Who of you needs an overview on the lecture?

So would you like to summarize the previous lecture because we're now starting a new chapter?

And should we do a summary of the previous lecture or do you say okay I got everything

in focus?

Rather not.

Rather not?

Rather yes.

Should we do a summary of the lecture what we've heard so far?

We can vote?

We can vote?

How about a vote?

Some new tool that we could explore in this lecture?

So who wants to summarize the lecture so far?

Who would rather skip a summary?

Why do you want to skip the summary?

Maybe we have next week another summary.

Who wants to do the summary next week?

That's a pretty clear vote.

So today we're starting a new chapter.

So, because so far, okay then let's do the summary.

So far we did everything we have discussed so far was of course interventional medical

image processing.

And what we heard so far were a couple of refreshers, right?

So we were talking about a couple of tools.

Who remembers one of these tools?

Yes you voted against the summary so you must know everything, right?

SVD.

See SVD that's the answer.

Okay so SVD.

Good.

What's that good for SVD?

For linear equations.

So can you give an example?

Yes, so what's a famous example for linear equations?

Yes awesome.

Ax equals, so this is a vector, equals to b.

And then exactly what you said we solved with the pseudo inverse.

And how do we do that?

Well yes.

SVD.

Ah yeah you take the SVD.

So you multiply from the left side with the pseudo inverse.

And then you get a pseudo inverse here times the vector b and there you go you have solved

for x.

Of course we need to say that x is the unknown.

So if b is the unknown this is pretty trivial.

Okay so what happens if you have something like this?

So this is a vector of zeros.

What do you do then?

Can you solve this with the pseudo inverse?