Disturbances in cerebral hemodynamics are one of the principal causes of cerebral damage in premature infants. Specifically, changes in cerebral blood flow might cause ischemia or hemorrhage that can lead to motor and developmental disabilities. Under normal circumstances, there are several mechanisms that act jointly to preserve cerebral hemodynamics homeostasis. However, in case that one of these mechanisms is disrupted the brain is exposed to damage. Premature infants are susceptible to variations in cerebral circulation due to their fragility. Therefore, monitoring cerebral hemodynamics is of vital importance in order to prevent brain damage in this population and avoid subsequent sequelae. This thesis is oriented to the development of signal processing techniques that can be of help in monitoring cerebral hemodynamics in neonates. There are several problems that hinder the use in clinical practice of monitoring cerebral hemodynamics. On one hand, ...

Every year, an estimated 15 million babies are born preterm, that is, before 37 weeks of gestation. This number is rising in all countries and currently represents more than 1 in 10 babies, affecting families all over the world. During the last decades, the survival rate of prematurely born neonates has steadily increased, mainly as a result of medical and technical progress in neonatal intensive care. The very preterm infants, which represent up to 10% of the preterm infants in the EU, remain at risk for adverse outcome and neurodevelopmental disability. These maladaptive outcomes have a severe effect on the children’s quality of life and a huge economic impact on society. In order to reduce this burden and improve neonatal care in general, appropriate tools need to be developed to identify the neonates with a higher risk of adverse outcomes. ...

A complex physical system like the human brain can only be comprehended by the use of a combination of various medical imaging techniques, each of which shed light on only a specific aspect of the neural processes that take place beneath the skull. Electroencephalography (EEG) and functional magnetic resonance (fMRI) are two such modalities, which enable the study of brain (dys)function. While the EEG is measured with a limited set of scalp electrodes which record rapid electrical changes resulting from neural activity, fMRI offers a superior spatial resolution at the expense of only picking up slow fluctuations of oxygen concentration that takes place near active brain cells. Hence, combining these very complementary modalities is an appealing, but complicated task due to their heterogeneous nature. In this thesis, we devise advanced signal processing techniques which integrate the multimodal data stemming from ...

Work-related musculoskeletal disorders are a growing problem in todays society. These musculoskeletal disorders are caused by, amongst others, repetitive movements and mental stress. Stress is defined as the mismatch between a perceived demand and the perceived capacities to meet this demand. Although stress has a subjective origin, several physiological manifestations (e.g. cardiovascular and muscular) occur during periods of perceived stress. New insight and algorithms to extract information, related to stress are beneficial. Therefore, two series of stress experiments are executed in a laboratory environment, where subjects underwent different tasks inducing physical strain, mental stress and a combination of both. In this manuscript, new and modified algorithms for electromyography signals are presented that improve the individual analysis of electromyography signals. A first algorithm removes the interference of the electrical activity of the heart on singlechannel electromyography measurements. This interference signal is ...

Hyperspectral (HS) imaging, which consists of acquiring a same scene in several hundreds of contiguous spectral bands (a three dimensional data cube), has opened a new range of relevant applications, such as target detection [MS02], classification [C.-03] and spectral unmixing [BDPD+12]. However, while HS sensors provide abundant spectral information, their spatial resolution is generally more limited. Thus, fusing the HS image with other highly resolved images of the same scene, such as multispectral (MS) or panchromatic (PAN) images is an interesting problem. The problem of fusing a high spectral and low spatial resolution image with an auxiliary image of higher spatial but lower spectral resolution, also known as multi-resolution image fusion, has been explored for many years [AMV+11]. From an application point of view, this problem is also important as motivated by recent national programs, e.g., the Japanese next-generation space-borne ...

Magnetic Resonance Imaging (MRI) constitutes a non-invasive medical imaging technique that allows the exploration of the inner anatomy, tissues, and physiological processes of the body. Among the different MRI applications, functional Magnetic Resonance Imaging (fMRI) has slowly become an essential tool for investigating the brain behavior and, nowadays, it plays a fundamental role in clinical and neurophysiological research. Due to its particular nature, specialized signal processing techniques are required in order to analyze the fMRI data properly. Among the various related techniques that have been developed over the years, the General Linear Model (GLM) is one of the most widely used approaches, and it usually appears as a default in many specialized software toolboxes for fMRI. On the other end, Blind Source Separation (BSS) methods constitute the most common alternative to GLM, especially when no prior information regarding the brain ...

During the last decades, information is being gathered and processed at an explosive rate. This fact gives rise to a very important issue, that is, how to effectively and precisely describe the information content of a given source signal or an ensemble of source signals, such that it can be stored, processed or transmitted by taking into consideration the limitations and capabilities of the several digital devices. One of the fundamental principles of signal processing for decades is the Nyquist-Shannon sampling theorem, which states that the minimum number of samples needed to reconstruct a signal without error is dictated by its bandwidth. However, there are many cases in our everyday life in which sampling at the Nyquist rate results in too many data and thus, demanding an increased processing power, as well as storage requirements. A mathematical theory that emerged ...

