Wavelet-based estimation of long-memory noise in Diffuse Optical Imaging

In this work we extend a previously proposed method for functional Magnetic Resonance Imaging (fMRI) to estimate the parameters of a linear model of Diffuse Optical Imaging (DOI) time series. The regression is performed in the wavelet domain to infer drift coefficients at different scales and to estimate the strength of the Hemodynamic Response Function (HRF). This multiresolution approach benefits from the whitening property of the Discrete Wavelet Transform (DWT), and we observe an estimation improvement which is then related to a quantitative measure of 1/ƒ noise. The performances of the method are evaluated against a standard spline-cosine GLM approach with simulated HRF and real background physiology.