Computer Vision News Computer Vision News 38 Congrats, Doctor Francesca! The motivation for Francesca’s thesis lies in the awareness that the increasing volume of archived medical data, especially medical imaging, represents an invaluable source of information to improve the diagnostic process. In her thesis, Francesca addressed specific challenges related to three different imaging modalities using DL methodologies to support clinicians in the diagnostic process and enhance patient care. The first application uses DL to enhance clinical endoscopy in otolaryngology, specifically for assessing vocal folds (VF) motility. VF are muscular structures located in the larynx, responsible for vocalization, breathing, and airway protection. Neurological and inflammatory diseases can impair VF movements, leading to paralysis. Diagnosing VF paralysis relies on visually examining endoscopic videos, a subjective and timeconsuming task. Francesca's DL approach automates the estimation of five clinically relevant keypoints of the larynx, providing a reliable and objective tool for VF motility assessment. Francesca Pia Villani recently completed her PhD in the VRAI (Vision Robotics and Artificial Intelligence) Lab at the University of Macerata, Italy. Her research focused on developing deep learning (DL) methods for analyzing medical images stored in healthcare archives. Francesca is now a postdoctoral researcher in the Department of Information Engineering at the Università Politecnica delle Marche in Ancona, Italy. Congrats, Doctor Francesca! Francesca (second from the left) after her PhD defense. On the far left and right are her supervisors (Emanuele Frontoni and Sara Moccia, respectively), whose guidance and mentorship were instrumental throughout her research journey. Also present is one of the defense committee members: Muhammad AbdulMageed from the University of British Columbia (second from the right), who provided valuable insights during the defense.
RkJQdWJsaXNoZXIy NTc3NzU=