L'objectif de cette thèse est de concevoir une architecture de services Web pour la dé-termination automatique d’un système de capteurs personnalisé, embarqué sous forme de vêtement intelligent dédié au self-care, afin de permettre à tout utilisateur profane d'enregistrer lui-même son propre électrocardiogramme (ECG), à tout moment et n’importe où. Le défi principal réside dans l'orchestration intelligente et dynamique de services métiers en fonction du contexte, pour qu’ils fournissent à l'utilisateur une solution personnalisée optimale tout en maîtrisant la complexité inhérente à la dépendance au contexte des interactions homme-machine, à l'extraction des connaissances des signes vitaux spécifiques à un sujet, et à l'automatisation de la reconfiguration des services. Une solution à ce défi est de créer une intelligence ambiante qui étend la notion d'informatique ubiquitaire et est capable d’offrir à l’instar d’un expert du domaine, une assistance intelligente personnalisée à tout citoyen. Nous proposons une méthodologie de construction d'une architecture DM-SOA orientée-services, dirigée à la fois par les données et par des modèles, pour la production de services métiers intelligents et tenant compte du contexte. Cette architecture permet l’automatisation d’activités sophistiquées et personnalisées qu’un expert mettrait en œuvre pour le traitement d’un cas individuel, à partir de ses connaissances professionnelles. La solution proposée est basée sur une nouvelle approche de modélisation dynamique de processus métiers, et l’implémentation de services reconfigurables automatiquement. Elle consiste à mettre en œuvre un environnement intelligent fondé sur une ontologie de processus métiers des concepts du domaine et de son contexte, et sur une base de règles pour l'orchestration contextuelle des services. Pour valider le bien-fondé de notre approche, nous avons proposé une ontologie pour l’automatisation de processus d’aide à la décision et nous l’avons évaluée dans le domaine de la cardiologie, en l’appliquant au problème de la sélection la plus adéquate possible d’un système de positionnement d’électrodes, spécifique à chaque individu, et capable de fournir un signal ECG de contenu diagnostique similaire à celui d'un ECG standard à 12 dérivations. Pour répondre aux besoins en situation de self-care tout en préservant la qualité diagnostique des signaux enregistrés à la demande, nous proposons la réalisation d’un nouveau système prototype de capture ECG-3D à trois modalités. Ce prototype a été testé sur huit jeunes adultes volontaires sains (4 hommes et 4 femmes) présentant diverses caractéristiques morphologiques. L'intelligence ambiante est fournie par un ensemble de services de qualité professionnelle, prêts à l’emploi par des utilisateurs profanes. Ces services peuvent être accessibles au travail, n’importe où, via des moyens classiquement utilisés chaque jour, et fournissent une aide appropriée aux utilisateurs non-compétents. / The aim of the thesis is to design a web services architecture that shall support the automatic determination of a personalised sensor-system, which is embedded in smart garments, and which shall be used in self-care in order to allow a profane user to record himself a personal electrocardiogram (ECG), at anytime and anywhere. The main challenge consists in the intelligent and dynamic orchestration of context-aware business services that supply the user with an optimal personalised solution, while mastering the system’s complexity: context dependent user and system interactions, knowledge extraction from subject-specific vital signs, services reconfiguration automation. The solution to this challenge is to create an Ambient Intelligence which goes beyond Ubiquitous Computing and is capable to replace an expert by proposing an Intelligent Assistance to any citizen. We propose a methodology expressed in terms of Data and Model driven Service Oriented Architecture (DM-SOA), which provides a framework for the production of context-aware intelligent business services. This architecture supports the automation of sophisticated and personalised expert activities, which apply professional knowledge to process an individual case. The proposed solution is based on a new dynamic business process modelling approach and in its implementation via automatically reconfigurable services. It consists in setting-up an intelligent environment based on a business process ontology of context-aware concepts and on related context handling rules for services orchestration. A core-ontology has been designed to support the automation of expert activities related to decision-making. The soundness of the method and of the underlying decision-making techniques has been demonstrated in the cardiology domain for selecting the most adequate subject-specific sensor-system, characterised by its ability to yield for an ECG signal of similar diagnostic content than a standard 12-lead ECG. We propose the design of a new three modalities sensor-system prototype as a response to the need of sensor-systems used on demand in self-care situations and ensuring diagnostic quality signals recordings. This prototype has been tested on a set of healthy volunteers presenting various characteristics in age, sex and morphology. In this context, Ambient Intelligence is demonstrated as an ensemble of professional quality services ready to use by profane users. These services are accessible pervasively, trough the objects people work with / wear on / use each day, and provide an appropriate guidance to the non-competent users. Such an Ambient Intelligence approach strongly supports the vision of an Information Society which corresponds to the long-term goal of the EU Information Society Technologies Research Programme that aims to ease knowledge access. The future healthcare will benefit from this approach and will significantly improve patient care.

Orientador: Alexandre César Rodrigues da Silva / Resumo: O reconhecimento de atividades humanas abrange diversas técnicas de classificação que permitem identificar padrões específicos do comportamento humano no momento da ocorrência. A identificação é realizada analisando dados gerados por diversos sensores corporais, entre os quais destaca-se o acelerômetro, pois responde tanto à frequência como à intensidade dos movimentos. A identificação de atividades é uma área bastante explorada. Porém, existem desafios que necessitam ser superados, podendo-se mencionar a necessidade de sistemas leves, de fácil uso e aceitação por parte dos usuários e que cumpram com requerimentos de consumo de energia e de processamento de grandes quantidades de dados. Neste trabalho apresenta-se o desenvolvimento do Sistema de Reconhecimento de atividades Humanas e Avaliação do Equilíbrio Postural, denominado SIRAH. O sistema está baseado no uso de um acelerômetro localizado na cintura do usuário. As duas fases do reconhecimento de atividades são apresentadas, fase Offline e fase Online. A fase Offline trata do treinamento de uma rede neural artificial do tipo perceptron de três camadas. No treinamento foram avaliados três estudos de caso com conjuntos de atributos diferentes, visando medir o desempenho do classificador na diferenciação de 3 posturas e 4 atividades. No primeiro caso o treinamento foi realizado com 15 atributos, gerados no domínio do tempo, com os que a rede neural artificial alcançou uma precisão de 94,40%. No segundo caso foram gerados 34 ... (Resumo completo, clicar acesso eletrônico abaixo) / Doutor

