Fig. 3: Isotropic restoration of super-resolution imaging data. | Light: Science & Applications

Fig. 3: Isotropic restoration of super-resolution imaging data.

From: Deep self-learning enables fast, high-fidelity isotropic resolution restoration for volumetric fluorescence microscopy

Fig. 3

a Volume rendering of the Z-STED (100% depletion power for Z), partial Z-STED imaging data, and Self-Net output of nanobeads. Scale bar: 1 μm. b XZ cross-sections indicated by the blue dashed boxes in a. Scale bar: 0.5 μm. c 3D resolution quantification of the data in (a). d Color-coded XY MIP of the raw iSIM data after deconvolution. Scale bar: 5 μm. e, f Color-coded XZ MIPs and their typical cross-sections indicated by the white dotted line in (d) of the raw data (e) and Self-Net-restored data (f). Scale bar: 5 μm. g, h Enlarged views of the corresponding regions indicated by the blue box in (e) and (f), respectively. The green arrowheads highlight the hollow structures of mitochondria, which were blurry in the original axial view and resolved in the Self-Net reconstructed image. Scale bar, 1 μm. i Optical resolutions of the raw data and Self-Net-restored data by measuring the FWHMs of sub-diffractive objects in the corresponding image volumes

Back to article page