| 12 | | (...) |
| | 12 | Techniques based on imaging serial sections of brain tissue provide |
| | 13 | insight into brain structure and function. However, to compare or |
| | 14 | combine it with results from three dimensional imaging methods such |
| | 15 | data requires reconstruction into a volumetric form. Currently, |
| | 16 | there are no tools for performing such a task in a streamlined way. |
| | 17 | |
| | 18 | Here we propose the Possum volumetric reconstruction framework which |
| | 19 | provides a selection of 2D to 3D image reconstruction routines |
| | 20 | allowing one to build workflows tailored to one's specific |
| | 21 | requirements. The main components include routines for |
| | 22 | reconstruction with or without using external reference and |
| | 23 | solutions for typical issues encountered during the reconstruction |
| | 24 | process, such as propagation of the registration errors due to |
| | 25 | distorted sections. We validate the implementation using synthetic |
| | 26 | datasets and actual experimental imaging data derived from publicly |
| | 27 | available resources. We also evaluate efficiency of a subset of the |
| | 28 | algorithms implemented. |
| | 29 | |
| | 30 | The Possum framework is distributed under MIT license and it |
| | 31 | provides researchers with a possibility of building reconstruction |
| | 32 | workflows from existing components, without the need for low-level |
| | 33 | implementation. As a consequence, it also facilitates sharing and |
| | 34 | data exchange between researchers and laboratories. |
