The following content has been provided by the University of Erlangen-Nürnberg.
So good morning. Good morning everybody. Today we want to open the chapter on 3D ultrasound
imaging and before we look into the algorithmic part I want to introduce to you at least the
rough concept on ultrasound and I also want to point out to a few historical remarks.
Okay? Good. So let's start with the history of ultrasound. Yesterday I asked already who
got an ultrasound already and it are quite a few of you who have received an ultrasound
imaging procedure and the nice thing with ultrasound is that it's a very cheap modality.
So it's mostly with all the doctors that are out in the field. It's not harming the human
body so there is no ionization, no X-ray, no nothing. It's just sound waves that are
performed so it's easy and the tough part starts once you have a look at the images
because ultrasound images are very noisy, low quality and they just give a rough intuition
what's going on in the inner of the human body. So reading ultrasound images is a very,
very complicated thing and the second point I want to mention is if you want to perform
image processing using ultrasound images you are also not on the sunny side of life because
this is very, very hard to do smart image processing on these noisy data. And the third
part I want to mention there is no other imaging modality that is so much dependent on the
person, that depends so much on the person that is acquiring the images because you are
using the ultrasound probe and you basically guide it over the human body and there is
no way to repeat this exactly this way. This is not a well defined procedure in industry
and research is also working on standardizing the acquisition procedures for ultrasound
imaging. And ultrasound was also used in the early days in Greece when they were looking
for gold, maybe they should reconsider this option today as well. They tried to find out
what is here in the ground by sending sound waves and listening to the reflections and
sound waves you know are partially reflected if you have a transition from one material
to another material and that allows you to analyze the structure once you know how the
speed behaves in the ultrasound or the sound waves behave in different materials and how
the speed is actually in different materials. And in 1942 medical ultrasound was basically
discovered by Dusik and in 1984 the first scanners were built that do not only provide
two dimensional information but also three dimensional information so you can use ultrasound
to do 3D imaging. And if you go to your doctor he might have an ultrasound device that is
already in the years and 20 years or more but they still provide images that are sufficient
to do certain diagnosis. And how are these sound waves generated basically? You have
a periodic motion and you generate by this periodic motion these longitudinal waves that
are sent into the human body and we will not go into the technical structure of such ultrasound
probes what material is used and what technology is basically or was basically developed for
this. This will be covered in lectures on medical physics. We just accept at this point
okay there are sound waves that are generated we know that sound waves are attenuated or
reflected dependent on the tissues and we use a few of these physical properties reflection
is the most important thing. So at the boundary of two media waves are not transmitted but
partially reflected. So I send them sound wave and it comes back. If you are in the
Alps and you shout the name of your girlfriend or boyfriend and then it comes back after
a while. And if you know what the sound speed is you can compute the distance. The only
complicated thing is that you have to divide by two. The result of the standard formula
gives you twice the distance because the sound waves goes to the mountain and gets reflected.
So factor of two of one half. Then you have refraction that basically means that the sound
waves are banded. They are banded. And absorption or attenuation is the third effect that we
have. So it's basically a weakening of the signal. And there are different ways of dealing
with ultrasound. For instance if you are familiar with the CT reconstruction methods we have
discussed in winter semester. If you send sound waves to the human body through the
human body you will receive an attenuated signal on the other side. And using these
Presenters
Zugänglich über
Offener Zugang
Dauer
00:43:05 Min
Aufnahmedatum
2012-05-15
Hochgeladen am
2012-05-21 12:13:29
Sprache
en-US
This lecture focuses on recent developments in image processing driven by medical applications. All algorithms are motivated by practical problems. The mathematical tools required to solve the considered image processing tasks will be introduced.