Statistical Machine Learning for Automatic Assessment of Physical Activity Intensity Using Multi-axial Accelerometry and Heart Rate

This work explores the automatic recognition of physical activity intensity patterns from multi-axial accelerometry and heart rate signals. Data collection was carried out in free-living conditions and in three controlled gymnasium circuits, for a total amount of 179.80 h of data divided into: sedentary situations (65.5%), light-to-moderate activity (17.6%) and vigorous exercise (16.9%). The proposed machine learning algorithms comprise the following steps: time-domain feature definition, standardization and PCA projection, unsupervised clustering (by k-means and GMM) and a HMM to account for long-term temporal trends. Performance was evaluated by 30 runs of a 10-fold cross-validation. Both k-means and GMM-based approaches yielded high overall accuracy (86.97% and 85.03%, respectively) and, given the imbalance of the dataset, meritorious F-measures (up to 77.88%) for non-sedentary cases. Classification errors tended to be concentrated around transients, what constrains their practical impact. Hence, we consider our proposal to be suitable for 24 h-based monitoring of physical activity in ambulatory scenarios and a first step towards intensity-specific energy expenditure estimators

Enhancing Activity Recognition by Fusing Inertial and Biometric Information

Activity recognition is an active research field nowadays, as it enables the development of highly adaptive applications, e.g. in the field of personal health. In this paper, a light high-level fusion algorithm to detect the activity that an individual is performing is presented. The algorithm relies on data gathered from accelerometers placed on different parts of the body, and on biometric sensors. Inertial sensors allow detecting activity by analyzing signal features such as amplitude or peaks. In addition, there is a relationship between the activity intensity and biometric response, which can be considered together with acceleration data to improve the accuracy of activity detection. The proposed algorithm is designed to work with minimum computational cost, being ready to run in a mobile device as part of a context-aware application. In order to enable different user scenarios, the algorithm offers best-effort activity estimation: its quality of estimation depends on the position and number of the available inertial sensors, and also on the presence of biometric information.

Influencia del ejercicio físico desarrollado durante el embarazo en la respuesta cardiaca fetal = Influence of a physical activity program performed during pregnancy on fetal heart response

