Learning Biomedical Signals in Bodytrack

• Excellent presentation on various ways to do time-series analysis. Skip to about page 40 (slide 240) to get to the interesting analysis.
• Learning that correlation is not always the best way to do things. In fact, why do we do correlation anyways? Don't we want to compute mean-square error?

Some pre-ventricular contractions to start off the day

• Paris Smaragdis presentation on matched filtering. On his course webpage

We got the following from Zeeshan, the guy who gave the talk about ECG analysis last month:

I wanted to send you a couple of references to the process we use to deal with ECG noise in our current work. Our specific approach is described in:

http://www.springerlink.com/content/9m715n688l42050t/fulltext.pdf

The other paper I think might be really useful in this context is:

Li Q, Mark RG, Clifford GD. Robust heart rate estimate fusion using signal quality indices and a Kalman filter. Physiol Meas. 2008; 29:15–32.

There's a lot of work proposed to deal with noise, but these are some of the things we've found to work best in practice. I hope that's helpful.