Ramsey, Nick F. ; Mattay, Venkata ; Hommer, Daniel W. ; Frank, Joseph A. ; Weinberger, Daniel. / Statistical analysis of functional MRI data in the wavelet domain. In: IEEE Transactions on Medical Imaging. 1998 ; Vol. 17, No. 2. pp. 142-154.
@article{f730c724470b4ca392b6e7f8a6ee9517,
title = "Statistical analysis of functional MRI data in the wavelet domain",
abstract = "The use of the wavelet transform is explored for the detection of differences between brain functional magnetic resonance images (fMRI's) acquired under two different experimental conditions. The method benefits from the fact that a smooth and spatially localized signal can be represented by a small set of localized wavelet coefficients, while the power of white noise is uniformly spread throughout the wavelet space. Hence, a statistical procedure is developed that uses the imposed decomposition orthogonality to locate wavelet-space partitions with large signal-to-noise ratio (SNR), and subsequently restricts the testing for significant wavelet coefficients to these partitions. This results in a higher SNR and a smaller number of statistical tests, yielding a lower detection threshold compared to spatial-domain testing and, thus, a higher detection sensitivity without increasing type I errors. The multiresolution approach of the wavelet method is particularly suited to applications where the signal bandwidth and/or the characteristics of an imaging modality cannot be well specified. The proposed method was applied to compare two different fMRI acquisition modalities. Differences of the respective useful signal bandwidths could be clearly demonstrated; the estimated signal, due to the smoothness of the wavelet representation, yielded more compact regions of neuroactivity than standard spatial-domain testing.",
keywords = "Functional magnetic resonance imaging, Multiresolution analysis, Statistical models, Wavelet transform",
author = "Ruttimann, {Urs E.} and Michael Unser and Rawlings, {Robert R.} and Daniel Rio and Ramsey, {Nick F.} and Venkata Mattay and Hommer, {Daniel W.} and Frank, {Joseph A.} and Daniel Weinberger",
year = "1998",
language = "English (US)",
volume = "17",
pages = "142--154",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Statistical analysis of functional MRI data in the wavelet domain

AU - Ruttimann, Urs E.

AU - Unser, Michael

AU - Rawlings, Robert R.

AU - Rio, Daniel

AU - Ramsey, Nick F.

AU - Mattay, Venkata

AU - Hommer, Daniel W.

AU - Frank, Joseph A.

AU - Weinberger, Daniel

PY - 1998

Y1 - 1998

N2 - The use of the wavelet transform is explored for the detection of differences between brain functional magnetic resonance images (fMRI's) acquired under two different experimental conditions. The method benefits from the fact that a smooth and spatially localized signal can be represented by a small set of localized wavelet coefficients, while the power of white noise is uniformly spread throughout the wavelet space. Hence, a statistical procedure is developed that uses the imposed decomposition orthogonality to locate wavelet-space partitions with large signal-to-noise ratio (SNR), and subsequently restricts the testing for significant wavelet coefficients to these partitions. This results in a higher SNR and a smaller number of statistical tests, yielding a lower detection threshold compared to spatial-domain testing and, thus, a higher detection sensitivity without increasing type I errors. The multiresolution approach of the wavelet method is particularly suited to applications where the signal bandwidth and/or the characteristics of an imaging modality cannot be well specified. The proposed method was applied to compare two different fMRI acquisition modalities. Differences of the respective useful signal bandwidths could be clearly demonstrated; the estimated signal, due to the smoothness of the wavelet representation, yielded more compact regions of neuroactivity than standard spatial-domain testing.

AB - The use of the wavelet transform is explored for the detection of differences between brain functional magnetic resonance images (fMRI's) acquired under two different experimental conditions. The method benefits from the fact that a smooth and spatially localized signal can be represented by a small set of localized wavelet coefficients, while the power of white noise is uniformly spread throughout the wavelet space. Hence, a statistical procedure is developed that uses the imposed decomposition orthogonality to locate wavelet-space partitions with large signal-to-noise ratio (SNR), and subsequently restricts the testing for significant wavelet coefficients to these partitions. This results in a higher SNR and a smaller number of statistical tests, yielding a lower detection threshold compared to spatial-domain testing and, thus, a higher detection sensitivity without increasing type I errors. The multiresolution approach of the wavelet method is particularly suited to applications where the signal bandwidth and/or the characteristics of an imaging modality cannot be well specified. The proposed method was applied to compare two different fMRI acquisition modalities. Differences of the respective useful signal bandwidths could be clearly demonstrated; the estimated signal, due to the smoothness of the wavelet representation, yielded more compact regions of neuroactivity than standard spatial-domain testing.

KW - Functional magnetic resonance imaging

KW - Multiresolution analysis

KW - Statistical models

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=0032034220&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032034220&partnerID=8YFLogxK

M3 - Article

C2 - 9688147

AN - SCOPUS:0032034220

VL - 17

SP - 142

EP - 154

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 2

ER -