Idle thought:

[cdave]

Colour is (badly) defined as the relative amount of one three frequencies of light waves reflected off a visible point.

Could you make a false colour ultrasound image by using three different frequencies of ultrasound? If so would they be of any use?

Comments

[the-magician.livejournal.com]

Red/Green 3D glasses would appear to be the obvious first answer to that.

And that doesn't necessarily need different frequencies if you can use two transmitters and time slice between then and combine the two to build a three d virtual image ...

Now it may be that different frequencies penetrate differently so you might be able to do the "looking under the skin" bit ... but I have no knowledge of that.

[cdave]

There's already 3D ultrasoundout there.

But it just still just shows a monochrome image of how the much of the ultrasound signal is reflected by boundaries between different organs.

What I was envisioning was that different boundries, may reflect with the same intensity at one frequency, and thus have the same brightness in the monochrome. But at different intensities in other frequencies, thus appearing as a different colour in the composite false colour image.

[andrewducker]

I don't believe your definition of colour is correct. Colour is a combination of different wavelengths of light. Because we have three receivers in our eyes we tend to define things in combinations of the colours that each of those produces, but one could as simply use different base colours - there's nothing magical about those particular three.

[cdave]

I disagree (except that I do think it's a bad definition), those frequencies are important.

But the receptors aren't only tuned to a single wavelength, or we would be nearly blind, only able to see things in a three single frequencies. The cells have a bell shaped curve (iirc) response to light around the red, green and blue frequency, and it's the relative levels where they overlap that "defines" the qualia.

Other animals have often have a different range of frequencies that their receptor cells are tuned too. Imagine a creature which can see into the infra-red watching a human TV. Stomatopods would need a twelve colour TV rather than three.

[andrewducker]

Sorry, yes - the absorption levels drop off as you vary further from "Red", "Green" and Blue".

Hmm - one type of cone gives you a single dimension of colour - (bright/dark)- two cones would give you two dimensions (a plane) - three cones gives you three dimensions of it. I wonder if a fourth dimension of colour would feel qualitatively different, or would simply feel like our current colour perception with more detail.

[sharikkamur.livejournal.com]

There is a fourth cone - some cones (relatively few, admittedly) are sensitive to yellow. Human tetrachromacy is a rare X-linked trait, it seems.

(Jakab Z, Wenzel K, 2004, "Detecting tetrachromacy in human subjects" Perception 33 ECVP Abstract Supplement, Wikipedia article on tetrachromacy)

[cdave]

That's not right. Our vision receptors are rods and cones, and one of them (never can remember which) is responsible for night vision, and is essentially monochromatic.

You can see in 2D, because you have a 2D retina. You can see in 3D because you have binocular vision and ... hang on. This is a metaphorical dimensions of colour isn't it?

Hmm. Interesting. Talking to someone who has gone colour blind later in life is the only test I can think of.

[andrewducker]

Yes, metaphorical dimensions :->

[idaho-smith.livejournal.com]

Yes to bell-curves for sensitvity vs. wavelength for each of the colour dependant chemicals "opsins".

_C_ones for _C_olour

_R_ods for ... erm ... _R_unning about in the woods at night without a gamekeeper seeing your torch.

[sharikkamur.livejournal.com]

You can define 'false colour' using any combination of colours; all you're doing is colour-coding particular areas using a defined code set.

The ultrasound approach is more like the astronomical approach of viewing something through multiple filters and then shifting the wavelengths to the visible. Astronomical images are often false-coloured because they are made up of a set of particular wavelengths - Hα comes immediately to mind.

Of course, ultrasound is a longitudinal wave form, not a transverse waveform. How about thinking of a multi-frequency ultrasound as a polyphonic piece of music instead of a single tone?

Hmm... there's an art idea - transforming your ultrasound baby scans into orchestral pieces.

[cdave]

I think the false colour scanner idea would be possible to make, but don't know enough about the reflective properties of organs to know whether there would be any benefits to it.

As for turning it into sound, Reminds me of an idea Douglas Adams had.

[sharikkamur.livejournal.com]

Yes, I'd forgotten that one. Just goes to prove what I've thought for a while - it's time to re-read Dirk Gently.