Recording neuron in actions at depth by unmixing scattered light


Fluorescence is one of the most common tools for biomedical imaging. In neuroscience, in particular, it is an essential tool that can be used in genetically modified animals to monitor the activity of neurons: when a particular neuron activates (emits an electrical pulse), it can generate fluorescent light, therefore allowing optical monitoring of the brain activity in real-time. For multiple active neurons, this can result in a myriad of blinking sources to be imaged. This is called functional imaging. However, neuronal tissues are highly heterogeneous and thus opaque to light. An extensive set of non-invasive or invasive techniques for scattered light rejection, optical sectioning, or localized excitation, have been developed, but non-invasive optical recording of activity through a highly scattering layer beyond the ballistic regime is to date impossible.

Propagation of light through a complex medium is subject to many scattering processes, but even for fluorescent light, it may retain some information about its source. Here, we show that functional signals from fluorescent time-varying sources located below a highly scattering tissue can be retrieved efficiently by exploiting matrix factorization algorithms to demix this information from low contrast fluorescence patterns. We performed a proof-of-principle experiment, mimicking neurons with fluorescent beads of the same size that we can excite at will, but hidden below a mouse skull (totally opaque). We showed that we were able to retrieve the temporal activity of multiple such  “neurons” (up to about 20) even though their fluorescence was blurred and overlapping. While still in its early stage, this technique opens a fascinating perspective in neuroscience, to record and study neuronal activity at unprecedented depth.


Arxiv :

Team Complex media optics lab: