You are not authenticated, login.
text: sort by
tags: modified
type: chronology
hide / / print
ref: -0 tags: spike sorting variational bayes PCA Japan date: 04-04-2012 20:16 gmt revision:1 [0] [head]

PMID-22448159 Spike sorting of heterogeneous neuron types by multimodality-weighted PCA and explicit robust variational Bayes.

  • Cutting edge windowing-then-sorting method.
  • projection multimodality-weighted principal component analysis (mPCA, novel).
    • Multimodality of a feature is by checking the informativeness using the KS test of a given feature.
  • Also investigate graph laplacian features (GLF), which projects high-dimensional data onto a low-dimensional space while preserving topological structure.
  • Clustering based on variational Bayes for Student's T mixture model (SVB).
    • Does not rely on MAP inference and works reliably over difficult-to sort data, e.g. bursting neurons and sparsely firing neurons.
  • Wavelet preprocessing improves spike separation.
  • open-source, available at http://etos.sourceforge.net/

hide / / print
ref: Schmidt-1984.12 tags: Schmidt spike sorting PCA date: 12-20-2011 23:34 gmt revision:1 [0] [head]

PMID-6396456[0] Computer separation of multi-unit neuroelectric data: a review

  • goes through the standard, by then well-established ideas: template matching, PCA, spike amplitude, peak-to-peak amplitude, Fourier analysis, curve fitting, spike area, rms value.
  • These are all useful features, though template matching seems the standard now..
  • Gerstein and Clark 1964 -- stored spikes on tape, then sampled the tape until a threshold was exceeded. 32 samples of the waveform around threshold crossing were stored for analysis on the computer; up to 7000 points could be saved.
  • also looked at cross-correlation of a spike with a template -- back in 1968 on a LINC-8!
  • Reviews a good number of other very clever spike sorting techniques for using the lmiited hardware available.
  • Talk about template realignment and resampling Mambrito and De Luca 1983


[0] Schmidt EM, Computer separation of multi-unit neuroelectric data: a review.J Neurosci Methods 12:2, 95-111 (1984 Dec)

hide / / print
ref: -0 tags: LDA PCA autoregressive EMG date: 07-29-2008 23:35 gmt revision:3 [2] [1] [0] [head]

Below, emg classification by computing the autoregressive coefficients and feeding them into linear discriminant analysis (LDA). LDA code from here; data in myopen svn. Nine classes of movement in the data, 4 repetitions of each. The input data is 16-dimensional: 4 AR coefficients per 4 channels. This is consistent with Blair Lock's thesis.

For reference, here is an imagesc() of the raw coefficients (the 4 different color bands correspond to the 4 different channels):

hide / / print
ref: bookmark-0 tags: neural_networks machine_learning matlab toolbox supervised_learning PCA perceptron SOM EM date: 0-0-2006 0:0 revision:0 [head]

http://www.ncrg.aston.ac.uk/netlab/index.php n.b. kinda old. (or does that just mean well established?)

hide / / print
ref: bookmark-0 tags: Bayes Baysian_networks probability probabalistic_networks Kalman ICA PCA HMM Dynamic_programming inference learning date: 0-0-2006 0:0 revision:0 [head]

http://www.cs.ubc.ca/~murphyk/Bayes/bnintro.html very, very good! many references, well explained too.