Development of optical technologies aiming to reverse engineer neural circuits has been flourishing over the past two decades. Multi-Photon Laser Scanning Microscopy (MPLMS) together with the development of fast kinetic fluorescent calcium dyes has revolutionised the world of modern neuroscience. This technology enables mesoscale functional imaging in deep scattering brain tissues of large two (2D) and three dimensional (3D) neural networks. With single cell sensitivity in vitro as well as in vivo, it is one of the main contenders for deciphering higher brain functions. My approach in this thesis is to develop and test new scanning techniques for fast functional calcium imaging aiming to enhance the temporal precision of the acquisition. To avoid the slow and sequential "point" raster scanning nature of these Galvanometric Scanners (GSs) based microscopes, I developed new 2D and 3D scanning algorithms. These algorithms were developed in MATLAB with a simulation platform that models the main mechanical elements of the MPLSM. Both my 2D Adaptive Spiral Scanning (SSA) algorithm and my 3D Orbital Scanning Trajectory (OST) algorithm were developed to minimize the inertial slowdowns of the GSs and Electrical Tunable Lens (ETL) and therefore increase the temporal resolution of the acquisition. In 2D, I tested the SSA algorithm on in vitro hippocampal brain slices loaded with the synthetic calcium dye Cal520. To assess the performance of the scanning technique, I used the Cramer-Rao Bound (CRB) as a metric for signal quality. The CRB estimates the time of occurrence of an Action Potential (AP) from the calcium imaging data, taking into account the sampling frequency and the SNR of the acquisition. In this thesis, I show that the use of scanning strategies enables sampling rates one order of magnitude higher than traditional frame scanning in functional calcium imaging. I also show that frame scanning needs considerably higher SNR values than scanning strategies to reach the same temporal precision. In 3D, I implemented the scanning algorithms into the software and hardware of the MPLSM and recorded the trajectory of the focal point with a high-speed camera as a proof of principle. More analyses regarding the precision of the paths needs to be carried out in 3D for functional calcium imaging in vitro or in vivo. These software-based scanning strategies are attractive as they are inexpensive, easily transferable from one setup to another and enable fast functional calcium imaging with standard commercial MPSLMs. Finally, through this implementation of scanning strategies, I recorded multiple data sets of spontaneous and evoked activity in populations of Dentate Granular Cells (DGCs). This lead to the new beginning of a larger in vitro investigation at the microcircuit level on the functionality of the DG.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:739655 |
| Date | January 2017 |
| Creators | Schuck, Renaud |
| Contributors | Schultz, Simon |
| Publisher | Imperial College London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/10044/1/57510 |
Page generated in 0.0066 seconds