The ethical use of high-performance computing and artificial intelligence: fighting COVID-19 at Barcelona Supercomputing Center.

The COVID-19 pandemic has created an extraordinary medical, economic and humanitarian emergency. Artificial intelligence, in combination with other digital technologies, is being used as a tool to support the fight against the viral pandemic that has affected the entire world since the beginning of 2020. Barcelona Supercomputing Center collaborates in the battle against the coronavirus in different areas: the application of bioinformatics for the research on the virus and its possible treatments, the use of artificial intelligence, natural language processing and big data techniques to analyse the spread and impact of the pandemic, and the use of the MareNostrum 4 supercomputer to enable massive analysis on COVID-19 data. Many of these activities have included the use of personal and sensitive data of citizens, which, even during a pandemic, should be treated and handled with care. In this work we discuss our approach based on an ethical, transparent and fair use of this information, an approach aligned with the guidelines proposed by the European Union.

The Fourth-Revolution in the Water Sector Encounters the Digital Revolution.

The so-called fourth revolution in the water sector will encounter the Big data and Artificial Intelligence (AI) revolution. The current data surplus stemming from all types of devices together with the relentless increase in computer capacity is revolutionizing almost all existing sectors, and the water sector will not be an exception. Combining the power of Big data analytics (including AI) with existing and future urban water infrastructure represents a significant untapped opportunity for the operation, maintenance, and rehabilitation of urban water infrastructure to achieve economic and environmental sustainability. However, such progress may catalyze socio-economic changes and cross sector boundaries (e.g., water service, health, business) as the appearance of new needs and business models will influence the job market. Such progress will impact the academic sector as new forms of research based on large amounts of data will be possible, and new research needs will be requested by the technology industrial sector. Research and development enabling new technological approaches and more effective management strategies are needed to ensure that the emerging framework for the water sector will meet future societal needs. The feature further elucidates the complexities and possibilities associated with such collaborations.

Artificial Intelligence to Identify Retinal Fundus Images, Quality Validation, Laterality Evaluation, Macular Degeneration, and Suspected Glaucoma.

PURPOSE: To assess the performance of deep learning algorithms for different tasks in retinal fundus images: (1) detection of retinal fundus images versus optical coherence tomography (OCT) or other images, (2) evaluation of good quality retinal fundus images, (3) distinction between right eye (OD) and left eye (OS) retinal fundus images,(4) detection of age-related macular degeneration (AMD) and (5) detection of referable glaucomatous optic neuropathy (GON). PATIENTS AND METHODS: Five algorithms were designed. Retrospective study from a database of 306,302 images, Optretina's tagged dataset. Three different ophthalmologists, all retinal specialists, classified all images. The dataset was split per patient in a training (80%) and testing (20%) splits. Three different CNN architectures were employed, two of which were custom designed to minimize the number of parameters with minimal impact on its accuracy. Main outcome measure was area under the curve (AUC) with accuracy, sensitivity and specificity. RESULTS: Determination of retinal fundus image had AUC of 0.979 with an accuracy of 96% (sensitivity 97.7%, specificity 92.4%). Determination of good quality retinal fundus image had AUC of 0.947, accuracy 91.8% (sensitivity 96.9%, specificity 81.8%). Algorithm for OD/OS had AUC 0.989, accuracy 97.4%. AMD had AUC of 0.936, accuracy 86.3% (sensitivity 90.2% specificity 82.5%), GON had AUC of 0.863, accuracy 80.2% (sensitivity 76.8%, specificity 83.8%). CONCLUSION: Deep learning algorithms can differentiate a retinal fundus image from other images. Algorithms can evaluate the quality of an image, discriminate between right or left eye and detect the presence of AMD and GON with a high level of accuracy, sensitivity and specificity.

Supraspinal Shaping of Adaptive Transitions in the State of Functional Connectivity Between Segmentally Distributed Dorsal Horn Neuronal Populations in Response to Nociception and Antinociception.