In this thesis we investigate centralized and distributed variants of sparse Bayesian learning (SBL), an effective probabilistic regression method used in machine learning. Since inference in an SBL model is not tractable in closed form, approximations are needed. We focus on the variational Bayesian approximation, as opposed to others used in the literature, for three reasons: First, it is a flexible general framework for approximate Bayesian inference that estimates probability densities including point estimates as a special case. Second, it has guaranteed convergence properties. And third, it is a deterministic approximation concept that is even applicable for high dimensional problems where non-deterministic sampling methods may be prohibitive. We resolve some inconsistencies in the literature involved in other SBL approximation techniques with regard to a proper Bayesian treatment and the incorporation of a very desired property, namely scale invariance. More specifically, ...

Understanding the cognitive processes that are going on in the human brain, requires the combination of several types of observations. For this reason, since several years, neuroscience research started to focus on multimodal approaches. One such multimodal approach is the combination of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI). The non-invasive character of these two modalities makes their combination not only harmless and painless, but also especially suited for widespread research in both clinical and experimental applications. Moreover, the complementarity between the high temporal resolution of the EEG and the high spatial resolution of the fMRI, allows obtaining a more complete picture of the processes under study. However, the combination of EEG and fMRI is challenging, not only on the level of the data acquisition, but also when it comes to extracting the activity of interest and interpreting the ...

This dissertation explores the intersection of data fusion, federated learning, and Bayesian methods, with a focus on their applications in indoor localization, GNSS, and image processing. Data fusion involves integrating data and knowledge from multiple sources. It becomes essential when data is only available in a distributed fashion or when different sensors are used to infer a quantity of interest. Data fusion typically includes raw data fusion, feature fusion, and decision fusion. In this thesis, we will concentrate on feature fusion. Distributed data fusion involves merging sensor data from different sources to estimate an unknown process. Bayesian framework is often used because it can provide an optimal and explainable feature by preserving the full distribution of the unknown given the data, called posterior, over the estimated process at each agent. This allows for easy and recursive merging of sensor data ...

The growing interest in neuroimaging technologies generates a massive amount of biomedical data that exhibit high dimensionality. Tensor-based analysis of brain imaging data has by now been recognized as an effective approach exploiting its inherent multi-way nature. In particular, the advantages of tensorial over matrix-based methods have previously been demonstrated in the context of functional magnetic resonance imaging (fMRI) source localization; the identification of the regions of the brain which are activated at specific time instances. However, such methods can also become ineffective in realistic challenging scenarios, involving, e.g., strong noise and/or significant overlap among the activated regions. Moreover, they commonly rely on the assumption of an underlying multilinear model generating the data. In the first part of this thesis, we aimed at investigating the possible gains from exploiting the 3-dimensional nature of the brain images, through a higher-order tensorization ...

This PhD thesis considers the problem of locating some target nodes in different wireless infrastructures such as wireless cellular radio networks and wireless sensor networks. To be as realistic as possible, mixed line-of-sight and non-line-of-sight (LOS/NLOS) localization environment is introduced. Both the conventional non-cooperative localization and the new emerging cooperative localization have been studied thoroughly. Owing to the random nature of the measurements, probabilistic methods are more advanced as compared to the old-fashioned geometric methods. The gist behind the probabilistic methods is to infer the unknown positions of the target nodes in an estimation process, given a set of noisy position related measurements, a probabilistic measurement model, and a few known reference positions. In contrast to the majority of the existing methods, harsh but practical constraints are taken into account: neither offline calibration nor non-line-of-sight state identification is equipped in ...

Most cognitive processes in the brain are reflected through several aspects simultaneously, allowing us to observe the same process from different biological phenomena. The diverse nature of these biological processes suggests that a better understanding of cerebral activity may be achieved through multimodal measurements. One of the possible multimodal brain recording settings is the combination of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI), which is one of the main topics of this thesis. Two groups of EEG-fMRI integration approaches are possible. The first group, commonly called model-based techniques, are very popular due to the fact that the results from such analyses confirm or disprove a specific hypothesis. However, such hypotheses are not always available, requiring a more explorative approach to analyze the data. This exploration is possible with the second group of approaches, the so-called data-driven methods. The data-driven ...

This thesis presents techniques for the modelling of musical signals, with particular regard to monophonic and polyphonic pitch estimation. Musical signals are modelled as a set of notes, each comprising of a set of harmonically-related sinusoids. An hierarchical model is presented that is very general and applicable to any signal that can be decomposed as the sum of basis functions. Parameter estimation is posed within a Bayesian framework, allowing for the incorporation of prior information about model parameters. The resulting posterior distribution is of variable dimension and so reversible jump MCMC simulation techniques are employed for the parameter estimation task. The extension of the model to time-varying signals with high posterior correlations between model parameters is described. The parameters and hyperparameters of several frames of data are estimated jointly to achieve a more robust detection. A general model for the ...

This thesis introduces new learning algorithms for Radial Basis Function (RBF) networks. RBF networks is a feed-forward two-layer neural network used for functional approximation or pattern classification applications. The proposed training algorithms are based on robust statistics. Their theoretical performance has been assessed and compared with that of classical algorithms for training RBF networks. The applications of RBF networks described in this thesis consist of simultaneously modeling moving object segmentation and optical flow estimation in image sequences and 3-D image modeling and segmentation. A Bayesian classifier model is used for the representation of the image sequence and 3-D images. This employs an energy based description of the probability functions involved. The energy functions are represented by RBF networks whose inputs are various features drawn from the images and whose outputs are objects. The hidden units embed kernel functions. Each kernel ...