Toward Millisecond Volumetric Microscopy
Fast Axial Scanning for Increased 3D Imaging Speed
- Fig. 1: Methods to acquire a z-stack. (a) Objective motion along the axial direction. (b) Multiplexing of the excitation beam to generate multiple foci at different focal planes. (c) Control of the beam divergence for remote axial focus shifting.
- Fig. 2: Fast imaging performance of the inertia-free light sheet microscope operated without synchronization of the acoustic lens. (a) 3D trajectories of beads flowing in a microchannel. Imaging was performed at 200 volumes per second over an axial extent of 100 μm. (b) In vivo imaging of a Paramecium acquired at 11 volumes per second. Volume size: 138 × 138 × 60 μm.
- Fig. 3: Synchronized acquisition of 2PE microscopy images of neuronal cells labelled with GFP from a 150 μm thick mouse brain slice. (a) A standard xy 2PE acquisition. (b) An extended depth-of-field image obtained by integrating the axial information at each pixel. (c) Three-dimensional reconstruction of the imaged volume from (b) using photon sorting. Scale bars are 20 μm. Volume size: 115 × 115 × 15 µm.
Optical microscopes are the tool-of-choice to study the dynamics of living organisms with subcellular resolution. However, current systems are optimized to acquire two-dimensional information from a single focal plane. While volumetric imaging is possible by mechanically moving the sample or focusing lens, it normally comes at the cost of sacrificing temporal resolution. Recent technological developments for fast axial focusing offer new opportunities to advance toward high-speed 3D microscopy.
Fast and non-invasive three-dimensional (3D) imaging is key to unveil the dynamics of biological processes as relevant as neuronal plasticity  or embryogenesis . Among existing characterization techniques, optical microscopes offer a unique combination of high resolution, speed, low phototoxicity and specificity that renders them particularly suitable for the fundamental study of living systems. Unfortunately, the rate at which volumetric information can be retrieved from a sample remains largely suboptimal, typically below the characteristic time scales of the phenomena under study. For instance, common microscopy modalities such as laser scanning microscopy (LSM)  or light-sheet fluorescence microscopy (LSFM)  struggle to exceed 5 volumes/second. In these cases, 3D imaging is performed by the sequential acquisition of sections from different focal planes, the so-called z-stack. Even if the time required to capture each section can be significantly decreased by using parallelization methods [5, 6] or fast scanning mirrors , the shifting of the focal plane is normally a much more time-consuming task. As shown in Figure 1a, the standard approach to capture a z-stack is based on the mechanical translation of the sample or objective, a step that requires long idle times for proper repositioning to occur. Also, vibrations caused by such a movement can result in image artifacts. Therefore, in instances when the photon budget is high, as with scattering signal or densely labeled samples, this asymmetry in imaging speed between the lateral and axial directions constitutes the main reason that prevents fast 3D microscopy.
Need for Faster z-focus Control
Several methods have been recently developed aimed at reducing the acquisition time of a z-stack.
They include the spatiotemporal multiplexing of the illumination to simultaneously acquire multiple focal planes [8-10], effectively increasing 3D imaging rate by an amount proportional to the planes imaged (fig. 1b). However, this approach can require complex setups, and more importantly, the number of imaged planes is fixed, thus prone to image artifacts when sample motion is present. Alternatively, the remote change of focus by controlling the divergence of the incident beam (fig. 1c) offers a straightforward implementation with intrinsic flexibility. Indeed, by simply combining an objective with a varifocal lens, namely a lens which focusing power can be varied, a user-controllable shift of the focal plane can be achieved. The rate at which the z-stack can be collected depends on the varifocal lens speed, which is strongly linked to the core technology of the particular lens used. In general, though, widely used varifocal liquid lenses based on the mechanical deformation of a membrane [11–13], on the electrowetting effect [14,15] or on liquid crystals  typically have a response time above 5 ms, only offering a slim advantage over traditional piezo stages used for objective motion. A technology capable of exceeding this speed by more than one order of magnitude could pave the way to a shift in the paradigm of 3D imaging. In this case, the fastest axis of the microscope would become the z-axis, opening the door to the reconstruction of a volume from a collection of sections parallel to the z-axis, that is, an x- or y-stack. Also, focus position could be selected on a pixel-by-pixel basis, enabling a voxel-based control of the regions of interest for further increase in imaging speed. The challenge is to provide and implement a technology that could reach such fast z-focus control.
Inertia-free Focusing with Sound
Our strategy consists of using a varifocal liquid lens that uses sound to induce periodic changes in the refractive index of a liquid. This lens, named a TAG lens , lacks of any mechanically moving parts, thus enabling axial focus scanning at frequencies as high as 1 MHz . The lens is a resonant device, which implies that it is continuously changing focus. Thus, the speed advantages of the lens come with a caveat: the direct selection of a given focal plane is not possible. Instead, the user can only select the axial range and speed over which the focus is scanned. So, how is it possible to perform 3D imaging with a TAG-enabled microscope? There are different approaches, which can be broadly classified into two groups based on the presence or not of synchronization between the detector/illumination sources with the lens oscillation.
Continuous z-focus scanning at microsecond time scales implies that, in the absence of synchronization, the information from multiple planes is integrated by the detector, whether it is a camera (typical exposure time of ms) or a single-point detector (typical pixel dwell time of hundreds of microseconds). As a result, an extended depth-of-field (EDOF) image is obtained [19–21]. An EDOF can preserve lateral resolution and is particularly interesting for the study of fast and sparse events such as calcium dynamics in brain , but the loss of axial position information is detrimental for accurate volume rendering. To solve this issue, we redesigned a light-sheet microscope to incorporate the TAG lens in the detection arm . Because a single plane of light (light-sheet) perpendicular to the camera is used for illumination during the acquisition of a frame, the above ambiguity can be solved: the z-position of the grabbed frame corresponds to that of the illuminated plane, which is known. Notably, as long as the light-sheet is within the TAG-induced EDOF, the image will look sharp (in focus), and thus the speed at which a z-stack can be collected only depends on the camera frame rate and on how fast the light-sheet can be scanned. By using a fast camera and acousto-optic deflectors for moving the light-sheet, we could obviate the need of any mechanical actuation and reach unsurpassed 3D imaging rates up to 200 volumes/second at diffraction-limited resolution . Figure 2 shows examples of two dynamic systems that were studied with our inertia-free LSFM.
Alternatively, the reconstruction of a z-stack with a TAG-enabled microscope is possible by using pulsed light synchronized with the lens oscillation [17,20]. Pulses can be sent that always encounter the same lens oscillation state, hence illuminating a fixed focal plane. A change of the delay time between light pulses enables the reconstruction of a z-stack. This approach works best in wide-field detection systems that use relatively slow detectors such as cameras. Instead, when using single point detectors, which typically present a response time of picoseconds, the use of synchronized pulse light can be avoided. In this case, it suffices to sort the collected photons based on their arrival time with respect to the lens position. In other words, information from multiple focal planes is continuously acquired, but provided the knowledge of lens oscillation state, a whole volume can be rendered in a post-processing step. This approach has been successfully demonstrated in LSM systems including confocal  and two-photon excitation (2PE) microscopy [25–27]. Interestingly, the beam shaping abilities of the TAG lens can also be used to enhance the signal to noise ratio of a 2PE microscope, providing a novel way to increase both 3D imaging speed and penetration depth . Figure 3 illustrates the possibilities offered by our 2PE microscope.
The use of an acoustic optofluidic lens for axial focus scanning significantly increases volumetric acquisition speed. Implemented in different microscopy architectures with either synchronized or unsychronized configurations, the achieved z-focus control enables imaging rates only limited by detector speed or sensitivity. With the development of faster and more sensitive detectors, high-resolution 3D imaging at even higher rates will be possible, offering researchers a unique tool for the study of natural processes with unprecedented temporal detail.
Simonluca Piazza1, Giuseppe Sancataldo2, Marti Duocastella1
1Nanophysics, Istituto Italiano di Tecnologia, Genoa, Italy
2European Laboratory for Non Linear
Spectroscopy (LENS), Italy
1Nanophysics, Istituto Italiano di Tecnologia, Genoa, Italy
2European Laboratory for Non Linear Spectroscopy (LENS), Italy
 N. Ji, J. Freeman, and S. L. Smith: Technologies for imaging neural activity in large volumes, Nat. Neurosci., 19(9): 1154–1164 (2016) doi: 10.1038/nn.4358
 J. Huisken and D. Y. R. Stainier: Selective plane illumination microscopy techniques in developmental biology, Development, 136(12): 1963–75, (2009) doi: 10.1242/dev.022426
 J. Mertz: Optical sectioning microscopy with planar or structured illumination, Nat. Methods, 8(10): 811–9 (2011) doi: 10.1038/nmeth.1709
 J. Huisken, J. Swoger, F. Del Bene, J. Wittbrodt, and E. H. K. Stelzer: Optical Sectioning Deep Inside Live Embryos by Selective Plane Illumination Microscopy, Science 305: 1007–1009 (2004) doi: 10.1126/science.1100035
 G. Zhu, J. van Howe, M. Durst, W. Zipfel, and C. Xu: Simultaneous spatial and temporal focusing of femtosecond pulses, Opt. Express, 13(6): 2153–2159 (2005) doi: org/10.1364/OPEX.13.002153
 D. Oron and Y. Silberberg: Temporal focusing microscopy, Cold Spring Harb. Protoc., 2015(2): 145–151 (2015) doi: 10.1101/pdb.top085928
 G. Y. Fan, H. Fujisaki, A. Miyawaki, R.-K. Tsay, R. Y. Tsien, and M. H. Ellisman: Video-Rate Scanning Two-Photon Excitation Fluorescence Microscopy and Ratio Imaging with Cameleons, Biophys. J., 76(5): 2412–2420 (1999) doi: 10.1016/S0006-3495(99)77396-0
 A. Cheng, J. T. Gonçalves, P. Golshani, K. Arisaka, and C. Portera-Cailliau: Simultaneous two-photon calcium imaging at different depths with spatiotemporal multiplexing, Nat. Methods, 8(2): 139–142 (2011) doi: 10.1038/nmeth.1552
 W. Amir, R. Carriles, E. E. Hoover, T. a Planchon, C. G. Durfee, and J. a Squier: Simultaneous imaging of multiple focal planes using a two-photon scanning microscope, Opt. Lett., 32(12): 1731–1733 (2007) doi: org/10.1364/OL.32.001731
 S. Abrahamsson et al.: Fast multicolor 3D imaging using aberration-corrected multifocus microscopy, Nat. Methods 10(1): 60–3 (2013) doi: 10.1038/nmeth.2277
 L. Maffl, S. Rosset, M. Ghilardi, F. Carpi, and H. Shea: Ultrafast All-Polymer Electrically Tunable Silicone Lenses, Adv. Funct. Mater., 25: 1656–1665 (2015) doi: org/10.1002/adfm.201570075
 F. O. Fahrbach, F. F. Voigt, B. Schmid, F. Helmchen, and J. Huisken: Rapid 3D light-sheet microscopy with a tunable lens, Opt. Express 21(18): 21010–26 (2013) doi: 10.1364/OE.21.021010.
