Automated registration of multisource and multiresolution images for brain mapping and 3D reconstructions.
Reference persons SANTA DI CATALDO
External reference persons Francesco Ponzio (firstname.lastname@example.org)
Federico Luzzati, NICO (http://www.nico.ottolenghi.unito.it/)
Research Groups GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA
Description Histological analysis is mainly a hierarchical process in which multiple analyses of the same specimen are performed at different scales. Typically, the experimenter manually identifies the region of interests where to perform low and high resolution analyses. Such manual interpretation is prone to errors, it is not easily standardized and causes a considerable loss of spatial information. Moreover, processed histological sections are subjected to considerable non-linear deformations, a major source of errors and imprecision for several applications. In collaboration with Prof. Federico Luzzati at Neuroscience Institute Cavalieri Ottolenghi (NICO) we will thus implement a novel automated method to improve precision and standardization of brain histology. This method, that we will call 3Dhist, is based on the acquisition of block-face images during sectioning to produce a reference 3D volume of the sample and the fast low magnification imaging of stained sections with a conventional or AXIOSCAN slide scanner. By registering stained sections to the block-face volume, each slide point should be precisely re-mapped to its original position in the un-sectioned brain. A major strength of this method is that it is fully compatible with conventional sample processing and will not burden researchers with additional time consuming steps.
In this project 3Dhist will be applied to resolve the connectivity of a novel population of neurons that is produced in response to excitotoxic lesions in the mice striatum (Nato et al. 2015, PMID:25655705). These cells were sparsely labelled with a retrovirus and their afferent inputs has been revealed with a rabies virus. Preliminary results indicate that these newborn neurons are connected with a wide range of neurons spanning multiple areas, suggesting a potential role in compensatory plasticity during recovery. The standardized 3Dhist maps of each sample will be mainly used 1) to compare the spatial distribution of projecting and receiving neurons in different samples 2) To perform 3D reconstructions of individual cells inputs.
This project will partially performed at NICO (http://www.nico.ottolenghi.unito.it/) under the supervision of Prof. Federico Luzzati.
Required skills Computer programming, Image Processing
Deadline 03/06/2020 PROPONI LA TUA CANDIDATURA