Learning joint representations for order and timing of perceptual-motor sequences: a dynamic neural field approach

Many of our everyday tasks require the control of the serial order and the timing of component actions. Using the dynamic neural field (DNF) framework, we address the learning of representations that support the performance of precisely time action sequences. In continuation of previous modeling work and robotics implementations, we ask specifically the question how feedback about executed actions might be used by the learning system to fine tune a joint memory representation of the ordinal and the temporal structure which has been initially acquired by observation. The perceptual memory is represented by a self-stabilized, multi-bump activity pattern of neurons encoding instances of a sensory event (e.g., color, position or pitch) which guides sequence learning. The strength of the population representation of each event is a function of elapsed time since sequence onset. We propose and test in simulations a simple learning rule that detects a mismatch between the expected and realized timing of events and adapts the activation strengths in order to compensate for the movement time needed to achieve the desired effect. The simulation results show that the effector-specific memory representation can be robustly recalled. We discuss the impact of the fast, activation-based learning that the DNF framework provides for robotics applications.

Facilitative effects of bi-hemispheric tDCS in cognitive deficits of Parkinson Disease patients

Parkinson’s disease (PD) is a progressive neurodegenerative disorder, primarily characterized by motor symptoms such as tremor, rigidity, bradykinesia, stiffness, slowness and impaired equilibrium. Although the motor symptoms have been the focus in PD, slight cognitive deficits are commonly found in non-demented and non-depressed PD patients, even in early stages of the disease, which have been linked to the subsequent development of pathological dementia. Thus, strongly reducing the quality of life (QoL). Both levodopa therapy and deep brain stimulation (DBS) have yield controversial results concerning the cognitive symptoms amelioration in PD patients. That does not seems to be the case with transcranial direct current stimulation (tDCS), although better stimulation parameters are needed. Therefore we hypothesize that simultaneously delivering cathodal tDCS (or ctDCS), over the right prefrontal cortex delivered with anodal tDCS (or atDCS) to left prefrontal cortex could be potentially beneficial for PD patients, either by mechanisms of homeostatic plasticity and by increases in the extracellular dopamine levels over the striatum.

Desenvolvimento de rede neuronal artificial para o diagnóstico de doenças neurodegenerativas

Dissertação de mestrado integrado em Engenharia Eletrónica Industrial e Computadores

Development of a thermoelectric generator for the exhaust of a vehicle

Dissertação de mestrado integrado em Engenharia Mecânica

Photosynthetic parameters and growth in seedlings of Bertholletia excelsa and Carapa guianensis in response to pre-acclimation to full sunlight and mild water stress

Light and water are important factors that may limit the growth and development of higher plants. The aim of this study was to evaluate photosynthetic parameters and growth in seedlings of Bertholletia excelsa and Carapa guianensis in response to pre-acclimation to full sunlight and mild water stress. I used six independent pre-acclimation treatments (0, 90 (11h15-12h45), 180 (10h30-13h30), 360 (09h00-15h00), 540 (07h30-16h30) and 720 min (06h00-18h00)) varying the time of exposure to full sunlight (PFS) during 30 days, followed by whole-day outdoor exposure for 120 days. Before PFS, the plants were kept in a greenhouse at low light levels (0.8 mol m-2 day-1). The PFS of 0 min corresponded to plants constantly kept under greenhouse conditions. From the beginning to the end of the experiment, each PFS treatment was submitted to two water regimes: moderate water stress (MWS, pre-dawn leaf water potential (ΨL) of -500 to -700 kPa) and without water stress (WWS, ΨL of -300 kPa, soil kept at field capacity). Plants under MWS received only a fraction of the amount of water applied to the well-watered ones. At the end of the 120-day-period under outdoor conditions, I evaluated light saturated photosynthesis (Amax), stomatal conductance (g s), transpiration (E) and plant growth. Both Amax and g s were higher for all plants under the PFS treatment. Stem diameter growth rate and Amax were higher for C. guianensis subjected to MWS than in well-watered plants. The contrary was true for B. excelsa. The growth of seedlings was enhanced by exposure to full sunlight for 180 minutes in both species. However, plants of B. excelsa were sensitive to moderate water stress. The higher photosynthetic rates and faster growth of C. guianensis under full sun and moderate water stress make this species a promissory candidate to be tested in reforestation programs.

Desenvolvimento de um Controlador com DSP para um Motor CA de 30 kW para o CEPIUM

Dissertação de mestrado integrado em Engenharia Eletrónica Industrial e Computadores

Influence of river ecological condition on changes in physico-chemical water parameters along rivers

Dissertação de mestrado em Ecology

Composition and structure variation for magnetron sputtered tantalum oxynitride thin films, as function of deposition parameters

Tantalum oxynitride thin films were produced by magnetron sputtering. The films were deposited usinga pure Ta target and a working atmosphere with a constant N2/O2ratio. The choice of this constant ratiolimits the study concerning the influence of each reactive gas, but allows a deeper understanding of theaspects related to the affinity of Ta to the non-metallic elements and it is economically advantageous.This work begins by analysing the data obtained directly from the film deposition stage, followed bythe analysis of the morphology, composition and structure. For a better understanding regarding theinfluence of the deposition parameters, the analyses are presented by using the following criterion: thefilms were divided into two sets, one of them produced with grounded substrate holder and the otherwith a polarization of −50 V. Each one of these sets was produced with different partial pressure of thereactive gases P(N2+ O2). All the films exhibited a O/N ratio higher than the N/O ratio in the depositionchamber atmosphere. In the case of the films produced with grounded substrate holder, a strong increaseof the O content is observed, associated to the strong decrease of the N content, when P(N2+ O2) is higherthan 0.13 Pa. The higher Ta affinity for O strongly influences the structural evolution of the films. Grazingincidence X-ray diffraction showed that the lower partial pressure films were crystalline, while X-rayreflectivity studies found out that the density of the films depended on the deposition conditions: thehigher the gas pressure, the lower the density. Firstly, a dominant -Ta structure is observed, for lowP(N2+ O2); secondly a fcc-Ta(N,O) structure, for intermediate P(N2+ O2); thirdly, the films are amorphousfor the highest partial pressures. The comparison of the characteristics of both sets of produced TaNxOyfilms are explained, with detail, in the text.

