SiROP opportunities [x-ray]

Advancing intensity-based iterative reconstruction for grating interferometry breast CT
Breast cancer is the most common cancer type in women. At ETH, a dedicated tomographic scanner (CT) based on grating interferometry (GI) is operated and its potential to improve early stage cancer detection is investigated. Unlike conventional CT scanners (based only on the attenuation of the x-ray beam), GI allows to additionally obtain information from the beam’s refraction and diffusion in the sample. In order to reconstruct the projection data from a GICT scan, dedicated reconstruction algorithms are needed. In our group, an intensity-based iterative algorithm was developed and successfully tested for circular scans. In order to allow to reconstruct helical scanning data from our GICT system, the forward and backward operators will have to be adjusted accordingly, which is the scope of this project. The code adjustments should then be tested with simulations and measurement data.
Web application for tracking micro-fabrication processes
The aim of this project is to develop a database and web-interface for the X-ray micro-fabrication team at PSI and ETHZ to track realized pieces, fabrication protocols and characterization of the products.
Quantitative thoracic dark-field imaging: simulation and experimental validation
Dark-field imaging is an emerging medical diagnostic tool, which allows to assess the diffusion of unresolved micro structures. This has yielded promising results in imaging organs like the lung more efficiently and could allow for new screening opportunities to improve treatment of diseases like the chronic obstructive pulmonary disease (COPD). The dark-field contrast, however, is difficult to model and little to no data is available to improve simulation and analysis capabilities. The goal of the thesis is to utilize a novel wave propagation simulation tool and validate its ability to quantitatively simulate and model dark-field images of known lung samples with results from measured data.