Abstract
Neuroimaging community usually employs spatial smoothing to denoise magnetic resonance imaging (MRI) data, e.g., Gaussian smoothing kernels. Such an isotropic diffusion (ISD) based smoothing is widely adopted for denoising purpose due to its easy implementation and efficient computation. Beyond these advantages, Gaussian smoothing kernels tend to blur the edges, curvature and texture of images. Researchers have proposed anisotropic diffusion (ASD) and non-local diffusion (NLD) kernels. We recently demonstrated the effect of these new filtering paradigms on preprocessing real degraded MRI images from three individual subjects. Here, to further systematically investigate the effects at a group level, we collected both structural and functional MRI data from 23 participants. We first evaluated the three smoothing strategies' impact on brain extraction, segmentation and registration. Finally, we investigated how they affect subsequent mapping of default network based on resting-state functional MRI (R-fMRI) data. Our findings suggest that NLD-based spatial smoothing maybe more effective and reliable at improving the quality of both MRI data preprocessing and default network mapping. We thus recommend NLD may become a promising method of smoothing structural MRI images of R-fMRI pipeline.
Highlights
Partial Differential Equation (PDE), a well-established mathematical theory, has given its advances on denoising images in terms of the strong theoretical framework with simple and efficient numerical strategies [1]
As we demonstrated in our previous work [20], applied to real degraded raw T1 image (RAW), the non-local means diffusion (NLD) outperforms both isotropic diffusion (ISD) and anisotropic diffusion equations (ASD) smoothers (Figure 1)
Our statistical analyses showed that NLD most significantly increases the partial volume estimation (PVE) amount within level-2 grey matter tissues (Figure 3D) while both ISD and ASD significantly increase PVE distribution within grey-white matter boundary (Figure 3B) and level-1 grey-matter (Figure 3C)
Summary
Partial Differential Equation (PDE), a well-established mathematical theory, has given its advances on denoising images in terms of the strong theoretical framework with simple and efficient numerical strategies [1]. The ASD attempts to avoid the drawback of ISD by smoothing the image at a pixel only in the direction orthogonal to its gradient (i.e., smoothing along with edges). Both ISD and ASD are local smoothers and hard to preserve some global features of images (e.g., texture or periodic pattern). To address this issue, the NLD smoothes an image by taking into account the similarity of the geometrical configuration in a whole neighborhood (i.e., a patch of the image)
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call