Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Biotech Medicine News Science

A High-Res 3D Video of the Embryonic Heartbeat 207

An anonymous reader writes "Researchers at the University of Houston, TX, adapted an imaging technique called optical coherence tomography to capture 3D video of the mammalian heart as it forms. They used the method to image a mouse embryo just 8.5 days past conception and about a day after it starts to form. In the remarkable video a normal heartbeat is visible. Normally optical coherence tomography is used for clinical imaging of the retina. Having such a high-resolution, non-invasive way to image the developing heart could perhaps help doctors treat congenital heart disorders in human babies."
This discussion has been archived. No new comments can be posted.

A High-Res 3D Video of the Embryonic Heartbeat

Comments Filter:
  • by jeffb (2.718) ( 1189693 ) on Sunday October 25, 2009 @07:24PM (#29867757)

    You can get arbitrarily good images of fixed (dead) embryos, but live imaging using any method is damn tough, and live 3D imaging at this resolution is, as far as I know, unprecedented. Motion makes it nearly impossible to do MR or CT 3D imaging. You can gate against the cardiac cycle to image a single animal, but nobody can yet gate against a fetal heartbeat in a mouse. I'm not even sure if that would be enough, because the maternal heartbeat contributes significant motion, too.

    One of our doctoral students did a 3D atlas of the embryonic mouse using MR microscopy [duke.edu]. These were fixed specimens, but they're isotropic (the same spatial resolution in all three dimensions), and nobody's come close to matching our resolution as far as I know. Part of her work was looking at cardiac septal defects, which you pretty much have to study in embryos, because they aren't compatible with live birth.

    One drawback of OCT is that it fails if you have to go through much tissue. Mice are tiny enough to make this work possible, but I don't think there's any way you could do it in humans, short of inserting a source/detector into the uterus, which kind of spoils the whole "non-invasive" feature.

    • You can go for a 3D video with an ultrasound transducer. That way you have your deep tissue imaging (unfortunately in 2D) coupled with superficial 3D imaging. If you are creative enough you can play around with your ultrasound transducers to get surround sound.
      • Yeah, and if you've got repeating motion, you can sort of patch together a 4D image (3D over time). Alex's atlas is actually 4D, although the different timepoints were necessarily from different specimens. (And obviously not isotropic on time -- to get a true 4D isotropic dataset, our 19.5 micron/voxel spatial resolution would demand 65-femtosecond temporal resolution, which is (a) physically impossible and (b) of no use to an anatomist.)

    • You can get arbitrarily good images of fixed (dead) embryos, but live imaging using any method is damn tough, and live 3D imaging at this resolution is, as far as I know, unprecedented.

      Not really. Well, only if you take a very mammal-centric view: chicken [youtube.com] and zebrafish [youtube.com] embryos are extremely easy to image in much, much higher resolution. It's also possible to do live imaging on partially dissected embryonic tissues. I make live slices of chicken embryos and watch subcellular organelles move around under an automated microscope with much better resolution than this study.

      Granted, that wouldn't be of any theraputic value in humans, but live imaging is not unprecedented.

      • That was a lot creepier than I meant it to be. "Humans" should have also been in bold.

      • I actually meant "mouse embryos", since that's what TFA and our lab's work was about. Yes, zebrafish are easier -- in fact, you can get varieties that stay transparent into adulthood [sciencedaily.com], which opens all sorts of possibilities. But there are a lot of things you can do in mouse models that you can't do in fish or whatnot.

  • ... but this is stereo(-2D), not (volumetric) 3D. Despite the layman definition.

  • Unless this video comes with some 3D glasses and I missed out on them, this is 2D. The image has height and width, but I can't rotate it around to see the heartbeat from the side or the back or the top or the bottom.
  • TFA has a Low res grainy 2D video, and the author readily admits: "Though it looks grainy, this and other video of the developing heart made by the Houston group are some of the best ever taken."

    I think, and probably, it is very impressive to make a videothe beating hart of a tiny mouse embryo, even if it is grainy and 2D.

    BUT WHERE THE HELL is the H.D. 3D video announced in the headline?

"You can have my Unix system when you pry it from my cold, dead fingers." -- Cal Keegan