When I was a kid, I always wondered how Superman knew how to stop his X-ray vision. In other words, how did he manage to look through the wall, but not continue looking through the other side of the house?
Now as a father, I wonder the same thing about ultrasound? How does it know how to stop looking through the various levels of the abdomen, and stop at at the fetus? I was once told that the sex of the fetus couldn’t be determined because he/she had its legs crossed. Why can’t the ultrasound see through the legs?
To the ultrasound question: the density of the material the ultrasound is looking at makes the difference. Bone is denser than skin and muscle tissue; thus the ultrasound passes through the skin and muscle, but does not pass through bone. If bones are blocking the view of the skin or muscle tissue (such as would be needed to determine a fetus’s sex…no “boner” jokes, please), you won’t see what’s below it.
Just a theory, in Superman’s case. In rare instances, when overloaded, Superman couldn’t turn off his X ray vision, and ended up seeing through everything. I’d say he knows how much “power” to use to see through a certain amout of material.
I’d apply the same thing to ultrasound. They have a certain power level, and can penetrate that. They can adjust it to see certain details.
Just like you can focus your eyes on near and far objects, Superman probably has a way of focusing his eyes on areas closer and farther inside the item he was X-raying. His telescopic vision probably worked along the same lines.
Ultrasound is used for these tasks? Not CAT scans for first, PET for second, or X-ray and a radioactive tracer?
Not that I know much about the techniques hospitals use for imaging, just didn’t think you could do that with ultrasound…
Maybe if they have the device up against the skull, the bone transmits the vibrations, and the interior can still be determined?
Ultrasound works the same way as radar. You transmit a pulse, and listen for a while. The farther the sound travels before it bounces off something, the longer the time until you hear it come back. You just stop listening after the farthest thing you want to see has time to bounce the signal back, wait a while longer for later returns to die out, then send the next pulse.
They do use ultrasound for scanning an adult heart. Ultrasound can take real-time movies. CAT scanners can only take snapshots, and since each snapshot takes several seconds I don’t think they are suited for fast-moving organs like the heart.
In processing seismic reflection data (an exploration technology analgous to medical ultrasound as a remote imaging method) we commonly make a spherical divergence correction. The assumption made is that the wave energy naturally decays as the wave propagates nearly spherically in the medium, with the wave energy being proportional to the square of the surface wave velocity. This assumption is, of course, not correct, because it assumes isotropic media and we encounter those neither in the earth nor in the human body. But, hey! It’s a starting point. Seismic signals are rectified in processing by multiplying the reflection amplitude of later samples by an exponential factor (2).
That being said, I’ll WAG ahead. Seismic data are typically recorded for 6 to 8 seconds in exploration work (engineering work uses much shorter records, with much higher sample rates). Since ultrasound exams, in my experience, are conducted in real time, I’ll guess that they have a target range (say 1" below surface for checking out your gall bladder or deeper to see the little baby) and they set the gain on the machine to focus at a particular depth of intrusion, using a generalized average velocity that may vary with the depth of inspection.
Upon further imbibing, er, uh, excuse me, thought, it occurs that there must be an integral frequency filter involved in real time imaging as well. Mind you all that I’m applying experience with remote imaging of the earth to same with the body and I’m WAGging. If you gain up a certain time interval you’ll see relative amplitudes for that depth of penetration, but you’ll also see a lot of shallow stuff that you’re not investigating at the moment, and, without a filter, those events’ll be hot (and will interefere with the interpretation).
The signal you put into the medium loses frequency content with depth of penetration, so, if you want to see something 3" in you’ll have to filter out the higher frequency signals that you’ll be picking up off of surface or shallow stuff, in a real time imaging environment. You need to analyze the data to determine the frequency spectrum in your target range. So, besides the gain I addressed earlier, I would imagine they must run a real time bandpass filter on the returned signal (to primarily cut out the high end).
You know what really bugs me? This (and my previous reply) is on topic, yet I know I need to rewrite it to make it more accessible.
OK, that’s my whack at it. Do you suppose we might actually 'net someone who knows?
Ugh! Break me out of the static data evaluation mode I live in. Upon further rumination it seems (still WAGing) that they must have a reflection time notch filter (or, i.e., they’re looking at a set time and nothing above or below that time, time being analgous to depth). I would like to take a closer look at a sonogram machine sometime. I’m guessing a bandpass frequency filter is still part of it, but the gain plus time window must be the parameters that drive such an investigation. Well, really, the time window would (seemingly) have to drive what results are garnered.
If I’d just go off and search the web, I could probably learn enough to say something, but I do that often enough in this forum and just post links to stuff I don’t really know. I’m pushing this one in the hopes one who knows will appear.
My experience with ultrasound exams includes my colleagues’ pre-birth images of their sons and daughters and my own singular experience (abdominal hernia mapping). IIRC, the gal driving the thing was using a Mr. Microphone sized source/receiver. The things I’d be curious about would include: 1) is it a point source/point receiver setup or is it 2) an array of sources/receivers?