Functional connectivity analysis of the neural bases of emotion regulation: A comparison of independent component method with density-based k-means clustering method

Abstract

Abstract. Functional magnetic resonance imaging (fMRI) is an important tool in neuroscience for assessing connectivity and interactions between distant areas of the brain. To find and characterize the coherent patterns of brain activity as a means of identifying brain systems for the cognitive reappraisal of the emotion task, both density-based k-means clustering and independent component analysis (ICA) methods can be applied to characterize the interactions between brain regions involved in cognitive reappraisal of emotion. Our results reveal that compared with the ICA method, the density-based k-means clustering method provides a higher sensitivity of polymerization. In addition, it is more sensitive to those relatively weak functional connection regions. Thus, the study concludes that in the process of receiving emotional stimuli, the relatively obvious activation areas are mainly distributed in the frontal lobe, cingulum and near the hypothalamus. Furthermore, density-based k-means clustering method creates a more reliable method for follow-up studies of brain functional connectivity.

Internet and Biometric Web Based Business Management Decision Support

“… Among regression algorithms the usual choices are: linear regression [173,174,175], Lasso Regression [176,177], Logistic Regression [178,179,180], Multivariate Regression [181,182], and Multiple Regression Algorithm [183,184].  Among clustering algorithms the most common choices in biometrics or neuroscience research are: K-Means Clustering [185,186,187], Fuzzy C-means Algorithm [188,189], Expectation-Maximization (EM) Algorithm [190], and Hierarchical Clustering Algorithm [188,191,192].  Among reinforcement learning algorithms the most common choices are: deep reinforcement learning [193,194,195] and inverse reinforcement learning [196].…”

A Review of AI Cloud and Edge Sensors, Methods, and Applications for the Recognition of Emotional, Affective and Physiological States

“… Among regression algorithms the usual choices are: linear regression [ 173 , 174 , 175 ], Lasso Regression [ 176 , 177 ], Logistic Regression [ 178 , 179 , 180 ], Multivariate Regression [ 181 , 182 ], and Multiple Regression Algorithm [ 183 , 184 ]. Among clustering algorithms the most common choices in biometrics or neuroscience research are: K-Means Clustering [ 185 , 186 , 187 ], Fuzzy C-means Algorithm [ 188 , 189 ], Expectation-Maximization (EM) Algorithm [ 190 ], and Hierarchical Clustering Algorithm [ 188 , 191 , 192 ]. Among reinforcement learning algorithms the most common choices are: deep reinforcement learning [ 193 , 194 , 195 ] and inverse reinforcement learning [ 196 ].…”

“…Among clustering algorithms the most common choices in biometrics or neuroscience research are: K-Means Clustering [ 185 , 186 , 187 ], Fuzzy C-means Algorithm [ 188 , 189 ], Expectation-Maximization (EM) Algorithm [ 190 ], and Hierarchical Clustering Algorithm [ 188 , 191 , 192 ].…”

Without insight accompanied with deteriorated brain functional alterations in healthy individuals with auditory verbal hallucinations: a pilot study

“…The functional connectivity of each voxel was calculated using an in-house Linux script as previously reported; briefly, Pearson's linear correlation was applied with a correlation coefficient threshold of R > 0.6 (Tomasi et al, 2010;Zou et al, 2016). The gFCD calculations were limited to those voxels within the cerebral gray matter mask, and the gFCD at any given voxel (×0) was calculated as the total number of functional connections, denoted as k (×0), between ×0 and all other voxels using a growth algorithm, which was repeated for all of the ×0 voxels.…”