Linear or non-linear multivariate calibration models? That is the question.

Concepts from data science, machine learning, deep learning and artificial neural networks are spreading in many disciplines. The general idea is to exploit the power of statistical tools to interpret complex and, in many cases, non-linear data. Specifically in analytical chemistry, many chemometrics tools are being developed. However, they tend to get more complex without necessarily improving the prediction ability, which conspires against parsimony. In this report, we show how non-linear analytical data sets can be solved with equal or better efficiency by easily interpretable modified linear models, based on the concept of local sample selection before model building. The latter activity is conducted by choosing a sub-set of samples located in the neighborhood of each unknown sample in the space spanned by the latent variables. Two experimental examples related to the use of near infrared spectroscopy for the analysis of target properties in food samples are examined. The comparison with seemingly more complex chemometric models reveals that local regression is able to achieve similar analytical performance, with considerably less computational burden.

Diagnóstico de la dinámica cardiaca mediante proporciones de la entropía / Heart dynamics diagnosis based on entropy proportions

Introducción: Los valores de la frecuencia cardiaca normales y anormales registrados en los equipos electrocardiográficos ambulatorios o continuos han sido caracterizados mediante nuevas metodologías diagnósticas, las cuales se han desarrollado en el contexto de la teoría de los sistemas dinámicos y la entropía. Objetivo: Evaluar la dinámica cardiaca de adultos, teniendo en cuenta su comportamiento en el contexto de la teoría de sistemas dinámicos y las proporciones de la entropía mediante un estudio ciego. Métodos: Se realizó una prueba diagnóstica mediante un estudio ciego de 500 Holter, aplicando una nueva metodología basada en las proporciones de la entropía del atractor numérico construido con los valores registrados en el Holter. Para esto, los valores máximos y mínimos de la frecuencia cardiaca para cada hora, así como el número de latidos, fueron tomados de cada Holter durante, mínimo, 18 horas. Con estos valores se generó un atractor numérico que cuantificó la probabilidad de parejas de frecuencias cardiacas consecutivas. Se evaluó cada dinámica a partir de los valores de entropía y de sus proporciones. Posteriormente, estos resultados fueron comparados con la evaluación clínica convencional estimando la sensibilidad y especificidad, así como el coeficiente Kappa. Resultados: Se encontraron diferencias entre la dinámica de casos normales y anormales, en las dinámicas cardiacas evaluadas en 18 horas, hallando valores de sensibilidad y especificidad del 100 por ciento y coeficiente Kappa de uno, respecto al diagnóstico clínico convencional. Conclusiones: Los valores de entropía y de sus proporciones permiten diferenciar cuantitativamente la normalidad de la enfermedad en la dinámica cardiaca, durante un mínimo de 18 horas(AU)

Introduction: The normal and abnormal heart rate values recorded on ambulatory or continuous electrocardiographic devices have been characterized with novel diagnostic methodologies developed in the context of dynamic systems and entropy theory. Objective: Through a blind study, evaluate the heart dynamics of adults taking into account their behavior in the context of dynamic systems theory and entropy proportions. Methods: A diagnostic test was conducted through a 500 Holter blind study, applying a novel methodology based on the entropy proportions of the numerical attractor constructed with the values registered on the Holter device. To achieve this end, maximum and minimum heart rate values for each hour, as well as the number of beats, were obtained from each Holter device for at least 18 hours. Based on these values, a numerical attractor was generated which quantified the probability of consecutive heart rate pairs. Each dynamic was evaluated in terms of entropy values and their proportions. These results were then compared with the conventional clinical evaluation, estimating the sensitivity and specificity as well as the kappa coefficient. Results: Differences were found between the dynamics of normal and abnormal cases, in the heart dynamics evaluated in 18 hours, finding sensitivity and specificity values of 100 percent and a kappa coefficient of 1, with respect to conventional clinical diagnosis. Conclusions: Entropy values and their proportions make it possible to quantitatively differentiate the normality of the disease in heart dynamics for a minimum of 18 hours(AU)

A novel strategy for dynamic identification in AC/DC microgrids based on ARX and Petri Nets.

This paper presents a new hybrid strategy which allows the dynamic identification of AC/DC microgrids (MG) by using algorithms such as Auto-Regressive with exogenous inputs (ARX) and Petri Nets (PN). The proposed strategy demonstrated in this study serves to obtain a dynamic model of the DC MG in isolated or connected modes. Given the non-linear nature of the system under study, the methodology divides the whole system in a bank of linearized models at different stable operating points, coordinated by a PN state machine. The bank of models obtained in state space, together with an adequate selection of models, can capture and reflect the non-linear dynamic properties of the AD/DC MGs and the different systems that it composes. The performance of the proposed algorithm has been tested using the Matlab/Simulink simulation platform.

Impedance Control for Robotic Rehabilitation: A Robust Markovian Approach.

The human-robot interaction has played an important role in rehabilitation robotics and impedance control has been used in the regulation of interaction forces between the robot actuator and human limbs. Series elastic actuators (SEAs) have been an efficient solution in the design of this kind of robotic application. Standard implementations of impedance control with SEAs require an internal force control loop for guaranteeing the desired impedance output. However, nonlinearities and uncertainties hamper such a guarantee of an accurate force level in this human-robot interaction. This paper addresses the dependence of the impedance control performance on the force control and proposes a control approach that improves the force control robustness. A unified model of the human-robot system that considers the ankle impedance by a second-order dynamics subject to uncertainties in the stiffness, damping, and inertia parameters has been developed. Fixed, resistive, and passive operation modes of the robotics system were defined, where transition probabilities among the modes were modeled through a Markov chain. A robust regulator for Markovian jump linear systems was used in the design of the force control. Experimental results show the approach improves the impedance control performance. For comparison purposes, a standard [Formula: see text] force controller based on the fixed operation mode has also been designed. The Markovian control approach outperformed the [Formula: see text] control when all operation modes were taken into account.