Single-cell analysis (SCA) has revolutionized approaches to understanding complex biological systems, especially on the role of cellular heterogeneity in tissue development, health, and disease. Despite its recognized impact in basic biological sciences and clinical diagnosis/prognosis from its early successes, further progress of SCA remains hampered by its fundamental challenge in obtaining measurement with high information content on a large number of independent cells at high throughput.  This unmet capability to provide comprehensive catalogue of single-cells holds the key to further our understanding on tissue development and diseases, such as on the role of stem/progenitor cells in cancer and degenerative disorders.

 Optical imaging is regarded as one of the most effective tools to visualize living cells with high spatiotemporal resolution. However, its full adoption for high-throughput SCA has been hampered by the intrinsic speed limit imposed by the prevalent image capture strategies, which involve the laser scanning technologies, (e.g. galvanometric mirrors), and/or the image sensors (e.g. charge-coupled device (CCD) and complementary metal-oxide-semiconductor (CMOS)). The laser scanning speed is fundamentally limited by the mechanical inertia of the mirrors whereas the image capture rate of CCD or CMOS is ultimately limited by the required image sensitivity, i.e. scaling the frame rate inevitably compromises the detected signal intensity. Notably, this speed-versus-sensitivity trade-off of the image sensor explains why the throughput of flow cytometry has to be scaled down from 100,000 cells/sec to 1,000 cells/sec when the imaging capability is incorporated [1].

 To address these challenges, we adopt two related techniques to enable single-cell imaging with the unprecedented combination of imaging resolution and speed. Sharing a common concept of all-optical laser-scanning by ultrafast spatiotemporal encoding of laser pulses, these techniques, time-stretch imaging [2-6] and free-space angular-chirp-enhanced delay (FACED) imaging [7] enable ultrahigh-throughput single-cell imaging with multiple image contrasts (e.g. quantitative phase and fluorescence imaging) at a line-scan rate beyond 10’s MHz (i.e. an imaging throughput up to ~100,000 cells/sec). Moreover, they also enable quantification of intrinsic biophysical markers of individual cells – a largely unexploited class of single-cell signatures that is known to be correlated with the overwhelmingly investigated biochemical markers. All in all, these ultrafast single-cell imaging platforms could find new potentials in automated (deep) learning complex biological processes from such an enormous size of image data (from molecular signatures to biophysical phenotypes), especially to unveil the unknown heterogeneity between different single cells and to detect (and even quantify) rare aberrant cells. 

References:

1. Andy K. S. Lau, Ho Cheung Shum, Kenneth K. Y. Wong and Kevin K. Tsia "Optofluidic time-stretch imaging – an emerging tool for high-throughput imaging flow cytometry ," Lab on a Chip, 10.1039/C5LC01458A (2016).
2. T. W. Wong, et. al., “Asymmetric-detection time-stretch optical microscopy (ATOM) for ultrafast high-contrast cellular imaging in flow,” Sci. Rep. 4, 3656 (2014).
3. Goda, et al., “High-throughput single-microparticle imaging flow analyzer,” Proc. Natl Acad. Sci. USA 109, 11630 (2012). 
4. L. Chen, et. al., Deep learning in lable-free cell classification, Sci. Rep., 6, 21471 (2016).
5. Anson H. L. Tang, P. Yeung, Godfrey C. F. Chan, Barbara P. Chan, Kenneth K. Y. Wong, and Kevin K. Tsia, "Time-stretch microscopy on a DVD for high-throughput imaging cell-based assay," Biomed. Opt. Express 8, 640-652 (2017).
6. Queenie T. K. Lai al., "High-throughput time-stretch imaging flow cytometry for multi-class classification of phytoplankton," Opt. Express 24, 28170-28184 (2016).
7. Jianglai Wu, al., "Ultrafast Laser-Scanning Time-Stretch Imaging at Visible Wavelengths," Light: Sci.& Appl. 6, e16196 (2017).