Fig. 2: Characterization of the system.

a, Theoretical simulations. Left top: PSFs of different systems. Systems with a high NA (0.3) have a high lateral resolution but a very small DOF (10 μm). Systems with low NA (0.1 NA) have low lateral resolution and a moderate DOF (90 μm). The MFIAS system achieves high NA and large DOF simultaneously. Left bottom: axial intensity distributions of different systems. Right: a simulation of a neuron with different levels of out-of-focus blur ranging from 0 to 225 μm. For high-NA systems, the image quality is acceptable within a 25-μm range of defocus, whereas for low-NA systems, the detection efficiency is reduced. Scale bar, 20 μm. b, Validation of MFIAS on RUSH system. Left: FWHM changes between 1.4 and 1.6 μm with the insertion of glass of different thicknesses, from 0 to 1.8 mm, measured with 0.5 μm fluorescent microspheres and 0.8 μm pixel size. Central black mark: median. Bottom and top edges: 25th and 75th percentiles. Whiskers extend to extreme points excluding outliers (1.5 times above or below the interquartile range). Right: FWHMs across the entire FOV (except for the four corner cameras, which are left blank in the figure) for glass thicknesses of 0 (minimal thickness), 0.98 mm (thickness for system design) and 1.8 mm (largest thickness). Beads number, n = 802, 961, 1,158, 1,269, 1,303, 1,294, 1,469, 1,435, 1,430 and 1,500 for each glass thickness. c, MFIAS enables the capture of a mouse brain image with FITC injection in vessels. Left: the brain image and its depth map. Middle: top projection of the mouse brain from MFIAS using a customized macroscope with SLR lenses. Three local zoom-ins on the right demonstrate that while conventional microscopes have out-of-focus blur, MFIAS enables globally all-in-focus imaging (n = 3). The dashed box indicates the in-focus area without MFIAS. Scale bars, 1 mm for the global views and 100 μm for the local views.