Brain imaging today mainly relies on techniques like X-ray computed tomography and magnetic resonance angiography. However, these methods cannot image structures several microns in size. They are also long and take minutes to produce an image – something that makes it difficult to monitor blood flow in real time.

Fluorescence-based brain imaging in the visible and near-infrared (NIR) regions of the electromagnetic spectrum (400–900 nm) is a good alternative but at the moment it requires skull-thinning, or worse still, craniotomy (where sections of the skull are removed and replaced with a transparent “window”) to work properly. What is more, light at these wavelengths can only penetrate the skull at around 1 mm at best because it is scattered by biological tissues in the brain.

The NIR-IIa window

Now, a team led by Hongjie Dai and Calvin Kuo at Stanford has developed a new through-scalp and through-skull fluorescence imaging technique that goes a long way in overcoming these problems. The method makes use of the intrinsic fluorescence of single-walled carbon nanotubes in the 1.3–1.4 µm range. “We define this wavelength as the NIR-IIa window, and it represents just about the longest wavelengths for fluorescence imaging reported thus far,” explains Dai.

“Photons at these wavelengths are much less scattered than those in the 400–900 nm window when traversing biological tissues and are not absorbed significantly by water either. All in all, this allows us to see deeper into the brain through intact scalp skin and bone than is possible with traditional fluorescence imaging (which is mostly done with <800 nm wavelength photons).”

Imaging single capillary blood vessels, and fast

"Compared to all other techniques for in vivo brain imaging (including magnetic resonance imaging MRI and CT), our technique affords higher spatial resolution", he tells “It allows us to image single capillary blood vessels that are just microns across and as deep as 3 mm inside the brain.”

And that is not all: the technique is also fast, at 200 ms per frame or less, which allows blood flow to be monitored in real time. Blood flow is drastically reduced, for example, when there are arterial occlusions in patients who have had a stroke.

The researchers, who have tested their technique on mice, say that they are now busy trying to image in 3D using their method. “We are also developing imaging agents for an even longer wavelength window to further minimize photon scattering,” adds Dai. “And, we are looking at making NIR-IIa fluorophores that might potentially be used in human clinical trials.”

The research is detailed in Nature Photonics.