Priya Aggarwal

Accelerated fMRI reconstruction using Matrix Completion with Sparse Recovery via Split Bregman

In this work, we propose a new method of accelerated functional MRI reconstruction, namely, Matrix Completion with Sparse Recovery (MCwSR). The proposed method combines low rank condition with transform domain sparsity for fMRI reconstruction and is solved using state-of-the-art Split Bregman algorithm. We compare results with state-of-the-art fMRI reconstruction algorithms. Experimental results demonstrate better performance of MCwSR method compared to the existing methods with reference to normalized mean squared error (NMSE) and other reconstruction quality metrics. In addition, the proposed method is able to preserve voxel activation maps on brain volume. None of the other existing methods is able to demonstrate this property. This shows that the proposed method is accurate and faster, and preserves the voxel activation maps that is the key to study fMRI data.

Multivariate brain network graph identification in functional MRI

HighlightsA Multivariate Vector Regression‐based Connectivity (MVRC) method is proposed for brain network identification.Proposed method estimates pairwise association between two regions with consideration of influence of other regions.MVRC employs both l1‐ and l2‐ norms constraints on adjacency matrix coefficients.Results on simulated and real fMRI dataset demonstrate that MVRC is able to build brain networks. Graphical abstract Figure. No caption available. ABSTRACT Motivated by recent interest in identification of functional brain networks, we develop a new multivariate approach for functional brain network identification and name it as Multivariate Vector Regression‐based Connectivity (MVRC). The proposed MVRC method regresses time series of all regions to those of other regions simultaneously and estimates pairwise association between two regions with consideration of influence of other regions and builds the adjacency matrix. Next, modularity method is applied on the adjacency matrix to detect communities or functional brain networks. We compare the proposed MVRC method with existing methods ranging from simple Pearson correlation to advanced Multivariate Adaptive Sparse Representation (ASR) methods. Experimental results on simulated and real fMRI dataset demonstrate that MVRC is able to extract functional brain networks that are consistent with the literature. Also, the proposed MVRC method is 650–750 times faster compared to the existing ASR method on 90 node network.

A guard interval assisted OFDM symbol-based channel estimation for rapid time-varying scenarios in IEEE 802.lip

IEEE 802.11p standard is a wireless vehicular communication standard meant for outdoor applications. This standard suffers from the challenge of robust channel estimation due to rapid time-varying nature of the channel This paper proposes a novel scheme of channel estimation by utilizing the guard interval of every orthogonal frequency division multiplexing (OFDM) symbol. For a typical vehicular wireless communication where the channel fades quite rapidly, inter-symbol-interference (ISI) may not be as significant a problem as time varying nature of the channel due to Doppler effect. Hence, the proposed scheme utilizes the redundant space of guard interval (GI) (other than that required for cyclic prefix (CP) to combat ISI) to insert pseudo-random sequence (PRS)for channel estimation. A decision-directed time-domain least squares channel estimation method is proposed using the inserted PRS with CP. Simulation results show that the proposed scheme can considerably improve the bit error rate (BER) performance compared to the existing techniques.

Joint Estimation of Hemodynamic Response Function and Voxel Activation in Functional MRI Data

This paper proposes a method of voxel-wise hemodynamic response function HRF estimation using sparsity and smoothing constraints on the HRF. The slow varying baseline drift at the voxel time-series is initially estimated via empirical mode decomposition EMD. This estimation is refined by two-stage optimization that estimates HRF and slow-varying noise iteratively. In addition, this paper proposes a novel method of finding voxel activation via projection of voxel time-series on signal subspace constructed using the prior estimates of HRF. The performance of the proposed method is demonstrated on both synthetic and real fMRI data.

Joint estimation of activity signal and HRF in fMRI using fused LASSO

In this paper, we propose a novel voxel-based method for joint estimation of underlying activity signal and hemodynamic response function (HRF) in functional magnetic resonance imaging (fMRI). In the proposed two stage iterative framework, fused-least absolute shrinkage and selection operator (Fused LASSO) penalty is utilized for activity detection and HRF estimation. Conditions of smoothness and sparsity are imposed on HRF for its estimation. The validity of the proposed method is demonstrated on both synthetic and real fMRI data.

A digitally controlled floating resistor using CMOS translinear cells

A CMOS based digitally controlled floating resistor (DCFR) is proposed using translinear cells. The proposed DCFR consists of two CMOS translinear cells along with a CMOS current-division network (CDN). The proposed circuit is designed in 0.35 μm technology and works at power supply of ±1.6V. DCFR is analyzed as well as SPICE simulation along with MATLAB is used to verify the proposed design. An application of proposed DCFR is shown in a high pass filter.

Dynamic Phase Synchrony based Ranked Spatio-Temporal Clustering for Tracking Time-Resolved Functional Brain Networks

Of late, there has been a growing interest in studying brain networks, particularly, for understanding spontaneous temporal changes in functional brain networks. Recently, phase synchrony based methods have been proposed to track instantaneous time-resolved functional connectivity without any need of windowing the data. This paper extends one such recently used phase synchrony measure in two steps. First, multiple temporal models are built from four-mode tensor that are further clustered to detect dynamic brain network communities. This clustering is based on spatio-temporal data and hence, is named as Spatio-Temporal Clustering (STC). Second, a method is proposed to rank all the communities allowing the proposed model to deal with multiple communities of differing time evolution. This helps in the comparison of network communities, especially, when available communities are too dense to provide relevant information for comparison. The ranking of communities allows for the dimensionality reduction of communities, while still maintaining the key brain networks. Intrinsic time-varying functional connectivity has been investigated for large scale brain networks, including default-mode network (DMN), visual network (VN), cognitive control network (CCN), auditory network (AN), etc. The proposed method provides a new complementary tool to investigate dynamic network states at a high temporal resolution and is tested on resting-state functional MRI data of 26 typically developing controls (TDC) and 35 autism spectrum disorder (ASD) subjects. Simulation results demonstrate that ASD subjects have altered dynamic brain networks compared to TDC.

Recursive Least Squares Channel Estimation for Rapidly Time-Varying Scenarios in IEEE 802.11p

Rapid time-varying channel estimation is one of the biggest challenges in IEEE 802.11p standard. It is a wireless vehicular communication standard which is used for outdoor applications. This paper proposes a novel decision-directed recursive least squares time-domain channel estimation method that utilizes the guard interval of every orthogonal frequency division multiplexing symbol. Simulation results show considerably improved bit error rate performance with the proposed method that enables robust channel equalization in rapidly time-varying channel with high Doppler spread.