EDoF-Miniscope: pupil engineering for extended depth-of-field imaging in a fluorescence miniscope

Extended depth of field (EDoF) microscopy has emerged as a powerful solution to greatly increase the access into neuronal populations in table-top imaging platforms. Here, we present EDoF-Miniscope, which integrates an optimized thin and lightweight binary diffractive optical element (DOE) onto the gradient refractive index (GRIN) lens of a head-mounted fluorescence miniature microscope, i.e. "miniscope". We achieve an alignment accuracy of 70 m to allow a 2.8X depth-of-field extension between the twin foci. We optimize the phase profile across the whole back aperture through a genetic algorithm that considers the primary GRIN lens aberrations, optical property of the submersion media, and axial intensity loss from tissue scattering in a Fourier optics forward model. Compared to other computational miniscopes, our EDoF-Miniscope produces high-contrast signals that can be recovered by a simple algorithm and can successfully capture volumetrically distributed neuronal signals without significantly compromising the speed, signal-to-noise, signal-to-background, and maintain a comparable 0.9-m lateral spatial resolution and the size and weight of the miniature platform. We demonstrate the robustness of EDoF-Miniscope against scattering by characterizing its performance in 5-m and 10-m beads embedded in scattering phantoms. We demonstrate that EDoF-Miniscope facilitates deeper interrogations of neuronal populations in a 100-m thick mouse brain sample, as well as vessels in a mouse brain. Built from off-the-shelf components augmented by a customizable DOE, we expect that this low-cost EDoF-Miniscope may find utility in a wide range of neural recording applications.