Programa de Doctorat en Tecnologies de la Informació i les Comunicacions: Envíos recientes
Mostrando ítems 1-20 de 394
High resolution 3D modelling of lumbar spine from MRI using deep learning
This thesis addresses the growing problem of degenerative disorders of the intervertebral discs (IVD). Intervertebral Disc Degeneration (IDD) represents a progressive degradation of the IVDs compromising their biomechanical ...
Industrial augmented reality prototypes in the field of art authoring, dissemination, and consumption tools for artistic content on mobile devices with a focus generation Z
The advent of spatial content and increasingly advanced mixed reality devices is set to revolutionise the manner in which we interact with digital content, becoming the predominant human-computer interface in the ...
Deep audio representation learning for music using weak supervision
Music audio tagging is the Music Information Retrieval task of assigning one or multiple labels to an audio signal. Current state-of-the-art music taggers rely on deep learning approaches, which offer high performance ...
Violin perfomance analysis using weak supervision
This thesis investigates the development of core Music Information Retrieval (MIR) technologies for Violin Performance Analysis, addressing both data and label scarcity. To mitigate data scarcity, the Violin Repertoire ...
Striking a balance global impact of the local regulation of excitatory-inhibitory balance in cortical networks
The study of complex systems has unraveled the basic principles underlying the emergence of collective behaviors in networks of interacting units. Following this framework, we suggest that excitatory-inhibitory balance ...
Normalizing flows for neuroimaging research
Psychotic disorders are characterized by poorly understood pathophysiological mechanisms, with a remarkable heterogeneity among patients, both cross-sectionally and longitudinally. This results in a lack of definitive ...
Cinematic volume rendering solutions for the advanced exploration of medical images
Medical visualization, commonly used in clinical routine for exploring cardiac data, is based on traditional techniques including multiplanar reformations, maximum intensity projection, and volume rendering that rely ...
Contributions to fully homomorphic encryption
In recent years, the field of Fully Homomorphic Encryption (FHE) has become one of the main pilars for privacy-preserving technologies. However, many challenges still remain to be tackled before this technology can be ...
Promoting socio-emotional interactions and regulation in computer supported collaborative learning: design, implementation and evaluation of technology features and affordances to facilitate the socio-emotional side of CSCL
This thesis, a compendium of publications, aims at contributing knowledge to better design CSCL tools for promoting socioemotional interactions and regulation. We started with a field study in which we observed groups ...
Beyond the spectrum: prosodic, noise-enhanced and self-supervised features for speech recognition
This dissertation investigates three methods for improving speech recognition in challenging conditions, focusing on feature enhancement and extending beyond spectral features. First, it demonstrates that incorporating ...
Blockchains in the real world: an interdisciplinary perspective on enchancing security and adoption
This thesis investigates the complex nature of the blockchain ecosystem, where technology, law, and economics intersect across layers and applications. By offering an interdisciplinary perspective on blockchains, we seek ...
Characterization of left atrial hemodynamics with 4D flow magnetic resonance imaging, in-silico models and deep learning
Although the left ventricle has traditionally held the spotlight as the primary determinant of cardiovascular disease prognosis, the characterization of left atrial hemodynamics is increasingly being recognized as a ...
Large scale off-policy and offline learning
Reinforcement Learning (RL), a subfield of machine learning and artifical intelligence, is a learning paradigm where an artificial agent learns to reach a predefined goal by trying to maximize a reward signal while ...
Prediction of mental health crises based on electronic health records: probabilistic and machine learning models for clinical applications
This thesis delves into the transformative potential of Electronic Health Records and Machine Learning in the prevention of mental health crisis relapses. Our work focuses on the prediction of these relapses throughout the ...
Toward a new model of borderline personality disorder: a psychophysiological approach on the link between emotional and behavioural dysregulation
Borderline personality disorder (BPD) is one of the most studied personality disorders, given its prevalence and the severity of its symptoms. In the last years, several models of the disorder were presented, however, no ...
The role of conflict in perception: case studies in binocular rivalry, cinema and flavour
Our everyday perception of the world is remarkably unambiguous, considering that it stems from a multitude of ambiguous sensory signals. This feat is believed to be achieved by a set of perceptual and cognitive mechanisms ...
A journey in representation learning textual emotion detection and structural graph encodings
Representation learning is the field of modern machine learning focused on automatically producing useful representations from data. This thesis studies the impact of data volumes, annotation perspectives, and structural ...
Computational analysis of fetal brain development with patient-specific modelling and deep learning techniques
This dissertation presents a novel computational pipeline for creating patient-specific models of brain development. Initially, the human brain develops from a smooth to a convoluted cortical surface, with abnormal ...
Computer-supported scripting mechanisms to support adolescents' development of social media self-protection skills
In recent years, there's been increasing concern over potential negative experiences for adolescents on social media. Despite efforts in social media literacy education, traditional approaches often neglect crucial ...
The Role of inhibitory plasticity in the formation and the long-term maintenance of neural assemblies and memories
Brain circuits display modular architecture at different scales of organization. Such neural assemblies are typically associated to functional specialization that favours both the segregation and the integration of ...
