|
Coherent, bright and focused light to resolve neural circuits| old_uid | 16752 |
|---|
| title | Coherent, bright and focused light to resolve neural circuits |
|---|
| start_date | 2018/11/22 |
|---|
| schedule | 11h30 |
|---|
| online | no |
|---|
| summary | Scientists have been seeking for over a century to uncover the neural circuits controlling brain functions. This requires visualization of complex nanometre size structures over millimetre size tissue volumes. Today only electron microscopes are capable of generating data with sufficient resolving power to access directly the neural connections. But mapping just one cubic mm of brain tissue would require years of sample sectioning and data collection, leading to prohibitive costs. For these reasons, this goal has not been reached yet and the largest brain ever mapped corresponds to a fruit fly, which took about five years of work. In this context, I will discuss the potential of phase contrast X-ray microscopy to probe large tissue samples at nanometre scale in a short time. X-ray holographic nano-tomography is a coherent imaging technique which is capable of generating exceptional contrast in soft tissue through phase shift retrieval. The penetrating power of X-rays and the full-field, free space propagation setup enable rapid multiscale 3D imaging of large samples which are opaque to visible light. By combining a highly brilliant X-ray nanoprobe, a carefully designed nanopositioning system, a cutting edge detection system and cryogenics for sample preservation, isotropic 3D spatial resolution better than 30 nm is conceivable. Additionally, I will discuss the possibility to use machine learning approaches, specifically convolutional neural networks, to extract the knowledge from highly complex and large 3D images. |
|---|
| responsibles | Sadoul |
|---|
| |
|