In the anesthetized cat the correlation between the ongoing cord dorsum potentials (CDPs) recorded from different lumbar spinal segments has a non-random structure, suggesting relatively stable patterns of functional connectivity between the dorsal horn neuronal ensembles involved in the generation of these potentials. During the nociception induced by the intradermic injection of capsaicin, the patterns of segmental correlation between the spontaneous CDPs acquire other non-random configurations that are temporarily reversed to their pre-capsaicin state by the systemic injection of lidocaine, a procedure known to decrease the manifestation of neuropathic pain in both animals and humans. We have now extended these studies and utilized machine learning for the automatic extraction and selection of particular classes of CDPs according to their shapes and amplitudes. By using a Markovian analysis, we disclosed the transitions between the different kinds of CDPs induced by capsaicin and lidocaine and constructed a global model based on the changes in the behavior of the CDPs generated along the whole set of lumbar segments. This allowed the identification of the different states of functional connectivity within the whole ensemble of dorsal horn neurones attained during nociception and their transitory reversal by systemic administration of lidocaine in preparations with the intact neuroaxis and after spinalization. The present observations provide additional information on the state of self-organized criticality that leads to the adaptive behavior of the dorsal horn neuronal networks during nociception and antinociception both shaped by supraspinal descending influences.

Recurrent inhibition in the cerebral cortex.

Neuronal activity can be modulated by endogenous control mechanisms that either facilitate or suppress it. With this idea in mind, we attempted to evaluate and correlate spinal neuronal activity with the amplitude of corticogram (ECoG) event related potentials (ERP) in the presence of nociceptive stimulation in rats. We evaluated the ERP in response to noxious stimuli, endogenous analgesic actions, different frequencies, and heterotopic nociceptive stimulation, as well as in conjunction with recordings from neurons in the spinal cord that are activated by noxious stimuli. Computational tasks enabled us to establish correlations between the amplitude of ERP and neuronal firing of cells in the spinal dorsal horn. Our results show that the ERP amplitude could be modified by previous activity in the cerebral cortex, but the activity in the spinal cord did not change. Previous activity could originate spontaneously or could be driven by sensory stimulation. A recurrent inhibitory cortical action is proposed that could explain the suppression of pain perception during electrical or magnetic transcranial stimulation, as well as during heterotopic stimulation. This study aims to uncover a local recurrent inhibitory cortical action that could modify the sensory information.

A Web-Based Platform for People With Memory Problems and Their Caregivers (CAREGIVERSPRO-MMD): Mixed-Methods Evaluation of Usability.

BACKGROUND: The increasing number of people with dementia (PwD) drives research exploring Web-based support interventions to provide effective care for larger populations. In this concept, a Web-based platform (CAREGIVERSPRO-MMD, 620911) was designed to (1) improve the quality of life for PwD, (2) reduce caregiver burden, (3) reduce the financial costs for care, and (4) reduce administration time for health and social care professionals. OBJECTIVE: The objective of this study was to evaluate the usability and usefulness of CAREGIVERSPRO-MMD platform for PwD or mild cognitive impairment (MCI), informal caregivers, and health and social care professionals with respect to a wider strategy followed by the project to enhance the user-centered approach. A secondary aim of the study was to collect recommendations to improve the platform before the future pilot study. METHODS: A mixed methods design was employed for recruiting PwD or MCI (N=24), informal caregivers (N=24), and professionals (N=10). Participants were asked to rate their satisfaction, the perceived usefulness, and ease of use of each function of the platform. Qualitative questions about the improvement of the platform were asked when participants provided low scores for a function. Testing occurred at baseline and 1 week after participants used the platform. The dropout rate from baseline to the follow-up was approximately 10% (6/58). RESULTS: After 1 week of platform use, the system was useful for 90% (20.75/23) of the caregivers and for 89% (5.36/6) of the professionals. When users responded to more than 1 question per platform function, the mean of satisfied users per function was calculated. These user groups also provided positive evaluations for the ease of use (caregivers: 82%, 18.75/23; professionals: 97%, 5.82/6) and their satisfaction with the platform (caregivers: 79%, 18.08/23; professionals: 73%, 4.36/6). Ratings from PwD were lower than the other groups for usefulness (57%, 13/23), ease of use (41%, 9.4/23), and overall satisfaction (47%, 11/23) with the platform (P<.05). Qualitative comments related to both improvements for functionality and the platform interface. CONCLUSIONS: Although caregivers and professionals were overall satisfied with the platform, further adaptations were recommended by PwD. This reiterates the importance of the involvement of end users in the development of Web-based interventions. Recommendations from users in this paper apply for the interface and functionality of a wider range of Web-based support interventions.

