Fig. 1: Absorption artifacts in light sheet imaging.

a Attenuation artifacts in simple vs. complex structures. Green: fluorescent regions; gray: attenuating regions; cyan arrows: illumination directions (for simplicity, the effects of attenuation on the emitted fluorescence are not shown); dark green: non/poorly illuminated regions. Left: When the attenuating region is relatively simple (geometrically), the artifacts can be corrected by multi-view reconstruction (in the case shown, two views are sufficient). Right: For more complex attenuating structures, there are generally regions in the sample that are not clearly illuminated by any view and thus are not properly corrected by standard multi-view reconstructions. b Surface rendering of a cleared embryonic stage E12.5 mouse head, immunolabeled for Tuj1 (class III β-tubulin, a neuronal marker). The sample was imaged using both LSFM (Tuj1, white surface) and tOPT (eye pigments, cyan surface). The retina contains pigmented cells that significantly absorb light and, therefore, create contrast to visualize the eyeball. Although absorption artifacts are present in the LSFM image, if they are not recognized for what they are, they may be misinterpreted as an intrinsically weaker signal. c A 130-µm-thick slice through the fluorescence image at the level of the red dashed line in b. The light-absorbing retina casts two shadows: the horizontal shadow on the right indicates where illumination (from the left) was considerably reduced, and the vertical shadow (below the eye) indicates the regions obscured from the view of the objective lens used for detection, which is above (ill = illumination, det = detection). d, e show the reconstruction of the eye pigmentation from a tOPT scan in the region indicated by the red box in (c) and the overlay of the pigmentation and the fluorescence signal imaged in the SPIM mode, respectively. The shadow artifacts in the fluorescence data are well aligned with the eye pigmentation. Scale bars: 500 µm