Search

The objective of this research was to predict the death of a newborn based on particular features. We propose a method called RIGOR to identify 68 deaths of children who did not complete a year of life. We conclude that the Brazilian government can prevent child deaths by emphasizing the need to monitor fetal weight.

Anxiety disorders are highly prevalent and disabling but seem particularly tractable to investigation with translational neuroscience methodologies. Neuroimaging has informed our understanding of the neurobiology of anxiety disorders, but research has been limited by small sample sizes and low statistical power, as well as heterogenous imaging methodology. The ENIGMA‐Anxiety Working Group has brought together researchers from around the world, in a harmonized and coordinated effort to address these challenges and generate more robust and reproducible findings. This paper elaborates on the concepts and methods informing the work of the working group to date, and describes the initial approach of the four subgroups studying generalized anxiety disorder, panic disorder, social anxiety disorder, and specific phobia. At present, the ENIGMA‐Anxiety database contains information about more than 100 unique samples, from 16 countries and 59 institutes. Future directions include examining additional imaging modalities, integrating imaging and genetic data, and collaborating with other ENIGMA working groups. The ENIGMA consortium creates synergy at the intersection of global mental health and clinical neuroscience, and the ENIGMA‐Anxiety Working Group extends the promise of this approach to neuroimaging research on anxiety disorders., Anxiety disorders are highly prevalent and disabling but seem particularly tractable to investigation with translational neuroscience methodologies. Neuroimaging has informed our understanding of the neurobiology of anxiety disorders, but research has been limited by small sample sizes and low statistical power, as well as heterogenous imaging methodology. The ENIGMA‐Anxiety Working Group has brought together researchers from around the world, in a harmonized and coordinated effort to address these challenges and generate more robust and reproducible findings. The ENIGMA consortium creates synergy at the intersection of global mental health and clinical neuroscience, and the ENIGMA‐Anxiety Working Group extends the promise of this approach to neuroimaging research on anxiety disorders.

Contains fulltext : 202693.pdf (Publisher’s version ) (Open Access) Schizophrenia is a severe mental disorder characterized by numerous subtle changes in brain structure and function. Machine learning allows exploring the utility of combining structural and functional brain magnetic resonance imaging (MRI) measures for diagnostic application, but this approach has been hampered by sample size limitations and lack of differential diagnostic data. Here, we performed a multi-site machine learning analysis to explore brain structural patterns of T1 MRI data in 2668 individuals with schizophrenia, bipolar disorder or attention-deficit/ hyperactivity disorder, and healthy controls. We found reproducible changes of structural parameters in schizophrenia that yielded a classification accuracy of up to 76% and provided discrimination from ADHD, through it lacked specificity against bipolar disorder. The observed changes largely indexed distributed grey matter alterations that could be represented through a combination of several global brain-structural parameters. This multi-site machine learning study identified a brain-structural signature that could reproducibly differentiate schizophrenia patients from controls, but lacked specificity against bipolar disorder. While this currently limits the clinical utility of the identified signature, the present study highlights that the underlying alterations index substantial global grey matter changes in psychotic disorders, reflecting the biological similarity of these conditions, and provide a roadmap for future exploration of brain structural alterations in psychiatric patients.

33rd European Photovoltaic Solar Energy Conference and Exhibition; 2485-2491, On regional scale, the estimation of the solar power generation from the environmental conditions and the solar power forecast is essential for Distribution System Operators, Transmission System Operator, energy traders, and aggregators. Indeed, high photovoltaic penetration results in a stochastic variability of the electric demand that could compromise the stability of the grid, increase the amount of energy reserve and the energy imbalance cost. In this context, two upscaling methods and several models were developed and used for estimation and forecast of the photovoltaic distributed generation in a small area of Italy with high photovoltaic penetration. The upscaling methods are based on chains of data-driven models (k-mean clustering, principal component analysis and ensemble of neural networks) that use as inputs satellite derived irradiance and numerical weather prediction. These models allow a complete assessment of the accuracy of the regional power forecast at different horizons, from 0 hour ahead (power estimation) to 48 hours ahead. The power estimation model provided a RMSE of 3% of installed capacity. Intra-day forecast (from 1 to 4 hours) obtained a RMSE of 5% - 7% and a skill score with respect to the clear sky persistence from -8% to 33.6%. The one and two days ahead forecast achieved a RMSE of 7% and 7.5% and a skill score of 39.2% and 45.7%. The ensemble smoothing effect on cluster scale was also studied. It was proved that the RMSE reduction achieved on cluster scale due to the ensemble smoothing follows the same exponential law found for single systems. In this case the smoothing effect reduces the RMSE of day-ahead forecast of 12% with respect to the mean single cluster value. It was also shown that the skill score increases linearly with the size of the region with a rate of 0.1% for km. Furthermore, a method to estimate the forecast error was also developed. It was based on an ensemble neural network model coupled with a probabilistic correction. It can provide a highly reliable computation of the prediction intervals.

