Project Summary/Abstract
Cognitive functions of the brain are underpinned by complex and highly dynamic neural activities at the
sub-cellular levels and millisecond time scales. To discover the normal/abnormal neuronal activities and thus
understand detailed mechanisms of neurological disorders and dysfunctions such as Alzheimer’s,
Parkinson’s disease and ALS (amyotrophic lateral sclerosis), measurement tools that offer sufficient
spatiotemporal resolution are needed. Fluorescence imaging/microscopy is one of the state-of-the-art
technologies for high spatial resolution recording of the activity of neuron populations. However, existing
fluorescence neural imaging technologies generally have limited speed, providing less than a few hundred
frames per second at most. High-speed imaging is particularly challenging for miniaturized, head-mounted
imagers used for in vivo studies on freely-behaving animals. The milliseconds or slower temporal resolution
substantially precludes measuring the precise timing of the generation and propagation of neuron spikes,
which is the key component of neural signaling. Moreover, current head-mounted fluorescence imagers use
epi-fluorescence illumination, which cannot reject out-of-focus background fluorescence, resulting in low
discrimination of voltage-sensitive signals from the thin membranes of individual neurons.
During this R&D program, Physical Sciences Inc. (PSI) and the Broad Institute of MIT and Harvard
propose to develop and demonstrate a high-speed (>kHz frame rate), head-mounted, confocal imager
that can optically capture neuronal electrical activity with high spatiotemporal details. Technology
innovations are proposed to enable this capability. First, an “active sensing” signal detection method
combines two complementary imaging channels to achieve parallel neuronal recording with both sub-micron
spatial resolution and sub-millisecond temporal resolution. Second, a novel hybrid fiber bundle scanning
approach achieves confocal imaging capability based on a miniaturized optical setup. During the proposed
Phase I, we will demonstrate the feasibility of the technology by imaging cultured neurons and brain slices
labeled with voltage fluorescent indicators. Then, we will upgrade the technology to a miniaturized imager to
further demonstrate its performance during in vivo imaging of mice. These test experiments will demonstrate
high spatiotemporal resolution recording of fast action potentials from individual neurons and sub-cellular
neuron structures (e.g., dendrites and synapses). This R&D project will result in a robust technology for non-
invasive recording of neuronal kinetics with high spatiotemporal resolution on freely-behaving animals,
offering a critical tool for neuroscience research.