Markovian Analysis of the Sequential Behavior of the Spontaneous Spinal Cord Dorsum Potentials Induced by Acute Nociceptive Stimulation in the Anesthetized Cat.

In a previous study we developed a Machine Learning procedure for the automatic identification and classification of spontaneous cord dorsum potentials (CDPs). This study further supported the proposal that in the anesthetized cat, the spontaneous CDPs recorded from different lumbar spinal segments are generated by a distributed network of dorsal horn neurons with structured (non-random) patterns of functional connectivity and that these configurations can be changed to other non-random and stable configurations after the noceptive stimulation produced by the intradermic injection of capsaicin in the anesthetized cat. Here we present a study showing that the sequence of identified forms of the spontaneous CDPs follows a Markov chain of at least order one. That is, the system has memory in the sense that the spontaneous activation of dorsal horn neuronal ensembles producing the CDPs is not independent of the most recent activity. We used this markovian property to build a procedure to identify portions of signals as belonging to a specific functional state of connectivity among the neuronal networks involved in the generation of the CDPs. We have tested this procedure during acute nociceptive stimulation produced by the intradermic injection of capsaicin in intact as well as spinalized preparations. Altogether, our results indicate that CDP sequences cannot be generated by a renewal stochastic process. Moreover, it is possible to describe some functional features of activity in the cord dorsum by modeling the CDP sequences as generated by a Markov order one stochastic process. Finally, these Markov models make possible to determine the functional state which produced a CDP sequence. The proposed identification procedures appear to be useful for the analysis of the sequential behavior of the ongoing CDPs recorded from different spinal segments in response to a variety of experimental procedures including the changes produced by acute nociceptive stimulation. They are envisaged as a useful tool to examine alterations of the patterns of functional connectivity between dorsal horn neurons under normal and different pathological conditions, an issue of potential clinical concern.

Reducing Fall Risk with Combined Motor and Cognitive Training in Elderly Fallers.

BACKGROUND: Falling is a major clinical problem in elderly people, demanding effective solutions. At present, the only effective intervention is motor training of balance and strength. Executive function-based training (EFt) might be effective at preventing falls according to evidence showing a relationship between executive functions and gait abnormalities. The aim was to assess the effectiveness of a motor and a cognitive treatment developed within the EU co-funded project I-DONT-FALL. METHODS: In a sample of 481 elderly people at risk of falls recruited in this multicenter randomised controlled trial, the effectiveness of a motor treatment (pure motor or mixed with EFt) of 24 one-hour sessions delivered through an i-Walker with a non-motor treatment (pure EFt or control condition) was evaluated. Similarly, a 24 one-hour session cognitive treatment (pure EFt or mixed with motor training), delivered through a touch-screen computer was compared with a non-cognitive treatment (pure motor or control condition). RESULTS: Motor treatment, particularly when mixed with EFt, reduced significantly fear of falling (F(1,478) = 6.786, p = 0.009) although to a limited extent (ES -0.25) restricted to the period after intervention. CONCLUSIONS: This study suggests the effectiveness of motor treatment empowered by EFt in reducing fear of falling.

Crossing the Death Valley to Transfer Environmental Decision Support Systems to the Water Market.

Environmental decision support systems (EDSSs) are attractive tools to cope with the complexity of environmental global challenges. Several thoughtful reviews have analyzed EDSSs to identify the key challenges and best practices for their development. One of the major criticisms is that a wide and generalized use of deployed EDSSs has not been observed. The paper briefly describes and compares four case studies of EDSSs applied to the water domain, where the key aspects involved in the initial conception and the use and transfer evolution that determine the final success or failure of these tools (i.e., market uptake) are identified. Those aspects that contribute to bridging the gap between the EDSS science and the EDSS market are highlighted in the manuscript. Experience suggests that the construction of a successful EDSS should focus significant efforts on crossing the death-valley toward a general use implementation by society (the market) rather than on development.

