Matched signal detection on graphs: Theory and application to brain imaging data classification

Hu, Chenhui; Sepulcre, Jorge; Johnson, Keith A.; Fakhri, Georges El; Lu, Yue M.; Li, Quanzheng

doi:10.1016/j.neuroimage.2015.10.026

Cited by 43 publications

(44 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hu et al . developed a MSD theory for signals derived from weighted graphs 46 . Graph Laplacian eigenvalues are regarded as frequencies of graph-signals and the signals are assumed to lie in a subspace spanned by the first few graph Laplacian eigenvectors that are associated with lower eigenvalues.…”

Section: Methodsmentioning

confidence: 99%

Partial volume correction for PET quantification and its impact on brain network in Alzheimer’s disease

Yang

Guo

et al. 2017

Sci Rep

Self Cite

View full text Add to dashboard Cite

Amyloid positron emission tomography (PET) imaging is a valuable tool for research and diagnosis in Alzheimer’s disease (AD). Partial volume effects caused by the limited spatial resolution of PET scanners degrades the quantitative accuracy of PET image. In this study, we have applied a method to evaluate the impact of a joint-entropy based partial volume correction (PVC) technique on brain networks learned from a clinical dataset of AV-45 PET image and compare network properties of both uncorrected and corrected image-based brain networks. We also analyzed the region-wise SUVRs of both uncorrected and corrected images. We further performed classification tests on different groups using the same set of algorithms with same parameter settings. PVC has sometimes been avoided due to increased noise sensitivity in image registration and segmentation, however, our results indicate that appropriate PVC may enhance the brain network structure analysis for AD progression and improve classification performance.

show abstract

Section: Methodsmentioning

confidence: 99%

Partial volume correction for PET quantification and its impact on brain network in Alzheimer’s disease

Yang

Guo

et al. 2017

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…The filtered noise back is used in the iterative regularization [27] to denoise the image at each iteration to improve the quality of the image based on the previous estimation using Eq. (10).…”

Section: Iterative Regularizationmentioning

confidence: 99%

Robust Image Restoration using Non-Local Graph Based Transform and Wave Atoms

Ganesh¹,

Kusagur²

2019

IJIES

View full text Add to dashboard Cite

The image restoration is the process of recovering the uncorrupted image from the noisy or blurred image. The aim of this method is to propose a hybrid method for effective image restoration process. The Hybrid involves in Non-Local Graph based Transform (NLGBT) learns basic vectors from the image and Wave atom technique is applied for harmonic computational analysis. In order to test the performance of the proposed system, inbuilt MATLAB image is considered and these images has been degraded with Gaussian blur, then added to the Gaussian noise. The degraded images are used to plot NLGBT and the basic vectors learn from this graph. Wave atoms coefficient has been set with the threshold values, which smoothen the image with provided vector. Instead of smoothening of the images, smoothening on the given basic vectors of the images from the NLGBT technique helps in achieving higher efficiency. The existing methods involves in smoothing of images for denoising and proposed hybrid method involves in smoothening of basic vectors from NLGBT. The result provided by the wave atom technique achieves the PSNR value of 18.57 dB, state-of-the-art method achieves 20.08 dB and proposed technique achieve up to 29.02. The experimental result shows that the NLGBT and wave atoms attains high quality of the image in the restoration process compared to other existing method.

show abstract

“…Although no proper treatment exists to cure this illness, its early detection is helpful in controlling the disease [28]. Various attempts have been made to detect the AD in the early stage called mild cognitive impairment (MCI) using different modalities like graph signal processing [29], machine learning [30], [31] and diffusion model [5] etc. Recently, GCNN has also been applied for this detection purpose [26], [27].…”

Section: Introductionmentioning

confidence: 99%

Graph Wavelet-Based Multilevel Graph Coarsening and Its Application in Graph-CNN for Alzheimer’s Disease Detection

2020

View full text Add to dashboard Cite

Along with the classical applications like graph partitioning, graph visualization, etc., graph coarsening has been recently applied in graph convolutional neural network (GCNN) architectures to perform the pooling operation in the graph domain. In this paper, we propose a novel two-stage graph coarsening method rooted on the graph signal processing with its application in the GCNN architecture. In the first stage of coarsening, the graph wavelet transform (GWT) based features are used to obtain a coarsened graph which preserves the topological characteristics of the original graph. In the second stage, the coarsening problem is formulated as an optimization problem where the reduced Laplacian operator at each level is obtained as a restriction of the original Laplacian operator to a specified subspace that also maximizes the topological similarity. The performance of the proposed coarsening algorithm is quantified in the general coarsening context using different graph coarsening quality measures. Its effectiveness as a pooling operator in GCNN is validated by applying it for the graph coarsening operation in the GCNN architecture. This modified GCNN architecture is then used as a graph signal classifier for the early detection of Alzheimer's disease. The results show that the proposed coarsening method outperforms state-of-the-art methods, both in the general coarsening context and as a pooling operator in the GCNN architecture. INDEX TERMS Graph coarsening, graph signal processing, convolutional neural network, Alzheimer's disease.

show abstract

Matched signal detection on graphs: Theory and application to brain imaging data classification

Cited by 43 publications

References 41 publications

Partial volume correction for PET quantification and its impact on brain network in Alzheimer’s disease

Partial volume correction for PET quantification and its impact on brain network in Alzheimer’s disease

Robust Image Restoration using Non-Local Graph Based Transform and Wave Atoms

Graph Wavelet-Based Multilevel Graph Coarsening and Its Application in Graph-CNN for Alzheimer’s Disease Detection

Contact Info

Product

Resources

About