Showing papers by "Majid Mirmehdi published in 2010"

16th Annual Meeting of the Organization for Human Brain Mapping

Abstract: LISTINGS 26TH ANNUAL MEETING OF THE ORGANIZATION FOR HUMAN BRAIN MAPPING

Archive film restoration based on spatiotemporal random walks

Abstract: We propose a novel restoration method for defects and missing regions in video sequences, particularly in application to archive film restoration. Our statistical framework is based on random walks to examine the spatiotemporal path of a degraded pixel, and uses texture features in addition to intensity and motion information traditionally used in previous restoration works. The degraded pixels within a frame are restored in a multiscale framework by updating their features (intensity, motion and texture) at each level with reference to the attributes of normal pixels and other defective pixels in the previous scale as long as they fall within the defective pixel's random walk-based spatiotemporal neighbourhood. The proposed algorithm is compared against two state-of-the-art methods to demonstrate improved accuracy in restoring synthetic and real degraded image sequences.

Empirical Mode Decomposition in Data-Driven fMRI Analysis

Abstract: Empirical Mode Decomposition has emerged in recent years as a promising data analysis method to adaptively decompose non-linear and non-stationary signals. Here we introduce multi-EMD, to be used where there are many thousands of signals to analyse and compare, such as is common in the analysis of functional neuroimages. The number of component signals found through Empirical Mode Decomposition varies at each location in the brain. We seek to rearrange these components so that they may be compared to others at a similar temporal scale. This is a data-driven process based on grouping those components which have similar dominant frequencies to target frequencies which have been found to be most common from the initial decomposition. This new set of rearranged components is then clustered so that regions behaving synchronously at each temporal scale may be discovered. Results are presented for both simulated and real data from a functional MRI experiment.

Initialisation-Free Active Contour Segmentation

Abstract: We present a region based active contour model which does not require any initialisation and is capable of modelling multi-modal image regions. Its external force is based on statistically learning and grouping of image primitives in multiscale, and its numerical solution is carried out using radial basis function interpolation and time dependent expansion coefficient updating. The initialisation-free property makes it attractive to applications such as detecting unkown number of objects with unkown topologies.

Fast dynamic texture detection

Abstract: Dynamic textures can be considered to be spatio-temporally varying visual patterns in image sequences with certain temporal regularity. We propose a novel and efficient approach to explore the violation of the brightness constancy assumption, as an indication of presence of dynamic texture, using simple optical flow techniques. We assume that dynamic texture regions are those that have poor spatio-temporal optical flow coherence. Further, we propose a second approach that uses robust global parametric motion estimators that effectively and efficiently detect motion outliers, and which we exploit as powerful cues to localize dynamic textures. Experimental and comparative studies on a range of synthetic and real-world dynamic texture sequences show the feasibility of the proposed approaches, with results which are competitive to or better than recent state-of-art approaches and significantly faster.

Automated detection of fMRI artefacts from Shannon entropy distributions

Abstract: As the number of subjects in modern fMRI experiments increases, the use of automated analysis pipelines is becoming more popular, leading to less manual inspection of the data. Here we promote the use of Shannon entropy distributions to discover those datasets in large studies suffering from various artefacts. Entropy distributions of 1444 resting state fMRI datasets from the 1000 Functional Connectomes Project are examined and mean distributions found after each of several different preprocessing steps. Empirically derived envelopes are generated so that significantly outlying datasets may be identified. This process of outlier detection may be automated such that those datasets with characteristic shifts in entropy caused by specific artefacts may be flagged for further manual examination or removed from further analysis. We conclude this technique will be a useful quality control method when dealing with data from large studies.

Automated detection of datasets with artefactual signal in large n studies using Shannon entropy

Abstract: The recent public release of more than 1200 resting state BOLD MRI (R-fMRI) datasets as part of the 1000 Functional Connectomes Project (Biswal, 2010) provides the community with the opportunity to apply and test analysis techniques on a much larger number of subjects than may be available locally. With the potential to examine data from many sources comes the issue of how the characteristics of this data vary between site, and also between studies at the same site.