Parcellation of fMRI datasets with ICA and PLS: a data driven approach

Inter-subject parcellation of functional Magnetic Resonance Imaging (fMRI) data based on a standard General Linear Model (GLM) and spectral clustering was recently proposed as a means to alleviate the issues associated with spatial normalization in fMRI. However, for all its appeal, a GLM-based parcellation approach introduces its own biases, in the form of a priori knowledge about the shape of Hemodynamic Response Function (HRF) and task-related signal changes, or about the subject behaviour during the task. In this paper, we introduce a data-driven version of the spectral clustering parcellation, based on Independent Component Analysis (ICA) and Partial Least Squares (PLS) instead of the GLM. First, a number of independent components are automatically selected. Seed voxels are then obtained from the associated ICA maps and we compute the PLS latent variables between the fMRI signal of the seed voxels (which covers regional variations of the HRF) and the principal components of the signal across all voxels. Finally, we parcellate all subjects data with a spectral clustering of the PLS latent variables. We present results of the application of the proposed method on both single-subject and multi-subject fMRI datasets. Preliminary experimental results, evaluated with intra-parcel variance of GLM t-values and PLS derived t-values, indicate that this data-driven approach offers improvement in terms of parcellation accuracy over GLM based techniques.

Learning data-driven models of non-verbal behaviors for building rapport using an intelligent virtual agent

There is a growing societal need to address the increasing prevalence of behavioral health issues, such as obesity, alcohol or drug use, and general lack of treatment adherence for a variety of health problems. The statistics, worldwide and in the USA, are daunting. Excessive alcohol use is the third leading preventable cause of death in the United States (with 79,000 deaths annually), and is responsible for a wide range of health and social problems. On the positive side though, these behavioral health issues (and associated possible diseases) can often be prevented with relatively simple lifestyle changes, such as losing weight with a diet and/or physical exercise, or learning how to reduce alcohol consumption. Medicine has therefore started to move toward finding ways of preventively promoting wellness, rather than solely treating already established illness.^ Evidence-based patient-centered Brief Motivational Interviewing (BMI) interventions have been found particularly effective in helping people find intrinsic motivation to change problem behaviors after short counseling sessions, and to maintain healthy lifestyles over the long-term. Lack of locally available personnel well-trained in BMI, however, often limits access to successful interventions for people in need. To fill this accessibility gap, Computer-Based Interventions (CBIs) have started to emerge. Success of the CBIs, however, critically relies on insuring engagement and retention of CBI users so that they remain motivated to use these systems and come back to use them over the long term as necessary.^ Because of their text-only interfaces, current CBIs can therefore only express limited empathy and rapport, which are the most important factors of health interventions. Fortunately, in the last decade, computer science research has progressed in the design of simulated human characters with anthropomorphic communicative abilities. Virtual characters interact using humans’ innate communication modalities, such as facial expressions, body language, speech, and natural language understanding. By advancing research in Artificial Intelligence (AI), we can improve the ability of artificial agents to help us solve CBI problems.^ To facilitate successful communication and social interaction between artificial agents and human partners, it is essential that aspects of human social behavior, especially empathy and rapport, be considered when designing human-computer interfaces. Hence, the goal of the present dissertation is to provide a computational model of rapport to enhance an artificial agent’s social behavior, and to provide an experimental tool for the psychological theories shaping the model. Parts of this thesis were already published in [LYL+12, AYL12, AL13, ALYR13, LAYR13, YALR13, ALY14].^

Methods for integrating machine learning and constrained optimization

In the framework of industrial problems, the application of Constrained Optimization is known to have overall very good modeling capability and performance and stands as one of the most powerful, explored, and exploited tool to address prescriptive tasks. The number of applications is huge, ranging from logistics to transportation, packing, production, telecommunication, scheduling, and much more. The main reason behind this success is to be found in the remarkable effort put in the last decades by the OR community to develop realistic models and devise exact or approximate methods to solve the largest variety of constrained or combinatorial optimization problems, together with the spread of computational power and easily accessible OR software and resources. On the other hand, the technological advancements lead to a data wealth never seen before and increasingly push towards methods able to extract useful knowledge from them; among the data-driven methods, Machine Learning techniques appear to be one of the most promising, thanks to its successes in domains like Image Recognition, Natural Language Processes and playing games, but also the amount of research involved. The purpose of the present research is to study how Machine Learning and Constrained Optimization can be used together to achieve systems able to leverage the strengths of both methods: this would open the way to exploiting decades of research on resolution techniques for COPs and constructing models able to adapt and learn from available data. In the first part of this work, we survey the existing techniques and classify them according to the type, method, or scope of the integration; subsequently, we introduce a novel and general algorithm devised to inject knowledge into learning models through constraints, Moving Target. In the last part of the thesis, two applications stemming from real-world projects and done in collaboration with Optit will be presented.

Data-driven Signal Processing per Jamming Detection in Sistemi Satellitari

In questo elaborato vengono analizzate differenti tecniche per la detection di jammer attivi e costanti in una comunicazione satellitare in uplink. Osservando un numero limitato di campioni ricevuti si vuole identificare la presenza di un jammer. A tal fine sono stati implementati i seguenti classificatori binari: support vector machine (SVM), multilayer perceptron (MLP), spectrum guarding e autoencoder. Questi algoritmi di apprendimento automatico dipendono dalle features che ricevono in ingresso, per questo motivo è stata posta particolare attenzione alla loro scelta. A tal fine, sono state confrontate le accuratezze ottenute dai detector addestrati utilizzando differenti tipologie di informazione come: i segnali grezzi nel tempo, le statistical features, le trasformate wavelet e lo spettro ciclico. I pattern prodotti dall’estrazione di queste features dai segnali satellitari possono avere dimensioni elevate, quindi, prima della detection, vengono utilizzati i seguenti algoritmi per la riduzione della dimensionalità: principal component analysis (PCA) e linear discriminant analysis (LDA). Lo scopo di tale processo non è quello di eliminare le features meno rilevanti, ma combinarle in modo da preservare al massimo l’informazione, evitando problemi di overfitting e underfitting. Le simulazioni numeriche effettuate hanno evidenziato come lo spettro ciclico sia in grado di fornire le features migliori per la detection producendo però pattern di dimensioni elevate, per questo motivo è stato necessario l’utilizzo di algoritmi di riduzione della dimensionalità. In particolare, l'algoritmo PCA è stato in grado di estrarre delle informazioni migliori rispetto a LDA, le cui accuratezze risentivano troppo del tipo di jammer utilizzato nella fase di addestramento. Infine, l’algoritmo che ha fornito le prestazioni migliori è stato il Multilayer Perceptron che ha richiesto tempi di addestramento contenuti e dei valori di accuratezza elevati.

Multimodel inference and multimodel averaging in empirical modeling of occupational exposure levels.

Empirical modeling of exposure levels has been popular for identifying exposure determinants in occupational hygiene. Traditional data-driven methods used to choose a model on which to base inferences have typically not accounted for the uncertainty linked to the process of selecting the final model. Several new approaches propose making statistical inferences from a set of plausible models rather than from a single model regarded as 'best'. This paper introduces the multimodel averaging approach described in the monograph by Burnham and Anderson. In their approach, a set of plausible models are defined a priori by taking into account the sample size and previous knowledge of variables influent on exposure levels. The Akaike information criterion is then calculated to evaluate the relative support of the data for each model, expressed as Akaike weight, to be interpreted as the probability of the model being the best approximating model given the model set. The model weights can then be used to rank models, quantify the evidence favoring one over another, perform multimodel prediction, estimate the relative influence of the potential predictors and estimate multimodel-averaged effects of determinants. The whole approach is illustrated with the analysis of a data set of 1500 volatile organic compound exposure levels collected by the Institute for work and health (Lausanne, Switzerland) over 20 years, each concentration having been divided by the relevant Swiss occupational exposure limit and log-transformed before analysis. Multimodel inference represents a promising procedure for modeling exposure levels that incorporates the notion that several models can be supported by the data and permits to evaluate to a certain extent model selection uncertainty, which is seldom mentioned in current practice.

Tactile-Proprioceptive Robotic Grasping

Robotic grasping has been studied increasingly for a few decades. While progress has been made in this field, robotic hands are still nowhere near the capability of human hands. However, in the past few years, the increase in computational power and the availability of commercial tactile sensors have made it easier to develop techniques that exploit the feedback from the hand itself, the sense of touch. The focus of this thesis lies in the use of this sense. The work described in this thesis focuses on robotic grasping from two different viewpoints: robotic systems and data-driven grasping. The robotic systems viewpoint describes a complete architecture for the act of grasping and, to a lesser extent, more general manipulation. Two central claims that the architecture was designed for are hardware independence and the use of sensors during grasping. These properties enables the use of multiple different robotic platforms within the architecture. Secondly, new data-driven methods are proposed that can be incorporated into the grasping process. The first of these methods is a novel way of learning grasp stability from the tactile and haptic feedback of the hand instead of analytically solving the stability from a set of known contacts between the hand and the object. By learning from the data directly, there is no need to know the properties of the hand, such as kinematics, enabling the method to be utilized with complex hands. The second novel method, probabilistic grasping, combines the fields of tactile exploration and grasp planning. By employing well-known statistical methods and pre-existing knowledge of an object, object properties, such as pose, can be inferred with related uncertainty. This uncertainty is utilized by a grasp planning process which plans for stable grasps under the inferred uncertainty.

Mobile systems for monitoring Parkinson's disease

A challenge for the clinical management of Parkinson's disease (PD) is the large within- and between-patient variability in symptom profiles as well as the emergence of motor complications which represent a significant source of disability in patients. This thesis deals with the development and evaluation of methods and systems for supporting the management of PD by using repeated measures, consisting of subjective assessments of symptoms and objective assessments of motor function through fine motor tests (spirography and tapping), collected by means of a telemetry touch screen device. One aim of the thesis was to develop methods for objective quantification and analysis of the severity of motor impairments being represented in spiral drawings and tapping results. This was accomplished by first quantifying the digitized movement data with time series analysis and then using them in data-driven modelling for automating the process of assessment of symptom severity. The objective measures were then analysed with respect to subjective assessments of motor conditions. Another aim was to develop a method for providing comparable information content as clinical rating scales by combining subjective and objective measures into composite scores, using time series analysis and data-driven methods. The scores represent six symptom dimensions and an overall test score for reflecting the global health condition of the patient. In addition, the thesis presents the development of a web-based system for providing a visual representation of symptoms over time allowing clinicians to remotely monitor the symptom profiles of their patients. The quality of the methods was assessed by reporting different metrics of validity, reliability and sensitivity to treatment interventions and natural PD progression over time. Results from two studies demonstrated that the methods developed for the fine motor tests had good metrics indicating that they are appropriate to quantitatively and objectively assess the severity of motor impairments of PD patients. The fine motor tests captured different symptoms; spiral drawing impairment and tapping accuracy related to dyskinesias (involuntary movements) whereas tapping speed related to bradykinesia (slowness of movements). A longitudinal data analysis indicated that the six symptom dimensions and the overall test score contained important elements of information of the clinical scales and can be used to measure effects of PD treatment interventions and disease progression. A usability evaluation of the web-based system showed that the information presented in the system was comparable to qualitative clinical observations and the system was recognized as a tool that will assist in the management of patients.

Measurement of the semileptonic charge asymmetry in B0 meson mixing with the D0 detector

We present a measurement of the semileptonic mixing asymmetry for B0 mesons, asld, using two independent decay channels: B0→μ +D -X, with D -→K +π -π -; and B0→μ +D *-X, with D * -→D ̄0π -, D ̄0→ K +π - (and charge conjugate processes). We use a data sample corresponding to 10.4fb -1 of pp̄ collisions at √s=1.96TeV, collected with the D0 experiment at the Fermilab Tevatron collider. We extract the charge asymmetries in these two channels as a function of the visible proper decay length of the B0 meson, correct for detector-related asymmetries using data-driven methods, and account for dilution from charge-symmetric processes using Monte Carlo simulation. The final measurement combines four signal visible proper decay length regions for each channel, yielding asld=[0.68±0.45(stat)±0.14(syst)]%. This is the single most precise measurement of this parameter, with uncertainties smaller than the current world average of B factory measurements. © 2012 American Physical Society.

Measurement of the differential cross section of tt pairs in pp collision at sqrt(s) = 7TeV with the ATLAS detector at the LHC

In this thesis three measurements of top-antitop differential cross section at an energy in the center of mass of 7 TeV will be shown, as a function of the transverse momentum, the mass and the rapidity of the top-antitop system. The analysis has been carried over a data sample of about 5/fb recorded with the ATLAS detector. The events have been selected with a cut based approach in the "one lepton plus jets" channel, where the lepton can be either an electron or a muon. The most relevant backgrounds (multi-jet QCD and W+jets) have been extracted using data driven methods; the others (Z+ jets, diboson and single top) have been simulated with Monte Carlo techniques. The final, background-subtracted, distributions have been corrected, using unfolding methods, for the detector and selection effects. At the end, the results have been compared with the theoretical predictions. The measurements are dominated by the systematic uncertainties and show no relevant deviation from the Standard Model predictions.

Search for high mass resonances decaying into electron-positron pairs in proton-proton collisions at sqrt(s) = 7 TeV with the ATLAS detector

The Standard Model of particle physics was developed to describe the fundamental particles, which form matter, and their interactions via the strong, electromagnetic and weak force. Although most measurements are described with high accuracy, some observations indicate that the Standard Model is incomplete. Numerous extensions were developed to solve these limitations. Several of these extensions predict heavy resonances, so-called Z' bosons, that can decay into an electron positron pair. The particle accelerator Large Hadron Collider (LHC) at CERN in Switzerland was built to collide protons at unprecedented center-of-mass energies, namely 7 TeV in 2011. With the data set recorded in 2011 by the ATLAS detector, a large multi-purpose detector located at the LHC, the electron positron pair mass spectrum was measured up to high masses in the TeV range. The properties of electrons and the probability that other particles are mis-identified as electrons were studied in detail. Using the obtained information, a sophisticated Standard Model expectation was derived with data-driven methods and Monte Carlo simulations. In the comparison of the measurement with the expectation, no significant deviations from the Standard Model expectations were observed. Therefore exclusion limits for several Standard Model extensions were calculated. For example, Sequential Standard Model (SSM) Z' bosons with masses below 2.10 TeV were excluded with 95% Confidence Level (C.L.).

Spatial and temporal characterisation of ground deformation recorded by geodetic techniques

A critical point in the analysis of ground displacements time series is the development of data driven methods that allow the different sources that generate the observed displacements to be discerned and characterised. A widely used multivariate statistical technique is the Principal Component Analysis (PCA), which allows reducing the dimensionality of the data space maintaining most of the variance of the dataset explained. Anyway, PCA does not perform well in finding the solution to the so-called Blind Source Separation (BSS) problem, i.e. in recovering and separating the original sources that generated the observed data. This is mainly due to the assumptions on which PCA relies: it looks for a new Euclidean space where the projected data are uncorrelated. The Independent Component Analysis (ICA) is a popular technique adopted to approach this problem. However, the independence condition is not easy to impose, and it is often necessary to introduce some approximations. To work around this problem, I use a variational bayesian ICA (vbICA) method, which models the probability density function (pdf) of each source signal using a mix of Gaussian distributions. This technique allows for more flexibility in the description of the pdf of the sources, giving a more reliable estimate of them. Here I present the application of the vbICA technique to GPS position time series. First, I use vbICA on synthetic data that simulate a seismic cycle (interseismic + coseismic + postseismic + seasonal + noise) and a volcanic source, and I study the ability of the algorithm to recover the original (known) sources of deformation. Secondly, I apply vbICA to different tectonically active scenarios, such as the 2009 L'Aquila (central Italy) earthquake, the 2012 Emilia (northern Italy) seismic sequence, and the 2006 Guerrero (Mexico) Slow Slip Event (SSE).

Dynamic Ontologies in Information Security Systems

Different types of ontologies and knowledge or metaknowledge connected to them are considered and analyzed aiming at realization in contemporary information security systems (ISS) and especially the case of intrusion detection systems (IDS) or intrusion prevention systems (IPS). Human-centered methods INCONSISTENCY, FUNNEL, CALEIDOSCOPE and CROSSWORD are algorithmic or data-driven methods based on ontologies. All of them interact on a competitive principle ‘survival of the fittest’. They are controlled by a Synthetic MetaMethod SMM. It is shown that the data analysis frequently needs an act of creation especially if it is applied to knowledge-poor environments. It is shown that human-centered methods are very suitable for resolutions in case, and often they are based on the usage of dynamic ontologies

Data-stream driven Fuzzy-granular approaches for system maintenance

Intelligent systems are currently inherent to the society, supporting a synergistic human-machine collaboration. Beyond economical and climate factors, energy consumption is strongly affected by the performance of computing systems. The quality of software functioning may invalidate any improvement attempt. In addition, data-driven machine learning algorithms are the basis for human-centered applications, being their interpretability one of the most important features of computational systems. Software maintenance is a critical discipline to support automatic and life-long system operation. As most software registers its inner events by means of logs, log analysis is an approach to keep system operation. Logs are characterized as Big data assembled in large-flow streams, being unstructured, heterogeneous, imprecise, and uncertain. This thesis addresses fuzzy and neuro-granular methods to provide maintenance solutions applied to anomaly detection (AD) and log parsing (LP), dealing with data uncertainty, identifying ideal time periods for detailed software analyses. LP provides deeper semantics interpretation of the anomalous occurrences. The solutions evolve over time and are general-purpose, being highly applicable, scalable, and maintainable. Granular classification models, namely, Fuzzy set-Based evolving Model (FBeM), evolving Granular Neural Network (eGNN), and evolving Gaussian Fuzzy Classifier (eGFC), are compared considering the AD problem. The evolving Log Parsing (eLP) method is proposed to approach the automatic parsing applied to system logs. All the methods perform recursive mechanisms to create, update, merge, and delete information granules according with the data behavior. For the first time in the evolving intelligent systems literature, the proposed method, eLP, is able to process streams of words and sentences. Essentially, regarding to AD accuracy, FBeM achieved (85.64+-3.69)%; eGNN reached (96.17+-0.78)%; eGFC obtained (92.48+-1.21)%; and eLP reached (96.05+-1.04)%. Besides being competitive, eLP particularly generates a log grammar, and presents a higher level of model interpretability.

GeoKernels: Modeling of Spatial Data on GeoManifolds

This paper presents a review of methodology for semi-supervised modeling with kernel methods, when the manifold assumption is guaranteed to be satisfied. It concerns environmental data modeling on natural manifolds, such as complex topographies of the mountainous regions, where environmental processes are highly influenced by the relief. These relations, possibly regionalized and nonlinear, can be modeled from data with machine learning using the digital elevation models in semi-supervised kernel methods. The range of the tools and methodological issues discussed in the study includes feature selection and semisupervised Support Vector algorithms. The real case study devoted to data-driven modeling of meteorological fields illustrates the discussed approach.

A comparison of methods for differential expression analysis of RNA-seq data.

BACKGROUND: Finding genes that are differentially expressed between conditions is an integral part of understanding the molecular basis of phenotypic variation. In the past decades, DNA microarrays have been used extensively to quantify the abundance of mRNA corresponding to different genes, and more recently high-throughput sequencing of cDNA (RNA-seq) has emerged as a powerful competitor. As the cost of sequencing decreases, it is conceivable that the use of RNA-seq for differential expression analysis will increase rapidly. To exploit the possibilities and address the challenges posed by this relatively new type of data, a number of software packages have been developed especially for differential expression analysis of RNA-seq data. RESULTS: We conducted an extensive comparison of eleven methods for differential expression analysis of RNA-seq data. All methods are freely available within the R framework and take as input a matrix of counts, i.e. the number of reads mapping to each genomic feature of interest in each of a number of samples. We evaluate the methods based on both simulated data and real RNA-seq data. CONCLUSIONS: Very small sample sizes, which are still common in RNA-seq experiments, impose problems for all evaluated methods and any results obtained under such conditions should be interpreted with caution. For larger sample sizes, the methods combining a variance-stabilizing transformation with the 'limma' method for differential expression analysis perform well under many different conditions, as does the nonparametric SAMseq method.