The following content has been provided by the University of Erlangen-Nürnberg.

Okay, until we wait for the guy from the technical, we can start with a short test about the wavelets.

So, the first question is, list the three dimensions of the signal that can be given simultaneously by the wavelet analysis.

So, do you recall? It was a few weeks ago, but...

Okay, someone? So, first we start talking about time domains, filters, then frequency domains, filters, and now the wavelets are the...

time domain analysis, so the three dimensions are intensity, frequency, and time.

Okay, the second question is, list three mother wavelets, and which two parameters play an important role in generating the daughter wavelets?

So, do you remember some wavelets names for mother wavelets?

One has very specific shape, like...

Yeah, Mexican hat.

Morellet, yes.

And maybe the most famous one is Debechev.

And what are the two parameters that plays an important role?

So, what do you change when you are making daughter's wavelets out of mother's? Do you remember those two parameters?

Yeah, translation, and the other one is...

So, translation or tau, or the other one which is related to the frequency, it's called scaling or S.

Does anybody want to draw the frequency response?

Okay, so what is on the x-axis? What...so it's frequency, and on the y-axis, it's intensity.

So, what is important here to know? Are the wavelets at lower frequencies wider or narrower than the wavelets in the higher frequencies?

That's actually an important thing to know in order to be able to plot it.

Lower frequencies are narrower.

Lower frequencies are narrower.

So, you will have something like this.

Okay, I need a new chalk.

So, the wavelets at the lower frequencies are narrower, and do you know what is the value of this? So, when you sum them all up,

you will get the flat line at which intensity, if they are orthogonal wavelets. It will be one.

Okay, and the last question is, mention three examples of biomedical signal analysis using wavelets.

So, do you remember some applications from the last lecture?

Do you remember that snowboarder, the picture of the snowboarder, so what were you examining there?

So, you were actually looking at the muscle activation.

Then, you were also, I think, talking about the measurement of the thumb.

So, in the finger muscle, what were you looking at? The force production of the muscle.

And, of course, because this is time frequency analyzed, you can actually look at the influences from the different sources,

okay.

Okay, good day, everybody. Sorry for this delay. Today, we will talk about event detection.

So, because we are a little bit late, we will skip the recap of the wavelets. We did the short test.

So, as I said, today we will start a new chapter that's event detection. And, as always, we will start with the definition.

So, event is a dynamic phenomenon whose behavior changes enough over time so as to be considered qualitatively significant change.

Each such change is an event, and the QRS complex is a good example of it.

So, the specific problem we address is of applying data mining techniques to identify the time points at which the changes or at which these events occurs.

And, actually, what does that mean? That in lots of biomedical signal processing applications, you will actually want to know the exact point in time

when some event happens in order to analyze some intervals more specifically.

So, the good example is the QRS detection, when you can actually analyze RR intervals, so the heart rate for lots of different applications.

But, besides the QRS complex, you can also in ECG analyze, for example, P wave if you want to see what's going on in the atrial of the heart.

So, you will actually, depending on your applications and depending on what you are interested in, you will look at the different events.

And, you can also analyze events in the sense that you want to see the time intervals in the signals for diagnostic purposes.

But, you can also use event detection if you want to communicate with a device, for example, for a therapy.

And, one example is, for example, patients who have problems with the fallen feet, so they cannot really, to say, move very well,

so they cannot lift their foot when they are walking, so actually they are just pulling their feet.

You can help them by using the functional electrical stimulation, so place the electrodes on the foot,