Cette thèse traite du problème du développement d'espaces intelligents par l'utilisateur final sous l'angle de l'Interaction Homme-Machine et de l'Intelligence Ambiante. Dans les processus actuels de développement, l'utilisateur est un consommateur contraint par un système pensé et réalisé par d'autres. L'objectif de cette thèse est de redonner le pouvoir à l'utilisateur final par le biais d'outils adaptés au développement d'espaces intelligents. Cette thèse retient l'habitat intelligent comme lieu de vie privilégié. Ses contributions incluent : (1) DisQo (Dispositifs du QuOtidien), une nouvelle méthode d'investigation des besoins, réalisable au domicile de familles, qui sollicite l'imagination et assure un juste équilibre entre contrôle expérimental, respect de la sphère privée et validité écologique des résultats ; (2) Un espace de classification pour une lecture comparative systématique et synthétique des outils portant sur le développement et la programmation d'habitats intelligents. Cette taxonomie met en évidence le peu d'avancées en édition multisyntaxe de même pour l'aide à la mise au point de programmes ; (3) KISS (Knit Your Ideas into Smart Spaces), un outil de programmation et de mise au point dont le langage de programmation est de type déclaratif orienté règles, avec potentiel d'égale opportunité syntaxique entre langue française pseudonaturelle (LPN) et langage visuel iconique. La technique d'interaction de construction des programmes LPN s'appuie sur l'utilisation de menus dont les options sont calculées dynamiquement assurant ainsi la découverte progressive du langage ainsi que l'extensibilité et la correction syntaxique et sémantique des programmes. La mise au point peut se pratiquer, au choix, dans le monde physique ou dans un monde dual numérique. L'évaluation de KISS dans DOMUS, un habitat intelligent d'expérimentation, montre que les utilisateurs parviennent à programmer un scénario réaliste de la vie réelle. / This dissertation addresses the problem of end-user development for smart spaces from a human-computer interaction perspective in the context of ambient intelligence. End-users are currently doomed to be consumers of systems that have been designed and implemented by others. The goal of this thesis is to provide end-users with tools that will enable them to develop their own smart spaces. This work focuses on the home as a key place for smartness. The contribution of this doctoral research is threefold: (1) DisQo, a new method for field studies that combines several techniques to reach a satisfying balance between experimental control, privacy issues, and ecological validity. Its key element is for observers to be able to “visit” people homes through the pictures of intimate objects taken by the participants themselves and to use these pictures as playful cultural probes to envision future use; (2) A problem space that makes explicit the functional coverage (as well as the limits) of the tools from the state of the art in the area of end-user development for smart homes. In particular, the problem space reveals a lack of support for multi-syntax editing as well as for testing and debugging programs; (3) KISS (Knit your Ideas into Smart Spaces), an end-user development tool that uses a declarative rule-based programming paradigm where programs are expressed in a French pseudo-natural language with potentiality for syntactic equal opportunity with an iconic visual language. Programs are constructed by selecting items from pull down menus that are dynamically updated with the functionalities of the smart home. By so doing, the end-user can learn the programming language incrementally and specify programs that are semantically and syntactically correct. Programs can be tested either in a virtual home or in a real home. The evaluation of KISS in the DOMUS experimental platform, shows that users are able to program a real-life scenario.

Hospital environments are ranked as one of the most stressful contemporary work environments for their employees, and this especially concerns nurses (Nejati et al. 2016). One of the core problems comprises the notion that the current technology adopted in hospitals does not support the mobile nature of medical work and the complex work environment, in which people and information are distributed (Bardram 2003). The employment of inadequate technology and the strenuous access to information results in a decrease in efficiency regarding the fulfilment of medical tasks, and puts a strain on the attention of the medical personnel. This thesis proposes a solution to the aforementioned problems through the design of ambient displays, that inform the medical personnel with the health statuses of patients whilst requiring minimal allocation of attention. The ambient displays concede a hierarchy of information, where the most essential information encompasses an overview of patients’ vital signs. Data regarding the vital signs are measured by biometric sensors and are embodied by shape-changing interfaces, of which the ambient displays consist. User-authentication permits the medical personnel to access a deeper layer within the hierarchy of information, entailing clinical data such as patient EMRs, after gesture-based interaction with the ambient display. The additional clinical information is retrieved on the user’s PDA, and can subsequently be viewed in more detail, or modified at any place within the hospital.In this thesis, prototypes of shape-changing interfaces were designed and evaluated in a hospital environment. The evaluation was focused on the interaction design and user-experience of the shape-changing interface, the capabilities of the ambient displays to inform users through peripheral awareness, as well as the remote communication between patient and healthcare professional through biometric data. The evaluations indicated that the required attention allocated for the acquisition of information from the shape-changing interface was minimal. The interaction with the ambient display, as well as with the PDA when accessing additional clinical data, was deemed intuitive, yet comprised a short learning curve. Furthermore, the evaluations in situ pointed out that for optimised communication through the ambient displays, an overview of the health statuses of approximately eight patients should be displayed, and placed in the corridors of the hospital ward.