Journal: Journal of neuroscience methods

Article Title: Decoding the Encoding of Functional Brain Networks: an fMRI Classification Comparison of Non-negative Matrix Factorization (NMF), Independent Component Analysis (ICA), and Sparse Coding Algorithms

doi: 10.1016/j.jneumeth.2017.03.008

Figure Lengend Snippet: Classification accuracy averaged across 304 traditionally preprocessed data scans in predicting whether a subject was viewing a video, listening to an audio stimuli, or resting, using 20 components. Chance accuracy is 50%.

Article Snippet: Within each scan, the algorithms in order from best to worst classification accuracy were: L 1 Regularization, K-SVD, Fast ICA, EBM ICA, JADE ICA, InfoMax ICA, NMF-PG, and NMF-ALS, as shown in using 20 spatial maps on artifact-cleaned data.

Techniques:

Journal: Journal of neuroscience methods

Article Title: Decoding the Encoding of Functional Brain Networks: an fMRI Classification Comparison of Non-negative Matrix Factorization (NMF), Independent Component Analysis (ICA), and Sparse Coding Algorithms

doi: 10.1016/j.jneumeth.2017.03.008

Figure Lengend Snippet: Classification accuracy averaged across 304 traditionally preprocessed scans in predicting whether a subject was viewing a video, listening to an audio stimuli, or resting, using 50 components. Chance accuracy is 50%.

Article Snippet: Within each scan, the algorithms in order from best to worst classification accuracy were: L 1 Regularization, K-SVD, Fast ICA, EBM ICA, JADE ICA, InfoMax ICA, NMF-PG, and NMF-ALS, as shown in using 20 spatial maps on artifact-cleaned data.

Techniques:

Journal: Journal of neuroscience methods

Article Title: Decoding the Encoding of Functional Brain Networks: an fMRI Classification Comparison of Non-negative Matrix Factorization (NMF), Independent Component Analysis (ICA), and Sparse Coding Algorithms

doi: 10.1016/j.jneumeth.2017.03.008

Figure Lengend Snippet: Classification accuracy averaged across 304 artifact-cleaned scans in predicting whether a subject was viewing a video, listening to an audio stimuli, or resting, using 20 components. Chance accuracy is 50%.

Article Snippet: Within each scan, the algorithms in order from best to worst classification accuracy were: L 1 Regularization, K-SVD, Fast ICA, EBM ICA, JADE ICA, InfoMax ICA, NMF-PG, and NMF-ALS, as shown in using 20 spatial maps on artifact-cleaned data.

Techniques:

Journal: Journal of neuroscience methods

Article Title: Decoding the Encoding of Functional Brain Networks: an fMRI Classification Comparison of Non-negative Matrix Factorization (NMF), Independent Component Analysis (ICA), and Sparse Coding Algorithms

doi: 10.1016/j.jneumeth.2017.03.008

Figure Lengend Snippet: Classification accuracy of BSS algorithm compared to Fast ICA, using 20 components extracted from artifact-cleaned scans, in order of performance from worst to best. Time series weights from InfoMax ICA, JADE ICA, NMF-PG, NMF-ALS predicted activity significantly worse than Fast ICA, while L1-Regularization did significantly better ( p < 0.001). Baseline is set to Fast ICA untuned SVM classification accuracy. Scan-ID and Subject-ID were included as random effects within a general linear mixed-effects regression model to adjust for multiple comparisons.

Article Snippet: Within each scan, the algorithms in order from best to worst classification accuracy were: L 1 Regularization, K-SVD, Fast ICA, EBM ICA, JADE ICA, InfoMax ICA, NMF-PG, and NMF-ALS, as shown in using 20 spatial maps on artifact-cleaned data.

Techniques: Activity Assay

Journal: Journal of neuroscience methods

Article Title: Decoding the Encoding of Functional Brain Networks: an fMRI Classification Comparison of Non-negative Matrix Factorization (NMF), Independent Component Analysis (ICA), and Sparse Coding Algorithms

doi: 10.1016/j.jneumeth.2017.03.008

Figure Lengend Snippet: Greater sparsity for an extracted spatial map was associated with a higher classification accuracy in predicting a subject’s task during scan time when using those spatial maps for encoding ( p < 0.001), holding constant the effect of the algorithm. Using 20 components extracted from artifact-cleaned scans, sparsity was measured using the negative averaged number of zero-valued voxels of all spatial maps, which is insensitive to the scaling of the individual algorithms. Baseline is set to Fast ICA untuned SVM classification accuracy. Scan-ID and Subject-ID were included as random effects within a general linear mixed-effects regression model to adjust for multiple comparisons.

Article Snippet: Within each scan, the algorithms in order from best to worst classification accuracy were: L 1 Regularization, K-SVD, Fast ICA, EBM ICA, JADE ICA, InfoMax ICA, NMF-PG, and NMF-ALS, as shown in using 20 spatial maps on artifact-cleaned data.

Techniques:

Journal: Journal of neuroscience methods

Article Title: Decoding the Encoding of Functional Brain Networks: an fMRI Classification Comparison of Non-negative Matrix Factorization (NMF), Independent Component Analysis (ICA), and Sparse Coding Algorithms

doi: 10.1016/j.jneumeth.2017.03.008

Figure Lengend Snippet: Encodings using spatial maps with high intensity in CSF regions had reduced classification accuracy, while spatial maps with variable extractions in white-matter and grey-matter regions had higher classification accuracy. Baseline is set to Fast ICA untuned SVM classification accuracy. Scan-ID and Subject-ID were included as random effects within a general linear mixed-effects regression model to adjust for multiple comparisons.

Article Snippet: Within each scan, the algorithms in order from best to worst classification accuracy were: L 1 Regularization, K-SVD, Fast ICA, EBM ICA, JADE ICA, InfoMax ICA, NMF-PG, and NMF-ALS, as shown in using 20 spatial maps on artifact-cleaned data.

Techniques: