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

Okay, hello everyone and welcome to the lecture on interventional image processing.

As you can see, Professor Hornegger is not here today.

I will give the lecture instead and the topic will be diagnostic and interventional perfusion imaging.

Maybe a few words about myself. My name is Andreas Fieselmann and I'm a PhD candidate in Professor Hornegger's group.

And I'm actually working on perfusion imaging.

So what will we be talking about today?

First, I will give an introduction into the topic.

Then I will tell something about diagnostic perfusion imaging, which is done with CT and MR.

And that covers image acquisition and image processing.

And it has a special focus on the deconvolution based image analysis.

So that will be the most part of the lecture.

The second part will then be about interventional perfusion imaging, where I will present the motivation and challenges and current approaches.

So why is perfusion imaging interesting?

You probably all know stroke.

It happens quite often and it's actually the third leading cause of death in Western countries.

And in most of the strokes, there is some major artery in the brain that is blocked by a clot, as you can see here in this image.

It's about 85% of all strokes.

And if this artery is blocked, then of course the tissue that is supplied by this artery does not get enough blood and does not get enough oxygen.

And eventually it will die.

So you can see it's a very major disease.

So what do you think?

Every minute a stroke is left untreated, how many neurons will die?

Is it 20,000, 200,000, 2 million or 20 million?

Who thinks A is correct?

B, C, okay, most of you think C is correct and D?

Actually it's about 2 million neurons that die every minute.

So this means you should be very fast in your stroke therapy.

You have to diagnose the stroke very fast and then you start therapy so you don't lose too many neurons.

And this is where perfusion imaging comes into the play.

Perfusion imaging is used to measure the blood flow at the capillary level, so really at the basic level where the oxygen exchange occurs.

And then it's used to diagnose stroke correctly such that the therapy can be started in a correct manner.

And especially one wants to find out are there tissue regions that profit from certain kinds of therapy because often the stroke has already evolved such that tissue regions cannot be treated anymore

and the therapy would even be dangerous to the patient.

So I really decide do I want to make the therapy or not?

This is why perfusion imaging is so important.

And when we talk about perfusion we often use the term cerebral blood flow or CBF.

And this refers to the rate of delivery of arterial blood to a capillary bed in tissue.

In units it's maybe easier to explain. It means how many millilitres of blood come per minute to a certain mass of tissue.

And in a healthy human brain it is about 60 ml per 100 gram per minute.

And if it falls below 10 ml per 100 gram per minute then the tissue will die.

So you can set some threshold values and decide when you see the CBF if you want to make therapy or not.

You can compute this parameter at various positions in the brain.

In one of the next slides I will present how a CT perfusion scan is actually done.

So you have some image data which is reconstructed at many voxel positions.

And for each of the voxel positions you can compute this parameter.

And then you usually display this parameter by using so-called colour map which basically maps one scalar value to a colour.

And in this image here you can see the CBF map of a 69 year old male stroke patient

which presented to the hospital with an acute right high grade hemiparesis on the right side.

This means the patient had some weakness on the right side and this is a typical sign for stroke.