INTRODUCCIÓN: El riesgo de padecer enfermedades cardiovasculares y los índices de obesidad infantil han ido en aumento durante los últimos años empobreciendo la salud de la población. La Teoría de Barker relaciona el estado de salud de la madre con el desarrollo fetal, asociando a un deficiente estado físico y hábitos de vida negativos de la mujer embarazada con el aumento del riesgo de padecer cardiopatías en la infancia y adolescencia, así como predisponer al recién nacido a padecer sobrepeso y/u obesidad en su vida posterior. Por otro lado los estudios efectuados sobre ejercicio físico durante el embarazo reportan beneficios para salud materna y fetal. Uno de los parámetros más utilizados para comprobar la salud fetal es su frecuencia cardiaca, mediante la que se comprueba el buen desarrollo del sistema nervioso autónomo. Si se observa este parámetro en presencia de ejercicio materno podría encontrarse una respuesta crónica del corazón fetal al ejercicio materno como consecuencia de una adaptación y mejora en el funcionamiento del sistema nervioso autónomo del feto. De esta forma podría mejorar su salud cardiovascular intrauterina, lo que podría mantenerse en su vida posterior descendiendo el riesgo de padecer enfermedades cardiovasculares en la edad adulta. OBJETIVOS: Conocer la influencia de un programa de ejercicio físico supervisado en la frecuencia cardiaca fetal (FCF) en reposo y después del ejercicio materno en relación con gestantes sedentarias mediante la realización de un protocolo específico. Conocer la influencia de un programa de ejercicio físico en el desarrollo del sistema nervioso autónomo fetal, relacionado con el tiempo de recuperación de la FCF. MATERIAL Y MÉTODO: Se diseñó un ensayo clínico aleatorizado multicéntrico en el que participaron 81 gestantes (GC=38, GE=43). El estudio fue aprobado por el comité ético de los hospitales que participaron en el estudio. Todas las gestantes fueron informadas y firmaron un consentimiento para su participación en el estudio. Las participantes del GE recibieron una intervención basada en un programa de ejercicio físico desarrollado durante la gestación (12-36 semanas de gestación) con una frecuencia de tres veces por semana. Todas las gestantes realizaron un protocolo de medida de la FCF entre las semanas 34-36 de gestación. Dicho protocolo consistía en dos test llevados a cabo caminando a diferentes intensidades (40% y 60% de la frecuencia cardiaca de reserva). De este protocolo se obtuvieron las principales variables de estudio: FCF en reposo, FCF posejercicio al 40 y al 60% de intensidad, tiempo de recuperación de la frecuencia cardiaca fetal en ambos esfuerzos. El material utilizado para la realización del protocolo fue un monitor de frecuencia cardiaca para controlar la frecuencia cardiaca de la gestante y un monitor fetal inalámbrico (telemetría fetal) para registrar el latido fetal durante todo el protocolo. RESULTADOS: No se encontraron diferencias estadísticamente significativas en la FCF en reposo entre grupos (GE=140,88 lat/min vs GC= 141,95 lat/min; p>,05). Se encontraron diferencias estadísticamente significativas en el tiempo de recuperación de la FCF entre los fetos de ambos grupos (GE=135,65 s vs GC=426,11 s esfuerzo al 40%; p<,001); (GE=180,26 s vs GC=565,61 s esfuerzo al 60%; p<,001). Se encontraron diferencias estadísticamente significativas en la FCF posejercicio al 40% (GE=139,93 lat/min vs GC=147,87 lat/min; p<,01). No se encontraron diferencias estadísticamente significativas en la FCF posejercicio al 60% (GE=143,74 lat/min vs GC=148,08 lat/min; p>,05). CONLUSIÓN: El programa de ejercicio físico desarrollado durante la gestación influyó sobre el corazón fetal de los fetos de las gestantes del GE en relación con el tiempo de recuperación de la FCF. Los resultados muestran un posible mejor funcionamiento del sistema nervioso autónomo en fetos de gestantes activas durante el embarazo. ABSTRACT INTRODUCTION: The risk to suffer cardiovascular diseases and childhood obesity index has grown in the last years worsening the health around the population. Barker´s Theory related maternal health with fetal development establishing an association between a poorly physical state and an unhealthy lifestyle in the pregnant woman with the risk to suffer heart disease during childhood and adolescence, childhood overweight and/or obese is related to maternal lifestyle. By the other way researches carried out about physical exercise and pregnancy show benefits in maternal and fetal health. One of the most studied parameters to check fetal health is its heart rate, correct fetal autonomic nervous system development and work is also corroborated by fetal heart rate. Looking at this parameter during maternal exercise a chronic response of fetal heart could be found due to an adaptation and improvement in the working of the autonomic nervous system. Therefore its cardiovascular health could be enhanced during its intrauterine life and maybe it could be maintained in its posterior life descending the risk to suffer cardiovascular diseases in adult life. OBJECTIVES: To know the influence of a supervised physical activity program in the fetal heart rate (FHR) at rest, FHR after maternal exercise related to sedentary pregnant women by a FHR assessment protocol. To know the influence of a physical activity program in the development of the autonomic nervous system related to FHR recovery time. MATERIAL AND METHOD: A multicentric randomized clinical trial was design in which 81 pregnant women participated (CG=38, EG=43). The study was approved by the ethics committee of all of the hospitals participating in the study. All of the participants signed an informed consent for their participation in the study. EG participants received an intervention based on a physical activity program carried out during gestation (12-36 gestation weeks) with a three days a week frequency. All of the participants were tested between 34-36 weeks of gestation by a specific FHR assessment protocol. The mentioned protocol consisted in two test performed walking and at a two different intensities (40% and 60% of the reserve heart rate). From this protocol we obtained the main research variables: FHR at rest, FHR post-exercise at 40% and 60% intensity, and FHR recovery time at both walking test. The material used to perform the protocol were a FH monitor to check maternal HR and a wireless fetal monitor (Telemetry) to register fetal beats during the whole protocol. RESULTS: There were no statistical differences in FHR at rest between groups (EG=140,88 beats/min vs CG= 141,95 beats/min; p>,05). There were statistical differences in FHR recovery time in both walking tests between groups (EG=135,65 s vs CG=426,11 s test at 40% intensity; p<,001); (EG=180,26 s vs CG=565,61 s test at 60% intensity; p<,001). Statistical differences were found in FHR post-exercise at 40% intensity between groups (EG=139,93 beats/min vs CG=147,87 beats/min; p<,01). No statistical differences were found in FHR at rest post-exercise at 60% intensity between groups (EG=143,74 beats/min vs CG=148,08 beats/min; p>,05). CONCLUSIONS: The physical activity program performed during gestation had an influence in fetal heart of the fetus from mother in the EG related to FHR recovery time. These results show a possible enhancement on autonomic nervous system working in fetus from active mothers during gestation.

V-substituted In2S3: an intermediate band material with photocatalytic activity in the whole visible light range

We proposed in our previous work V-substituted In2S3 as an intermediate band (IB) material able to enhance the efficiency of photovoltaic cells by combining two photons to achieve a higher energy electron excitation, much like natural photosynthesis. Here this hyper-doped material is tested in a photocatalytic reaction using wavelength-controlled light. The results evidence its ability to use photons with wavelengths of up to 750 nm, i.e. with energy significantly lower than the bandgap (=2.0 eV) of non-substituted In2S3, driving with them the photocatalytic reaction at rates comparable to those of non-substituted In2S3 in its photoactivity range (λ ≤ 650 nm). Photoluminescence spectra evidence that the same bandgap excitation as in V-free In2S3 occurs in V-substituted In2S3 upon illumination with photons in the same sub-bandgap energy range which is effective in photocatalysis, and its linear dependence on light intensity proves that this is not due to a nonlinear optical property. This evidences for the first time that a two-photon process can be active in photocatalysis in a single-phase material. Quantum calculations using GW-type many-body perturbation theory suggest that the new band introduced in the In2S3 gap by V insertion is located closer to the conduction band than to the valence band, so that hot carriers produced by the two-photon process would be of electron type; they also show that the absorption coefficients of both transitions involving the IB are of significant and similar magnitude. The results imply that V-substituted In2S3, besides being photocatalytically active in the whole visible light range (a property which could be used for the production of solar fuels), could make possible photovoltaic cells of improved efficiency.

Automatic physical activity pattern recognition and meal intake modelling in type 1 diabetes for their inclusion in artificial pancreas systems

La diabetes comprende un conjunto de enfermedades metabólicas que se caracterizan por concentraciones de glucosa en sangre anormalmente altas. En el caso de la diabetes tipo 1 (T1D, por sus siglas en inglés), esta situación es debida a una ausencia total de secreción endógena de insulina, lo que impide a la mayoría de tejidos usar la glucosa. En tales circunstancias, se hace necesario el suministro exógeno de insulina para preservar la vida del paciente; no obstante, siempre con la precaución de evitar caídas agudas de la glucemia por debajo de los niveles recomendados de seguridad. Además de la administración de insulina, las ingestas y la actividad física son factores fundamentales que influyen en la homeostasis de la glucosa. En consecuencia, una gestión apropiada de la T1D debería incorporar estos dos fenómenos fisiológicos, en base a una identificación y un modelado apropiado de los mismos y de sus sorrespondientes efectos en el balance glucosa-insulina. En particular, los sistemas de páncreas artificial –ideados para llevar a cabo un control automático de los niveles de glucemia del paciente– podrían beneficiarse de la integración de esta clase de información. La primera parte de esta tesis doctoral cubre la caracterización del efecto agudo de la actividad física en los perfiles de glucosa. Con este objetivo se ha llevado a cabo una revisión sistemática de la literatura y meta-análisis que determinen las respuestas ante varias modalidades de ejercicio para pacientes con T1D, abordando esta caracterización mediante unas magnitudes que cuantifican las tasas de cambio en la glucemia a lo largo del tiempo. Por otro lado, una identificación fiable de los periodos con actividad física es un requisito imprescindible para poder proveer de esa información a los sistemas de páncreas artificial en condiciones libres y ambulatorias. Por esta razón, la segunda parte de esta tesis está enfocada a la propuesta y evaluación de un sistema automático diseñado para reconocer periodos de actividad física, clasificando su nivel de intensidad (ligera, moderada o vigorosa); así como, en el caso de periodos vigorosos, identificando también la modalidad de ejercicio (aeróbica, mixta o de fuerza). En este sentido, ambos aspectos tienen una influencia específica en el mecanismo metabólico que suministra la energía para llevar a cabo el ejercicio y, por tanto, en las respuestas glucémicas en T1D. En este trabajo se aplican varias combinaciones de técnicas de aprendizaje máquina y reconocimiento de patrones sobre la fusión multimodal de señales de acelerometría y ritmo cardíaco, las cuales describen tanto aspectos mecánicos del movimiento como la respuesta fisiológica del sistema cardiovascular ante el ejercicio. Después del reconocimiento de patrones se incorpora también un módulo de filtrado temporal para sacar partido a la considerable coherencia temporal presente en los datos, una redundancia que se origina en el hecho de que en la práctica, las tendencias en cuanto a actividad física suelen mantenerse estables a lo largo de cierto tiempo, sin fluctuaciones rápidas y repetitivas. El tercer bloque de esta tesis doctoral aborda el tema de las ingestas en el ámbito de la T1D. En concreto, se propone una serie de modelos compartimentales y se evalúan éstos en función de su capacidad para describir matemáticamente el efecto remoto de las concetraciones plasmáticas de insulina exógena sobre las tasas de eleiminación de la glucosa atribuible a la ingesta; un aspecto hasta ahora no incorporado en los principales modelos de paciente para T1D existentes en la literatura. Los datos aquí utilizados se obtuvieron gracias a un experimento realizado por el Institute of Metabolic Science (Universidad de Cambridge, Reino Unido) con 16 pacientes jóvenes. En el experimento, de tipo ‘clamp’ con objetivo variable, se replicaron los perfiles individuales de glucosa, según lo observado durante una visita preliminar tras la ingesta de una cena con o bien alta carga glucémica, o bien baja. Los seis modelos mecanísticos evaluados constaban de: a) submodelos de doble compartimento para las masas de trazadores de glucosa, b) un submodelo de único compartimento para reflejar el efecto remoto de la insulina, c) dos tipos de activación de este mismo efecto remoto (bien lineal, bien con un punto de corte), y d) diversas condiciones iniciales. ABSTRACT Diabetes encompasses a series of metabolic diseases characterized by abnormally high blood glucose concentrations. In the case of type 1 diabetes (T1D), this situation is caused by a total absence of endogenous insulin secretion, which impedes the use of glucose by most tissues. In these circumstances, exogenous insulin supplies are necessary to maintain patient’s life; although caution is always needed to avoid acute decays in glycaemia below safe levels. In addition to insulin administrations, meal intakes and physical activity are fundamental factors influencing glucose homoeostasis. Consequently, a successful management of T1D should incorporate these two physiological phenomena, based on an appropriate identification and modelling of these events and their corresponding effect on the glucose-insulin balance. In particular, artificial pancreas systems –designed to perform an automated control of patient’s glycaemia levels– may benefit from the integration of this type of information. The first part of this PhD thesis covers the characterization of the acute effect of physical activity on glucose profiles. With this aim, a systematic review of literature and metaanalyses are conduced to determine responses to various exercise modalities in patients with T1D, assessed via rates-of-change magnitudes to quantify temporal variations in glycaemia. On the other hand, a reliable identification of physical activity periods is an essential prerequisite to feed artificial pancreas systems with information concerning exercise in ambulatory, free-living conditions. For this reason, the second part of this thesis focuses on the proposal and evaluation of an automatic system devised to recognize physical activity, classifying its intensity level (light, moderate or vigorous) and for vigorous periods, identifying also its exercise modality (aerobic, mixed or resistance); since both aspects have a distinctive influence on the predominant metabolic pathway involved in fuelling exercise, and therefore, in the glycaemic responses in T1D. Various combinations of machine learning and pattern recognition techniques are applied on the fusion of multi-modal signal sources, namely: accelerometry and heart rate measurements, which describe both mechanical aspects of movement and the physiological response of the cardiovascular system to exercise. An additional temporal filtering module is incorporated after recognition in order to exploit the considerable temporal coherence (i.e. redundancy) present in data, which stems from the fact that in practice, physical activity trends are often maintained stable along time, instead of fluctuating rapid and repeatedly. The third block of this PhD thesis addresses meal intakes in the context of T1D. In particular, a number of compartmental models are proposed and compared in terms of their ability to describe mathematically the remote effect of exogenous plasma insulin concentrations on the disposal rates of meal-attributable glucose, an aspect which had not yet been incorporated to the prevailing T1D patient models in literature. Data were acquired in an experiment conduced at the Institute of Metabolic Science (University of Cambridge, UK) on 16 young patients. A variable-target glucose clamp replicated their individual glucose profiles, observed during a preliminary visit after ingesting either a high glycaemic-load or a low glycaemic-load evening meal. The six mechanistic models under evaluation here comprised: a) two-compartmental submodels for glucose tracer masses, b) a single-compartmental submodel for insulin’s remote effect, c) two types of activations for this remote effect (either linear or with a ‘cut-off’ point), and d) diverse forms of initial conditions.