Parallel coordinates: a tool for visualizing multi-dimensional geometry
VIS '90 Proceedings of the 1st conference on Visualization '90
Performance evaluation of PCA-based spike sorting algorithms
Computer Methods and Programs in Biomedicine
Hi-index | 0.01 |
In vitro neuronal networks are known to fire in synchronized bursting events (SBEs), with characteristic temporal width of 100ms. We treat these events as the principal data atoms of the network. Applying singular value decomposition (SVD) (or Principal component analysis, PCA) to the spatial information, i.e. activity of neurons per burst, we demonstrate the characteristic changes that take place over time scales of hours. We consider this as an evidence for synaptic plasticity. We discover clusters of SBEs in the reduced SVD space, representing behavior of the experiments at different times. We find two interesting characteristics of SVD analysis of these data, which may be helpful to future users of SVD and PCA.