Robust Unmixing of Dynamic Sequences Using Regions of Interest

@article{Filippi2018RobustUO,
  title={Robust Unmixing of Dynamic Sequences Using Regions of Interest},
  author={Marc Filippi and Michel Desvignes and Eric Moisan},
  journal={IEEE Transactions on Medical Imaging},
  year={2018},
  volume={37},
  pages={306-315}
}
In dynamic planar imaging, extraction of signals specific to structures is complicated by structures superposition. Due to overlapping, signals extraction with classic regions of interest (ROIs) methods suffers from inaccuracy, as extracted signals are a mixture of targeted signals. Partial volume effect raises the same issue in dynamic tomography. Source separation methods, such as factor analysis of dynamic sequences, have been developed to unmix such data. However, the underlying problem is… 

Figures and Tables from this paper

Spatially Constrained Online Dictionary Learning for Source Separation
TLDR
A spatially constrained dictionary learning source separation algorithm that uses e.g. high resolution segmentation map or regions of interest defined by an expert to regularise the source contribution estimation and is easily adaptable to different applications by adding or modifying the constraints on the sources properties in the optimisation problem.
Bayesian Non-Negative Matrix Factorization With Adaptive Sparsity and Smoothness Prior
TLDR
This work forms the problem using a probabilistic model and estimates all its parameters using the variational Bayes method, using the fact that the assumptions of sparsity and smoothness are different forms of prior covariance matrix modeling to avoid ad hoc combination of these assumptions using weighting coefficient.
Low-Element Image Restoration Based on an Out-of-Order Elimination Algorithm
TLDR
An image restoration method for underdetermined blind-source separation based on an out-of-order elimination algorithm and the fast independent component analysis (FastICA) algorithm is proposed, which can effectively and completely extract all the source signals.
Séparation de sources en imagerie nucléaire
En imagerie nucleaire (scintigraphie, TEMP, TEP), les diagnostics sont frequemment faits a l'aide des courbes d'activite temporelles des differents organes et tissus etudies. Ces courbes representent

References

SHOWING 1-10 OF 32 REFERENCES
Factor Analysis of Dynamic Sequence with Spatial Prior for 2D Cardiac Spect Sequences Analysis
TLDR
A new objective function using soft spatial prior knowledge is developed for unmixing SPECT sequences and its efficiency and robustness to the choice of priors are illustrated with tests on synthetic dataset.
Bayesian Blind Separation and Deconvolution of Dynamic Image Sequences Using Sparsity Priors
TLDR
A novel blind source separation method based on probabilistic model of dynamic image sequences assuming each source dynamics as convolution of an input function and a source specific kernel (modeling organ impulse response or retention function) and solved by the Variational Bayes method.
Automatic regions of interest in factor analysis for dynamic medical imaging
  • V. Šmídl, O. Tichý
  • Medicine
    2012 9th IEEE International Symposium on Biomedical Imaging (ISBI)
  • 2012
TLDR
A new probabilistic model for functional analysis with inherent estimation of Probabilistic ROI is proposed using the Variational Bayes method which provides also automatic relevance determination of the estimated factors.
Spatial regularization applied to factor analysis of medical image sequences (FAMIS).
TLDR
A spatial regularization method for computing factor images (REG-FAMIS) is proposed to remove irregularities due to noise in the original data while preserving discontinuities between structures.
Reconstruction of 4-D Dynamic SPECT Images From Inconsistent Projections Using a Spline Initialized FADS Algorithm (SIFADS)
TLDR
An algorithm of extracting voxel-by-voxel time activity curves directly from inconsistent projections applied in dynamic cardiac SPECT to solve the inverse problem of dynamic SPECT with slow gantry rotation is proposed.
Quantitative Evaluation of Two-Factor Analysis Applied to Hepatic Perfusion Study Using Contrast-enhanced Ultrasound
TLDR
The results indicate that two-factor analysis has the potential to perform quantitative analysis of hepatic perfusion, which would be helpful to the differential diagnosis of FLLs.
Tissue-Specific Compartmental Analysis for Dynamic Contrast-Enhanced MR Imaging of Complex Tumors
TLDR
A convex analysis of mixtures (CAM) algorithm to explicitly mitigate PVE by expressing the kinetics in each pixel as a nonnegative combination of underlying compartments and subsequently identifying pure volume pixels at the corners of the clustered pixel time series scatter plot simplex is proposed.
Hepatic Perfusion Imaging Using Factor Analysis of Contrast Enhanced Ultrasound
TLDR
This paper presents a method of using factor analysis for extracting distinct time-intensity curves by clustering of measured bolus curves and their projection onto a positivity domain to obtain nonnegative curves.
Dynamic single photon emission computed tomography--basic principles and cardiac applications.
TLDR
A review of advancements toward the realization of the potential of dynamic SPECT imaging, especially in light of new developments that enable measurement of dynamic processes directly from projection measurements.
Sparsity in Bayesian Blind Source Separation and Deconvolution
TLDR
The assumption of sparseness is shown to suppress artifacts of unconstrained separation method and performance of the algorithm is demonstrated on the problem of decomposition of a sequence of medical data.
...
...