32nd European Photovoltaic Solar Energy Conference and Exhibition; 1475-1481, This work aims to compare several data-driven models using different Numerical Weather Prediction (NWP) input data and then to build up an outperforming Multi-Model Ensemble (MME) and its prediction intervals. Statistic, stochastic and hybrid machine-learning algorithms were developed and the NWP data from IFS and WRF models were used as input. It was found that the same machine learning algorithm differs in performance even if it make use of NWP data with comparable accuracy. This apparent inconsistency depends on the capability of the machine learning model to correct the bias error of the input data. The stochastic and the hybrid model using the same WRF input, as well as the stochastic and the non-linear statistic models using the same IFS input, produce very similar results. The MME resulting from the averaging of the best data-driven forecasts, improves the accuracy of the outperforming member of the ensemble, bringing the skill score from 42% to 46%. To reach this performance, the ensemble should include forecasts with similar accuracy but generated with the higher variety of different data-driven technology and NWP input.

A identificação de padrões em fluxos de dados contínuos tem despertado o interesse científico, seja na detecção de falhas em sistemas, identificação de operações fraudulentas em transações bancárias, propagação de doenças ou ainda na preservação do meio ambiente. A categorização destes dados, concomitante com a ampliação do sensoriamento e monitoramento de diversos outros domínios, motiva a busca por soluções práticas e eficientes que auxiliem na busca por padrões recorrentes. A extração de conhecimento dos dados, quando dependentes do tempo, exige um tratamento especial e a mineração dos dados apresenta-se como uma atividade valiosa. Neste trabalho, é proposta uma abordagem chamada DSP-Class para classificação de séries temporais utilizando Descritores de Textura aplicados em Gráficos de Recorrência (RP). São utilizados 14 conjuntos de dados reais relacionados a vocalizações de aves, identificação de insetos, categorização de reações químicas, dentre outros. O objetivo desta pesquisa é verificar a utilização das características texturais de RPs em algoritmos de aprendizagem, tais como Support Vector Machine (SVM) e C5:0, aplicando a Decomposição de Modo Empírico (EMD) na classificação de séries temporais. Também é analisada a influência estocástica-determinística presentes nos fluxos. Verifica-se desempenho ruim do algoritmo 1NN, considerado estado-da-arte, em séries predominantemente estocásticas ou determinísticas e desempenho 67:66% superior da abordagem DSP-Class, uma vez que as características texturais distinguem classes de séries temporais mais satisfatoriamente que a busca por similaridade utilizada no algoritmo 1NN nos dados analisados. Verifica-se inclusive, resultados 18;67% superiores àqueles obtidos por pesquisas semelhantes que utilizam outras características presentes em séries temporais. ABSTRACT - Identify patterns in continuous data streams has attracted scientific interest for detecting system failures, identify fraudulent transactions in banks, the spread of diseases and also in the preservation of the environment. The increased volume of data produced in the last decade, concomitant with the number of sensors motivates the search for practical and efficient solutions that help data categorization. The extraction of knowledge about them, when time-dependent, requires special treatment and mining data becomes a valuable activity. In this paper, we propose an approach called DSP-Class for time series classification using texture descriptors applied Recurrence Charts (RP). We adopt 14 real datasets related to sound recognition, signals processing, chemical reactions and another produced by Big Data which demand high processing capacity. The purpose of this research is to verify the use of textural features in machine learning algorithms, such as SVM and C 5:0, applying Decomposition Empirical Mode (EMD) in time series classification. It is also analysed the stochastic-deterministic influence present in data streams. It presented poor results of 1-Nearest Neighbour algorithm when the data stream is mostly deterministic or stochastic. Our approach outperforms a traditional preprocessing approach applied on an audio stream using coefficients as features in around 18;67% of average accuracy and in around 67;66% the state-of-art algorithm that uses distance as measure.