CETpD - Technical Research Centre for Dependency Care and Autonomous Living
Total activity: 876
Type
Research group
Type of group
Specific research center
UPC research group
Acronym
CETpD
Address
Rambla de l'Exposició, 61-69 Edifici NEAPOLIS Pl.2 Open in new window
City
Vilanova i la Geltrú
URL
http://www.epsevg.upc.edu/cetpd//index.php Open in new window
Objectives
CETpD's main research goal consists in the knowledge creation within the field of technology addressed to the dependency care area. In this way, the main interest is to improve the quality of life of people with dependency care assistance needs. Particularly, one of the main objectives relies on searching for new paradigms related to human movement analysis, usability and social robotics. CETpD directly cooperates with clinicians from several hospitals around Barcelona and some hospitals along Europe, developing research activities and projects in the dependency care area.

CETpD main research lines are:

Ambient Assisted Living and Pervasive Computing: create smart devices for both home and ambulatory monitoring of elderly people and patients with chronic diseases with tele-medicine and tele-care needs in order to assess gait, balance and falls. In this sense, CETpD has developed a wearable measurement unit for long-term monitoring of human movement. This measurement unit is being used to analyze human movement in the dependency care field.
Computational paradigms: develop soft-computing models, machine learning techniques and pervasive computing approaches addressed to the dependency care area.
Epidemiologic studies: CETpD performs, together with medical centers, clinical and epidemiologic studies on gait pathologies with the aim of knowing its causes and find new therapies that help to prevent its consequences in an effective way.
Social robotics: The priorities of this research line are three: (1) to develop new architectures of control and forms of signal processing which the robots allow to capture knowledge and being cognitive through the human-machine interaction; (2) to research in the more suitable forms of interaction, so they are motivating and satisfying for the user; and (3) to develop in an effective way robotic systems that are capable of supplying with personalised individual assistance to people with dependence (permanent, rehabilitation or convalescence).
Keywords
Inertial sensors, Intelligent algorithms, Motor symptoms, Parkinson's disease, REMPARK info, Smartphone, Wearable devices

Graphic summary
  • Show / hide key
  • Information


Scientific and technological production
  •  

1 to 50 of 876 results
  • Access to the full text
    Beyond cognition and affect: sensing the unconscious  Open access

     Ivonin, Leonid; Huang-Ming, Chang; Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias
    Behaviour & information technology
    Vol. 34, num. 3, p. 220-238
    DOI: 10.1080/0144929X.2014.912353
    Date of publication: 2015-03-04
    Journal article

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    In the past decade, research on human-computer interaction has embraced psychophysiological user interfaces that enhance awareness of computers about conscious cognitive and affective states of users and increase their adaptive capabilities. Still, human experience is not limited to the levels of cognition and affect but extends further into the realm of universal instincts and innate behaviours that form the collective unconscious. Patterns of instinctual traits shape archetypes that represent images of the unconscious. This study investigated whether seven various archetypal experiences of users lead to recognisable patterns of physiological responses. More specifically, the potential of predicting the archetypal experiences by a computer from physiological data collected with wearable sensors was evaluated. The subjects were stimulated to feel the archetypal experiences and conscious emotions by means of film clips. The physiological data included measurements of cardiovascular and electrodermal activities. Statistical analysis indicated a significant relationship between the archetypes portrayed in the videos and the physiological responses. Data mining methods enabled us to create between-subject prediction models that were capable of classifying four archetypes with an accuracy of up to 57.1%. Further analysis suggested that classification performance could be improved up to 70.3% in the case of seven archetypes by using within-subject models.

    In the past decade, research on human–computer interaction has embraced psychophysiological user interfaces that enhance awareness of computers about conscious cognitive and affective states of users and increase their adaptive capabilities. Still, human experience is not limited to the levels of cognition and affect but extends further into the realm of universal instincts and innate behaviours that form the collective unconscious. Patterns of instinctual traits shape archetypes that represent images of the unconscious. This study investigated whether seven various archetypal experiences of users lead to recognisable patterns of physiological responses. More specifically, the potential of predicting the archetypal experiences by a computer from physiological data collected with wearable sensors was evaluated. The subjects were stimulated to feel the archetypal experiences and conscious emotions by means of film clips. The physiological data included measurements of cardiovascular and electrodermal activities. Statistical analysis indicated a significant relationship between the archetypes portrayed in the videos and the physiological responses. Data mining methods enabled us to create between-subject prediction models that were capable of classifying four archetypes with an accuracy of up to 57.1%. Further analysis suggested that classification performance could be improved up to 70.3% in the case of seven archetypes by using within-subject models.

  • Energy Aware Networking for Home Enviroments  Open access

     Khan, Rafiullah
    Department of Automatic Control, Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    A study by Lawrence Berkeley National Laboratory (LBNL) revealed that about 60% of the office PCs are left powered-up 24/7 only to maintain the network connectivity for remote access, Voice-over-IP (VOIP) clients, Instant Messaging (IM) and otheradministrative management reasons. The Advanced Configuration and Power Interface (ACPI) proposed several low power states for PCs as effective mechanism to reduce energy waste, but unfortunately they are seldomly used due to their incapability to maintain network presence. Thus, Billions of dollars of electricity is wasted every year to keep idle or unused network devices fully powered-up only to maintain the network connectivity.This dissertation addresses the Network Connectivity Proxy (NCP), a concept recently been proposed as an optimal strategyto reduce energy waste due to idle network devices. The NCP is a software entity running on a low power network device (such as home gateway, switch or router) and impersonates presence for high power devices (such as PCs) during theirsleeping periods. It wakes-up a sleeping device only when its resources are required. In short, the NCP impersonates link layer, network layer, transport layer and application layer presence on behalf of sleeping devices.In this dissertation, we presented the design and implementation of our NCP prototype. The NCP concept faces several issues and challenges that we tried to address in the most effective way in our implementations. Knowing when to start orstop proxying presence on behalf of sleeping devices is critically important for the NCP operations. To achieve this objectivein a seamless way without requiring any user intervention, we developed a kernel module that monitors the power state transitions of the device and immediately informs the NCP over a suitable communication protocol in case of any update. Animportant challenge for the NCP is its ability to proxy a huge and ever increasing number of applications and networking protocols on behalf of sleeping devices. To tackle with this challenge in an efficient way, we implemented a quite generalizedset of behavioral rules in our NCP framework that can be suitable for any protocol or application. We also incorporated deployment flexibility in our NCP software that enables us to operate it on on-board NIC, switch/router or on a standalone PC.On-board NIC and switch/router are the optimal locations for the NCP software in home/small office environment (very limited number of devices) or a standalone PC with enough resources is a good choice if high scalability is desirable e.g., mediumor large size organizations. A communication protocol is required for information exchange between the NCP and client devices e.g., for power state notifications, registration/de-registration of client devices/behavioral rules etc. To avoid anyconfiguration issues, we developed a flexible and reliable communication framework based on the Universal Plug & Play (UPnP) architecture that provides interesting features such as auto-discovery, zero-configuration and seamlesscommunication between the NCP and client devices.We expanded the NCP coverage beyond LAN boundaries in order to exploit its full potential in terms of energy savings by covering for thousands of client devices. A single global powerful NCP instance located anywhere in the Internet can makeeasier the implementation of complex tasks and boosts up the energy savings by also shutting down the unused access links and the packets forwarding equipments whenever possible. Further, we also extended the NCP concept for mobile devices to help in improving the battery life. Another important contribution of this dissertation includes the extensive evaluation of the NCP performance on different low power hardwares. We performed large number of experiments and evaluated the effectiveness of NCP prototype in different

    Cotutela Universitat Politècnica de Catalunya i Università degli Studi di Genova

    A study by Lawrence Berkeley National Laboratory (LBNL) revealed that about 60% of the office PCs are left powered-up 24/7 only to maintain the network connectivity for remote access, Voice-over-IP (VOIP) clients, Instant Messaging (IM) and other administrative management reasons. The Advanced Configuration and Power Interface (ACPI) proposed several low power states for PCs as effective mechanism to reduce energy waste, but unfortunately they are seldomly used due to their incapability to maintain network presence. Thus, Billions of dollars of electricity is wasted every year to keep idle or unused network devices fully powered-up only to maintain the network connectivity.This dissertation addresses the Network Connectivity Proxy (NCP), a concept recently been proposed as an optimal strategy to reduce energy waste due to idle network devices. The NCP is a software entity running on a low power network device (such as home gateway, switch or router) and impersonates presence for high power devices (such as PCs) during their sleeping periods. It wakes-up a sleeping device only when its resources are required. In short, the NCP impersonates link layer, network layer, transport layer and application layer presence on behalf of sleeping devices. In this dissertation, we presented the design and implementation of our NCP prototype. The NCP concept faces several issues and challenges that we tried to address in the most effective way in our implementations. Knowing when to start or stop proxying presence on behalf of sleeping devices is critically important for the NCP operations. To achieve this objective in a seamless way without requiring any user intervention, we developed a kernel module that monitors the power state transitions of the device and immediately informs the NCP over a suitable communication protocol in case of any update. An important challenge for the NCP is its ability to proxy a huge and ever increasing number of applications and networking protocols on behalf of sleeping devices. To tackle with this challenge in an efficient way, we implemented a quite generalized set of behavioral rules in our NCP framework that can be suitable for any protocol or application. We also incorporated deployment flexibility in our NCP software that enables us to operate it on on-board NIC, switch/router or on a standalone PC. On-board NIC and switch/router are the optimal locations for the NCP software in home/small office environment (very limited number of devices) or a standalone PC with enough resources is a good choice if high scalability is desirable e.g., medium or large size organizations. A communication protocol is required for information exchange between the NCP and client devices e.g., for power state notifications, registration/de-registration of client devices/behavioral rules etc. To avoid any configuration issues, we developed a flexible and reliable communication framework based on the Universal Plug & Play (UPnP) architecture that provides interesting features such as auto-discovery, zero-configuration and seamless communication between the NCP and client devices. We expanded the NCP coverage beyond LAN boundaries in order to exploit its full potential in terms of energy savings by covering for thousands of client devices. A single global powerful NCP instance located anywhere in the Internet can make easier the implementation of complex tasks and boosts up the energy savings by also shutting down the unused access links and the packets forwarding equipments whenever possible. Further, we also extended the NCP concept for mobile devices to help in improving the battery life. Another important contribution of this dissertation includes the extensive evaluation of the NCP performance on different low power hardwares. We performed large number of experiments and evaluated the effectiveness of NCP prototype in different

  • Adapted step length estimators for patients with Parkinson's disease using a lateral belt worn accelerometer

     Sayeed, Taufique; Sama Monsonis, Albert; Català Mallofré, Andreu; Rodríguez Molinero, Alejandro; Cabestany Moncusi, Joan
    Technology and health care
    Vol. 23, num. 2, p. 179-194
    Date of publication: 2014-12-02
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    BACKGROUND: Parkinson's disease (PD) is a neurodegenerative disease that predominantly alters patients' motor performance. Reduced step length and inability of step are important symptoms associated with PD. Assessing patients' motor state monitoring step length helps to detect periods in which patients suffer lack of medication effect. OBJECTIVE: Evaluate the adaption of existing step length estimation methods based on accelerometer sensors to a new position on left lateral side of waist in 28 PD patients. METHODS: In this paper, a user-friendly position, the lateral side of the waist, is selected to place a tri-axial accelerometer. A newly developed step detection algorithm – Sliding Window Averaging Technique (SWAT) is evaluated in detecting steps using signals from this location. The detected steps are then used to estimate step length using four proposed correction factors for Zijlstra's, Gonzalez's and Weinberg's methods that were originally developed for the signals from lower back. RESULT: Results obtained from 28 PD patients are discussed and the effects of calibrating in each motor state are compared. A generic correction factor is also proposed and compared with the best method to use instead of individual calibration. Despite variable gait speed and different motor state, SWAT achieved overall accuracy of 96.76% in step detection. Among the different step length estimators, the Zijlstra method performs better with multiplying individual correction factors that consider left and right step length separately providing average error of 0.033 m. CONCLUSIONS: Zijlstra's method with individual correction factor that considers left and right step length separately and obtained from during ON state of a PD patients provide most accurate estimation among the others. As training session is during ON state, data from induced OFF state to train the methods are not required. A generic correction factor is also proposed to apply with Zijlstra's method to avoid individual calibration process.