Desenvolvimento de um Posto de Carregamento Lento com BMS e Interface com o Cockpit para o CEPIUM

Dissertação de mestrado integrado em Engenharia Eletrónica Industrial e de Computadores

Controlador com FPGA para um motor-in-wheel CA de fluxo axial

Dissertação de mestrado integrado em Engenharia Eletrónica Industrial e de Computadores

Hydrocarbonoclastic bacteria: from bioremediation to bioenergy feedstock

Tese de Doutoramento em Engenharia Química e Biológica.

Automatically estimating iSAX parameters

The Symbolic Aggregate Approximation (iSAX) is widely used in time series data mining. Its popularity arises from the fact that it largely reduces time series size, it is symbolic, allows lower bounding and is space efficient. However, it requires setting two parameters: the symbolic length and alphabet size, which limits the applicability of the technique. The optimal parameter values are highly application dependent. Typically, they are either set to a fixed value or experimentally probed for the best configuration. In this work we propose an approach to automatically estimate iSAX’s parameters. The approach – AutoiSAX – not only discovers the best parameter setting for each time series in the database, but also finds the alphabet size for each iSAX symbol within the same word. It is based on simple and intuitive ideas from time series complexity and statistics. The technique can be smoothly embedded in existing data mining tasks as an efficient sub-routine. We analyze its impact in visualization interpretability, classification accuracy and motif mining. Our contribution aims to make iSAX a more general approach as it evolves towards a parameter-free method.

Cognitive processes affect the gait of subjects with Parkinson’s and Alzheimer’s disease in dual tasks

Objective To investigate the relation between gait parameters and cognitive impairments in subjects with Parkinson’s disease (PD) and Alzheimer’s disease (AD) during the performance of dual tasks. Methods This was a cross-sectional study involving 126 subjects divided into three groups: Parkinson group (n = 43), Alzheimer group (n = 38), and control group (n = 45). The subjects were evaluated using the Timed Up and Go test administered with motor and cognitive distracters. Gait analyses consisted of cadence and speed measurements, with cognitive functions being assessed by the Brief Cognitive Screening Battery and the Clock Drawing Test. Statistical procedures included mixed-design analyses of variance to observe the gait patterns between groups and tasks and the linear regression model to investigate the influence of cognitive functions in this process. A 5% significant level was adopted. Results Regarding the subjects’ speed, the data show a significant difference between group vs task interaction (p = 0.009), with worse performance of subjects with PD in motor dual task and of subjects with AD in cognitive dual task. With respect to cadence, no statistical differences was seen between group vs task interaction (p = 0.105), showing low interference of the clinical conditions on such parameter. The linear regression model showed that up to 45.79%, of the variance in gait can be explained by the interference of cognitive processes. Conclusion Dual task activities affect gait pattern in subjects with PD and AD. Differences between groups reflect peculiarities of each disease and show a direct interference of cognitive processes on complex tasks.

Searching for therapeutic strategies in a mouse modelo of Machado-Joseph disease: targeting proteostases

Tese de Doutoramento em Ciências da Saúde

Influence of emphysema distribution on pulmonary function parameters in COPD patients

Objective: To evaluate the impact that the distribution of emphysema has on clinical and functional severity in patients with COPD. Methods: The distribution of the emphysema was analyzed in COPD patients, who were classified according to a 5-point visual classification system of lung CT findings. We assessed the influence of emphysema distribution type on the clinical and functional presentation of COPD. We also evaluated hypoxemia after the six-minute walk test (6MWT) and determined the six-minute walk distance (6MWD). Results: Eighty-six patients were included. The mean age was 65.2 ± 12.2 years, 91.9% were male, and all but one were smokers (mean smoking history, 62.7 ± 38.4 pack-years). The emphysema distribution was categorized as obviously upper lung-predominant (type 1), in 36.0% of the patients; slightly upper lung-predominant (type 2), in 25.6%; homogeneous between the upper and lower lung (type 3), in 16.3%; and slightly lower lung-predominant (type 4), in 22.1%. Type 2 emphysema distribution was associated with lower FEV1 , FVC, FEV1 /FVC ratio, and DLCO. In comparison with the type 1 patients, the type 4 patients were more likely to have an FEV1 < 65% of the predicted value (OR = 6.91, 95% CI: 1.43-33.45; p = 0.016), a 6MWD < 350 m (OR = 6.36, 95% CI: 1.26-32.18; p = 0.025), and post-6MWT hypoxemia (OR = 32.66, 95% CI: 3.26-326.84; p = 0.003). The type 3 patients had a higher RV/TLC ratio, although the difference was not significant. Conclusions: The severity of COPD appears to be greater in type 4 patients, and type 3 patients tend to have greater hyperinflation. The distribution of emphysema could have a major impact on functional parameters and should be considered in the evaluation of COPD patients.