HIGEA: An Intelligent Conversational Agent to Detect Caregiver Burden.

Mental health disorders increasingly affect people worldwide. As a consequence, more families and relatives find themselves acting as caregivers. Most often, these are untrained people who experience loneliness, abandonment, and often develop signs of depression (i.e., caregiver burden syndrome). In this work, we present HIGEA, a digital system based on a conversational agent to help to detect caregiver burden. The conversational agent naturally embeds psychological test questions into informal conversations, which aim at increasing the adherence of use and avoiding user bias. A proof-of-concept is developed based on the popular Zarit Test, which is widely used to assess caregiver burden. Preliminary results show the system is useful and effective.

Optimization of Deep Architectures for EEG Signal Classification: An AutoML Approach Using Evolutionary Algorithms.

Electroencephalography (EEG) signal classification is a challenging task due to the low signal-to-noise ratio and the usual presence of artifacts from different sources. Different classification techniques, which are usually based on a predefined set of features extracted from the EEG band power distribution profile, have been previously proposed. However, the classification of EEG still remains a challenge, depending on the experimental conditions and the responses to be captured. In this context, the use of deep neural networks offers new opportunities to improve the classification performance without the use of a predefined set of features. Nevertheless, Deep Learning architectures include a vast number of hyperparameters on which the performance of the model relies. In this paper, we propose a method for optimizing Deep Learning models, not only the hyperparameters, but also their structure, which is able to propose solutions that consist of different architectures due to different layer combinations. The experimental results corroborate that deep architectures optimized by our method outperform the baseline approaches and result in computationally efficient models. Moreover, we demonstrate that optimized architectures improve the energy efficiency with respect to the baseline models.

COVID-19 y embarazo: caso clínico de presentación crítica, inflamación placentaria y transmisión vertical fetal demostrada / COVID-19 and pregnancy: clinical case of critical presentation, placental inflammation and demonstrated fetal vertical transmission

ANTECEDENTES: La pandemia global de COVID-19 llega al continente americano en marzo del año 2020 y en menos de dos meses reúne a más de la mitad de los casos a nivel global. OBJETIVO: Caso clínico de una mujer embarazada con una presentación crítica de COVID-19 y embarazo a las 25 semanas de gestación, en el contexto del peak de la pandemia en Chile en el otoño del año 2020. CASO CLÍNICO: El 20 de junio de 2020, una mujer de 34 años, con 25 semanas de embarazo, es trasladada desde Hospital de San Bernardo a Clínica Las Condes en Santiago, Chile, con un cuadro de 10 días de evolución de COVID-19, que evoluciona a una situación crítica con insuficiencia respiratoria severa. Ingresa a unidad de cuidados intensivos para ventilación mecánica. Las imágenes de radiología simple y de tomografía axial computarizada de tórax demuestran una neumopatía bilateral con imágenes características opacidades en vidrio esmerilado, asociado a engrosamiento intersticial, imágenes descritas previamente como características para COVID-19. La paciente permanece en unidad de cuidados intensivos en ventilación mecánica por siete días, con evolución favorable posterior, mejoría del cuadro séptico y alta después de 22 días de hospitalización. El parto ocurre en forma espontánea a las 38 semanas, la madre y el recién nacido evolucionan en buen estado general. El examen histopatológico placentario demuestra compromiso inflamatorio vellositario y los exámenes de anticuerpos en sangre del recién nacido demuestran la presencia de anticuerpos del tipo IgG e IgM. Se trata de uno de los pocos casos demostrados reportados de transmisión transplacentaria vía sanguínea de SARS-CoV-2 de la madre al recién nacido.

BACKGROUND: The global COVID-19 pandemic reaches the American continent in March 2020 and in less than two months it brings together more than half of the cases globally.OBJECTIVE: The clinical case of a 25-week pregnant woman with a critical presentation of COVID-19 and pregnancy at 25 weeks of gestation, is presented in the context of the peak of the pandemic in Chile in the fall of 2020. CLINICAL CASE: On June 20, 2020, a 34-year-old woman, 25 weeks pregnant, is transferred from Hospital de San Bernardo to Clinica Las Condes in Santiago, Chile, with a ten-day evolution of a COVID-19 that evolves to critical with severe respiratory failure. She is admitted to the intensive care unit for mechanical ventilation. Chest computerized axial tomography images demonstrate bilateral pneumopathy with characteristic images of ground-glass opacities, associated with interstitial thickening, images previously described as characteristics for COVID-19. The patient remains in the intensive care unit on mechanical ventilation for seven days, with subsequent favorable evolution, improvement of the septic condition, and discharge after 22 days of hospitalization. Delivery occurs at 38 weeks, the mother and the newborn evolve in good general condition. The placental histopathological examination demonstrates villous inflammatory involvement, and the newborn's blood tests show the presence of IgG and IgM antibodies. It is one of the few reported cases of transplacental transmission of SARS-CoV-2 from the mother to the newborn.

JMAC Protocol: A Cross-Layer Multi-Hop Protocol for LoRa.

The emergence of Low-Power Wide-Area Network (LPWAN) technologies allowed the development of revolutionary Internet Of Things (IoT) applications covering large areas with thousands of devices. However, connectivity may be a challenge for non-line-of-sight indoor operation or for areas without good coverage. Technologies such as LoRa and Sigfox allow connectivity for up to 50,000 devices per cell, several devices that may be exceeded in many scenarios. To deal with these problems, this paper introduces a new multi-hop protocol, called JMAC, designed for improving long range wireless communication networks that may support monitoring in scenarios such smart cities or Industry 4.0. JMAC uses the LoRa radio technology to keep low consumption and extend coverage area, and exploits the potential mesh behaviour of wireless networks to improve coverage and increase the number of supported devices per cell. JMAC is based on predictive wake-up to reach long lifetime on sensor devices. Our proposal was validated using the OMNeT++ simulator to analyze how it performs under different conditions with promising results.

Deep learning for EEG-based Motor Imagery classification: Accuracy-cost trade-off.

Electroencephalography (EEG) datasets are often small and high dimensional, owing to cumbersome recording processes. In these conditions, powerful machine learning techniques are essential to deal with the large amount of information and overcome the curse of dimensionality. Artificial Neural Networks (ANNs) have achieved promising performance in EEG-based Brain-Computer Interface (BCI) applications, but they involve computationally intensive training algorithms and hyperparameter optimization methods. Thus, an awareness of the quality-cost trade-off, although usually overlooked, is highly beneficial. In this paper, we apply a hyperparameter optimization procedure based on Genetic Algorithms to Convolutional Neural Networks (CNNs), Feed-Forward Neural Networks (FFNNs), and Recurrent Neural Networks (RNNs), all of them purposely shallow. We compare their relative quality and energy-time cost, but we also analyze the variability in the structural complexity of networks of the same type with similar accuracies. The experimental results show that the optimization procedure improves accuracy in all models, and that CNN models with only one hidden convolutional layer can equal or slightly outperform a 6-layer Deep Belief Network. FFNN and RNN were not able to reach the same quality, although the cost was significantly lower. The results also highlight the fact that size within the same type of network is not necessarily correlated with accuracy, as smaller models can and do match, or even surpass, bigger ones in performance. In this regard, overfitting is likely a contributing factor since deep learning approaches struggle with limited training examples.

A Power-Performance Perspective to Multiobjective Electroencephalogram Feature Selection on Heterogeneous Parallel Platforms.

This article provides an insight on the power-performance issues related with the CPU-GPU (Central Processing Unit-Graphics Processing Unit) parallel implementations of problems that frequently appear in the context of applications on bioinformatics and biomedical engineering. More specifically, we analyze the power-performance behavior of an evolutionary parallel multiobjective electroencephalogram feature selection procedure that evolves subpopulations of solutions with time-demanding fitness evaluation. The procedure has been implemented in OpenMP to dynamically distribute either subpopulations or individuals among devices, and uses OpenCL to evaluate the fitness of the individuals. The development of parallel codes usually implies to maximize the code efficiency, thus optimizing the achieved speedups. To follow the same trend, this article extends and provides a more complete analysis of our previous works about the power-performance characteristics in heterogeneous CPU-GPU platforms considering different operation frequencies and evolutionary parameters, such as distribution of individuals, etc. This way, different experimental configurations of the proposed procedure have been evaluated and compared with respect to a master-worker approach, not only in runtime but also considering energy consumption. The experimental results show that lower operating frequencies does not necessarily mean lower energy consumptions since energy is the product of power and time. Thus, we have observed that parallel processing not only reduces the runtime, but also the energy consumed by the application despite a higher instantaneous power. Particularly, the workload distribution among both CPU and GPU cores provides the best runtime and very low energy consumption compared with the values achieved by the same alternatives executed by only CPU threads.

La persistencia de delirium a los 3 meses del egreso de los pacientes ancianos hospitalizados en una clínica universitaria / Persistent Delirium in Elderly patients Three Months After Hospital Discharge from a University Clinic

RESUMEN Objetivos: El delirium es muy prevalente entre los pacientes ancianos con enfermedad general. Si no se revierte en el momento del alta hospitalaria, se lo considera «delirium persistente¼ (DP). El propósito del estudio es describir la prevalencia y las características de los pacientes con DP 3 meses después del egreso hospitalario de la Clínica Universitaria Bolivariana (CUB). Métodos: Se realizó un estudio descriptivo longitudinal para evaluar la prevalencia y las características de los pacientes de 65 o más arios del servicio de hospitalización de la CUB que cumplieran criterios de delirium del DSM-5 al ingreso, el egreso y 3 meses después. Se determinaron las variables sociodemográficas y se aplicaron las escalas CGI-S y DRS-R98. Resultados: Se evaluó a 30 pacientes con diagnóstico de delirium con interconsulta por psiquiatría de enlace entre abril y octubre de 2013, y se excluyó a 6 por no cumplir los criterios de inclusión. Se incluyó en el estudio a 24 pacientes, de los que 9 fallecieron durante la hospitalización (37,5%). De los 15 sobrevivientes, 5 (el 20,8% de la muestra) presentaron remisión del delirium al egreso y 10 (41,6%) continuaron con síntomas y conformaron el grupo de DP. Del grupo de DP, 5 (20,8%) presentaron DP completo y los otros 5 (20,8%), DP subsindrómico (DPSS). A los 3 meses del egreso, solo 2 pacientes (8,3%) continuaron con DP completo y otros 2 (8,3%), con DPSS. En el grupo de pacientes con DP, la prevalencia fue del 30% (diagnóstico de delirium al ingreso) y una incidencia del 70% (aparición del delirium durante la hospitalización). Conclusiones: Un grupo importante de pacientes con delirium continúan sintomáticos 3 meses después del alta. El 40% de los pacientes con síntomas persistentes en el seguimiento a 3 meses indica una trayectoria de mejoría gradual del delirium, lo cual tiene implicaciones en la práctica clínica.

ABSTRACT Objective: The purpose of the study was to determine the prevalence and characteristics of patients with persistent delirium (PD) at three months after hospital discharge. Methodology: Longitudinal descriptive study to assess the prevalence and characteristics of in-patients aged 65 years and older in the Clinica Universitaria Bolivariana who met DSM-5 criteria for delirium at admission, at discharge, and at a 3-month follow up assessment. Socio-demographic features were determined, and CGI-S and DRS-R98 scales used. Results: A total of 30 patients were evaluated between April and October 2013, but 6 did not fulfil the inclusion criteria. The study included 24 patients, with 9 (37.5%) dying during hospitalisation. Of the 15 surviving patients, five (20.8% of the total sample) had their delirium resolved at discharge, and ten (41.6% of the sample) continued with symptoms. These established the PD group, of whom five of them (20.8%) had full PD, and the other five (20.8%) sub-syndromal PD (SSPD). At the final assessment, only two patients (8.3%) continued with full PD, and another two (8.3%) with SSPD. Among the PD group, 30% had a full delirium at admission (prevalence), and 70% developed full delirium during hospitalization (incidence). Conclusions: A significant number of patients did not recover from delirium at leaving hospital, and remained symptomatic three months after discharge. The study findings suggest a course of gradual improvement of delirium, with a persistence of symptoms over time in 40% of the patients, which would have implications for the clinical practice.

Persistent Delirium in Elderly patients Three Months After Hospital Discharge from a University Clinic. / La persistencia de delirium a los 3 meses del egreso de los pacientes ancianos hospitalizados en una clínica universitaria.

OBJECTIVE: The purpose of the study was to determine the prevalence and characteristics of patients with persistent delirium (PD) at three months after hospital discharge. METHODOLOGY: Longitudinal descriptive study to assess the prevalence and characteristics of in-patients aged 65 years and older in the Clinica Universitaria Bolivariana who met DSM-5 criteria for delirium at admission, at discharge, and at a 3-month follow up assessment. Socio-demographic features were determined, and CGI-S and DRS-R98 scales used. RESULTS: A total of 30 patients were evaluated between April and October 2013, but 6 did not fulfil the inclusion criteria. The study included 24 patients, with 9 (37.5%) dying during hospitalisation. Of the 15 surviving patients, five (20.8% of the total sample) had their delirium resolved at discharge, and ten (41.6% of the sample) continued with symptoms. These established the PD group, of whom five of them (20.8%) had full PD, and the other five (20.8%) sub-syndromal PD (SSPD). At the final assessment, only two patients (8.3%) continued with full PD, and another two (8.3%) with SSPD. Among the PD group, 30% had a full delirium at admission (prevalence), and 70% developed full delirium during hospitalization (incidence). CONCLUSIONS: A significant number of patients did not recover from delirium at leaving hospital, and remained symptomatic three months after discharge. The study findings suggest a course of gradual improvement of delirium, with a persistence of symptoms over time in 40% of the patients, which would have implications for the clinical practice.

Binuclear palladium(I) and palladium(II) complexes of ortho-functionalized 1,3-bis(aryl)triazenido ligands.

Three new bis(aryl)triazene ligands, Ar-NNNH-Ar' [Ar = o-C(6)H(4)-CO(2)Me, Ar' = p-C(6)H(4)-CH(3) (2); Ar = Ar' = o-C(6)H(4)-CO(2)Me (3); Ar = o-C(6)H(4)-SMe, Ar' = p-C(6)H(4)-CH(3)) (4)], have been synthesized. The reaction of 1-4 with PdCl(2)(NCCH(3))(2) in the presence of a base afforded a series of binuclear diamagnetic palladium complexes. In these reactions, ligands 1-3 afforded the palladium(I) complexes [Pd(I)(o-MeO(2)C-C(6)H(4)-NNN-o-C(6)H(4)-CO(2)Me)](2) (5, monoclinic, space group P21/c, a = 8.6070(10) Angstrom, b = 14.3220(10) Angstrom, c = 12.7310(10) Angstrom, beta = 100.2950(10) degrees, Z = 2), [Pd(I)(o-MeO-C(6)H(4)-NNN-o-C(6)H(4)-OMe)](2) (6, triclinic, space group P, a = 6.6288(5) Angstrom, b = 10.2631(10) Angstrom, c = 11.0246(11) Angstrom, alpha = 85.579(6) degrees, beta = 80.885(6) degrees, gamma = 74.607(6) degrees, Z = 1), and [Pd(I)(o-MeO(2)C-C(6)H(4)-NNN-p-C(6)H(4)-CH(3))](2) (7, tetragonal, space group I41/a, a = 20.866(3) Angstrom, b = 20.866(3) Angstrom, c = 13.156(2) Angstrom, Z = 8). In contrast, the reaction of ligand 4 with PdCl(2)(NCCH(3))(2) resulted in the formation of a palladium(II) dimer, [Pd(II)(o-MeS-C(6)H(4)-NNN-p-C(6)H(4)-CH(3))Cl](2) (8, orthorhombic, space group P2(1)2(1)2, a = 10.4058(5) Angstrom, b = 16.2488(8) Angstrom, c = 9.9500(5) Angstrom, Z = 2).

Prolapso valvular mitral y embarazo: análisis de 90 casos / Mitral valve prolapse and pregnancy: analysis of 90 cases

El prolapso valvular mitral (PVM) es actualmente la cardiopatía congénita más frecuente, con una incidencia cercana al 6%. Nuestra serie analiza 90 embarazos en 39 pacientes con PVM diagnosticado como hallazgo auscultatorio y ecocardiográfico. Los resultados señalan una excelente tolerancia al embarazo en todas las pacientes sin complicaciones cardíacas relevantes (arrtmia severa, insuficiencia cardíaca, endocarditis bacteriana subaguda (EBSA) o embolias). No se observaron diferencias estadísticamente significativas (P < 0,05) en la morbilidad materna, en la morbimortalidad perinatal, al comparar el grupo control con el grupo PVM.