  • Smartphone-based human activity recognition  Open access

     Reyes Ortiz, Jorge Luis
    Department of Automatic Control, Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    El Reconocimiento de Actividades Humanas (RAH) es un campo de investigación multidisciplinario que busca recopilar información sobre el comportamiento de las personas y su interacción con el entorno con el propósito de ofrecer información contextual de alta significancia sobre las acciones que ellas realizan. Recientemente, el RAH ha contribuido en el desarrollo de áreas de estudio enfocadas a la mejora de la calidad de vida del hombre tales como: la intel¿ligència ambiental (Ambient Intelligence) y la vida cotidiana asistida por el entorno para personas dependientes (Ambient Assisted Living).El primer paso para conseguir el RAH consiste en realizar observaciones mediante el uso de sensores fijos localizados en el ambiente, o bien portátiles incorporados de forma vestible en el cuerpo humano. Sin embargo, para el segundo caso, aún se dificulta encontrar dispositivos poco invasivos, de bajo consumo energético, que permitan ser llevados a cualquier lugar, y de bajo costo. En esta tesis, nosotros exploramos el uso de teléfonos móviles inteligentes (Smartphones) como una alternativa para el RAH. Estos dispositivos, de uso cotidiano y fácilmente asequibles en el mercado, están dotados de sensores embebidos, potentes capacidades de cómputo y diversas tecnologías de comunicación inalámbrica que los hacen apropiados para esta aplicación.Nuestro trabajo presenta una serie de contribuciones en relación al desarrollo de sistemas para el RAH con Smartphones. En primera instancia proponemos un sistema que permite la detección de seis actividades físicas en tiempo real y que, además, tiene en cuenta las transiciones posturales que puedan ocurrir entre ellas. Con este fin, hemos contribuido en distintos ámbitos que van desde el procesamiento de señales y la selección de características, hasta algoritmos de Aprendizaje Automático (AA). Nosotros utilizamos dos sensores inerciales (el acelerómetro y el giroscopio) para la captura de las señales de movimiento de los usuarios. Estas han de ser procesadas a través de técnicas de filtrado para la reducción de ruido, segmentación y obtención de características relevantes en la detección de actividad. También hacemos énfasis en el estudio de Máquinas de soporte vectorial (MSV) que son uno de los algoritmos de AA más usados en la actualidad. Para ello reformulamos varios de sus métodos estándar (lineales y no lineales) con el propósito de encontrar la mejor combinación de variables que garanticen un buen desempeño del sistema en cuanto a precisión, coste computacional y requerimientos de energía, los cuales son aspectos esenciales en dispositivos portátiles con suministro de energía mediante baterías. En concreto, proponemos dos MSV multiclase para la clasificación de actividad: un algoritmo lineal que permite el balance entre la reducción de la dimensionalidad y la precisión del sistema; y asimismo presentamos un algoritmo no lineal conveniente para dispositivos con limitaciones de hardware que solo utiliza aritmética de punto fijo en la fase de predicción y que permite reducir la complejidad del modelo de aprendizaje mientras mantiene el rendimiento del sistema.La eficacia del sistema propuesto es verificada a través de una experimentación extensiva sobre la base de datos RAH que hemos generado y hecho pública en la red. Esta contiene la información inercial obtenida de un grupo de 30 participantes que realizaron una serie de actividades de la vida cotidiana en un ambiente controlado mientras tenían sujeto a su cintura un smartphone que capturaba su movimiento.Los resultados obtenidos en esta investigación demuestran que es posible realizar el RAH en tiempo real con una precisión cercana al 97%. De esta manera, podemos emplear la metodología propuesta en aplicaciones de alto nivel que requieran el RAH tales como monitorizaciones ambulatorias para personas dependientes (ej. ancianos o discapacitados) durante periodos mayores a cinco días sin la necesidad de recarga de baterías.

    Cotutela Universitat Politècnica de Catalunya i Università degli Studi di Genova

    Human Activity Recognition (HAR) is a multidisciplinary research field that aims to gather data regarding people's behavior and their interaction with the environment in order to deliver valuable context-aware information. It has nowadays contributed to develop human-centered areas of study such as Ambient Intelligence and Ambient Assisted Living, which concentrate on the improvement of people's Quality of Life. The first stage to accomplish HAR requires to make observations from ambient or wearable sensor technologies. However, in the second case, the search for pervasive, unobtrusive, low-powered, and low-cost devices for achieving this challenging task still has not been fully addressed. In this thesis, we explore the use of smartphones as an alternative approach for performing the identification of physical activities. These self-contained devices, which are widely available in the market, are provided with embedded sensors, powerful computing capabilities and wireless communication technologies that make them highly suitable for this application. This work presents a series of contributions regarding the development of HAR systems with smartphones. In the first place we propose a fully operational system that recognizes in real-time six physical activities while also takes into account the effects of postural transitions that may occur between them. For achieving this, we cover some research topics from signal processing and feature selection of inertial data, to Machine Learning approaches for classification. We employ two sensors (the accelerometer and the gyroscope) for collecting inertial data. Their raw signals are the input of the system and are conditioned through filtering in order to reduce noise and allow the extraction of informative activity features. We also emphasize on the study of Support Vector Machines (SVMs), which are one of the state-of-the-art Machine Learning techniques for classification, and reformulate various of the standard multiclass linear and non-linear methods to find the best trade off between recognition performance, computational costs and energy requirements, which are essential aspects in battery-operated devices such as smartphones. In particular, we propose two multiclass SVMs for activity classification:one linear algorithm which allows to control over dimensionality reduction and system accuracy; and also a non-linear hardware-friendly algorithm that only uses fixed-point arithmetic in the prediction phase and enables a model complexity reduction while maintaining the system performance. The efficiency of the proposed system is verified through extensive experimentation over a HAR dataset which we have generated and made publicly available. It is composed of inertial data collected from a group of 30 participants which performed a set of common daily activities while carrying a smartphone as a wearable device. The results achieved in this research show that it is possible to perform HAR in real-time with a precision near 97\% with smartphones. In this way, we can employ the proposed methodology in several higher-level applications that require HAR such as ambulatory monitoring of the disabled and the elderly during periods above five days without the need of a battery recharge. Moreover, the proposed algorithms can be adapted to other commercial wearable devices recently introduced in the market (e.g. smartwatches, phablets, and glasses). This will open up new opportunities for developing practical and innovative HAR applications.

    El Reconocimiento de Actividades Humanas (RAH) es un campo de investigación multidisciplinario que busca recopilar información sobre el comportamiento de las personas y su interacción con el entorno con el propósito de ofrecer información contextual de alta significancia sobre las acciones que ellas realizan. Recientemente, el RAH ha contribuido en el desarrollo de áreas de estudio enfocadas a la mejora de la calidad de vida del hombre tales como: la inteligència ambiental (Ambient Intelligence) y la vida cotidiana asistida por el entorno para personas dependientes (Ambient Assisted Living). El primer paso para conseguir el RAH consiste en realizar observaciones mediante el uso de sensores fijos localizados en el ambiente, o bien portátiles incorporados de forma vestible en el cuerpo humano. Sin embargo, para el segundo caso, aún se dificulta encontrar dispositivos poco invasivos, de bajo consumo energético, que permitan ser llevados a cualquier lugar, y de bajo costo. En esta tesis, nosotros exploramos el uso de teléfonos móviles inteligentes (Smartphones) como una alternativa para el RAH. Estos dispositivos, de uso cotidiano y fácilmente asequibles en el mercado, están dotados de sensores embebidos, potentes capacidades de cómputo y diversas tecnologías de comunicación inalámbrica que los hacen apropiados para esta aplicación. Nuestro trabajo presenta una serie de contribuciones en relación al desarrollo de sistemas para el RAH con Smartphones. En primera instancia proponemos un sistema que permite la detección de seis actividades físicas en tiempo real y que, además, tiene en cuenta las transiciones posturales que puedan ocurrir entre ellas. Con este fin, hemos contribuido en distintos ámbitos que van desde el procesamiento de señales y la selección de características, hasta algoritmos de Aprendizaje Automático (AA). Nosotros utilizamos dos sensores inerciales (el acelerómetro y el giroscopio) para la captura de las señales de movimiento de los usuarios. Estas han de ser procesadas a través de técnicas de filtrado para la reducción de ruido, segmentación y obtención de características relevantes en la detección de actividad. También hacemos énfasis en el estudio de Máquinas de soporte vectorial (MSV) que son uno de los algoritmos de AA más usados en la actualidad. Para ello reformulamos varios de sus métodos estándar (lineales y no lineales) con el propósito de encontrar la mejor combinación de variables que garanticen un buen desempeño del sistema en cuanto a precisión, coste computacional y requerimientos de energía, los cuales son aspectos esenciales en dispositivos portátiles con suministro de energía mediante baterías. En concreto, proponemos dos MSV multiclase para la clasificación de actividad: un algoritmo lineal que permite el balance entre la reducción de la dimensionalidad y la precisión del sistema; y asimismo presentamos un algoritmo no lineal conveniente para dispositivos con limitaciones de hardware que solo utiliza aritmética de punto fijo en la fase de predicción y que permite reducir la complejidad del modelo de aprendizaje mientras mantiene el rendimiento del sistema. La eficacia del sistema propuesto es verificada a través de una experimentación extensiva sobre la base de datos RAH que hemos generado y hecho pública en la red. Esta contiene la información inercial obtenida de un grupo de 30 participantes que realizaron una serie de actividades de la vida cotidiana en un ambiente controlado mientras tenían sujeto a su cintura un smartphone que capturaba su movimiento. Los resultados obtenidos en esta investigación demuestran que es posible realizar el RAH en tiempo real con una precisión cercana al 97%. De esta manera, podemos emplear la metodología propuesta en aplicaciones de alto nivel que requieran el RAH tales como monitorizaciones ambulatorias para personas dependientes (ej. ancianos o discapacitados) durante periodos mayores a cinco días sin la necesidad de recarga de baterías.

  • A Computational creativity system to support chocolate designers decisions

     Ruiz Vegas, Francisco Javier; Raya Giner, Cristobal; Sama Monsonis, Albert; Agell Jané, Núria
    DOI: 10.3233/978-1-61499-452-7-289
    Date of publication: 2014-10-23
    Book chapter

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    In the present work, the Fuzzy Inductive Reasoning methodology (FIR) is used to improve coherence among beat patterns, structured in a musical A-B form. Patterns were generated based on a probability matrix, encoding a particular musical style, designed by experts. Then, all possible patterns were generated and the most probables were selected. A-B musical forms were created and the coherence of the sequence was evaluated by experts by using linguistic quantities. The output pairs (A-B pattern and its qualification) were used as inputs to train a FIR system, and the variables that produce “coherent” outputs and the relations among them where identified as rules. The extracted rules are discussed in the context of the musical form and from the psychological perception.

  • Reconeixement del Consell Social UPC als guardonats durant l'any 2013

     Cabestany Moncusi, Joan; Català Mallofré, Andreu
    Award or recognition

     Share

  • Access to the full text
    Develoment of an inverse model for honing processes by means of neural networks  Open access

     Sivatte Adroer, Maurici; Llanas Parra, Francesc Xavier; Buj Corral, Irene; Vivancos Calvet, Joan
    International Reseach/Expert Conference "Trends in the Development of Machinery and Associated Technology"
    p. 9-12
    Presentation's date: 2014-09-11
    Presentation of work at congresses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    In a previous paper, artificial neural networks were employed for modelling average roughness Ra in rough honing processes as a function of process variables, namely grain size, density, linear speed, tangential speed and pressure, by means of the direct problem. In addition, neural network model was compared to statistical models for modelling roughness. In the present paper the inverse problem was studied and analyzed by means of neural networks, in which given a certain average roughness Ra value, the model predicts process variables to be employed. This is not possible with statistical models. Two different approaches were considered: use of a single network or use of five networks.

  • Human activity recognition on smartphones with awareness of basic activities and postural transitions

     Reyes Ortiz, Jorge Luis; Oneto, Luca; Ghio, Alessandro; Sama Monsonis, Albert; Anguita, Davide; Parra Perez, Xavier
    International Conference on Artificial Neural Networks
    p. 177-184
    DOI: 10.1007/978-3-319-11179-7-23
    Presentation's date: 2014-09
    Presentation of work at congresses

    Read the abstract Read the abstract  Share Reference managers Reference managers Open in new window

    Postural Transitions (PTs) are transitory movements that describe the change of state from one static posture to another. In several Human Activity Recognition (HAR) systems, these transitions cannot be disregarded due to their noticeable incidence with respect to the duration of other Basic Activities (BAs). In this work, we propose an online smartphone-based HAR system which deals with the occurrence of postural transitions. If treated properly, the system accuracy improves by avoiding fluctuations in the classifier. The method consists of concurrently exploiting Support Vector Machines (SVMs) and temporal filters of activity probability estimations within a limited time window. We present the benefits of this approach through experiments over a HAR dataset which has been updated with PTs and made publicly available. We also show the new approach performs better than a previous baseline system, where PTs were not taken into account.

