Computer Vision News - May 2023

41 Maria Chiara Fiorentino to assess fetus growth and its assessment is strongly dependent from clinicians expertise posing issues of both intra- and inter-operator variability and time consuming. Experimental results have demonstrated that distance field produces precise delineations of Head Circumference (HC) boundaries invariably from shapes, heterogeneous appearances and gestational age of fetus head (Figure 1). Concerning rheumatology domain, Deep Learning in musculoskeletal Ultrasound has been receiving significant attention and it is expected to be pivotal in the development of next- generation cutting-edge Ultrasound imaging systems.  In this regard, I have developed a framework able to automatically estimate cartilage-thickness from Ultrasound images, paving the way for future research in the field. Cartilage damage is one of the most relevant determinants of physical disability in patients who suffer from Rheumatoid Arthritis (RA), a chronic disease characterized by erosive symmetrical polyarthritis. From a clinical perspective, the proposed framework proved to be a valuable tool to support RA clinical examination (Figure 2). My thesis had also technology transfer implications . In fact, I had the opportunity to join Esaote R&D , one of the world’s leading producers of medical diagnostic systems, and Vicomtech eHealth and Medical Devices department, a technological center set up as a private non-profit Foundation. Within these periods, I had the opportunity to focus on implementing and validate AI-based solutions for Ultrasound and MRI.