Contribución a la localización en interiores basada en comunicaciones inalámbricas

Determinar con buena precisión la posición en la que se encuentra un terminal móvil, cuando éste se halla inmerso en un entorno de interior (centros comerciales, edificios de oficinas, aeropuertos, estaciones, túneles, etc), es el pilar básico sobre el que se sustentan un gran número de aplicaciones y servicios. Muchos de esos servicios se encuentran ya disponibles en entornos de exterior, aunque los entornos de interior se prestan a otros servicios específicos para ellos. Ese número, sin embargo, podría ser significativamente mayor de lo que actualmente es, si no fuera necesaria una costosa infraestructura para llevar a cabo el posicionamiento con la precisión adecuada a cada uno de los hipotéticos servicios. O, igualmente, si la citada infraestructura pudiera tener otros usos distintos, además del relacionado con el posicionamiento. La usabilidad de la misma infraestructura para otros fines distintos ofrecería la oportunidad de que la misma estuviera ya presente en las diferentes localizaciones, porque ha sido previamente desplegada para esos otros usos; o bien facilitaría su despliegue, porque el coste de esa operación ofreciera un mayor retorno de usabilidad para quien lo realiza. Las tecnologías inalámbricas de comunicaciones basadas en radiofrecuencia, ya en uso para las comunicaciones de voz y datos (móviles, WLAN, etc), cumplen el requisito anteriormente indicado y, por tanto, facilitarían el crecimiento de las aplicaciones y servicios basados en el posicionamiento, en el caso de poderse emplear para ello. Sin embargo, determinar la posición con el nivel de precisión adecuado mediante el uso de estas tecnologías, es un importante reto hoy en día. El presente trabajo pretende aportar avances significativos en este campo. A lo largo del mismo se llevará a cabo, en primer lugar, un estudio de los principales algoritmos y técnicas auxiliares de posicionamiento aplicables en entornos de interior. La revisión se centrará en aquellos que sean aptos tanto para tecnologías móviles de última generación como para entornos WLAN. Con ello, se pretende poner de relieve las ventajas e inconvenientes de cada uno de estos algoritmos, teniendo como motivación final su aplicabilidad tanto al mundo de las redes móviles 3G y 4G (en especial a las femtoceldas y small-cells LTE) como al indicado entorno WLAN; y teniendo siempre presente que el objetivo último es que vayan a ser usados en interiores. La principal conclusión de esa revisión es que las técnicas de triangulación, comúnmente empleadas para realizar la localización en entornos de exterior, se muestran inútiles en los entornos de interior, debido a efectos adversos propios de este tipo de entornos como la pérdida de visión directa o los caminos múltiples en el recorrido de la señal. Los métodos de huella radioeléctrica, más conocidos bajo el término inglés “fingerprinting”, que se basan en la comparación de los valores de potencia de señal que se están recibiendo en el momento de llevar a cabo el posicionamiento por un terminal móvil, frente a los valores registrados en un mapa radio de potencias, elaborado durante una fase inicial de calibración, aparecen como los mejores de entre los posibles para los escenarios de interior. Sin embargo, estos sistemas se ven también afectados por otros problemas, como por ejemplo los importantes trabajos a realizar para ponerlos en marcha, y la variabilidad del canal. Frente a ellos, en el presente trabajo se presentan dos contribuciones originales para mejorar los sistemas basados en los métodos fingerprinting. La primera de esas contribuciones describe un método para determinar, de manera sencilla, las características básicas del sistema a nivel del número de muestras necesarias para crear el mapa radio de la huella radioeléctrica de referencia, junto al número mínimo de emisores de radiofrecuencia que habrá que desplegar; todo ello, a partir de unos requerimientos iniciales relacionados con el error y la precisión buscados en el posicionamiento a realizar, a los que uniremos los datos correspondientes a las dimensiones y realidad física del entorno. De esa forma, se establecen unas pautas iniciales a la hora de dimensionar el sistema, y se combaten los efectos negativos que, sobre el coste o el rendimiento del sistema en su conjunto, son debidos a un despliegue ineficiente de los emisores de radiofrecuencia y de los puntos de captura de su huella. La segunda contribución incrementa la precisión resultante del sistema en tiempo real, gracias a una técnica de recalibración automática del mapa radio de potencias. Esta técnica tiene en cuenta las medidas reportadas continuamente por unos pocos puntos de referencia estáticos, estratégicamente distribuidos en el entorno, para recalcular y actualizar las potencias registradas en el mapa radio. Un beneficio adicional a nivel operativo de la citada técnica, es la prolongación del tiempo de usabilidad fiable del sistema, bajando la frecuencia en la que se requiere volver a capturar el mapa radio de potencias completo. Las mejoras anteriormente citadas serán de aplicación directa en la mejora de los mecanismos de posicionamiento en interiores basados en la infraestructura inalámbrica de comunicaciones de voz y datos. A partir de ahí, esa mejora será extensible y de aplicabilidad sobre los servicios de localización (conocimiento personal del lugar donde uno mismo se encuentra), monitorización (conocimiento por terceros del citado lugar) y seguimiento (monitorización prolongada en el tiempo), ya que todos ellas toman como base un correcto posicionamiento para un adecuado desempeño. ABSTRACT To find the position where a mobile is located with good accuracy, when it is immersed in an indoor environment (shopping centers, office buildings, airports, stations, tunnels, etc.), is the cornerstone on which a large number of applications and services are supported. Many of these services are already available in outdoor environments, although the indoor environments are suitable for other services that are specific for it. That number, however, could be significantly higher than now, if an expensive infrastructure were not required to perform the positioning service with adequate precision, for each one of the hypothetical services. Or, equally, whether that infrastructure may have other different uses beyond the ones associated with positioning. The usability of the same infrastructure for purposes other than positioning could give the opportunity of having it already available in the different locations, because it was previously deployed for these other uses; or facilitate its deployment, because the cost of that operation would offer a higher return on usability for the deployer. Wireless technologies based on radio communications, already in use for voice and data communications (mobile, WLAN, etc), meet the requirement of additional usability and, therefore, could facilitate the growth of applications and services based on positioning, in the case of being able to use it. However, determining the position with the appropriate degree of accuracy using these technologies is a major challenge today. This paper provides significant advances in this field. Along this work, a study about the main algorithms and auxiliar techniques related with indoor positioning will be initially carried out. The review will be focused in those that are suitable to be used with both last generation mobile technologies and WLAN environments. By doing this, it is tried to highlight the advantages and disadvantages of each one of these algorithms, having as final motivation their applicability both in the world of 3G and 4G mobile networks (especially in femtocells and small-cells of LTE) and in the WLAN world; and having always in mind that the final aim is to use it in indoor environments. The main conclusion of that review is that triangulation techniques, commonly used for localization in outdoor environments, are useless in indoor environments due to adverse effects of such environments as loss of sight or multipaths. Triangulation techniques used for external locations are useless due to adverse effects like the lack of line of sight or multipath. Fingerprinting methods, based on the comparison of Received Signal Strength values measured by the mobile phone with a radio map of RSSI Recorded during the calibration phase, arise as the best methods for indoor scenarios. However, these systems are also affected by other problems, for example the important load of tasks to be done to have the system ready to work, and the variability of the channel. In front of them, in this paper we present two original contributions to improve the fingerprinting methods based systems. The first one of these contributions describes a method for find, in a simple way, the basic characteristics of the system at the level of the number of samples needed to create the radio map inside the referenced fingerprint, and also by the minimum number of radio frequency emitters that are needed to be deployed; and both of them coming from some initial requirements for the system related to the error and accuracy in positioning wanted to have, which it will be joined the data corresponding to the dimensions and physical reality of the environment. Thus, some initial guidelines when dimensioning the system will be in place, and the negative effects into the cost or into the performance of the whole system, due to an inefficient deployment of the radio frequency emitters and of the radio map capture points, will be minimized. The second contribution increases the resulting accuracy of the system when working in real time, thanks to a technique of automatic recalibration of the power measurements stored in the radio map. This technique takes into account the continuous measures reported by a few static reference points, strategically distributed in the environment, to recalculate and update the measurements stored into the map radio. An additional benefit at operational level of such technique, is the extension of the reliable time of the system, decreasing the periodicity required to recapture the radio map within full measurements. The above mentioned improvements are directly applicable to improve indoor positioning mechanisms based on voice and data wireless communications infrastructure. From there, that improvement will be also extensible and applicable to location services (personal knowledge of the location where oneself is), monitoring (knowledge by other people of your location) and monitoring (prolonged monitoring over time) as all of them are based in a correct positioning for proper performance.

A technological framework based on automatic messaging for improving adherence of diabetic patients

En todo el mundo se ha observado un crecimiento exponencial en la incidencia de enfermedades crónicas como la hipertensión y enfermedades cardiovasculares y respiratorias, así como la diabetes mellitus, que causa un número de muertes cada vez mayor en todo el mundo (Beaglehole et al., 2008). En concreto, la prevalencia de diabetes mellitus (DM) está aumentando de manera considerable en todas las edades y representa un serio problema de salud mundial. La diabetes fue la responsable directa de 1,5 millones de muertes en 2012 y 89 millones de años de vida ajustados por discapacidad (AVAD) (OMS, 2014). Uno de los principales dilemas que suelen asociarse a la gestión de EC es la adherencia de los pacientes a los tratamientos, que representa un aspecto multifactorial que necesita asistencia en lo relativo a: educación, autogestión, interacción entre los pacientes y cuidadores y compromiso de los pacientes. Medir la adherencia del tratamiento es complicado y, aunque se ha hablado ampliamente de ello, aún no hay soluciones “de oro” (Reviews, 2002). El compromiso de los pacientes, a través de la participación, colaboración, negociación y a veces del compromiso firme, aumentan las oportunidades para una terapia óptima en la que los pacientes se responsabilizan de su parte en la ecuación de adherencia. Comprometer e involucrar a los pacientes diabéticos en las decisiones de su tratamiento, junto con expertos profesionales, puede ayudar a favorecer un enfoque centrado en el paciente hacia la atención a la diabetes (Martin et al., 2005). La motivación y atribución de poder de los pacientes son quizás los dos factores interventores más relevantes que afectan directamente a la autogestión de la atención a la diabetes. Se ha demostrado que estos dos factores desempeñan un papel fundamental en la adherencia a la prescripción, así como en el fomento exitoso de un estilo de vida sana y otros cambios de conducta (Heneghan et al., 2013). Un plan de educación personalizada es indispensable para proporcionarle al paciente las herramientas adecuadas que necesita para la autogestión efectiva de la enfermedad (El-Gayar et al. 2013). La comunicación efectiva es fundamental para proporcionar una atención centrada en el paciente puesto que influye en las conductas y actitudes hacia un problema de salud ((Frampton et al. 2008). En este sentido, la interactividad, la frecuencia, la temporalización y la adaptación de los mensajes de texto pueden promover la adherencia a un régimen de medicación. Como consecuencia, adaptar los mensajes de texto a los pacientes puede resultar ser una manera de hacer que las sugerencias y la información sean más relevantes y efectivas (Nundy et al. 2013). En este contexto, las tecnologías móviles en el ámbito de la salud (mHealth) están desempeñando un papel importante al conectar con pacientes para mejorar la adherencia a medicamentos recetados (Krishna et al., 2009). La adaptación de los mensajes de texto específicos de diabetes sigue siendo un área de oportunidad para mejorar la adherencia a la medicación y ofrecer motivación a adultos con diabetes. Sin embargo, se necesita más investigación para entender totalmente su eficacia. Los consejos de texto personalizados han demostrado causar un impacto positivo en la atribución de poder a los pacientes, su autogestión y su adherencia a la prescripción (Gatwood et al., 2014). mHealth se puede utilizar para ofrecer programas de asistencia de autogestión a los pacientes con diabetes y, al mismo tiempo, superar las dificultades técnicas y financieras que supone el tratamiento de la diabetes (Free at al., 2013). El objetivo principal de este trabajo de investigación es demostrar que un marco tecnológico basado en las teorías de cambios de conducta, aplicado al campo de la mHealth, permite una mejora de la adherencia al tratamiento en pacientes diabéticos. Como método de definición de una solución tecnológica, se han adoptado un conjunto de diferentes técnicas de conducta validadas denominado marco de compromiso de retroacción conductual (EBF, por sus siglas en inglés) para formular los mensajes, guiar el contenido y evaluar los resultados. Los estudios incorporan elementos del modelo transteórico (TTM, por sus siglas en inglés), la teoría de la fijación de objetivos (GST, por sus siglas en inglés) y los principios de comunicación sanitaria persuasiva y eficaz. Como concepto general, el modelo TTM ayuda a los pacientes a progresar a su próxima fase de conducta a través de mensajes de texto motivados específicos y permite que el médico identifique la fase actual y adapte sus estrategias individualmente. Además, se adoptan las directrices del TTM para fijar objetivos personalizados a un nivel apropiado a la fase de cambio del paciente. La GST encierra normas que van a ponerse en práctica para promover la intervención educativa y objetivos de pérdida de peso. Finalmente, los principios de comunicación sanitaria persuasiva y eficaz aplicados a la aparición de los mensajes se han puesto en marcha para aumentar la efectividad. El EBF tiene como objetivo ayudar a los pacientes a mejorar su adherencia a la prescripción y encaminarlos a una mejora general en la autogestión de la diabetes mediante mensajes de texto personalizados denominados mensajes de retroacción automáticos (AFM, por sus siglas en inglés). Después de una primera revisión del perfil, consistente en identificar características significativas del paciente basadas en las necesidades de tratamiento, actitudes y conductas de atención sanitaria, el sistema elige los AFM personalizados, los aprueba el médico y al final se transfieren a la interfaz del paciente. Durante el tratamiento, el usuario recopila los datos en dispositivos de monitorización de pacientes (PMD, por sus siglas en inglés) de una serie de dispositivos médicos y registros manuales. Los registros consisten en la toma de medicación, dieta y actividad física y tareas de aprendizaje y control de la medida del metabolismo. El compromiso general del paciente se comprueba al estimar el uso del sistema y la adherencia del tratamiento y el estado de los objetivos del paciente a corto y largo plazo. El módulo de análisis conductual, que consiste en una serie de reglas y ecuaciones, calcula la conducta del paciente. Tras lograr el análisis conductual, el módulo de gestión de AFM actualiza la lista de AFM y la configuración de los envíos. Las actualizaciones incluyen el número, el tipo y la frecuencia de mensajes. Los AFM los revisa periódicamente el médico que también participa en el perfeccionamiento del tratamiento, adaptado a la fase transteórica actual. Los AFM se segmentan en distintas categorías y niveles y los pacientes pueden ajustar la entrega del mensaje de acuerdo con sus necesidades personales. El EBF se ha puesto en marcha integrado dentro del sistema METABO, diseñado para facilitar al paciente diabético que controle sus condiciones relevantes de una manera menos intrusiva. El dispositivo del paciente se vincula en una plataforma móvil, mientras que una interfaz de panel médico permite que los profesionales controlen la evolución del tratamiento. Herramientas específicas posibilitan que los profesionales comprueben la adherencia del paciente y actualicen la gestión de envíos de AFM. El EBF fue probado en un proyecto piloto controlado de manera aleatoria. El principal objetivo era examinar la viabilidad y aceptación del sistema. Los objetivos secundarios eran también la evaluación de la eficacia del sistema en lo referente a la mejora de la adherencia, el control glucémico y la calidad de vida. Se reclutaron participantes de cuatro centros clínicos distintos en Europa. La evaluación del punto de referencia incluía datos demográficos, estado de la diabetes, información del perfil, conocimiento de la diabetes en general, uso de las plataformas TIC, opinión y experiencia con dispositivos electrónicos y adopción de buenas prácticas con la diabetes. La aceptación y eficacia de los criterios de evaluación se aplicaron para valorar el funcionamiento del marco tecnológico. El principal objetivo era la valoración de la eficacia del sistema en lo referente a la mejora de la adherencia. En las pruebas participaron 54 pacientes. 26 fueron asignados al grupo de intervención y equipados con tecnología móvil donde estaba instalado el EBF: 14 pacientes tenían T1DM y 12 tenían T2DM. El grupo de control estaba compuesto por 25 pa cientes que fueron tratados con atención estándar, sin el empleo del EBF. La intervención profesional tanto de los grupos de control como de intervención corrió a cargo de 24 cuidadores, entre los que incluían diabetólogos, nutricionistas y enfermeras. Para evaluar la aceptabilidad del sistema y analizar la satisfacción de los usuarios, a través de LimeSurvey, se creó una encuesta multilingüe tanto para los pacientes como para los profesionales. Los resultados también se recopilaron de los archivos de registro generados en los PMD, el panel médico profesional y las entradas de la base de datos. Los mensajes enviados hacia y desde el EBF y los archivos de registro del sistema y los servicios de comunicación se grabaron durante las cinco semanas del estudio. Se entregaron un total de 2795 mensajes, lo que supuso una media de 107,50 mensajes por paciente. Como se muestra, los mensajes disminuyen con el tiempo, indicando una mejora global de la adherencia al plan de tratamiento. Como se esperaba, los pacientes con T1DM recibieron más consejos a corto plazo, en relación a su estado. Del mismo modo, al ser el centro de T2DM en cambios de estilo de vida sostenible a largo plazo, los pacientes con T2DM recibieron más consejos de recomendación, en cuanto a dietas y actividad física. También se ha llevado a cabo una comparación de la adherencia e índices de uso para pacientes con T1DM y T2DM, entre la primera y la segunda mitad de la prueba. Se han observado resultados favorables para el uso. En lo relativo a la adherencia, los resultados denotaron una mejora general en cada dimensión del plan de tratamiento, como la nutrición y las mediciones de inserción de glucosa en la sangre. Se han llevado a cabo más estudios acerca del cambio a nivel educativo antes y después de la prueba, medidos tanto para grupos de control como de intervención. Los resultados indicaron que el grupo de intervención había mejorado su nivel de conocimientos mientras que el grupo de control mostró una leve disminución. El análisis de correlación entre el nivel de adherencia y las AFM ha mostrado una mejora en la adherencia de uso para los pacientes que recibieron los mensajes de tipo alertas, y unos resultados no significativos aunque positivos relacionados con la adherencia tanto al tratamiento que al uso correlacionado con los recordatorios. Por otra parte, los AFM parecían ayudar a los pacientes que no tomaban suficientemente en serio su tratamiento en el principio y que sí estaban dispuestos a responder a los mensajes recibidos. Aun así, los pacientes que recibieron demasiadas advertencias, comenzaron a considerar el envío de mensajes un poco estresante. El trabajo de investigación llevado a cabo al desarrollar este proyecto ofrece respuestas a las cuatro hipótesis de investigación que fueron la motivación para el trabajo. • Hipótesis 1 : es posible definir una serie de criterios para medir la adherencia en pacientes diabéticos. • Hipótesis 2: es posible diseñar un marco tecnológico basado en los criterios y teorías de cambio de conducta mencionados con anterioridad para hacer que los pacientes diabéticos se comprometan a controlar su enfermedad y adherirse a planes de atención. • Hipótesis 3: es posible poner en marcha el marco tecnológico en el sector de la salud móvil. • Hipótesis 4: es posible utilizar el marco tecnológico como solución de salud móvil en un contexto real y tener efectos positivos en lo referente a indicadores de control de diabetes. La verificación de cada hipótesis permite ofrecer respuesta a la hipótesis principal: La hipótesis principal es: es posible mejorar la adherencia diabética a través de un marco tecnológico mHealth basado en teorías de cambio de conducta. El trabajo llevado a cabo para responder estas preguntas se explica en este trabajo de investigación. El marco fue desarrollado y puesto en práctica en el Proyecto METABO. METABO es un Proyecto I+D, cofinanciado por la Comisión Europea (METABO 2008) que integra infraestructura móvil para ayudar al control, gestión y tratamiento de los pacientes con diabetes mellitus de tipo 1 (T1DM) y los que padecen diabetes mellitus de tipo 2 (T2DM). ABSTRACT Worldwide there is an exponential growth in the incidence of Chronic Diseases (CDs), such as: hypertension, cardiovascular and respiratory diseases, as well as diabetes mellitus, leading to rising numbers of deaths worldwide (Beaglehole et al. 2008). In particular, the prevalence of diabetes mellitus (DM) is largely increasing among all ages and constitutes a major worldwide health problem. Diabetes was directly responsible for 1,5 million deaths in 2012 and 89 million Disability-adjusted life year (DALYs) (WHO 2014). One of the key dilemmas often associated to CD management is the patients’ adherence to treatments, representing a multi-factorial aspect that requires support in terms of: education, self-management, interaction between patients and caregivers, and patients’ engagement. Measuring adherence is complex and, even if widely discussed, there are still no “gold” standards ((Giardini et al. 2015), (Costa et al. 2015). Patient’s engagement, through participation, collaboration, negotiation, and sometimes compromise, enhance opportunities for optimal therapy in which patients take responsibility for their part of the adherence equation. Engaging and involving diabetic patients in treatment decisions, along with professional expertise, can help foster a patient-centered approach to diabetes care (Martin et al. 2005). Patients’ motivation and empowerment are perhaps the two most relevant intervening factors that directly affect self-management of diabetes care. It has been demonstrated that these two factors play an essential role in prescription adherence, as well as for the successful encouragement of a healthy life-style and other behavioural changes (Heneghan et al. 2013). A personalised education plan is indispensable in order to provide the patient with the appropriate tools needed for the effective self-management of the disease (El-Gayar et al. 2013). Effective communication is at the core of providing patient-centred care since it influences behaviours and attitudes towards a health problem (Frampton et al. 2008). In this regard, interactivity, frequency, timing, and tailoring of text messages may promote adherence to a medication regimen. As a consequence, tailoring text messages to patients can constitute a way of making suggestions and information more relevant and effective (Nundy et al. 2013). In this context, mobile health technologies (mHealth) are playing significant roles in improving adherence to prescribed medications (Krishna et al. 2009). The tailoring of diabetes-specific text messages remains an area of opportunity to improve medication adherence and provide motivation to adults with diabetes but further research is needed to fully understand their effectiveness. Personalized text advices have proven to produce a positive impact on patients’ empowerment, self-management, and adherence to prescriptions (Gatwood et al. 2014). mHealth can be used for offering self-management support programs to diabetes patients and at the same time surmounting the technical and financial difficulties involved in diabetes treatment (Free et al. 2013). The main objective of this research work is to demonstrate that a technological framework, based on behavioural change theories, applied to mHealth domain, allows improving adherence treatment in diabetic patients. The framework, named Engagement Behavioural Feedback Framework (EBF), is built on top of validated behavioural techniques to frame messages, guide the definition of contents and assess outcomes: elements from the Transtheoretical Model (TTM), the Goal-Setting Theory (GST), Effective Health Communication (EHC) guidelines and Principles of Persuasive Technology (PPT) were incorporated. The TTM helps patients to progress to a next behavioural stage, through specific motivated text messages, and allow clinician’s identifying the current stage and tailor its strategies individually. Moreover, TTM guidelines are adopted to set customised goals at a level appropriate to the patient’s stage of change. The GST was used to build rules to be applied for enhancing educational intervention and weight loss objectives. Finally, the EHC guidelines and the PPT were applied to increase the effectiveness of messages. The EBF aims to support patients on improving their prescription adherence and persuade them towards a general improvement in diabetes self-management, by means of personalised text messages, named Automatic Feedback Messages (AFM). After a first profile screening, consisting in identifying meaningful patient characteristics based on treatment needs, attitudes and health care behaviours, customised AFMs are selected by the system, approved by the professional, and finally transferred into the patient interface. During the treatment, the user collects the data into a Patient Monitoring Device (PMD) from a set of medical devices and from manual inputs. Inputs consist in medication intake, diet and physical activity, metabolic measurement monitoring and learning tasks. Patient general engagement is checked by estimating the usage of the system and the adherence of treatment and patient goals status in the short and the long term period. The Behavioural Analysis Module, consisting in a set of rules and equations, calculates the patient’s behaviour. After behavioural analysis is accomplished, the AFM library and the dispatch setting are updated by the AFM Manager module. Updates include the number, the type and the frequency of messages. The AFMs are periodically supervised by the professional who also participates to the refinement of the treatment, adapted to the current transtheoretical stage. The AFMs are segmented in different categories and levels and patients can adjust message delivery in accordance with their personal needs. The EBF was integrated to the METABO system, designed to facilitate diabetic patients in managing their disease in a less intrusive approach. Patient device corresponds in a mobile platform, while a medical panel interface allows professionals to monitoring the treatment evolution. Specific tools allow professional to check patient adherence and to update the AFMs dispatch management. The EBF was tested in a randomised controlled pilot. The main objective was to examine the feasibility and acceptance of the system. Secondary objectives were also the assessment of the effectiveness of system in terms of adherence improvement, glycaemic control, and quality of life. Participants were recruited from four different clinical centres in Europe. The baseline assessment included demographics, diabetes status, profile information, knowledge about diabetes in general, usage of ICT platforms, opinion and experience about electronic devices and adoption of good practices with diabetes. Acceptance and the effectiveness evaluation criteria were applied to evaluate the performance of the technological framework. The main objective was the assessment of the effectiveness of system in terms of adherence improvement. Fifty-four patients participated on the trials. Twenty-six patients were assigned in the intervention group and equipped with mobile where the EBF was installed: 14 patients were T1DM and 12 were T2DM. The control group was composed of 25 patients that were treated through a standard care, without the usage of the EBF. Professional’s intervention for both intervention and control groups was carried out by 24 care providers, including endocrinologists, nutritionists, and nurses. In order to evaluate the system acceptability and analyse the users’ satisfaction, an online multi-language survey, using LimeSurvey, was produced for both patients and professionals. Results were also collected from the log-files generated in the PMDs, the professional medical panel and the entries of the data base. The messages sent to and from the EBF and the log-files of the system and communication services were recorded over 5 weeks of the study. A total of 2795 messages were submitted, representing an average of 107,50 messages per patient. As demonstrated, messages decrease over time indicating an overall improvement of the care plan’s adherence. As expected, T1DM patients were more loaded with short-term advices, in accordance with their condition. Similarly, being the focus of T2DM on long-term sustainable lifestyle changes, T2DM received more reminders advices, as for diet and physical activity. Favourable outcomes were observed for treatment and usage adherences of the intervention group: for both the adherence indices, results denoted a general improvement on each care plan’s dimension, such as on nutrition and blood glucose input measurements. Further studies were conducted on the change on educational level before and after the trial, measured for both control and intervention groups. The outcomes indicated the intervention group has improved its level of knowledge, while the control group denoted a low decrease. The correlation analysis between the level of adherences and the AFMs showed an improvement in usage adherence for patients who received warnings message, while non-significantly yet even positive indicators related to both treatment and usage adherence correlated with the Reminders. Moreover, the AFMs seemed to help those patients who did not take their treatment seriously enough in the beginning and who were willing to respond to the messages they received. Even though, patients who received too many Warnings, started to consider the message dispatch to be a bit stressful. The research work carried out in developing this research work provides responses to the four research hypothesis that were the motivation for the work: •Hypothesis 1: It is possible to define a set of criteria to measure adherence in diabetic patients. •Hypothesis 2: It is possible to design a technological framework, based on the aforementioned criteria and behavioural change theories, to engage diabetic patients in managing their disease and adhere to care plans. •Hypothesis 3: It is possible to implement the technological framework in the mobile health domain. •Hypothesis 4: It is possible to use the technological framework as a mobile health solution in a real context and have positive effects in terms of diabetes management indicators. The verification of each hypothesis allowed us to provide a response to the main hypothesis: The Main Hypothesis is: It is possible to improve diabetic adherence through a mHealth technological framework based on behavioural change theories. The work carried out to answer these questions is explained in this research work. The framework was developed and applied in the METABO project. METABO is an R&D project, co-funded by the European Commission (METABO 2008) that integrates mobile infrastructure for supporting the monitoring, management, and treatment of type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) patients.

Singularities and undefinitions in the calibration functions of sonic anemometers

A mathematical model of the process employed by a sonic anemometer to build up the measured wind vector in a steady flow is presented to illustrate the way the geometry of these sensors as well as the characteristics of aerodynamic disturbance on the acoustic path can lead to singularities in the transformation function that relates the measured (disturbed) wind vector with the real (corrected) wind vector, impeding the application of correction/calibration functions for some wind conditions. An implicit function theorem allows for the identification of those combinations of real wind conditions and design parameters that lead to undefined correction/ calibration functions. In general, orthogonal path sensors do not show problematic combination of parameters. However, some geometric sonic sensor designs, available in the market, with paths forming smaller angles could lead to undefined correction functions for some levels of aerodynamic disturbances and for certain wind directions. The parameters studied have a strong influence on the existence and number of singularities in the correction/ calibration function as well as on the number of singularities for some combination of parameters. Some conclusions concerning good design practices are included.

Calibration of accelerometers for the measurement of microvibrations

The accelerometers used for the measurement of microvibrations or microgravity applications, such as active control of space structures, attitude control, scientific payloads, or even on-Earth testing of structures at very low-excitation levels, require a dedicated calibration procedure that includes the gravitational effects. Otherwise, on-Earth calibrations can be inaccurate due to the collateral projection of the local gravity onto the sensitive axis. An on-Earth calibration technique for the 107102s amplitude range and 0-100-Hz frequency range is described. Special attention has been given to the modeling of gravitational effects on the response of the calibration device and the accelerometer itself. The sensitivity and resolution tests performed on piezoelectric accelerometers showthe accuracy andthe potential of thistechnique. Typical scale factorun certainty, which hasbeen carefully analyzed, is of the order of 2% at acceleration levels of 10sg.

Multi-textured 3D humanoid reconstruction through passive and active automatic techniques = Reconstrucción de humanoides 3D multi-texturados mediante técnicas pasivas y activas automáticas

The proliferation of video games and other applications of computer graphics in everyday life demands a much easier way to create animatable virtual human characters. Traditionally, this has been the job of highly skilled artists and animators that painstakingly model, rig and animate their avatars, and usually have to tune them for each application and transmission/rendering platform. The emergence of virtual/mixed reality environments also calls for practical and costeffective ways to produce custom models of actual people. The purpose of the present dissertation is bringing 3D human scanning closer to the average user. For this, two different techniques are presented, one passive and one active. The first one is a fully automatic system for generating statically multi-textured avatars of real people captured with several standard cameras. Our system uses a state-of-the-art shape from silhouette technique to retrieve the shape of subject. However, to deal with the lack of detail that is common in the facial region for these kind of techniques, which do not handle concavities correctly, our system proposes an approach to improve the quality of this region. This face enhancement technique uses a generic facial model which is transformed according to the specific facial features of the subject. Moreover, this system features a novel technique for generating view-independent texture atlases computed from the original images. This static multi-texturing system yields a seamless texture atlas calculated by combining the color information from several photos. We suppress the color seams due to image misalignments and irregular lighting conditions that multi-texturing approaches typically suffer from, while minimizing the blurring effect introduced by color blending techniques. The second technique features a system to retrieve a fully animatable 3D model of a human using a commercial depth sensor. Differently to other approaches in the current state of the art, our system does not require the user to be completely still through the scanning process, and neither the depth sensor is moved around the subject to cover all its surface. Instead, the depth sensor remains static and the skeleton tracking information is used to compensate the user’s movements during the scanning stage. RESUMEN La popularización de videojuegos y otras aplicaciones de los gráficos por ordenador en el día a día requiere una manera más sencilla de crear modelos virtuales humanos animables. Tradicionalmente, estos modelos han sido creados por artistas profesionales que cuidadosamente los modelan y animan, y que tienen que adaptar específicamente para cada aplicación y plataforma de transmisión o visualización. La aparición de los entornos de realidad virtual/mixta aumenta incluso más la demanda de técnicas prácticas y baratas para producir modelos 3D representando personas reales. El objetivo de esta tesis es acercar el escaneo de humanos en 3D al usuario medio. Para ello, se presentan dos técnicas diferentes, una pasiva y una activa. La primera es un sistema automático para generar avatares multi-texturizados de personas reales mediante una serie de cámaras comunes. Nuestro sistema usa técnicas del estado del arte basadas en shape from silhouette para extraer la forma del sujeto a escanear. Sin embargo, este tipo de técnicas no gestiona las concavidades correctamente, por lo que nuestro sistema propone una manera de incrementar la calidad en una región del modelo que se ve especialmente afectada: la cara. Esta técnica de mejora facial usa un modelo 3D genérico de una cara y lo modifica según los rasgos faciales específicos del sujeto. Además, el sistema incluye una novedosa técnica para generar un atlas de textura a partir de las imágenes capturadas. Este sistema de multi-texturización consigue un atlas de textura sin transiciones abruptas de color gracias a su manera de mezclar la información de color de varias imágenes sobre cada triángulo. Todas las costuras y discontinuidades de color debidas a las condiciones de iluminación irregulares son eliminadas, minimizando el efecto de desenfoque de la interpolación que normalmente introducen este tipo de métodos. La segunda técnica presenta un sistema para conseguir un modelo humano 3D completamente animable utilizando un sensor de profundidad. A diferencia de otros métodos del estado de arte, nuestro sistema no requiere que el usuario esté completamente quieto durante el proceso de escaneado, ni mover el sensor alrededor del sujeto para cubrir toda su superficie. Por el contrario, el sensor se mantiene estático y el esqueleto virtual de la persona, que se va siguiendo durante el proceso, se utiliza para compensar sus movimientos durante el escaneado.

Automatic segmentation of rivers as a tool for assessing river responses. Case study: the Porma and Curueño rivers, nw Spain.

Automatic segmentation using univariate and multivariate techniques provides more objective and efficient segmentations of the river systems (Alber & Piégay, 2011) and can be complementary to the expert criteria traditionally used (Brenden et al., 2008) INTEREST: A powerful tool to objectively segment the continuity of rivers, which is required for diagnosing problems associated to human impacts OBJECTIVE: To evaluate the potentiality of univariate and multivariate methods in the assessment of river adjustments produced by flow regulation

A fully automatic evolutionary classification of protein folds: Dali Domain Dictionary version 3

The Dali Domain Dictionary (http://www.ebi.ac.uk/dali/domain) is a numerical taxonomy of all known structures in the Protein Data Bank (PDB). The taxonomy is derived fully automatically from measurements of structural, functional and sequence similarities. Here, we report the extension of the classification to match the traditional four hierarchical levels corresponding to: (i) supersecondary structural motifs (attractors in fold space), (ii) the topology of globular domains (fold types), (iii) remote homologues (functional families) and (iv) homologues with sequence identity above 25% (sequence families). The computational definitions of attractors and functional families are new. In September 2000, the Dali classification contained 10 531 PDB entries comprising 17 101 chains, which were partitioned into five attractor regions, 1375 fold types, 2582 functional families and 3724 domain sequence families. Sequence families were further associated with 99 582 unique homologous sequences in the HSSP database, which increases the number of effectively known structures several-fold. The resulting database contains the description of protein domain architecture, the definition of structural neighbours around each known structure, the definition of structurally conserved cores and a comprehensive library of explicit multiple alignments of distantly related protein families.

Substitution rate calibration of small subunit ribosomal RNA identifies chlorarachniophyte endosymbionts as remnants of green algae.

Chlorarachniophytes are amoeboid algae with chlorophyll a and b containing plastids that are surrounded by four membranes instead of two as in plants and green algae. These extra membranes form important support for the hypothesis that chlorarachniophytes have acquired their plastids by the ingestion of another eukaryotic plastid-containing alga. Chlorarachniophytes also contain a small nucleus-like structure called the nucleomorph situated between the two inner and the two outer membranes surrounding the plastid. This nucleomorph is a remnant of the endosymbiont's nucleus and encodes, among other molecules, small subunit ribosomal RNA. Previous phylogenetic analyses on the basis of this molecule provided unexpected and contradictory evidence for the origin of the chlorarachniophyte endosymbiont. We developed a new method for measuring the substitution rates of the individual nucleotides of small subunit ribosomal RNA. From the resulting substitution rate distribution, we derived an equation that gives a more realistic relationship between sequence dissimilarity and evolutionary distance than equations previously available. Phylogenetic trees constructed on the basis of evolutionary distances computed by this new method clearly situate the chlorarachniophyte nucleomorphs among the green algae. Moreover, this relationship is confirmed by transversion analysis of the Chlorarachnion plastid small subunit ribosomal RNA.

Aladdin's Magic Lamp: Developing Methods for Calibration and Geolocation Accuracy Assessment of the DMSP OLS

Nighttime satellite imagery from the Defense Meteorological Satellite Program (DMSP) Operational Linescan System (OLS) has a unique capability to observe nocturnal light emissions from sources including cities, wild fires, and gas flares. Data from the DMSP OLS is used in a wide range of studies including mapping urban areas, estimating informal economies, and estimating urban populations. Given the extensive and increasing list of applications a repeatable method for assessing geolocation accuracy, performing inter-calibration, and defining the minimum detectable brightness would be beneficial. An array of portable lights was designed and taken to multiple field sites known to have no other light sources. The lights were operated during nighttime overpasses by the DMSP OLS and observed in the imagery. A first estimate of the minimum detectable brightness is presented based on the field experiments conducted. An assessment of the geolocation accuracy was performed by measuring the distance between the GPS measured location of the lights and the observed location in the imagery. A systematic shift was observed and the mean distance was measured at 2.9km. A method for in situ radiance calibration of the DMSP OLS using a ground based light source as an active target is presented. The wattage of light used by the active target strongly correlates with the signal measured by the DMSP OLS. This approach can be used to enhance our ability to make inter-temporal and inter-satellite comparisons of DMSP OLS imagery. Exploring the possibility of establishing a permanent active target for the calibration of nocturnal imaging systems is recommended. The methods used to assess the minimum detectable brightness, assess the geolocation accuracy, and build inter-calibration models lay the ground work for assessing the energy expended on light emitted into the sky at night. An estimate of the total energy consumed to light the night sky globally is presented.

Análise de incerteza de um sistema automatizado para ensaios de aspersores de irrigação

Ensaios de distribuição de água de aspersores são convencionalmente realizados manualmente, requerendo tempo e mão de obra treinada. A automação desses ensaios proporciona redução da demanda por esses recursos e apresenta potencial para minimizar falhas e/ou desvios de procedimento. Atualmente, laboratórios de ensaio e calibração acreditados junto a organismos legais devem apresentar em seus relatórios a incerteza de medição de seus instrumentos e sistemas de medição. Além disso, normas de ensaio e calibração apresentam especificação de incerteza aceitável, como a norma de ensaios de distribuição de água por aspersores, ISO 15886-3 (2012), a qual exige uma incerteza expandida de até 3% em 80% dos coletores. Os objetivos deste trabalho foram desenvolver um sistema automatizado para os ensaios de aspersores em laboratório e realizar a análise de incerteza de medição, para sua quantificação nos resultados de ensaio e para dar suporte ao dimensionamento dos tubos de coleta. O sistema automático foi constituído por um subsistema de gerenciamento, por meio de um aplicativo supervisório, um de pressurização e um de coleta, por meio de módulos eletrônicos microprocessados desenvolvidos. De acordo com instruções do sistema de gerenciamento o sistema de pressurização ajustava a pressão no aspersor por meio do controle da rotação da motobomba, e o sistema de coleta realizava a medição da intensidade de precipitação de água ao longo do raio de alcance do aspersor. A água captada por cada coletor drenava para um tubo de coleta, que estava conectado a uma das válvulas solenoides de um conjunto, onde havia um transmissor de pressão. Cada válvula era acionada individualmente numa sequência para a medição do nível de água em cada tubo de coleta, por meio do transmissor. Por meio das análises realizadas, as menores incertezas foram obtidas para os menores diâmetros de tubo de coleta, sendo que se deve utilizar o menor diâmetro possível. Quanto ao tempo de coleta, houve redução da incerteza de medição ao se aumentar a duração, devendo haver um tempo mínimo para se atingir a incerteza-alvo. Apesar de cada intensidade requer um tempo mínimo para garantir a incerteza, a diferença mínima de nível a ser medida foi a mesma. Portanto, para os ensaios visando atender a incerteza, realizou-se o monitoramento da diferença de nível nos tubos, ou diferença de nível, facilitando a realização do ensaio. Outra condição de ensaio considerou um tempo de coleta para 30 voltas do aspersor, também exigido pela norma ISO 15886-3 (2012). A terceira condição considerou 1 h de coleta, como tradicionalmente realizado. As curvas de distribuição de água obtidas por meio do sistema desenvolvido foram semelhantes às obtidas em ensaios convencionais, para as três situações avaliadas. Para tempos de coleta de 1 h ou 30 voltas do aspersor o sistema automático requereu menos tempo total de ensaio que o ensaio convencional. Entretanto, o sistema desenvolvido demandou mais tempo para atingir a incerteza-alvo, o que é uma limitação, mesmo sendo automatizado. De qualquer forma, o sistema necessitava apenas que um técnico informasse os parâmetros de ensaio e o acionasse, possibilitando que o mesmo alocasse seu tempo em outras atividades.

Automatic extraction of syntactic semantic patterns for multilingual resources

In this paper we present an automatic system for the extraction of syntactic semantic patterns applied to the development of multilingual processing tools. In order to achieve optimum methods for the automatic treatment of more than one language, we propose the use of syntactic semantic patterns. These patterns are formed by a verbal head and the main arguments, and they are aligned among languages. In this paper we present an automatic system for the extraction and alignment of syntactic semantic patterns from two manually annotated corpora, and evaluate the main linguistic problems that we must deal with in the alignment process.

Machine learning techniques for automatic opinion detection in non-traditional textual genres

This paper presents a preliminary study in which Machine Learning experiments applied to Opinion Mining in blogs have been carried out. We created and annotated a blog corpus in Spanish using EmotiBlog. We evaluated the utility of the features labelled firstly carrying out experiments with combinations of them and secondly using the feature selection techniques, we also deal with several problems, such as the noisy character of the input texts, the small size of the training set, the granularity of the annotation scheme and the language object of our study, Spanish, with less resource than English. We obtained promising results considering that it is a preliminary study.

Calibration of the length of a chain of single gold atoms

Using a scanning tunnelling microscope or mechanically controllable break junction it has been shown that it is possible to control the formation of a wire made of single gold atoms. In these experiments an interatomic distance between atoms in the chain of ∼3.6 Å was reported which is not consistent with recent theoretical calculations. Here, using precise calibration procedures for both techniques, we measure the length of the atomic chains. Based on the distance between the peaks observed in the chain length histogram we find the mean value of the interatomic distance before chain rupture to be 2.5±0.2 Å. This value agrees with the theoretical calculations for the bond length. The discrepancy with the previous experimental measurements was due to the presence of He gas, that was used to promote the thermal contact, and which affects the value of the work function that is commonly used to calibrate distances in scanning tunnelling microscopy and mechanically controllable break junctions at low temperatures.

Exploring automatic feature selection for transition-based dependency parsing

En este artículo se investigan técnicas automáticas para encontrar un modelo óptimo de características en el caso de un analizador de dependencias basado en transiciones. Mostramos un estudio comparativo entre algoritmos de búsqueda, sistemas de validación y reglas de decisión demostrando al mismo tiempo que usando nuestros métodos es posible conseguir modelos complejos que proporcionan mejores resultados que los modelos que siguen configuraciones por defecto.

Portable 3D laser-camera calibration system with color fusion for SLAM

Nowadays, the use of RGB-D sensors have focused a lot of research in computer vision and robotics. These kinds of sensors, like Kinect, allow to obtain 3D data together with color information. However, their working range is limited to less than 10 meters, making them useless in some robotics applications, like outdoor mapping. In these environments, 3D lasers, working in ranges of 20-80 meters, are better. But 3D lasers do not usually provide color information. A simple 2D camera can be used to provide color information to the point cloud, but a calibration process between camera and laser must be done. In this paper we present a portable calibration system to calibrate any traditional camera with a 3D laser in order to assign color information to the 3D points obtained. Thus, we can use laser precision and simultaneously make use of color information. Unlike other techniques that make use of a three-dimensional body of known dimensions in the calibration process, this system is highly portable because it makes use of small catadioptrics that can be placed in a simple manner in the environment. We use our calibration system in a 3D mapping system, including Simultaneous Location and Mapping (SLAM), in order to get a 3D colored map which can be used in different tasks. We show that an additional problem arises: 2D cameras information is different when lighting conditions change. So when we merge 3D point clouds from two different views, several points in a given neighborhood could have different color information. A new method for color fusion is presented, obtaining correct colored maps. The system will be tested by applying it to 3D reconstruction.