  • Access to the full text
    Enacting archetypes in movies: grounding the unconscious mind in emotion-driven media  Open access

     Huang-Ming, Chang; Ivonin, Leonid; Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias
    Digital creativity (Exeter)
    DOI: 10.1080/14626268.2014.939985
    Date of publication: 2014-08-13
    Journal article

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    The concept of emotion-driven media integrates affective computing into developing new types of digital media by including the human spectator as an essential part of the whole system. Enactive media, as a landmark in this direction, PROVIDES a theoretical basis that is rooted in the enactivist approach in cognitive science. Based on this thesis, it is required to establish a practical procedure for developing the media content through integrating psychological approaches. We demonstrate a systematic approach for developing the content for emotion-driven media. The theory of archetypes, one of the psychoanalytical approaches, was applied for editing movie clips as the media content, and coupled the content with emotional responses by using experimental psychological methodologies. The results showed that it is promising to utilise archetypal movie clips as the content of emotion-driven media and archetypal symbolism can be a useful resource for developing the content of media systems.

  • Observation of chaotic behavior in automatic tuning loops for continuous-time filters

     Martínez García, Herminio; Cosp Vilella, Jordi; Manzanares Brotons, Manuel
    IEEE International Midwest Symposium on Circuits and Systems
    p. 742-745
    DOI: 10.1109/MWSCAS.2013.6674756
    Presentation's date: 2014-08
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    he appropriate linear dynamic modeling of continuous-time filters (CTFs) with automatic tuning loops should be obtained to assure stability in case an improved design of the loop controllers is to be carried out. With this aim, starting from a general and systematic analysis in order to obtain an equivalent small-signal linearized incremental model, from which transfer functions between output variables and control voltages are derived, the subsequent design of compensated loops with enhanced stability and dynamic performance is required. This systematic procedure allows obtaining improved controllers for the two involved control loops. However, CTFs with automatic tuning loops are nonlinear feedback systems with potential instability. What is more, nonlinear phenomena, which cannot be predicted by a design-oriented small signal modeling approach, are observed in this kind of tuning systems. The purpose of this work is to highlight that when control parameters are varied, the system could present different kinds of dynamical nonlinear phenomena such as bifurcations and chaotic behavior, which cannot be predicted by the small signal design-oriented model.

  • Unspoken emotions in movies: The basis of emotion-driven storytelling systems

     Chang, Huang-ming; Ivonin, Leonid; Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias
    Informatik-Spektrum
    DOI: 10.1007/s00287-014-0823-3
    Date of publication: 2014-07-28
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Stories are the core of many types of entertainment.Although developers and designers strive to develop new forms of entertainment that deliver richer audiovisual experiences to the audience, one of the key elements – a good story – remains the same. Storytelling is intertwined in people’s lives and has deeply influenced human societies for generations. It ranges from the simplest forms, such as bedtime stories for children, to more complex forms such as plays, movies, and games. Just as children love often-told bedtime stories, sometimes adults enjoy watching classic movies over and over again and never tire of them. Storytelling is not only a matter of how well the stories are memorized but also the overall emotional experience over the course of the storyline. The link between the story and the corresponding emotions is an intriguing topic for developing new interactive storytelling systems.

  • Proposal and development of a highly modular and scalable self-adaptive hardware architecture with parallel processing capability  Open access

     Soto Vargas, Javier Evandro
    Department of Electronic Engineering, Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    Esta tesis doctoral describe una arquitectura de hardware auto-adaptable novedosa y no convencional con capacidad de procesamiento en paralelo. Por razones de escalabilidad, esta arquitectura bioinspirada está basada en una matriz regular de células homogéneas. La arquitectura propuesta es programable, e implementa de manera distribuida diversas capacidades auto-adaptables incluyendo el auto-emplazamiento y auto-enrutamiento, los cuales debido a su diseño intrínseco, permiten el desarrollo de sistemas reconfigurables en tiempo de ejecución, así como de sistemas auto-reparables y/o con capacidades de tolerancia a fallos. La implementación física de esta arquitectura esta compuesta de dos capas, que incluyen células interconectadas en el primer nivel y matrices de conmutación y pines en el segundo nivel. La célula es el elemento básico de la arquitectura propuesta. Cualquier aplicación que se quiera programar en el sistema debe estar organizada en componentes, donde cada componente está compuesto por una o más células interconectadas. La interconexión de células dentro de un componente es realizado en el mismo nivel de la matriz de células, mientras que la interconexión de componentes es realizada en la segunda capa. Adicionalmente, se definen dos capas conceptuales que son usadas con propósitos organizativos en aplicaciones de propósito general, estas son: el SANE y el SANE-assembly (o conjunto de SANEs). La entidad auto-adaptable interconectada o SANE está compuesta por un grupo de componentes. Este es el sistema de computación auto-adaptable básico, el cual tiene la habilidad de monitorizar su entorno local y su proceso de computación interno.Las capacidades de procesamiento de la célula están incluidas en su unidad funcional (FU). Esta puede ser definida como un multicomputador configurable con cuatro núcleos, los cuales son agrupados o no dependiendo del modo de configuración. La FU tiene doce modos de configuración programables, por lo que cada célula permite seleccionar entre uno y cuatro procesadores trabajando en paralelo con diversas capacidades en las memorias de programa y datos. Las capacidades auto-adaptables de la célula son ejecutadas principalmente por la unidad de configuración de la célula (CCU). El algoritmo de auto-emplazamiento es el encargado de encontrar la posición mas adecuada dentro de la matriz de células para insertar la nueva célula de un componente. El algoritmo de auto-enrutamiento permite interconectar los puertos de las FU de dos células. Los procesos de auto-emplazamiento y auto-enrutamiento permiten realizar en tiempo real cambios funcionales complejos; estos procesos dotan al sistema de una mayor funcionalidad, permitiendo que el sistema cambie por si mismo, lo que permite la implementación de la auto-configuración en tiempo real, sin la necesidad de ningún gestor de configuración.La arquitectura propuesta incluye dos mecanismos de tolerancia a fallos. Uno de estos es una técnica escalonada y dinámica de tolerancia a fallos, que tiene la habilidad de crear y eliminar copias redundantes de la unidad funcional (o de cómputo) de una aplicación específica. El otro mecanismo de tolerancia a fallos es el Sistema de Tolerancia a Fallos dedicado o estático. Este provee capacidades de procesamiento redundante que están en funcionamiento continuamente. Cuando un fallo en la ejecución de un programa es detectado, los procesadores de la célula son detenidos y los procesos de auto-eliminación y auto-replicación se inician para la célula (o células) implicada en el fallo. Se desarrolló un prototipo basado en FPGAs y una herramienta de software para comprobar la funcionalidad del sistema. El prototipo incluye todas las características de los sistemas auto-adaptable descritas en este trabajo. El SANE Project developer (SPD) es un ambiente integrado de desarrollo (IDE) que permite generar y descargar la memoria de inicialización de datos para el Microprocesador de Control dentro del prototipo.

    This dissertation describes a novel unconventional self-adaptive hardware architecture with capacity for parallel processing. For scalability issues, this bioinspired architecture is based on a regular array of homogeneous cells. The proposed programmable architecture implements in a distributed way self-adaptive capabilities including self-placement and self-routing which, due to its intrinsic design, enable the development of systems with runtime reconfiguration, self-repair and/or fault tolerance capabilities. The physical implementation of this architecture is composed of two-layers, interconnected cells in the first level and interconnected switch and pin matrices in the second level. The cell is the basic element of the proposed self-adaptive architecture. Any application scheduled to the system has to be organized in components, where each component is composed by one or more interconnected cells. The interconnection of cells inside a component is made at cell level (first layer), while the physical interconnections of components are made in the second layer. Additionally, two layers are defined as conceptual organization for the implementation of general purpose applications: the SANE and the SANE assembly. The Self-Adaptive Networked Entity (SANE) is composed by a group of components. This is the basic self-adaptive computing system. It has the ability to monitor its local environment and its internal computation process. The SANE-Assembly (SANE-ASM) is composed by a group of interconnected SANEs. The processing capabilities of the cell are included in its Functional Unit (FU), which can be described as a four-core configurable multicomputer. The FU includes twelve programmable configuration modes, i.e., each cell permits to select from one to four processors working in parallel, with different size of program and data memories. The self-adaptive capabilities of the cell are executed mainly by the Cell Configuration Unit (CCU). The self-placement algorithm is responsible for finding out the most suitable position in the cell array to insert the new cell of a component. The self-routing algorithm permits interconnecting the ports of the FU of two cells through the cell ports. The self-placement and self-routing processes allow for performing complex functionality changes in real time, these processes endow the system with enhanced functionality, enabling the system to change itself, this allows for the implementation of run-time self-configuration, without the need for any configuration manager. The architecture proposed includes two mechanisms of fault tolerance. One of these is the Dynamic Fault Tolerance Scaling Technique, that has the ability to create and eliminate the redundant copies of the functional section of a specific application. The other mechanism of fault tolerance is a dedicated or static Fault Tolerance System. It provides redundant processing capabilities that are working continuously. When a failure in the execution of a program is detected, the processors of the cell are stopped and the self-elimination and self-replication processes start for the cell (or cells) involved in the failure. An FPGA-based prototype and a software tool have been built for demonstration purposes. The prototype includes all the self-adaptive capabilities described in this dissertation. With the purpose of having a complete development system, the software tool SANE Project Developer (SPD) has been implemented. The SPD is an Integrated Development Environment (IDE) that allows generating the memory initialization data for the control microprocessor inside the prototype.

    Esta tesis doctoral describe una arquitectura de hardware auto-adaptable novedosa y no convencional con capacidad de procesamiento en paralelo. Por razones de escalabilidad, esta arquitectura bioinspirada está basada en una matriz regular de células homogéneas. La arquitectura propuesta es programable, e implementa de manera distribuida diversas capacidades auto-adaptables incluyendo el auto-emplazamiento y auto-enrutamiento, los cuales debido a su diseño intrínseco, permiten el desarrollo de sistemas reconfigurables en tiempo de ejecución, así como de sistemas autoreparables y/o con capacidades de tolerancia a fallos. La implementación física de esta arquitectura esta compuesta de dos capas, que incluyen células interconectadas en el primer nivel y matrices de conmutación y pines en el segundo nivel. La célula es el elemento básico de la arquitectura propuesta. Cualquier aplicación que se quiera programar en el sistema debe estar organizada en componentes, donde cada componente está compuesto por una o más células interconectadas. La interconexión de células dentro de un componente es realizado en el mismo nivel de la matriz de células, mientras que la interconexión de componentes es realizada en la segunda capa. Adicionalmente, se definen dos capas conceptuales que son usadas con propósitos organizativos en aplicaciones de propósito general, estas son: el SANE y el SANE-assembly (o conjunto de SANEs). La entidad auto-adaptable interconectada o SANE está compuesta por un grupo de componentes. Este es el sistema de computación auto-adaptable básico, el cual tiene la habilidad de monitorizar su entorno local y su proceso de computación interno. Las capacidades de procesamiento de la célula están incluidas en su unidad funcional (FU). Esta puede ser definida como un multicomputador configurable con cuatro núcleos, los cuales son agrupados o no dependiendo del modo de configuración. La FU tiene doce modos de configuración programables, por lo que cada célula permite seleccionar entre uno y cuatro procesadores trabajando en paralelo con diversas capacidades en las memorias de programa y datos. Las capacidades auto-adaptables de la célula son ejecutadas principalmente por la unidad de configuración de la célula (CCU). El algoritmo de auto-emplazamiento es el encargado de encontrar la posición mas adecuada dentro de la matriz de células para insertar la nueva célula de un componente. El algoritmo de auto-enrutamiento permite interconectar los puertos de las FU de dos células. Los procesos de auto-emplazamiento y auto-enrutamiento permiten realizar en tiempo real cambios funcionales complejos; estos procesos dotan al sistema de una mayor funcionalidad, permitiendo que el sistema cambie por si mismo, lo que permite la implementación de la auto-configuración en tiempo real, sin la necesidad de ningún gestor de configuración. La arquitectura propuesta incluye dos mecanismos de tolerancia a fallos. Uno de estos es una técnica escalonada y dinámica de tolerancia a fallos, que tiene la habilidad de crear y eliminar copias redundantes de la unidad funcional (o de cómputo) de una aplicación específica. El otro mecanismo de tolerancia a fallos es el Sistema de Tolerancia a Fallos dedicado o estático. Este provee capacidades de procesamiento redundante que están en funcionamiento continuamente. Cuando un fallo en la ejecución de un programa es detectado, los procesadores de la célula son detenidos y los procesos de auto-eliminación y auto-replicación se inician para la célula (o células) implicada en el fallo. Se desarrolló un prototipo basado en FPGAs y una herramienta de software para comprobar la funcionalidad del sistema. El prototipo incluye todas las características de los sistemas auto-adaptable descritas en este trabajo. El SANE Project developer (SPD) es un ambiente integrado de desarrollo (IDE) que permite generar y descargar la memoria de inicialización de datos para el Microprocesador de Control dentro del prototipo.

  • Quasi-digital front-ends for current measurement in integrated circuits with giant magnetoresistance technology

     De Marcellis, A.; Cubells Beltrán, M. Dolores; Reig Escriva, Abilio Càndid; Madrenas Boadas, Jordi; zadov, Boris; Paperno, Eugene; Cardoso, S.; Freitas, P.P.
    IET circuits, devices and systems
    Vol. 8, num. 4, p. 291-300
    DOI: 10.1049/iet-cds.2013.0348
    Date of publication: 2014-07-01
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    In this study, the authors report on two different electronic interfaces for low-power integrated circuits electric current monitoring through current-to-frequency (I-f) conversion schemes. This proposal displays the intrinsic advantages of the quasi-digital systems regarding direct interfacing and self-calibrating capabilities. In addition, as current-sensing devices, they have made use of the giant magnetoresistance (GMR) technology because of its high sensitivity and compatibility with standard complementary metal oxide semiconductor processes. Single elements and Wheatstone bridges based on spin-valves and magnetic tunnel junctions have been considered. In this sense, schematic-level simulations for integration in Austria Microsystems 0.35 mu m technology have been corroborated by means of experimental measurements with the help of printed circuit board prototypes and real GMR devices. Tables with relevant parameters (silicon area, power consumption, sensitivity etc.) have been constructed as practical tools for designers. Electric currents down to 2 mu A have been resolved in this way.

  • Access to the full text
    A double closed loop to enhance the quality of life of Parkinson's disease patients: REMPARK system  Open access

     Sama Monsonis, Albert; Perez Lopez, Carlos; Rodriguez Martin, Daniel Manuel; Moreno Arostegui, Juan Manuel; Rovira, Jordi; Ahlrichs, Claas; Castro, Rui; Cevada, Joao; Cabestany Moncusi, Joan
    KES International Conference on Innovation in Medicine and Healthcare
    p. 115-124
    DOI: 10.3233/978-1-61499-474-9-115
    Presentation's date: 2014-07
    Presentation of work at congresses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    This paper presents REMPARK system, a novel approach to deal with Parkinson's Disease (PD). REMPARK system comprises two closed loops of actuation onto PD. The first loop consists in a wearable system that, based on a belt-worn movement sensor, detects movement alterations that activate an auditory cueing system controlled by a smartphone in order to improve patient's gait. The belt-worn sensor analyzes patient's movement through real-time learning algorithms that were developed on the basis of a database previously collected from 93 PD patients. The second loop consists in disease management based on the data collected during long periods and that enables neurologists to tailor medication of their PD patients and follow the disease evolution. REMPARK system is going to be tested in 40 PD patients in Spain, Ireland, Italy and Israel. This paper describes the approach followed to obtain this system, its components, functionalities and trials in which the system will be validated.

  • Efficient multiprocessing architectures for Spiking Neural Network emulation based on configurable devices

     Sanchez Rivera, Giovanny
    Department of Electronic Engineering, Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract  Share Reference managers Reference managers Open in new window

    L'estudi de la dinàmica de les xarxes neuronals bio-inspirades ha permès als neurocientífics entendre alguns processos i estructures del cervell . Les implementacions electròniques d'aquestes xarxes neuronals són eines útils per dur a terme aquest tipus d'estudi . No obstant això, l'alta complexitat de les xarxes neuronals requereix d'una arquitectura apropiada que pugui simular aquest tipus de xarxes. Emular aquest tipus de xarxes en dispositius configurables és possible a causa del seu extraordinari desenvolupament respecte a la seva disponibilitat de recursos, velocitat i capacitat de reconfiguració (FPGAs ) .En aquesta tesi es proposa una arquitectura maquinari paral · lela i configurable per emular les complexes i realistes xarxes neuronals tipus spiking en temps real . S'han estudiat i analitzat alguns models de neurones tipus spiking rellevants i les seves implementacions en maquinari , amb la finalitat de crear una arquitectura que suporti la implementació d'aquests models de manera eficient .S'han tingut en compte diversos factors clau, incloent flexibilitat en la programació d'algorismes, processament d'alt rendiment, baix consum d'energia i àrea. S'han aplicat diverses tècniques en l'arquitectura desenvolupada amb el propòsit d'augmentar la seva capacitat de processament. Aquestes tècniques són: mapejat de temps a espai, virtualització de les neurones, mapeig flexible de neurones i sinapsis, modes d'execució, i aprenentatge específic, entre d'altres. A més, s'ha desenvolupat una unitat d'interfície de dades per tal de construir un sistema bio-inspirat, que pot processar informació sensorial del medi ambient. Aquest sistema basat en neurones tipus spiking combina implementacions analògiques i digitals. S'han desenvolupat diverses aplicacions usant aquest sistema com a prova de concepte, per tal de mostrar les capacitats de l'arquitectura proposada per al processament d'aquest tipus d'informació.

  • Modified current-mode one-cycle control for linear–assisted DC/DC regulator

     Martínez García, Herminio; Cosp Vilella, Jordi
    IEEE International Symposium on Circuits and Systems
    p. 2688-2691
    DOI: 10.1109/ISCAS.2014.6865727
    Presentation's date: 2014-06-04
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Linear-assisted DC/DC converters are structures that allow to take advantages of the two classic alternatives in the design of power supply systems: Voltage linear regulators (classic NPN topology or LDO –low dropout–) and switching DC/DC converters. This paper shows the proposal of a modified current-mode one-cycle control for linear-assisted DC/DC converters. The modified current-mode one-cycle control technique is proposed in order to obtain the duty cycle of the linear-assisted converter switch. The proposed structure can provide an output with suitable load and line regulations. The paper shows the design of the proposed modified current-mode one-cycle linear-assisted regulator and simulation results that validate the aforementioned proposal.

  • Knee osteoarthritis and periarticular structure quantified by ultrasound. A case-control study

     Nuñez, M.; Sastre, S.; Nuñez, E.; Saulo, A.; Segur, JM.; Maurits, NM.; Moreno Arostegui, Juan Manuel; Cabestany Moncusi, Joan; Segarra, J.; Segura, V.; Lozano, L.; Alemany, X.; Maculé, F.; Suso, S.
    Annals of the rheumatic diseases
    Vol. 73, num. 2, p. 751
    DOI: 10.1136/annrheumdis-2014-eular.3136
    Date of publication: 2014-06
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Background Assessment of pain and physical function is complex in patients with knee osteoarthritis (OA), as standard criteria are lacking.A previous study examining correlations between functional capacity and pain (WOMAC) and anthropometric characteristics and periarticular knee structure (quantified by ultrasound imaging) in females with knee OA found increased quadriceps muscle density was associated with higher functional disability and pain scores, suggesting that not only joint wear and symptom severity are involved and more objective measures are necessary. Objectives To determine and compare the periarticular knee structure in obese patients with knee OA and a healthy control group. Methods Analytical case-control study. Study group. Patients diagnosed with knee OA. Control group. Adults with no history of knee involvement, able to walk normally, with no pain or functional difficulties on examination and no history of surgery in other lower limb joints. Controls were matched for age, sex and body mass index (BMI). Sociodemographic, clinical, functional (Timed Up and Go test [TUG]) and anthropometric (weight, height, BMI, waist circumference, and lower limb [suprapatellar and infrapatellar indices]) data were collected. Periarticular knee structure was assessed by ultrasound (thickness of subcutaneous fat [distance from skin to fascia, in mm] and quadriceps/rectus femoris [distance between fascia and femur, in mm]) and appearance [density on digital image analysis according to Maurits et al]). Statistical Analysis. Groups were compared using the t test for continuous variables and ¿2 test for categorical variables. Results 66 lower limbs from 14 patients (mean age 62.7 [SD 8.6]) years, BMI 30.4 (SD 5.9) and 19 matched controls (mean age 62.6 [SD 8.1] years, BMI 30.1 [SD 4.7]) were evaluated. Comparison between groups: no significant differences in anthropometric measures were found. TUG took a mean 13.7s (6.7) and 9.9s (2.4) in patients and controls, respectively, p=0.002. Mean subcutaneous fat was 18.7 (SD 9.8) mm and 15.2 (4.41) in patients and controls, respectively, p=0.028. Mean quadriceps muscle density was 61.1 (25.9) and 41.7 (13.7), respectively, p=0.001. Conclusions Between-group differences were found in the periarticular knee structure. Patients with knee OA had increased subcutaneous fat thickness and quadriceps muscle density was observed compared with controls. These findings suggest that the assessment of periarticular structures in these patients analyzed by digital image derived from ultrasound could add a variable to determine more objectively uniform methods in the classification of patients and evaluation of results. References Maurits NM et al. Muscle ultrasound analysis: normal values and differentiation between myopathies and neuropathies. Ultrasound Med Biol 2003;29:215-25.

  • Contribución al análisis del movimiento humano aplicado a la identificación de posturas y bloqueos de la marcha en pacientes con Parkinson  Open access

     Rodriguez Martin, Daniel Manuel
    Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    La siguiente disertación presenta las aportaciones del autor en el campo de la monitorización del movimiento humano y, especialmente, en la enfermedad del Parkinson. Las recientes tecnologías han permitido desarrollar sensores inerciales de tamaño muy reducido para monitorizar el movimiento humano. En consecuencia, el uso de las llamadas unidades de medida inercial, que consisten en unos dispositivos de pequeño tamaño capaces de medir el movimiento, se ha extendido ampliamente empleándose, entre otros, en campos como la medicina, el deporte, la automoción o los videojuegos. En la primera parte de la presente tesis, se describe el desarrollo de una unidad de medida inercial capaz de adquirir datos con una larga autonomía. La principal novedad aportada consiste en la posibilidad de implementar clasificadores basados en inteligencia artificial en tiempo real a través de un hardware y un firmware diseñados específicamente para que el sistema inercial sea flexible y capaz de implementar las dos operaciones. Esta herramienta se ha validado en diferentes proyectos europeos y diferentes trabajos llevados a cabo en el Centro de Estudios Tecnológicos para la Atención de la Dependencia y la Vida Autónoma de la Universidad Politécnica de Catalunya (CETpD-UPC)En la segunda parte de la tesis se describen dos algoritmos para identificar diferentes tipos de posturas. Los datos empleados para poder desarrollar los algoritmos se han adquirido con el sistema inercial presentado en la primera parte de la tesis. Los algoritmos desarrollados se han validado en personas sanas y en pacientes con la enfermedad del Parkinson. El algoritmo se focaliza en la detección de posturas con un único sistema inercial localizado en la cintura, consiguiendo una mayor comodidad para los usuarios. Una aportación clave es la metodología ofrecida para poder detectar transiciones posturales (movimiento para pasar de una postura a otra) a través de `Máquinas de vectores soporte¿, permitiendo tanto distinguirlas como detectarlas. La detección de transiciones posturales se integra en un sistema jerárquico de clasificación, al cual se añade un detector de actividades básicas donde se identifican posturas estáticas como estar sentado, de pie, o tumbado, y posturas dinámicas, como caminar. En la parte final de la tesis se presenta una aportación a la identificación de uno de los síntomas de la enfermedad del Parkinson más incapacitantes, el bloqueo de la marcha. El algoritmo de posturas se aplica en señales captadas por la unidad inercial descrita en la primera parta de la tesis para complementar a los algoritmos actuales de detección de bloqueos de la marcha. Además, se demuestra cómo los valores de evaluación del algoritmo de detección del bloqueo de la marcha mejoran al aplicar el algoritmo de posturas. De esta forma, se contextualiza el algoritmo, pudiendo discriminar aquellos episodios detectados provocados en situaciones donde no es posible que ocurra un episodio de bloqueo de la marcha.

    The following dissertation presents the contributions of the author in the field of human movement analysis and, specifically, in Parkinson's disease. Recent technologies have allowed developing reduced inertial sensors capable of monitoring human movement. This, along with the reduced prices of these inertial sensors, the so-called inertial measurement units, which consists in small devices capable to measure movement by means of inertial sensors, have widely spread. Inertial measurement units have been employed among others, in fields such as medicine, sports, automotive and gaming. In the first part of the present thesis, a wearable long-term monitoring inertial measurement unit is presented as the first main contribution in human movement analysis. The unit is capable of acquiring data and provides the possibility of implementing artificial intelligence-based classifiers in real time. A specific hardware and firmware has been developed in order to implement both operations. This tool has been validated in different European projects and studies carried out in the Technical Research Centre for Dependency Care and Autonomous Living of the Universitat Politècnica de Catalunya (CETpD-UPC). The second part of the thesis addresses the analysis of human posture based on accelerometry measurements. To this end, data acquired from the inertial system described at the first part of the thesis have been used. Two methodologies are presented that have been validated on healthy people and patients with Parkinson's disease. The algorithms developed are focused on the detection of positions with a single inertial system located at the waist thereby achieving an enhanced comfort and acceptance by the users. A key contribution is the methodology provided to detect postural transitions, which consist in the movement performed to achieve a position from another one. The algorithm is based on support vector machines applied to the inertial data coming from a single measurement unit. Basic activity recognition is performed recognizing static postures such as sitting, standing, or lying with a hierarchical classification system. Moreover, dynamic postures such as walking and different postural transitions are also recognized. Finally, the posture detection methodologies are employed to enhance the identification of one of the most annoying symptoms of Parkinson's disease, the so-called Freezing of Gait. This contribution relies on the posture algorithm which has been validated in Parkinson's disease patients. Furthermore, it is shown how the introduction of the posture detection improves the evaluation values of the FOG algorithms

    La siguiente disertación presenta las aportaciones del autor en el campo de la monitorización del movimiento humano y, especialmente, en la enfermedad del Parkinson. Las recientes tecnologías han permitido desarrollar sensores inerciales de tamaño muy reducido para monitorizar el movimiento humano. Estos sensores se emplean en las llamadas unidades de medida inercial, que son unos dispositivos de pequeño tamaño capaces de medir el movimiento, se ha extendido ampliamente empleándose, entre otros, en campos como la medicina, el deporte, la automoción o los videojuegos. En la primera parte de la presente tesis, se describe el desarrollo de una unidad de medida inercial optimizada para realizar la adquisición y el tratamiento de los datos con un mínimo consumo y, de esa forma, conseguir una larga autonomía. La principal novedad aportada consiste en la posibilidad de implementar en tiempo real clasificadores basados en inteligencia artificial a través de un hardware y un firmware diseñados específicamente para que el sistema inercial sea flexible y capaz de implementar las dos operaciones, captura de datos y ejecución de algoritmos. Esta herramienta se ha validado en diferentes proyectos europeos y diferentes trabajos llevados a cabo en el Centro de Estudios Tecnológicos para la Atención de la Dependencia y la Vida Autónoma de la Universidad Politécnica de Catalunya (CETpDUPC). En la segunda parte de la tesis se describen dos algoritmos para identificar diferentes tipos de posturas. Los datos empleados para poder desarrollar los algoritmos se han adquirido con el sistema inercial presentado en la primera parte de la tesis. Los algoritmos desarrollados se han validado en personas sanas y en pacientes con la enfermedad del Parkinson. Estos algoritmos se focalizan en la detección de posturas con un único sistema inercial localizado en la cintura. Esta localización permite una mayor comodidad para los usuarios, entre otras ventajas. Una aportación importante es la metodología utilizada para poder detectar transiciones posturales (movimiento para pasar de una postura a otra). Se ha empleado el paradigma de las máquinas de vectores soporte para poder distinguir las diferentes posturas del usuario o paciente. La detección de transiciones posturales se integra en un sistema jerárquico de clasificación, al cual se añade un detector de actividades básicas donde se identifican posturas estáticas, como estar sentado, de pie, o tumbado, y posturas dinámicas, como caminar. En la parte final de la tesis se presenta una aportación a la identificación de uno de los síntomas de la enfermedad del Parkinson más incapacitantes: el bloqueo de la marcha. El algoritmo de posturas se aplica en señales captadas por la unidad inercial descrita en la primera parta de la tesis para complementar a los algoritmos actuales de detección de bloqueos de la marcha. Además, se demuestra cómo los valores de evaluación del algoritmo de detección del bloqueo de la marcha mejoran al aplicar el algoritmo de posturas. De esta forma, el nuevo algoritmo contextualiza la detección del bloqueo de la marcha, omitiendo los casos positivos que ocurran en situaciones donde no es posible que un bloqueo de la marcha ocurra.

  • Speech scrambling based on imitation of a target speech signal with non-confidential content

     Ballesteros Larrotta, Dora Maria; Moreno Arostegui, Juan Manuel
    Circuits systems and signal processing
    Vol. 33, num. 11, p. 3475-3498
    DOI: 10.1007/s00034-014-9810-9
    Date of publication: 2014-05-15
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    This paper shows a new approach of speech scrambling using imitation to produce a scrambled speech signal with intelligible content. The secret message imitates a target speech signal with non-confidential content through an adaptation mechanism.Unlike the classical approach, the key is not an input of the system because it is created in the adaptation process. Several tests are conducted in order to validate adaptation as an efficient key generator, and the robustness of the scrambled speech signal against signal manipulation attacks. The advantages of our proposal are (1) the scrambled speech signal is an intelligible speech signal so that it does not generate suspicious about the existence of the secret message, (2) the system does not require an external key generator because the elements are permutated according to the adaptation process, (3) the length of the key is long enough so that it guarantees that the key is not found by brute-force attack, and (4) it works with perfect secrecy because the a priori probability of the secret message is the same as the a posteriori probability, given the scrambled speech signal.

  • Digitizing archetypal human experience through pshysiological signals

     Ivonin, Leonid
    Eindhoven University of Technology
    Theses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    The problem of capturing human experience is relevant in many application domains. In fact, the process of describing and sharing individual experience lies at the heart of human culture. Throughout the courses of our lives we learn a great deal of information about the world from other people’s experience. Besides the ability to share utilitarian experience such as whether a particular plant is poisonous, humans have developed a sophisticated competency of social signaling that enables us to express and decode emotional experience. The natural way of sharing emotional experiences requires those who share to be co-present during this event. However, people have overcome the limitation of physical presence by creating a symbolic system of representations. This advancement came at a price of losing some of the multidimensional aspects of primary, bodily experience during its projection into the symbolic form. Recent research in the field of affective computing has addressed the question of digitization and transmission of emotional experience through monitoring and interpretation of physiological signals. Although the outcomes of this research represent a great step forward in developing a technology that supports sharing of emotional experiences, they do not seem to help in preserving the original phenomenological experience during the aforementioned projection. This circumstance is explained by the fact that in affective computing the focus of investigation has been aimed at emotional experiences which can be consciously evaluated and described by individuals themselves. Therefore, generally speaking, applying an affective computing technique for capturing emotions of an individual is not a deeper or more precise way to project her experience into the symbolic form than asking this person to write down a description of her emotions on a piece of paper. One can say that so far the research in affective computing has aimed at delivering Technology that could automate the projection but it has not considered the problem of improving the projection in order to preserve more of the multidimensional aspects of human experience. This dissertation examines whether human experience, which individuals are not able to consciously transpose into the symbolic representation, can still be captured using the techniques of affective computing. First, a theoretical framework for description of human experience which is not accessible for conscious awareness was formulated. This framework was based on the work of Carl Jung who introduced a model of a psyche that includes three levels: consciousness, the personal unconscious and the collective unconscious. [...]

  • Access to the full text
    Mood boards as a universal tool for investigating emotional experience  Open access

     Huang-Ming, Chang; Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias
    Lecture notes in computer science
    Vol. 8520 Design, User Experie, p. 220-231
    DOI: 10.1007/978-3-319-07638-6_22
    Date of publication: 2014-04-03
    Journal article

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    Emotion is an essential part of user experience. While researchers are striving for new research tools for evaluate emotional experiences in design, designers have been using experience-based tools for studying emotions in practice, such as mood boards. Mood boards were developed for communicating emotional qualities between designers and clients, but have not yet been considered as an evaluation tool for investigating emotional experience. In this study we examined whether design students and non-design students have similar criteria in evaluating these mood boards. The results showed that the inter-rater reliability among all participants were considerably high, which suggested that mood boards are potential to be used as an evaluation tool for research on emotion.

  • Unifying interaction across distributed controls in a smart environment : Using anthropology-based computing to make human-computer interaction "Calm"

     Brown, John Neal Abram
    DOI: 10.13140/2.1.5166.4645
    Alpen-Adria-Universität Klagenfurt
    Theses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Rather than adapt human behavior to suit a life surrounded by computerized systems, is it possible to adapt the systems to suit humans? Mark Weiser called for this fundamental change to the design and engineering of computer systems nearly twenty years ago. We believe it is possible and offer a series of related theoretical developments and practical experiments designed in an attempt to build a system that can meet his challenge without resorting to black box design principles or Wizard of Oz protocols. This culminated in a trial involving 32 participants, each of whom used two different multimodal interactive techniques, based on our novel interaction paradigm, to intuitively control nine distributed devices in a smart home setting. The theoretical work and practical developments have led to our proposal of seven contributions to the state of the art.

  • Realistic model of compact VLSI FitzHugh-Nagumo oscillators

     Cosp Vilella, Jordi; Binczak, Stéphane; Madrenas Boadas, Jordi; Ruiz Fernández, Daniel
    International journal of electronics
    Vol. 101, num. 2, p. 220-230
    DOI: 10.1080/00207217.2013.780263
    Date of publication: 2014-02-01
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    In this article, we present a compact analogue VLSI implementation of the FitzHugh-Nagumo neuron model, intended to model large-scale, biologically plausible, oscillator networks. As the model requires a series resistor and a parallel capacitor with the inductor, which is the most complex part of the design, it is possible to greatly simplify the active inductor implementation compared to other implementations of this device as typically found in filters by allowing appreciable, but well modelled, nonidealities. We model and obtain the parameters of the inductor nonideal model as an inductance in series with a parasitic resistor and a second order low-pass filter with a large cut-off frequency. Post-layout simulations for a CMOS 0.35 m double-poly technology using the MOSFET Spice BSIM3v3 model confirm the proper behaviour of the design. © 2013 Taylor & Francis.

  • prórroga del TIN2010-20966-C02-02 - APRENDIZAJE AUTOMATICO DE CAPACIDADES SENSORIALES MEDIANTE MAQUINAS DE SOPORTE VECTORIAL

     Sanchez Soler, Monica; Ruiz Vegas, Francisco Javier; Prats Duaygues, Francesc; Aguado Chao, Juan Carlos; Sama Monsonis, Albert
    Competitive project

     Share

  • Centre d'Estudis Tecnològics per a l'atenció a la Dependència i vida autònoma

     Cabestany Moncusi, Joan; Moreno Arostegui, Juan Manuel; Madrenas Boadas, Jordi; Cosp Vilella, Jordi; Llanas Parra, Francesc Xavier; Sama Monsonis, Albert; Perez Lopez, Carlos; Rodriguez Martin, Daniel Manuel; Reyes Ortiz, Jorge Luis; Sayeed, Taufique; Takac, Boris; Khan, Rafiullah; Huang-Ming, Chang; Bano, Sophia; Català Mallofré, Andreu
    Competitive project

     Share

  • Contribució en la modelització de la rugositat superficial obtinguda en els processos de honing utilitzant xarxes neuronals artificials

     Sivatte Adroer, Maurici
    Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract  Share Reference managers Reference managers Open in new window

    A partir de dades experimentals obtingudes en una màquina d'assaigs construïda a l'efecte,l'objectiu d'aquesta tesi és modelitzar el procés de brunyit dels interiors de cilindres d'acer(procés de honing), utilitzant les xarxes neuronals artificials.Es seleccionen les cinc variables de procés que es consideren més rellevants i es defineix elPerceptró Multicapa amb l'algoritme d'aprenentatge backpropagation i el sistema de validaciócross-validation, com a xarxa neuronal base.En una primera fase es dissenya un model neuronal, de forma què a partir de les dades de procésseleccionades, dóna el paràmetre de rugositat superficial "Ra" que hi correspondria.En una segona fase es modelitza el problema invers. A partir del valor de la rugositat superficial"Ra" que es desitja, el model neuronal explicita els valors de les variables de procés que lafarien possible.Finalment, es dissenya un model iteratiu, utilitzant les xarxes seleccionades en les fasesanteriors, per millorar la modelització inversa.

  • Access to the full text
    Design of an on-chip linear-assisted DC-DC voltage regulator  Open access

     Cosp Vilella, Jordi; Martínez García, Herminio
    IEEE International Conference on Electronics, Circuits and Systems
    p. 353-356
    Presentation's date: 2013-12-10
    Presentation of work at congresses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    This article shows the design of an on-chip CMOS linear-assisted DC-DC regulator. It results a good alternative topology to classic switching DC-DC power converters. In the presented technique, an auxiliary linear regulator is used to cancel the output voltage ripple and provides fast responses for load and line variations. On the other hand, a switching converter, connected in parallel, allows supplying almost the whole output current demanded by the load. The objective of this linear-assisted regulator or hybrid topology is to achieve a high efficiency of switching converters, with suitable load and line regulation features, typical of linear regulators. In this kind of on-chip applications, CMOS is the current prevailing technology. Thus, in order to implement on-chip power supply systems and on-chip power management systems with low-to-medium current consumption, this structure has good features.

    Postprint (author’s final draft)

  • A self-adaptive hardware architecture with fault tolerance capabilities

     Soto, Javier; Moreno Arostegui, Juan Manuel; Cabestany Moncusi, Joan
    Neurocomputing
    Vol. 121, p. 25-31
    DOI: 10.1016/j.neucom.2012.10.038
    Date of publication: 2013-12-09
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    This paper describes a Fault Tolerance System (FTS) implemented in a new self-adaptive hardware architecture. This architecture is based on an array of cells that implements in a distributed way self-adaptive capabilities. The cell includes a configurable multiprocessor, so it can have between one and four processors working in parallel, with a programmable configuration mode that allows selecting the size of program and data memories. The self-elimination and self-replication capabilities of cell(s) are performed when the FTS detects a failure in any of the processors that include it, so that this cell(s) will be self-discarded for future implementations. Other adaptive capabilities of the system are self-routing, self-placement and runtime self-configuration. Additionally, it is described as an example application and a software tool that has been implemented to facilitate the development of applications to test the system.

  • SVM-based posture identification with a single waist-located triaxial accelerometer

     Rodriguez Martin, Daniel Manuel; Sama Monsonis, Albert; Perez Lopez, Carlos; Català Mallofré, Andreu; Cabestany Moncusi, Joan; Rodríguez Molinero, Alejandro
    Expert systems with applications
    Vol. 40, num. 18, p. 7203-7211
    DOI: 10.1016/j.eswa.2013.07.028
    Date of publication: 2013-12
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Analysis of human body movement is an important research area, specially for health applications. In order to assess the quality of life of people with mobility problems like Parkinson’s disease o stroke patients, it is crucial to monitor and assess their daily life activities. The main goal of this work is the characterization of basic activities using a single triaxial accelerometer located at the waist. This paper presents a novel postural detection algorithm based in SVM methods which is able to detect and identify Walking, Stand, Sit, Lying, Sit to Stand, Stand to sit, Bending up/down, Lying from Sit and Sit from Lying transitions with a sensitivity of 97% and specificity of 84% with 2884 postures analyzed from 31 healthy volunteers. Parameters and models found have been tested in another dataset from Parkinson’s disease patients, achieving results of 98% of sensitivity and 78% of specificity in postural transitions. The proposed algorithm has been optimized to be easily implemented in real-time system for on-line monitoring applications.

  • Access to the full text
    A wearable inertial measurement unit for long-term monitoring in the dependency care area  Open access

     Rodriguez Martin, Daniel Manuel; Perez Lopez, Carlos; Sama Monsonis, Albert; Cabestany Moncusi, Joan; Català Mallofré, Andreu
    Sensors
    Vol. 13, num. 10, p. 14079-14104
    DOI: 10.3390/s131014079
    Date of publication: 2013-10-18
    Journal article

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    Human movement analysis is a field of wide interest since it enables the assessment of a large variety of variables related to quality of life. Human movement can be accurately evaluated through Inertial Measurement Units (IMU), which are wearable and comfortable devices with long battery life. The IMU's movement signals might be, on the one hand, stored in a digital support, in which an analysis is performed a posteriori. On the other hand, the signal analysis might take place in the same IMU at the same time as the signal acquisition through online classifiers. The new sensor system presented in this paper is designed for both collecting movement signals and analyzing them in real-time. This system is a flexible platform useful for collecting data via a triaxial accelerometer, a gyroscope and a magnetometer, with the possibility to incorporate other information sources in real-time. A SD card can store all inertial data and a Bluetooth module is able to send information to other external devices and receive data from other sources. The system presented is being used in the real-time detection and analysis of Parkinson's disease symptoms, in gait analysis, and in a fall detection system.

    Human movement analysis is a field of wide interest since it enables the assessment of a large variety of variables related to quality of life. Human movement can be accurately evaluated through Inertial Measurement Units (IMU), which are wearable and comfortable devices with long battery life. The IMU’s movement signals might be, on the one hand, stored in a digital support, in which an analysis is performed a posteriori. On the other hand, the signal analysis might take place in the same IMU at the same time as the signal acquisition through online classifiers. The new sensor system presented in this paper is designed for both collecting movement signals and analyzing them in real-time. This system is a flexible platform useful for collecting data via a triaxial accelerometer, a gyroscope and a magnetometer, with the possibility to incorporate other information sources in real-time. A μSD card can store all inertial data and a Bluetooth module is able to send information to other external devices and receive data from other sources. The system presented is being used in the real-time detection and analysis of Parkinson’s disease symptoms, in gait analysis, and in a fall detection system

  • Comparative and adaptation of step detection and step length estimators to a lateral belt worn accelerometer

     Sayeed, Taufique; Sama Monsonis, Albert; Català Mallofré, Andreu; Cabestany Moncusi, Joan
    IEEE International Conference on e-Health Networking, Applications and Services
    p. 105-109
    Presentation's date: 2013-10-10
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Parkinson’s Disease (PD) is a neurodegenerative disease that predominantly alter patients’ motor performance and compromises the speed, the automaticity and fluidity of natural movements. The patients fluctuate between periods in which they can move almost normally for some hours (ON state) and periods with motor disorders (OFF state). Gait properties are affected by the motor state of a patient: reduced stride length, reduced gait speed, increased stride width etc. The ability to assess the motor states (ON/OFF) on a continuous basis for long time without disturbing the patients’ daily life activities is an important component of PD management. An accurate report of motor states could allow clinics to adjust the medication regimen to avoid OFF periods. The real-time monitoring will also allow an online treatment by combining, for instance, with automatic drug-administration pump doses. Many studies have attempted to extract gait properties through a belt-worn single tri-axial accelerometer. In this paper, a user friendly position is proposed to place the accelerometer and three step detection methods and three step length estimators are compared considering the proposed sensor placement in signals obtained from healthy volunteers and PD patients. Adaptation methods to these step length estimators are also proposed and compared. The comparison shows that the adapted estimators improve the performance with the new proposed step detection method and reduce errors in respect of the original methods.

  • Human movement analysis by means of accelerometers: Aplication to human gait and motor symptoms of Parkinson's Disease  Open access

     Sama Monsonis, Albert
    Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    La tesis que este documento recoge es una aportación al análisis del movimiento humano a partir de las señalescapturadas por acelerómetros. Estos sensores permiten convertir la aceleración producida por algunas partes del cuerpoen señales eléctricas susceptibles de un análisis posterior. El progresivo perfeccionamiento y miniaturización de losacelerómetros ha permitido construir sensores poco invasivos que pueden ser usados de forma ambulatoria para registrarlos movimientos realizados en las actividades de la vida diaria del sujeto.La tesis se focaliza en dos ámbitos. El primero, fundamentalmente clínico, se ha centrado en el análisis del movimiento enenfermos de la Enfermedad de Parkinson (EP). El segundo ámbito, más general, ha consistido en relacionar las señalesacelerométricas con características de la marcha. Con este fin, se han desarrollado métodos para la obtención de lavelocidad de la marcha y la longitud del paso, así como para la identificación de personas. En ambos ámbitos se haempleado un único acelerómetro situado en un lado de la cintura.La EP es una enfermedad neurodegenerativa que produce primordialmente trastornos del movimiento en los pacientes quela sufren. Los principales sóntomas motores de la enfermedad son 1) los temblores, 2) la lentitud de movimientos obradicinesia, 3) la congelación de la marcha (freezing of gait FoG) y 4) los movimientos involuntarios o discinesias. Los tresprimeros síntomas aparecen cuando la medicación no ha alcanzado un efecto terapeútico efectivo. Estos periodos seconocen habitualmente como período o estado motor OFF. Los periodos en los cuales la medicación hace efecto y lospacientes presentan una movilidad normal (o casi normal) son los periodos o estados motores ON. Las discinesiasaparecen principalmente cuando el nivel de medicación en sangre es excesivo. Tanto las discinesias como los estadosOFF son consecuencia de un defecto en la administración de la medicación. Un dispositivo no invasivo que detecte yregistre las discinesias y ambos periodos ON y OFF supone una importante herramienta que permite al médico prescribircon mayor precisión la dosis de medicamento adecuada a su paciente.El trabajo realizado en esta tesis en el ámbito de la EP ha consistido en el desarrollo de algoritmos de detección dediscinesias y periodos OFF. Estos algoritmos han sido adaptados para proporcionar una detección en tiempo real, de formaque se han empleado ya en un estudio piloto en los que el ajuste de medicación suministrada por una bomba de infusiónsubcutánea se realiza de forma automática en función de la presencia de discinesias y el estado motor del paciente.La experiencia ganada en el tratamiento de la señal acelerométrica proveniente de enfermos de Parkinson ha permitidocontribuir en el campo del análisis de la marcha y realizar una aportación que relaciona varios parámetros de la misma conla señal que suministra un único acelerómetro situado en la cintura. No solo la EP puede beneficiarse de este estudio, sinotambién otras enfermedades como la Diabetes o algunas enfermedades ortopédicas y traumatológicas puedenaprovecharse de sus resultados.Por útimo, usando algunas de las ténicas de los estudios anteriores, se ha realizado una importante contribución en elámbito de la identificación biométrica de personas. Se ha puesto de manifiesto que la señal proveniente de un únicoacelerómetro situado en la cintura no solo permite obtener algunas de las características de la marcha sino tambiénpermite identificar a la persona a través del patrón de su marcha. La principal contribución teórica de esta tesis ha sido eldesarrollo de técnicas basadas en la reconstrucción de atractores. Se ha evidenciado que un número muy reducido decaracterísticas procedentes del atractor reconstruido a partir de una serie temporal de medidas de aceleración permite laextracción de los parámetros de la marcha y la identificación de personas.

    This thesis presents the original contributions of the author on the field of human movement analysis from signals captured by accelerometers. These sensors are capable of converting acceleration from some body parts into electric signals for further analysis. The progressive refinement and miniaturization of accelerometers has allowed the development of minimally invasive devices that can be used to ambulatory monitor human movements during daily live activities. The study's contributions mainly fall under two heads: first, the analysis of movement in Parkinson's disease (PD); and, second, the relationship between accelerometer signals and characteristics of gait. To this end, new methods for obtaining speed and length of strides and, also, for identifying people have been developed. In all these studies, a single sensor fixed to the patient's waist has been used. PD is a neurodegenerative disease characterized by movement alterations. The main motor symptoms of PD are 1) tremor, 2) bradykinesia or slowness of movements, 3) freezing of gait and 4) dyskinesia or abnormal involuntary movements. The first three symptoms primarily occur when the medication has not yet reached an effective therapeutic effect. These periods are commonly known as OFF periods or OFF motor state. On the other hand, periods when the patient is suitably responding to the medication are known as ON periods or ON motor state. Dyskinesias mainly appear when the medication blood level is excessive. Both dyskinesias and OFF motor states are caused by a defect in the medication administration. In this sense, a wearable device capable of detecting and recording dyskinesias and OFF periods represents an important tool that enables clinics to more accurately prescribe the medication regimen of a patient. The work done in the field of PD consisted in developing algorithms able to detect dyskinesias and both ON and OFF periods. These algorithms have been adapted to provide real-time detection, which enabled their employment in a pilot study. This clinical study has tested, for the first time, the automatic adjustment of medication performed by means of a subcutaneous infusion pump according to the dyskinesias appearance and motor state of PD patients. The experience gained in the treatment of accelerometric signals from PD has led to contribute in the field of gait analysis. First, new methods for obtaining speed and length of strides from a single sensor fixed to the patient's waist have been obtained. Not only the PD can benefit from this study, but other diseases such as diabetes or some orthopedotraumatological diseases can also benefit from its results. Finally, using some of the techniques of the previous studies, another important contribution has been made in the field of biometric person identification. The work presented shows how the signal obtained from a single accelerometer located at the waist not only enables the extraction of some gait characteristics but also permits the identification of a person through its gait pattern. The main theoretical contribution of this thesis has been the development of techniques based on the reconstruction of attractors. It has been shown that the usage of only a small number of features that characterize the reconstructed attractor obtained from a time series of acceleration measurements makes possible the extraction of important parameters of gait and the person identification.