Overground walking training with the i-Walker, a robotic servo-assistive device, enhances balance in patients with subacute stroke: a randomized controlled trial.

BACKGROUND: Patients affected by mild stroke benefit more from physiological overground walking training than walking-like training performed in place using specific devices. The aim of the study was to evaluate the effects of overground robotic walking training performed with the servo-assistive robotic rollator (i-Walker) on walking, balance, gait stability and falls in a community setting in patients with mild subacute stroke. METHODS: Forty-four patients were randomly assigned to two different groups that received the same therapy in two daily 40-min sessions 5 days a week for 4 weeks. Twenty sessions of standard therapy were performed by both groups. In the other 20 sessions the subjects enrolled in the i-Walker-Group (iWG) performed with the i-Walker and the Control-Group patients (CG) performed the same amount of conventional walking oriented therapy. Clinical and instrumented gait assessments were made pre- and post-treatment. The follow-up observation consisted of recording the number of fallers in the community setting after 6 months. RESULTS: Treatment effectiveness was higher in the iWG group in terms of balance improvement (Tinetti: 68.4 ± 27.6 % vs. 48.1 ± 33.9 %, p = 0.033) and 10-m and 6-min timed walking tests (significant interaction between group and time: F(1,40) = 14.252, p = 0.001; and F(1,40) = 7.883, p = 0.008, respectively). When measured, latero-lateral upper body accelerations were reduced in iWG (F = 4.727, p = 0.036), suggesting increased gait stability, which was supported by a reduced number of falls at home. CONCLUSIONS: A robotic servo-assisted i-Walker improved walking performance and balance in patients affected by mild/moderate stroke, leading to increased gait stability and reduced falls in the community. TRIAL REGISTRATION: This study was registered on anzctr.org.au (July 1, 2015; ACTRN12615000681550 ).

Situated Agents and Humans in Social Interaction for Elderly Healthcare: From Coaalas to AVICENA.

Assistive Technologies (AT) are an application area where several Artificial Intelligence techniques and tools have been successfully applied to support elderly or impeded people on their daily activities. However, approaches to AT tend to center in the user-tool interaction, neglecting the user's connection with its social environment (such as caretakers, relatives and health professionals) and the possibility to monitor undesired behaviour providing both adaptation to a dynamic environment and early response to potentially dangerous situations. In previous work we have presented COAALAS, an intelligent social and norm-aware device for elderly people that is able to autonomously organize, reorganize and interact with the different actors involved in elderly-care, either human actors or other devices. In this paper we put our work into context, by first examining what are the desirable properties of such a system, analysing the state-of-the-art on the relevant topics, and verifying the validity of our proposal in a larger context that we call AVICENA. AVICENA's aim is develop a semi-autonomous (collaborative) tool to promote monitored, intensive, extended and personalized therapeutic regime adherence at home based on adaptation techniques.

A machine learning methodology for the selection and classification of spontaneous spinal cord dorsum potentials allows disclosure of structured (non-random) changes in neuronal connectivity induced by nociceptive stimulation.

Previous studies aimed to disclose the functional organization of the neuronal networks involved in the generation of the spontaneous cord dorsum potentials (CDPs) generated in the lumbosacral spinal segments used predetermined templates to select specific classes of spontaneous CDPs. Since this procedure was time consuming and required continuous supervision, it was limited to the analysis of two specific types of CDPs (negative CDPs and negative positive CDPs), thus excluding potentials that may reflect activation of other neuronal networks of presumed functional relevance. We now present a novel procedure based in machine learning that allows the efficient and unbiased selection of a variety of spontaneous CDPs with different shapes and amplitudes. The reliability and performance of the present method is evaluated by analyzing the effects on the probabilities of generation of different classes of spontaneous CDPs induced by the intradermic injection of small amounts of capsaicin in the anesthetized cat, a procedure known to induce a state of central sensitization leading to allodynia and hyperalgesia. The results obtained with the selection method presently described allowed detection of spontaneous CDPs with specific shapes and amplitudes that are assumed to represent the activation of functionally coupled sets of dorsal horn neurones that acquire different, structured configurations in response to nociceptive stimuli. These changes are considered as responses tending to adequate transmission of sensory information to specific functional requirements as part of homeostatic adjustments.

[Hemolytic anemia secondary to the placement of a portosystemic stented shunt]. / Anemia hemolítica secundaria a la colocación de Stent transyugular intrahepático porto sistémico.

INTRODUCTION: portal hypertension and variceal hemorrhage are common complications of hepatic cirrhosis, both associated with a high morbimortality. Portal system decompression by the placement of a transjugular intrahepatic portosystemic stented shunt, can reduce portal venus pressure and is effective controling complications of portal hypertension, like variceal hemorrhage and ascitis. The aim of this document is to describe a case of hemolytic anemia secondary to the placement of a transjugular intrahepatic portosystemic stented shunt. CLINICAL CASE: patient with portal hypertension secondary to liver cirrosis was given a transjugular intrahepatic portosystemic stented shunt for recurrent variceal hemorrhage. After the procedure, hemoglobin decreased 2 g/dL, associated with reticulocitosis, hipohaptoglobinemia, elevated lactic dehydrogenase and indirect hyperbilirrubinemia with negative Coombs test. The peripheral blood smear showed abnormal erythrocytes, with the prevalence of schistocytes. The final diagnosis was hemolytic anemia secondary to transjugular intrahepatic portosystemic stented shunt. CONCLUSIONS: the hemolytic anemia secondary to Transjugular Intrahepatic Portosystemic Stented Shunt is a rare complication. Usually, it has a benign prognosis, and it is self-limited once the stent is endothelialized.

[Variations of vital signs and peripheral oxygen saturation in critically ill preterm newborn, after sponge bathing]. / Variaciones de los signos vitales y saturación periférica de oxígeno en el recién nacido de pretérmino críticamente enfermo después del baño de esponja.

OBJECTIVE: To determine the variability of the vital signs (temperature, heart rate and respiratory frequency), skin coloration and peripheral oxygen saturation in critically ill preterm newborns (CI PTNB) before, during and after sponge bathing as well as to determine the possible presence of secondary complications of this procedure. MATERIAL AND METHODS: We performed a quasi-experimental study (experimental, prospective, comparative and clinical study with intervention) May to December 2008, in a Neonatal Intensive Care Unit. We included CI PTNB of 0 to 28 days of extrauterine life who have practiced in the routine sponge bathing. Area of significance was considered when p < 0.05. RESULTS: During or after the events in any of the patients presented any complications after 12 h of monitoring, but it was necessary to increase the inspired fraction of oxygen and temperature in the incubator or radiant heat cradle temporarily. CONCLUSIONS: We conclude that the sponge bath is not safe for a CI PTNB and this should be performed in the shortest time possible, and the medical must be very alert to the possibility that patients require more support than they had prior to sponge bathing, mainly in the temperature of the incubator or radiant heat cradle and inspired fraction of oxygen for the required time according to the evolution of these variables.

Fragilidad y sarcopenia / Frailty and sarcopenia

La fragilidad es un síndrome geriátrico caracterizado por pérdida de peso, cansancio, debilidad, marcha lenta y disminución de la actividad física. Es más común en mujeres, obesos y diabéticos. Es secundaria a disregulación endócrina y a un estado proinflamatorio y protrombótico. La sarcopenia, pérdida de la masa muscular, es característica de la fragilidad. El tratamiento está encaminado a incrementar la masa y fuerza muscular mediante un mejor aporte calórico-protéico y un programa de ejercicios. El presente trabajo tiene como objetivo revisar conceptos actuales relacionados a la definición, epidemiología, fisiopatología y tratamiento del síndrome de fragilidad y la sarcopenia, así como su impacto en la población geriátrica.

Frailty is a geriatric syndrome characterized by weight loss, fatigue, weakness, slow walking and reduced physical activity. It is more common among women, obese people and diabetic patients. It is secondary to endocrine dys-regulation and a proinflammatory, prothrombotic status. Sarcopenia, loss of muscle mass, is characteristic of frailty. The treatment is focused on increasing muscle mass and strength by improving caloric-proteic intake and implementing a physical exercise program. The aim of the present article is to review the current concepts related to the definition, epidemiology, pathophysiology and treatment of frailty syndrome and sarcopenia, as well as its impact on geriatric population.

Wheelchair collaborative control for disabled users navigating indoors.

OBJECTIVE: Mobility is of key importance for autonomous living. Persons with severe disabilities may be assisted by robotic wheelchairs when manual control is not possible. However, these persons should contribute to control as much as they can to avoid loss of residual skills and frustration. Traditionally, wheelchair shared control approaches either give control to person or robot depending on the situation. METHODS AND MATERIALS: We propose a new shared control technique where robot and person contribute simultaneously to control. Their commands are weighted according to their respective local efficiencies and then combined via a reactive navigation strategy. Thus, assistance adapts to the user's needs. We refer to this approach as collaborative control. RESULTS: Collaborative control was tested in a home environment in Fondazione Santa Lucia (Rome) by 18 volunteers presenting different degrees of physical and cognitive disability. All of them successfully finished a complex test path with assistance. Both users and caregivers' opinion on the system was very positive. Acceptance was very good according to the psychosocial impact of assistive devices scale. CONCLUSIONS: Collaborative control adapts to the person's needs and assists him/her when necessary, locally compensating any problem related to specific disabilities. An ANOVA returned a p-value of 0.016, meaning that there is significant improvement in task performance when collaborative control is used.

[Continuity of care: assistive technology at home]. / La continuità assistenziale: una visione orientata all'uso delle nuove tecnologie.

The purpose of this article is to evaluate the possibility of introducing the Assistive Technology for elderly persons at home in order to provide intermediate care as the range of services aimed at facilitating the transfer from hospital, and the transition from a situation of dependency on the medical staff to a situation of functional independence. Improvements in this area would allow an approach focused on the user and reduce the waste of economic resources. Once it achieves the objectives of strictly medical care, the discharge of patients can be anticipated. We believe that one possible solution is represented by the introduction and use of the intelligent agents for the support of the activities of daily living. We report some examples.

Connectivity for healthcare and well-being management: examples from six European projects.

Technological advances and societal changes in recent years have contributed to a shift in traditional care models and in the relationship between patients and their doctors/carers, with (in general) an increase in the patient-carer physical distance and corresponding changes in the modes of access to relevant care information by all groups. The objective of this paper is to showcase the research efforts of six projects (that the authors are currently, or have recently been, involved in), CAALYX, eCAALYX, COGKNOW, EasyLine+, I2HOME, and SHARE-it, all funded by the European Commission towards a future where citizens can take an active role into managing their own healthcare. Most importantly, sensitive groups of citizens, such as the elderly, chronically ill and those suffering from various physical and cognitive disabilities, will be able to maintain vital and feature-rich connections with their families, friends and healthcare providers, who can then respond to, and prevent, the development of adverse health conditions in those they care for in a timely manner, wherever the carers and the people cared for happen to be.

Situated robotics: from learning to teaching by imitation.

This paper presents an approach to imitation learning in robotics focusing on low level behaviours, so that they do not need to be encoded into sets and rules, but learnt in an intuitive way. Its main novelty is that, rather than trying to analyse natural human actions and adapting them to robot kinematics, humans adapt themselves to the robot via a proper interface to make it perform the desired action. As an example, we present a successful experiment to learn a purely reactive navigation behaviour using robotic platforms. Using Case Based Reasoning, the platform learns from a human driver how to behave in the presence of obstacles, so that no kinematics studies or explicit rules are required.

[Artificial intelligence in medicine: project of a mobile platform in an intelligent environment for the care of disabled and elderly people]. / Intelligenza artificiale in medicina: progetto di una piattaforma mobile inserita in un ambiente intelligente per l'assistenza ai disabili e agli anziani.

A project based on the integration of new technologies and artificial intelligence to develop a device--e-tool--for disabled patients and elderly people is presented. A mobile platform in intelligent environments (skilled-care facilities and home-care), controlled and managed by a multi-level architecture, is proposed to support patients and caregivers to increase self-dependency in activities of daily living.