Division of Genomic Sciences and Cancer, The John Curtin School of Medical Research, College of Health and Medicine, The Australian National University, Canberra, ACT 2601, Australia
Hanqi Lin
Division of Genomic Sciences and Cancer, The John Curtin School of Medical Research, College of Health and Medicine, The Australian National University, Canberra, ACT 2601, Australia
Yean J. Lim
Division of Genomic Sciences and Cancer, The John Curtin School of Medical Research, College of Health and Medicine, The Australian National University, Canberra, ACT 2601, Australia
Philip R. Nicovich
Cajal Neuroscience, 1616 Eastlake Ave. E, Seattle, Washington 98102, USA
Katharina Gaus
EMBL Australia Node in Single Molecule Science, School of Medical Sciences, University of New South Wales, Sydney, Australia
Woei Ming Lee
Division of Genomic Sciences and Cancer, The John Curtin School of Medical Research, College of Health and Medicine, The Australian National University, Canberra, ACT 2601, Australia
Single-objective scanning light sheet (SOLS) imaging has fueled major advances in volumetric bioimaging because it supports low phototoxic, high-resolution imaging over an extended period. The remote imaging unit in the SOLS does not use a conventional epifluorescence image detection scheme (a single tube lens). In this paper, we propose a technique called the computational SOLS (cSOLS) that achieves light sheet imaging without the remote imaging unit. Using a single microlens array after the tube lens (lightfield imaging), the cSOLS is immediately compatible with conventional epifluorescence detection. The core of cSOLS is a Fast Optical Ray (FOR) model. FOR generates 3D imaging volume (40 × 40 × 14 µm3) using 2D lightfield images taken under SOLS illumination within 0.5 s on a standard central processing unit (CPU) without multicore parallel processing. In comparison with traditional lightfield retrieval approaches, FOR reassigns fluorescence photons and removes out-of-focus light to improve optical sectioning by a factor of 2, thereby achieving a spatial resolution of 1.59 × 1.92 × 1.39 µm3. cSOLS with FOR can be tuned over a range of oblique illumination angles and directions and, therefore, paves the way for next-generation SOLS imaging. cSOLS marks an important and exciting development of SOLS imaging with computational imaging capabilities.