    La tesis que este documento recoge es una aportaci on al an alisis del movimiento humano a partir de las señales capturadas por aceler ometros. Estos sensores permiten convertir la aceleraci on producida por algunas partes del cuerpo en señales el ectricas susceptibles de un an alisis posterior. El progresivo perfeccionamiento y miniaturización de los aceler ometros ha permitido construir sensores poco invasivos y que pueden ser usados de forma ambulatoria para registrar los movimientos realizados en las actividades de la vida diaria del sujeto. La tesis se focaliza en dos ambitos. El primero, fundamentalmente cl inico, se ha centrado en el an alisis del movimiento en enfermos de la Enfermedad de Parkinson (EP). El segundo ambito, m as general, ha consistido en relacionar las señales acelerom etricas con caracter sticas de la marcha. Con este fin, se han desarrollado m etodos para la obtenci on de la velocidad de la marcha y la longitud del paso, as í como para la identificaci on de personas. En ambos ambitos se ha empleado un unico aceler ometro situado en un lado de la cintura. La EP es una enfermedad neurodegenerativa que produce primordialmente trastornos del movimiento en los pacientes que la sufren. Los principales sí ntomas motores de la enfermedad son 1) los temblores, 2) la lentitud de movimientos o bradicinesia, 3) la congelaci on de la marcha (freezing of gait FoG) y 4) los movimientos involuntarios o discinesias. Los tres primeros s ntomas aparecen cuando la medicaci on no ha alcanzado un efecto terape utico efectivo. Estos periodos se conocen habitualmente como per odo o estado motor OFF. Los periodos en los cuales la medicaci on hace efecto y los pacientes presentan una movilidad normal (o casi normal) son los periodos o estados motores ON. Las discinesias aparecen principalmente cuando el nivel de medicaci on en sangre es excesivo. Tanto las discinesias como los estados OFF son consecuencia de un defecto en la administraci on de la medicaci on. Un dispositivo no invasivo que detecte y registre las discinesias y ambos periodos ON y OFF supone una importante herramienta que permite al m edico prescribir con mayor precisi on la dosis de medicamento adecuada a su paciente. El trabajo realizado en esta tesis en el ambito de la EP ha consistido en el desarrollo de algoritmos de detecci on de discinesias y periodos OFF. Estos algoritmos han sido adaptados para proporcionar una detecci on en tiempo real, de forma que se han empleado ya en un estudio piloto en los que el ajuste de medicaci on suministrada por una bomba de infusi on subcut anea se realiza de forma autom atica en funci on de la presencia de discinesias y el estado motor del paciente. La experiencia ganada en el tratamiento de la señal acelerom etrica proveniente de enfermos de Parkinson ha permitido contribuir en el campo del an alisis de la marcha y realizar una aportaci on que relaciona varios par ametros de la misma con la señal que suministra un unico aceler ometro situado en la cintura. No solo la EP puede bene ciarse de este estudio, sino tambi en otras enfermedades como la Diabetes o algunas enfermedades ortop edicas y traumatol ogicas pueden aprovecharse de sus resultados. Por ultimo, usando algunas de las t ecnicas de los estudios anteriores, se ha realizado una importante contribuci on en el ambito de la identi caci on biom etrica de personas. Se ha puesto de mani esto que la señal proveniente de un unico aceler ometro situado en la cintura no solo permite obtener algunas de las caracter sticas de la marcha sino tambi en permite identi car a la persona a trav es del patr on de su marcha. La principal contribuci on te orica de esta tesis ha sido el desarrollo de t ecnicas basadas en la reconstrucci on de atractores. Se ha evidenciado que un n umero muy reducido de caracter sticas procedentes del atractor reconstruido a partir de una serie temporal de medidas de aceleraci on permite la extracci on de los par ametros de la marcha y la identi caci on de personas

  • Comparison and adaptation of step length and gait speed estimators from single belt worn accelerometer positioned on lateral side of the body

     Sayeed, Taufique; Sama Monsonis, Albert; Català Mallofré, Andreu; Cabestany Moncusi, Joan
    IEEE International Symposium on Intelligent Signal Processing
    p. 14-20
    Presentation's date: 2013-09-17
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Parkinson’s Disease (PD) is a neurodegenerative disease that predominantly alter patients’ motor performance and compromises the speed, the automaticity and fluidity of natural movements. The patients fluctuate between periods in which they can move almost normally for some hours (ON state) and periods with motor disorders (OFF state). Gait properties are affected by the motor state of a patient: reduced stride length, reduced gait speed, increased stride width etc. The ability to assess the motor states (ON/OFF) on a continuous basis for long time without disturbing the patients’ daily life activities is an important component of PD management. An accurate report of motor states could allow clinics to adjust the medication regimen to avoid OFF periods. The real-time monitoring will also allow an online treatment by combining, for instance, with automatic drugadministration pump doses. Many studies have attempted to extract gait properties through a belt-worn single tri-axial accelerometer. In this paper, a user friendly position is proposed to place the accelerometer and six step length estimators are compared considering the proposed sensor placement in a preliminary database of healthy volunteers. Adaptation methods to some of these estimators are also proposed and compared. The comparison shows that the adapted estimators improve the performance and reduce errors in respect of the original methods applied in the new sensor location.

  • Selection of a neural network for modelling the honing process

     Sivatte Adroer, Maurici; Llanas Parra, Francesc Xavier; Buj Corral, Irene; Vivancos Calvet, Joan
    International Reseach/Expert Conference "Trends in the Development of Machinery and Associated Technology"
    p. 1-4
    Presentation's date: 2013-09-11
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Roughness obtained in honing process depends on many different process parameters, such as grain size of abrasive stones, pressure of stones on the workpiece’s surface, density of abrasive, tangential speed of the honing head and linear speed of the honing head. This fact makes it difficult to study the process from an analytical point of view. For this reason, use of empirical methods or use of artificial intelligence is recommended in this case. In the present paper, results about use of neural networks for obtaining average roughness Ra as a function of honing parameters are presented. Best neural network was chosen among different possibilities. For doing this, experimental results were divided into three groups: 70 % of results were used for training, 15 % of results were used for validation and 15 % of results were used as test to compare networks with other models. The best neural network was considered to be the one with lowest errors using the validation experimental results.

  • On chaotic behavior in automatic tuning loops for continuous-time filters

     Cosp Vilella, Jordi; Martínez García, Herminio
    European Conference on Circuit Theory and Design
    p. 1-4
    DOI: 10.1109/ECCTD.2013.6662232
    Presentation's date: 2013-09-10
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Continuous-time filters (CTFs) with automatic tuning loops are nonlinear feedback systems with potential instability. Thus, their appropriate linear dynamic modeling should be obtained to assure stability in case of an improved design of the loop controllers is to be carried out. A systematic approach using a small signal model would allow obtaining these controllers. However, bifurcations and nonlinear phenomena may appear which cannot be predicted by this analysis. This leads to potential instability, semiperiodic or chaotic behavior and, thus, circuit malfunction. The aim of this paper is to show by means of simulations and experimental results that nonlinear phenomena, which cannot be predicted by the common small signal analysis, may appear in this kind of circuits when circuit parameters are varied.

    Continuous–time filters (CTFs) with automatic tuning loops are nonlinear feedback systems with potential instability. Thus, their appropriate linear dynamic modeling should be obtained to assure stability in case of an improved design of the loop controllers is to be carried out. A systematic approach using a small signal model would allow obtaining these controllers. However, bifurcations and nonlinear phenomena may appear which cannot be predicted by this analysis. This leads to potential instability, semiperiodic or chaotic behavior and, thus, circuit malfunction. The aim of this paper is to show by means of simulations and experimental results that nonlinear phenomena, which cannot be predicted by the common small signal analysis, may appear in this kind of circuits when circuit parameters are varied.

  • Access to the full text
    What do we feel about archetypes: self-reports and physiological signals  Open access

     Chang, Huang-ming; Ivonin, Leonid; Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias
    European Signal Processing Conference
    p. 1-5
    Presentation's date: 2013-09-10
    Presentation of work at congresses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    In research on emotion, presenting affective stimuli has been believed to be an effective and reliable technique for emotion elicitation. Instead of collecting stimuli for pre- defined emotions, we propose to develop stimuli based on their symbolic meanings. We adopted archetypal symbolism as a standard to edit eight movie clips of archetypes as a new set of affective stimuli. These stimuli were used in an experiment for emotion elicitation. Participants’ emotional responses toward these stimuli of archetypes were measured by the self-report technique and the physiological measurement. The results of linear discriminant analysis show that physiological measurement is more robust than the self-report techniques in recognizing emotions toward stimuli of archetypes. However, it is still unclear which technique reflects the ground truth of human emotion. We discuss alternative implications of these results, and provide more research questions for future studies on emotion recognition and model development.

    In research on emotion, presenting affective stimuli has been believed to be an effective and reliable technique for emotion elicitation. Instead of collecting stimuli for predefined emotions, we propose to develop stimuli based on their symbolic meanings. We adopted archetypal symbolism as a standard to edit eight movie clips of archetypes as a new set of affective stimuli. These stimuli were used in an experiment for emotion elicitation. Participants’ emotional responses toward these stimuli of archetypes were measured by the self-report technique and the physiological measurement. The results of linear discriminant analysis show that physiological measurement is more robust than the self-report techniques in recognizing emotions toward stimuli of archetypes. However, it is still unclear which technique reflects the ground truth of human emotion. We discuss alternative implications of these results, and provide more research questions for future studies on emotion recognition and model development.

  • Access to the full text
    From mythology to psychology: Identifying archetypal symbols in movies  Open access

     Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias; Ivonin, Leonid; Huang-Ming, Chang
    Technoetic arts: a journal of speculative research
    Vol. 11, num. 2, p. 99-113
    DOI: 10.1386/tear.11.2.99_1
    Date of publication: 2013-09-06
    Journal article

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    In this article, we introduce the theory of archetype, which explains the connection between ancient myths and the human mind. Based on the assumption that archetypes are in the deepest level of human mind, we propose that archetypal symbolism is a kind of knowledge that supports the cognitive process for creating subjective world-view towards the physical world we live in. According to archetypal symbolism, we conducted an empirical study to identify archetypal symbols in modern movies. A new collection of movie clips was developed to represent eight essential archetypes: anima, animus, mentor, mother, shadow, hero’s departure, hero’s initiation and hero’s return, which can be used in future studies on human emotion. In order to investigate the emotions towards these archetypal symbols, we provide suggestions from the psychological point of view. The present study demonstrates how to identify symbolic meanings in movies, and indicates a new direction for future studies in psychology.

  • Access to the full text
    Experiencing the world with archetypal symbols: A new form of aesthetics.  Open access

     Huang-Ming, Chang; Ivonin, Leonid; Diaz Boladeras, Marta; Català Mallofré, Andreu; Chen, Wei; Rauterberg, Mattias
    Lecture notes in computer science
    Vol. 8028 Distributed, Ambient, p. 205-214
    DOI: 10.1007/978-3-642-39351-8_23
    Date of publication: 2013-09-04
    Journal article

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    According to the theories of symbolic interactionism, phenomenology of perception and archetypes, we argue that symbols play the key role in translating the information from the physical world to the human experience, and archetypes are the universal knowledge of cognition that generates the background of human experience (the life-world). Therefore, we propose a conceptual framework that depicts how people experience the world with symbols, and how archetypes relate the deepest level of human experience. This framework indicates a new direction of research on memory and emotion, and also suggests that archetypal symbolism can be a new resource of aesthetic experience design.