 G. Sancataldo et al.: Three-dimensional multiple-particle tracking with nanometric precision over tunable axial ranges, Optica 4(3): 367, 2017 doi: org/10.1364/OPTICA.4.000367
 B. Berge and J. Peseux: “Variable focal lens controlled by an external voltage: An application of electrowetting,” Eur. Phys. J. E, 3(2): 159–163 (2000) doi: org/10.1007/s101890070029
 L. Li, D. Wang, C. Liu, and Q.-H. Wang: Zoom microscope objective using electrowetting lenses, Opt. Express, 24(3): 2931 (2016) doi: 10.1364/OE.24.002931.
 M. Martinez-Corral, P.-Y. Hsieh, A. Doblas, E. Sanchez-Ortiga, G. Saavedra, and Y.-P. Huang: Fast axial-scanning widefield microscopy with constant magnification and resolution, J. Disp. Technol., 11(11): 913 (2015) doi: 10.1109/JDT.2015.2404347
 A. Mermillod-Blondin, E. McLeod, and C. B. Arnold: High-speed varifocal imaging with a tunable acoustic gradient index of refraction lens, Opt. Lett. 33: 2146–2148 (2008) doi: org/10.1364/OL.33.002146
 M. Duocastella and C. B. Arnold: Transient response in ultra-high speed liquid lenses, J. Phys. D. Appl. Phys., 46: 071502 (2013) doi: 10.1088/0022-3727/46/7/075102
 N. Olivier, A. Mermillod-Blondin, C. B. Arnold, and E. Beaurepaire: Two-photon microscopy with simultaneous standard and extended depth of field using a tunable acoustic gradient-index lens, Opt. Lett., 34(11): 1684–1686 (2009) doi: org/10.1364/OL.34.001684
 M. Duocastella, B. Sun, and C. B. Arnold: Simultaneous imaging of multiple focal planes for three-dimensional microscopy using ultra-high-speed adaptive optics, J. Biomed. Opt., 17(5): 050505 (2012) DOI: 10.1117/1.JBO.17.5.050505
 M. Duocastella and C. B. Arnold: Enhanced depth of field laser processing using an ultra-high-speed axial scanner, Appl. Phys. Lett., 102(6): 061113 (2013) doi: org/10.1063/1.4791593
 R. Lu et al.: Video-rate volumetric functional imaging of the brain at synaptic resolution, Nat. Neurosci., 20(4): 620–628 (2017) doi: 10.1038/nn.4516
 M. Duocastella, G. Sancataldo, P. Saggau, P. Ramoino, P. Bianchini, and A. Diaspro: Fast Inertia-Free Volumetric Light-Sheet Microscope, ACS Photonics, 4: 1797–1804 (2017) doi: 10.1021/acsphotonics.7b00382
 M. Duocastella, G. Vicidomini, and A. Diaspro: Simultaneous multiplane confocal microscopy using acoustic tunable lenses, Opt. Express, 22(16) 19293-19301 (2014) doi: org/10.1364/OE.22.019293
 L. Kong et al.: Continuous volumetric imaging via an optical phase-locked ultrasound lens, Nat. Methods, 12: 759–762 (2015) doi: 10.1038/nmeth.3476
 L. Kong, J. Tang, and M. Cui: Multicolor multiphoton in vivo imaging flow cytometry, Opt. Express, 24(6): 6126–6135 (2016) doi: 10.1364/OE.24.006126
 S. Piazza, P. Bianchini, C. Sheppard, A. Diaspro, and M. Duocastella: Enhanced volumetric imaging in 2-photon microscopy via acoustic lens beam shaping, J. Biophotonics, 11: e201700050 (2018) doi: 10.1002/jbio.201700050