Indice degli argomenti
My main research field is medical imaging, in which I am involved in two main aspects: the realization and characterization of new imaging techniques, in particular regarding X-rays, and the process, analysis and classification of medical images.
I worked in the realization of a fast detection system, able to characterize the spectral distribution of x-rays produced by diagnostic machines (FLUXEN/XPRESS projects).
My research activity also investigated the feasibility of new x-ray sources, based on Thomson back-scattering, that can provide quasi-monochromatic, energy-tunable beams. This
activity was conducted in the framework of MAMBO/BEATS projects (funded by INFN), in collaboration with the Brookhaven National Laboratory (USA) and European Synchrotron Radiation Facility (France). A detailed investigation of image quality, dosimetric and phase contrast aspects was conducted.
In 2012 I leaded a project aimed at developing a system for dual-energy angiography and computed angiotomography, exploiting convetional X-ray tubes and K-edge filters.
Currently (since 214), I am involved in a project for the realization of the first world facility for the breast tomography using synchrotron radiation (Syrma-CT/Syrma-3D experiments), to be realized at the Elettra synchrotron in Trieste.
I also have a long experience in Computer Aided Detection systems, in particular for mammography and lung CT (projects CALMA,GP-CALMA, MAGIC-5 and PRIN 2005). I participated in the development of several algorithms for image segmentations, identification of regions of interest (ROIs) and feature extractions. A massive use of state-of-the art classification techniques and machine learning tools was made, in order to optimize detection chances of the systems.
Recently I started to investigate image classification in structural MRI (NextMR project), in particular concerning the Autistic Spectrum Disorder, both in toddlers and adults. From MRI scans, brain is segmented into atlas-based regions of interest from which structural features (volume, surface, thickness) are extracted. Features are then investigated my means of machine learning techniques, looking for differences between pathological and normal groups.
I am also involved in the development of innovative techniques for the characterization of cultural heritages: in the TANDEM project I am working in the development of a muon spectroscopic technique for material characterization. In collaboration with RIKEN/RAL facilty (UK).