简介：

Tagging different cell populations and tracking them over time in-vitro and in-vivo is fundamental to understand important biological processes, from tumor metastasis to embryo development. Techniques based on staining with fluorescent molecules with different emission wavelengths are generally used for this purpose; however, the large bandwidth of such emitters (50-100 nm) in most cases reduces the number of different populations that can unambiguously be tagged at the same time to very few. And most of these emitters are designed in the visible wavelength range, which is hard to be detected in deep tissue with strong scattering.

In this talk, I will first introduce the fabrication of large numbers of laser particles with a variability of emission wavelength as novel contrast agents for biomedical research. Then I will talk about the laser particle stimulated emission (LASE) two-photon microscopy we have recently developed for deep tissue cell tracking with hundreds of different channels. Both the longer emission wavelength and the non-linear power dependence of stimulated emission from laser particles provide a much longer penetration depth than that of the conventional two-photon microscopy.

In the meantime, with the rapid development of the genetically encoded voltage indicators (GEVI) of neuronal activities, people can observe the changes of cellular membrane voltage at sub-millisecond scale. However current microscopy meets a great challenge in high-speed imaging at the low-light condition and restricts the widespread applications of GEVIs, especially for high-speed large-scale neural recordings. The low quantum yield of most GEVIs and small exposure time for high-speed detection make the signal noise ratio (SNR) of voltage imaging much lower than conventional fluorescence imaging. To maintain the sufficient sampling rates of the voltage imaging, we can only use the scientific cameras with a selected Region of Interest(ROI) and pixel binning due to its limited pixel readout rate, which is a large sacrifice of the pixel number. I will present a scanless random access microscopy by adaptive sampling. The summation of each ROI is conducted directly by optics to increase the SNR and imaging speed. With the capability of our system, we demonstrate large-scale optical recording of the voltage signal on ~50 self-spiking HEK cells, transfected with Archon as GEVI and stably expressing Na_v1.3 and K_ir2.1, at over 1000 fps in a single trial, which, to our best knowledge, has never been observed before.

报告人简介：

Jiamin Wu (吴嘉敏) is a PhD student at the Tsinghua University, Beijing, where he is a member of Broadband Network & Digital Media Lab, under the supervisor of Professor Qionghai Dai. He received his bachelor’s degree in the Department of Automation from Tsinghua University in 2014. His current research interests focus on computational microscopy and high-speed imaging, with a particular emphasis on developing computation-based optical setups for observing complicated biological dynamics.