Abstract
This study presents a new three-dimensional discrete wavelet transform (3D-DWT)-based denoising method for functional magnetic resonance images (fMRI). This method is called weighted three-dimensional discrete wavelet transform (w-3D-DWT), and it is based on the principle of weighting the volume subbands which are obtained by 3D-DWT. Briefly, classical DWT denoising consists of wavelet decomposition, thresholding, and image reconstruction steps. In the thresholding algorithm, the thresholding value for each image cannot be chosen exclusively. Namely, a specific thresholding value is chosen and it is used for all images. The proposed algorithm in this study can be considered as a data-driven denoising model for fMRI. It consists of three-dimensional wavelet decomposition, subband weighting, and image reconstruction. The purposes of subband weighting algorithm are to increase the effect of the subband which represents the image better and to decrease the effect of the subband which represents the image in the worst way and thus to reduce the noises of the image adaptively. fMRI is one of the popular methods used to understand brain functions which are often corrupted by noises from various sources. The traditional denoising method used in fMRI is smoothing images with a Gaussian kernel. This study suggests an adaptive approach for fMRI filtering different from Gaussian smoothing and 3D-DWT thresholding. In this study, w-3D-DWT denoising results were evaluated with mean-square error (MSE), peak signal/noise ratio (PSNR), and structural similarity (SSIM) metrics, and the results were compared with Gaussian smoothing and 3D-DWT thresholding methods. According to this comparison, w-3D-DWT gave low-MSE and high-PSNR results for fMRI data.
Similar content being viewed by others
References
Onut I V, Ghorbani AA (2004) Classifying cognitive states from fMRI data using neural networks. Ieee. Conf. Neural Networks, Vols 1–4, Proc. 2871–2875\n3302
Billor N, Godwin J (2015) Variable selection for functional logistic regression in fMRI data analysis. Turkiye Klinikleri Journal of Biostatistics 7(1):1–10
Lindquist MA., Wager TD (2014) Principles of functional magnetic resonance imaging. Handbook of neuroimaging data analysis, London Chapman &Hall, 3–48
Yue YR, Loh JM, Lindquist M (2010) Adaptive spatial smoothing of fMRI images. Stat Interface 3:3–13
Khullar S, Michael A, Correa N, Adali T, Baum SA, Calhoun VD (2011) Wavelet-based fMRI analysis: 3-D denoising, signal separation, and validation metrics. NeuroImage 54:2867–2884
Wink AM, Roerdink JBTM (2004) Denoising functional MR images: a comparison of wavelet denoising and Gaussian smoothing. IEEE Trans Med Imaging 23(3):374–387
Yang X, Fei B (2011) A wavelet multiscale denoising algorithm for magnetic resonance (MR) images. Meas Sci Technol 22(2):025803
Donoho DL, Johnstone JM (1994) Ideal spatial adaptation by wavelet shrinkage. Biometrika 81(3):425–455
Wang Y, Che X, Ma S (2012) Nonlinear filtering based on 3D wavelet transform for MRI denoising. EURASIP J Adv Signal Process 2012:40
Kadah YM, Member S (2004) Adaptive denoising of event-related functional using spectral subtraction. 51, 1944–1953
Tabelow K, Polzehl J, Voss HU, Spokoiny V (2006) Analyzing fMRI experiments with structural adaptive smoothing procedures. NeuroImage 33:55–62
Monir SMG, Siyal MY (2009) Denoising functional magnetic resonance imaging time-series using anisotropic spatial averaging. Biomed Signal Process Control 4:16–25
Gonzalez RC, Woods RE (2002) Digital image processing, Prentice Hall Inc., 349–408
Prochazka A, Grafova L, Vysata O (2011) Three-dimensional wavelet transform in multidimensional biomedical volume processing. Proc. of the IASTED Int. Conf. Graphics and Virtual Reality, Cambridge, UK
Yavariabdi A, Bartoli A (2011) 3D medical image enhancement based on wavelet transforms. ALCOV-ISIT Fr. 3–7
Brammer MJ (2008) Multidimensional wavelet analysis of functional magnetic resonance images. 18, 773–778
Gibbs P, Buckley DL, BlackbandSJ HA (1996) Tumour volume determination from MR images by morphological segmentation. Phys Med Biol 41:2437–2446
Materka A, Strzelecki M (1998) Texture analysis methods—a review. Methods 11:1–33
Skea J, Anderson D, GreenT GR, Heptonstall P, Leach M (2008) Intermittent renewable generation and maintaining power system reliability. IET generation, transmission & distribution 2(1):82–89
Panda R, Khobragade PS, Jambhule PD, Jengthe SN, Pal PR, Gandhi TK (2010) Classification of EEG signal using wavelet transform and support vector machine for epileptic seizure diction. 2010 Int. Conf. Syst. Med. Biol. 405–408
Sidhu KS, Khaira BS, Virk IS (2012) Medical image denoising in the wavelet domain using haar and db3 filtering International Refereed Journal of Engineering and Science 1.1, 001–008. 1, 1–8
Rees G, Friston K, Single subject epoch (block) auditory fMRI activation data. http://www.fil.ion.ucl.ac.uk/spm/data/ Accessed 10 Oct 2016
Hu J, Pu Y, Wu X, Zhang Y, Zhou J (2012) Improved DCT-based nonlocal means filter for MR images denoising Computational and Mathematical Methods in Medicine
Wink AM, Roerdink JBTM (2006) BOLD noise assumptions in fMRI International Journal of Biomedical Imaging
Burrell LS (2008) Feature analysis of functional MRI data for mapping epileptic networks Feature Analysis Proquest. 36–37
Kong H (2010) A Bayesian model selection approach to fMRI activation detection Abd-Krim Seghouane and Ju Lynn Ong National ICT Australia, Canberra Research Laboratory The Australian National University, College of Engineering and Computer Science. Analysis. 4401–440
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no competing interests.
Rights and permissions
About this article
Cite this article
Özmen, G., Özşen, S. A new denoising method for fMRI based on weighted three-dimensional wavelet transform. Neural Comput & Applic 29, 263–276 (2018). https://doi.org/10.1007/s00521-017-2995-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-017-2995-7