High-spatiotemporal-throughput video microscopy of freely moving organisms with diffractive multiplexing across a sensor array

Conventional microscopes exhibit a tradeoff among spatial resolution, field of view (FOV), and frame rate. To overcome this tradeoff, we designed a highly parallelized and multiplexed computational microscope that increases the spatiotemporal throughput by two orders of magnitude, achieving tens of gigapixels per second. Specifically, our microscope features a non-contiguous array of 48 camera sensors, with the information in the gaps multiplexed across the sensor array via an optimized diffractive pupil in a single synchronized snapshot. We applied our computational microscope to image dozens of freely-moving C. elegans in parallel at few-micron resolution over a multi-cm2 FOV at up to 40 fps.