Hacker News new | past | comments | ask | show | jobs | submit login

>In practice, this doesn't work as well for more complicated images such as brains or abdomens. There is just much more to draw, and the images cease to be sparse.

Compressed sensing works just fine for brains, abdomens etc.. because the images need to be sparse in some basis (e.g. wavelets), not in the original space.




I haven't seen any good compressed sensing reconstruction for brains, though I'm not 100% up to date on the literature. Can you give me a cite?

The best I've seen is Candes/Donoho's reconstruction based on curvelets, but they don't offer any real improvement over regular reconstruction.


Check out Lustig's thesis. He finds that you can keep the same brain image quality while accelerating the sampling considerably.

http://yonit.stanford.edu/cgi-bin/axs/ax.pl?http://www-mrsrl...


Has someone been able to improve the (spatial or temporal) resolution of fMRI scans using this technique?


Thanks.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: