Skip to content

ICASSP 2010 Non-stationarity detection

2010/03/15

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2010 just started today. I have a paper titled Quantification of Inter-trial Non-stationarity in Spike Trains from Periodically Stimulated Neural Cultures

 

Graph showing drastic non-stationarity at some point in time

One of the slides from the presentation (long term plasticity). Y axis is divergence of a moving window to the fixed window of data indicated as the dark boxes below.

The slide above is the example I show for detecting long-term plasticity from a neural culture. Here we are stimulating the culture electrically for 5 minutes, followed by 5 minutes of recovery period is given, and the same stimulus is given for 5 minutes again. The above divergence plot shows that the culture went through a dramatic change in its firing pattern, and the Hellinger divergence on point process that we propose in this paper captures the change in the underlying probability law. The change was not recovered during the resting period, hence we claim that it was a long-term effect. For details of our method, please come to the presentation. (MATLAB implementation available as IOCANE)

Unfortunately, I will not be presenting the paper myself, but fortunately my advisor Jose C. Principe will be. It’s going to be presented during the special session SS-L2: Multivariate and Multimodal Analysis of Brain Signals, Room A3, Tuesday, March 16, 1:50 pm.

Please visit my colleagues’ presentations and posters as well.

  • Steven Van Vaerenbergh, Ignacio Santamaria, Weifeng Liu, Jose C. Principe. Fixed-budget kernel recursive least-squares (Learning Theory and Models I, Wednesday, March 17, 13:30 – 13:50)
  • Antonio Paiva, Tolga Tasdizen. Fast semi-supervised image segmentation by novelty selection (Poster Area G, Wednesday, March 17, 10:00 – 12:00)
  • Sohan Seth, Jose C. Principe. A conditional distribution function based approach to design nonparametric tests of independence and conditional independence (Learning Theory and Models III, Wednesday, March 17, 16:00 – 18:00)
  • Abhishek Singh, Jose Principe. Kernel width adaptation in information theoretic cost functions (Learning Theory and Models III, Wednesday, March 17, 16:00 – 18:00)
  • Abhishek Singh, Jose Principe. A closed form recursive solution for maximum correntropy training (Learning Theory and Models III, Time: Wednesday, March 17, 16:00 – 18:00)
  • Abhishek Singh, Tejaswi Tamminedi, Guy Yosiphon, Anurag Ganguli, Jacob Yadegar. Hidden markov models for modeling blood pressure data to predict acute hypotension (Bioinformatics and Biomedical Signal Processing, Tuesday, March 16, 16:00 – 18:00)
Advertisements
3 Comments leave one →
  1. 2010/03/15 4:30 pm

    Nice post! It reminded me that I should also wrap up a toolbox with my code some day.

  2. smh permalink
    2010/03/15 6:47 pm

    is it a plasticity result? sounds to me the stimulus just kills or run-down the neurons. :p any other evidence that this is THAT case?

    • memming permalink*
      2010/03/15 7:12 pm

      Hi, thanks for your comment.
      I didn’t really show the activity of the culture here, but they are still alive and firing after that change. Remember what I am plotting is not the firing rate or activity, it is the change in statistics. It is probably not a simple synaptic plasticity that we usually expect, but it was as if the whole system went into a new dynamic state. There’s always a possibility that this was a chemical process that damaged some neurons, so your point is valid. This culture was alive for days after the experiment.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: