On Linearly Constrained Minimum Variance Beamforming
Authors: Jian Zhang, Chao Liu
JMLR 2015 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The performances of the proposed beamformers are assessed by simulation studies. Superior performances of the proposed beamformers are obtained when the signalto-noise ratio is low. We apply the proposed procedure to real MEG data sets derived from five sessions of a human face-perception experiment, finding several highly active areas in the brain. |
| Researcher Affiliation | Academia | Jian Zhang EMAIL Chao Liu EMAIL School of Mathematics, Statistics and Actuarial Science University of Kent Canterbury, Kent CT2 7NF, UK |
| Pseudocode | No | The paper describes the methodology and procedures in detail (e.g., Section 2: Methodology, Subsection 2.1: Thresholding the sensor covariance matrix, Subsection 2.2: Beamforming), but it does not present these steps in a structured pseudocode or algorithm block format. |
| Open Source Code | No | The paper mentions a third-party software: "The software SPM8 is available at http://www.fil.ion.ucl.ac.uk/spm/software/spm8/". However, it does not provide explicit access (link, statement, or supplementary material) to the authors' own implementation code for the methodology described in this paper. |
| Open Datasets | Yes | We applied the proposed methodology to human MEG data acquired in five sessions by Wakeman and Henson (Henson et al., 2011). |
| Dataset Splits | No | For simulation studies, the paper mentions sampling "30 data sets of Y(t)" and for real data, "For each session, we applied the neuroimaging software SPM8 to read and preprocess the recorded data, and to epoch and average the data generated from the face stimulus trials and the scrambled face stimulus trials respectively." However, it does not provide specific training, validation, or test dataset splits, percentages, or methodology for the evaluation of the beamformers. |
| Hardware Specification | No | The paper describes the equipment used for data acquisition: "The data were collected with a Neuromag Vector View system, containing a magnetometer and two orthogonal, planar gradiometers located at each of 102 positions within a hemispherical array situated in a light, magnetically shielded room." However, it does not specify the computational hardware (e.g., CPU, GPU models) used for running the simulations or data analysis. |
| Software Dependencies | Yes | We started with specifying the following two head models (Sarvas, 1987)... created by using the software Field Trip (Oostenveld et al., 2010)... For each session, we applied the neuroimaging software SPM8 to read and preprocess the recorded data. |
| Experiment Setup | Yes | We considered various combinations of (n, p) = (91, 2222) and (102, 1487), and J = 500, 1000, 2000, and 3000. Note that p is substantially larger than n and that the sources are sparse in the sense that there are only two or three non-null sources among p candidates. We first applied the proposed procedures ma, mi and sh to each data set. We calculated the maximum indices over the grids and the L1-biases of the maximum location estimates to two sources respectively. For each combination of (n, p, J) and the SNR, we then summarized these values in the form of a box-whisker plot as in Figures 2, 3, 4, and 5 corresponding to Scenarios 1, 2, 3, and 4 respectively. We considered two values of the signal-to-noise-ratio (SNR): 0.04 and 1/0.64 = 1.5625. In simulations, we will show that c0 D0 = {0, 0.5, 1, 1.5, 2} has covered its useful range. |