High-speed volumetric multiphoton microscopy with deep-learning contrast enhancement for brain function investigation

To unveil how a brain functions, it is necessary to comprehend the underlying hardware and software, similar to understanding a computer. However, the main obstacle is lacking a suitable tool with enough spatiotemporal resolution to observe the physiological dynamics of an intact brain. We are developing an advanced optical imaging platform for high-speed, high-resolution, and large volume observation to decode brain functional connectome. In the conference I will report our recent progress on contrast improvement of multiphoton system, including (1) novel laser development to boost the signal strength of SRS and CARS; (2) a tunable acoustic gradient (TAG) lens-based two-photon video-rate volumetric imaging with a TAG-SPARK deep-learning denoising algorithm via spatial redundancy. The corresponding applications on Drosophila and mouse brain functions will be exemplified in the presentation.