  • The imbalance network and incremental evolution for mobile robot nervous system design

     Olivier, Paul; Moreno Arostegui, Juan Manuel
    International Conference on Artificial Neural Networks
    p. 519-526
    DOI: 10.1007/978-3-642-40728-4_65
    Presentation's date: 2013-09
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Automatic design of neurocontrollers (as in Evoluationary Robotics) utilizes incremental evolution to solve for more complex behaviors. Also manual design techniques such as task decomposition are employed. Manual design itself can benefit from focusing on using incremental evolution to add more automatic design. The imbalance network is a neural network that integrates incremental evolution with an incremental design process without the need for task decomposition. Instead, the imbalance network uses the mechanism of the equilibrium-action cycle to structure the network while emphasizing behavior emergence. An example 11-step design (including a 5-step evolutionary process) is briefly mentioned to help ground the imbalance network concepts.

    Automatic design of neurocontrollers (as in Evoluationary Robotics) utilizes incremental evolution to solve for more complex behaviors. Also manual design techniques such as task decomposition are employed. Manual design itself can benefit from focusing on using incremental evolution to add more automatic design. The imbalance network is a neural network that integrates incremental evolution with an incremental design process without the need for task decomposition. Instead, the imbalance network uses the mechanism of the equilibrium-action cycle to structure the network while emphasizing behavior emergence. An example 11-step design (including a 5-step evolutionary process) is briefly mentioned to help ground the imbalance network concepts.

  • Observation of chaotic behavior in automatic tuning loops for continuous–time filters

     Martínez García, Herminio; Cosp Vilella, Jordi; Manzanares Brotons, Manuel
    IEEE International Midwest Symposium on Circuits and Systems
    p. 742-745
    Presentation's date: 2013-08-06
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    The appropriate linear dynamic modeling of continuous–time filters (CTFs) with automatic tuning loops should be obtained to assure stability in case an improved design of the loop controllers is to be carried out. With this aim, starting from a general and systematic analysis in order to obtain an equivalent small–signal linearized incremental model, from which transfer functions between output variables and control voltages are derived, the subsequent design of compensated loops with enhanced stability and dynamic performance is required. This systematic procedure allows obtaining improved controllers for the two involved control loops. However, CTFs with automatic tuning loops are nonlinear feedback systems with potential instability. What is more, nonlinear phenomena, which cannot be predicted by a design-oriented small signal modeling approach, are observed in this kind of tuning systems. The purpose of this work is to highlight that when control parameters are varied, the system could present different kinds of dynamical nonlinear phenomena such as bifurcations and chaotic behavior, which cannot be predicted by the small signal design-oriented model.

  • METHODS OF COVERT COMMUNICATION OF SPEECH SIGNALS BASED ON A BIO-INSPIRED PRINCIPLE  Open access

     Ballesteros Larrotta, Dora Maria
    Department of Electronic Engineering, Universitat Politècnica de Catalunya
    Theses

    Read the abstract Read the abstract Access to the full text Access to the full text Open in new window  Share Reference managers Reference managers Open in new window

    Esta tesis presenta dos métodos de comunicación encubierta de señales de voz utilizando un concepto bio-inspirado, conocido como la habilidad de adaptación de señales de voz. Si se tienen dos señales de voz con diferente texto, la primera un mensaje secreto y la segunda una señal de voz con contenido no confidencial (señal objetivo), es posible que el mensaje secreto se manipule para suene como la señal objetivo. La habilidad de adaptación de las señales de voz se utiliza para encriptar señales de voz en el dominio wavelet. A diferencia de los esquemas tradicionales, la señal de voz encriptada es una señal con texto legible y el proceso de permutación se basa en la adaptación entre los coeficientes wavelet del mensaje secreto y los coeficientes wavelet de la señal de voz objetivo. Entonces, el sistema puede ser visto como un caso especial de encriptación tiempo-frecuencia (TFS). Este es un sistema perfectamente secreto porque la longitud de la clave es igual a la longitud del mensaje secreto, hay tantos mensajes secretos como señales encriptadas y el mapeo entre las entradas y las salidas es uno a uno. Se concluye que el sistema supera el ataque de fuerza bruta. Adicionalmente, el mensaje secreto puede ser recuperado por el usuario autorizado aún si la señal de voz encriptada es manipulada utilizando ataques de compresión, filtrado o re-muestreo.En el caso de esteganografía, se proponen dos esquemas. El primero se denomina enmascaramiento wavelet eficiente (EWM) y el segundo EWM mejorado (iEWM). Los dos aprovechan la propiedad de enmascaramiento del sistema auditivo humano (HAS) utilizando un proceso eficiente de enmascaramiento basado en la adaptación del mensaje secreto a la señal de voz huésped. EWM utiliza una sustitución directa basada en el parámetro Pd el cual relaciona la amplitud de los coeficientes wavelet de la señal huésped con los coeficientes wavelet del mensaje secreto adaptado. Los 5 bits menos significativos (LSBs) de los coeficientes wavelet de la señal huésped se reemplazan con el valor de Pd. A diferencia de EWM, iEWM utiliza sustitución directa. En este caso, los coeficientes wavelet del mensaje secreto adaptado se ocultan directamente en los coeficientes wavelet de la señal huésped. El número de bits varía de acuerdo a la amplitud de los coeficientes wavelet de la señal huésped, a mayor amplitud mayor es el número de LSBs que son modificados. No obstante, los bits más significativos (MSBs) se mantienen intactos de acuerdo al parámetro de control SBH. Mientras EWM demostró ser más transparente (es decir que la señal estego no genera sospechas sobre la existencia del mensaje secreto), iEWM es más robusto frente a manipulaciones de señal como compresión y re-muestreo.Dado que los dos modelos propuestos necesitan conocer de antemano tanto el mensaje secreto como la señal objetivo (o la señal huésped), no se pueden utilizar en aplicaciones de tiempo real. De tal forma, se presenta un esquema de esteganografía que trabaja sobre dispositivos hardware en el cual la señal estego se obtiene casi inmediatamente como se pronuncia el mensaje secreto.

    This work presents two speech hiding methods based on a bio-inspired concept known as the ability of adaptation of speech signals. A cryptographic model uses the adaptation to transform a secret message to a non-sensitive target speech signal, and then, the scrambled speech signal is an intelligible signal. The residual intelligibility is extremely low and it is appropriate to transmit secure speech signals. On the other hand, in a steganographic model, the adapted speech signal is hidden into a host signal by using indirect substitution or direct substitution. In the first case, the scheme is known as Efficient Wavelet Masking (EWM), and in the second case, it is known as improved-EWM (iEWM). While EWM demonstrated to be highly statistical transparent, the second one, iEWM, demonstrated to be highly robust against signal manipulations. Finally, with the purpose to transmit secure speech signals in real-time operation, a hardware-based scheme is proposed

    Esta tesis presenta dos métodos de comunicación encubierta de señales de voz utilizando un concepto bio-inspirado, conocido como la “habilidad de adaptación de señales de voz”. El modelo de criptografía utiliza la adaptación para transformar un mensaje secreto a una señal de voz no confidencial, obteniendo una señal de voz encriptada legible. Este método es apropiado para transmitir señales de voz seguras porque en la señal encriptada no quedan rastros del mensaje secreto original. En el caso de esteganografía, la señal de voz adaptada se oculta en una señal de voz huésped, utilizando sustitución directa o indirecta. En el primer caso el esquema se denomina EWM y en el segundo caso iEWM. EWM demostró ser altamente transparente, mientras que iEWM demostró ser altamente robusto contra manipulaciones de señal. Finalmente, con el propósito de transmitir señales de voz seguras en tiempo real, se propone un esquema para dispositivos hardware.

  • Position and orientation tracking in a ubiquitous monitoring system for Parkinson disease patients with freezing of gait symptom

     Takac, Boris; Català Mallofré, Andreu; Rodriguez Martin, Daniel Manuel; Van Der Aa, Nico; Chen, Wei; Rauterberg, Mattias
    JMIR mobile and ubiquitous health (mhealth & uhealth)
    Vol. 15, num. 7
    DOI: 10.2196/mhealth.2539
    Date of publication: 2013-07
    Journal article

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Background: Freezing of gait (FoG) is one of the most disturbing and least understood symptoms in Parkinson disease (PD). Although the majority of existing assistive systems assume accurate detections of FoG episodes, the detection itself is still an open problem. The specificity of FoG is its dependency on the context of a patient, such as the current location or activity. Knowing the patient's context might improve FoG detection. One of the main technical challenges that needs to be solved in order to start using contextual information for FoG detection is accurate estimation of the patient's position and orientation toward key elements of his or her indoor environment. Objective: The objectives of this paper are to (1) present the concept of the monitoring system, based on wearable and ambient sensors, which is designed to detect FoG using the spatial context of the user, (2) establish a set of requirements for the application of position and orientation tracking in FoG detection, (3) evaluate the accuracy of the position estimation for the tracking system, and (4) evaluate two different methods for human orientation estimation. Methods: We developed a prototype system to localize humans and track their orientation, as an important prerequisite for a context-based FoG monitoring system. To setup the system for experiments with real PD patients, the accuracy of the position and orientation tracking was assessed under laboratory conditions in 12 participants. To collect the data, the participants were asked to wear a smartphone, with and without known orientation around the waist, while walking over a predefined path in the marked area captured by two Kinect cameras with non-overlapping fields of view. Results: We used the root mean square error (RMSE) as the main performance measure. The vision based position tracking algorithm achieved RMSE = 0.16 m in position estimation for upright standing people. ..

  • Identification of postural transitions using a waist-located inertial sensor

     Rodriguez Martin, Daniel Manuel; Sama Monsonis, Albert; Perez Lopez, Carlos; Català Mallofré, Andreu; Cabestany Moncusi, Joan; Rodríguez Molinero, Alejandro
    International Work-Conference on Artificial Neural Networks
    p. 142-149
    DOI: 10.1007/978-3-642-38682-4
    Presentation's date: 2013-06-13
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    Analysis of human movement is an important research area, specially for health applications. In order to assess the quality of life of people with mobility problems like Parkinson’s disease (PD) or stroke patients, it is crucial to monitor their daily life activities. The main goal of this work is to characterize basic activities and their transitions using a single sensor located at the waist. This paper presents a novel postural detection algorithm which is able to detect and identify 6 different postural transitions, sit to stand, stand to sit, bending up/down and lying to sit and sit to lying transitions with a sensitivity of 86.5% and specificity of 95%. The algorithm has been tested on 31 healthy volunteers and 8 PD patients who performed a total of 545 and 176 transitions respectively. The proposed algorithm is suitable to be implemented in real-time systems for on-line monitoring applications.

  • Millor projecte europeu de recerca. European Projects Awards 2013

     Cabestany Moncusi, Joan; Rovira, Jordi; Rodríguez, Alejandro
    Award or recognition

    View View Open in new window  Share

  • Preface

     Rojas, Ignacio; Joya, Gonzalo; Cabestany Moncusi, Joan
    International Work-Conference on Artificial Neural Networks
    p. V-VII
    Presentation's date: 2013-06
    Presentation of work at congresses

     Share Reference managers Reference managers Open in new window

  • REMPARK: when AI and technology meet parkinson disease assessment

     Cabestany Moncusi, Joan; Perez Lopez, Carlos; Sama Monsonis, Albert; Moreno Arostegui, Juan Manuel; Bayes Rusiñol, Maria Ángels; Rodríguez Molinero, Alejandro
    International Conference Mixed Design of Integrated Circuits and Systems
    p. 562-567
    Presentation's date: 2013-06
    Presentation of work at congresses

    Read the abstract Read the abstract View View Open in new window  Share Reference managers Reference managers Open in new window

    REMPARK project objective is to develop a personal health system with closed loop detection, response and action capabilities for the assessment and possible management of Parkinson's Disease (PD) patients. The project is developing a wearable monitoring system able to identify in real time the motor status of the PD patients and evaluating ON/OFF/Dyskinesia status with a very high sensitivity and specificity degree (>80%) in operation during ambulatory conditions. Identification of the motor status is based on the knowledge included in a large database obtained with the collaboration of a number of volunteer PD patients, according a specific defined protocol in ambulatory conditions. Artificial Intelligence (AI) methods are applied to the database information for the automatic detection of motor symptoms.