A few science stories I had a hand in recently:
A method called laser scanning fluorescence lifetime microscopy (FLIM) looks good for peering into certain tissues, using different decay rates of fluorophores within a sample to compute an image. But the fields of view possible have so far been tiny. German specialists Becker & Hickl have tweaked the technique so that it can scan an 18 millimeter area, which is much more like it.
Also: ultrasound is commonplace, but the necessary electronics in the technique can interfere with the subtle ultrasound signals you’re looking to find. Using optical components rather than electronics and detecting the signals with a fibre-optics sensor would be better, and UCL has designed an all-optical system that can deliver high enough frame rates to make it clinically useful.
And: heavy water is already used to label proteins and lipids, since it’s close enough to normal H2O to be incorporated into metabolic processes but can then be followed with a mass spectrometer. But mass specs are big and cumbersome, so being able to do it optically would be easier. Raman spectroscopy might hold the answer, and offer a way to effectively identify the exact boundaries of cancer tumors, a perennial problem—so perennial that the FLIM technology mentioned above is being put to the same use too.