Automatic musical instrument recognition for multimedia indexing

Trabalho apresentado no âmbito do Mestrado em Engenharia Informática, como requisito parcial para obtenção do grau de Mestre em Engenharia Informática

Espresso beverages of pure origin coffee: Mineral characterization, contribution for mineral intake and geographical discrimination

Espresso coffee beverages prepared from pure origin roasted ground coffees from the major world growing regions (Brazil, Ethiopia, Colombia, India, Mexico, Honduras, Guatemala, Papua New Guinea, Kenya, Cuba, Timor, Mussulo and China) were characterized and compared in terms of their mineral content. Regular consumption of one cup of espresso contributes to a daily mineral intake varying from 0.002% (sodium; Central America) to 8.73% (potassium; Asia). The mineral profiles of the espresso beverages revealed significant inter- and intra-continental differences. South American pure origin coffees are on average richer in the analyzed elements except for calcium, while samples from Central America have generally lower mineral amounts (except for manganese). Manganese displayed significant differences (p < 0.05) among the countries of each characterized continent. Intercontinental and inter-country discrimination between the major world coffee producers were achieved by applying canonical discriminant analysis. Manganese and calcium were found to be the best chemical descriptors for origin.

A área curricular de matemática nos programas educativos individuais

Em acordo com o Dec. Lei nº 3/2008 de 7 de janeiro e para alunos com necessidades educativas especiais a medida currículo específico individual é considerada a mais restritiva de todas as medidas educativas. A área disciplinar da matemática, pela sua aplicabilidade no quotidiano, assume primordial importância no Programa Educativo Individual (PEI) destes alunos. Assim, o presente estudo visa analisar a área curricular de matemática dos PEI de alunos a frequentar o 2º e 3º ciclo de ensino básico ao abrigo da medida educativa currículo específico individual (CEI); visa igualmente constatar que seleção de conteúdos programáticos são percecionados como prioritários para a equipa que elabora o PEI. Em suma, o estudo visa compreender alguns aspetos que, de forma direta ou indireta, interagem com a elaboração do currículo. Tem, ainda, um caráter exploratório e está apoiado numa metodologia de natureza qualitativa e quantitativa (numa dimensão descritiva) que procede à análise documental de excertos (área curricular de matemática) dos Programas Educativos Individuais (PEI). Para o efeito foram analisados 50 PEI que identificaram regularidades relativas aos diferentes conteúdos e à extensão de cada conteúdo. Os resultados evidenciam uma escolha maioritária de conteúdos matemáticos associados ao programa do 1º ano do 1º ciclo do ensino básico e, simultaneamente, de descritores associados aos números e operações. Os resultados permitem extrapolar acerca da interação entre níveis de programação e de funcionalidade dos alunos em CEI e requerem mais estudos que sustentem aquelas evidências e clarifiquem variáveis que interagem na elaboração do currículo.

Automatic extraction of concepts from texts and applications

The extraction of relevant terms from texts is an extensively researched task in Text- Mining. Relevant terms have been applied in areas such as Information Retrieval or document clustering and classification. However, relevance has a rather fuzzy nature since the classification of some terms as relevant or not relevant is not consensual. For instance, while words such as "president" and "republic" are generally considered relevant by human evaluators, and words like "the" and "or" are not, terms such as "read" and "finish" gather no consensus about their semantic and informativeness. Concepts, on the other hand, have a less fuzzy nature. Therefore, instead of deciding on the relevance of a term during the extraction phase, as most extractors do, I propose to first extract, from texts, what I have called generic concepts (all concepts) and postpone the decision about relevance for downstream applications, accordingly to their needs. For instance, a keyword extractor may assume that the most relevant keywords are the most frequent concepts on the documents. Moreover, most statistical extractors are incapable of extracting single-word and multi-word expressions using the same methodology. These factors led to the development of the ConceptExtractor, a statistical and language-independent methodology which is explained in Part I of this thesis. In Part II, I will show that the automatic extraction of concepts has great applicability. For instance, for the extraction of keywords from documents, using the Tf-Idf metric only on concepts yields better results than using Tf-Idf without concepts, specially for multi-words. In addition, since concepts can be semantically related to other concepts, this allows us to build implicit document descriptors. These applications led to published work. Finally, I will present some work that, although not published yet, is briefly discussed in this document.

Health-related quality of life in patients with Chagas disease: a review of the evidence

Chagas disease (ChD), a neglected tropical disease caused by infection with the parasite Trypanosoma cruzi (T. cruzi), remains a serious public health issue in Latin America and is an emerging disease in several non-endemic countries, where knowledge of the condition and experience with its clinical management are limited. Regionally, the disease is the major cause of disability secondary to tropical diseases in young adults. Health-related quality of life (HRQoL) impairment is common in patients with ChD, especially in those with Chagas dilated cardiomyopathy, the most severe manifestation of the disease, which frequently leads to heart failure. The aim of this review was to conduct a literature search for studies that have evaluated the determining factors of HRQoL in ChD patients. We included cross-sectional, case-control, cohort, and experimental studies, as well as clinical trials that evaluated the HRQoL in ChD patients aged 18 to 60 years and are presenting an explicit description of statistical analysis. Using a combination of keywords based on Descriptors in Health Sciences (DeCS) and Medical Subject Headings (MeSH) for searches in PubMed and the Scientific Electronic Library Online (SciELO), 148 studies were found. After exclusions, 12 studies were selected for analysis. Three main findings were extracted from these studies: 1) cardiac involvement is associated with a worse HRQoL in ChD patients; 2) HRQoL is associated with the patients' functional capacity; and 3) simple and inexpensive therapeutic measures are effective for improving HRQoL in ChD patients. Hence, ChD patients' functional capacity, the effectiveness of non-surgical conservative treatment, and cardiac involvement are important determining factors for the HRQoL in ChD patients.

Pain evaluation of patients with fibromyalgia, osteoarthritis, and low back pain

The purpose of this study was to evaluate and compare pain as reported by outpatients with fibromyalgia, osteoarthritis, and low back pain, in view of designing more adequate physical therapy treatment. PATIENTS AND METHODS: A Portuguese version of the McGill Pain Questionnaire - where subjects are asked to choose, from lists of pre-categorized words, one or none that best describes what they feel - was used to assess pain intensity and quality of 64 patients, of which 24 had fibromyalgia, 22 had osteoarthritis, and 18 had low back pain. The pre-categorized words were organized into 4 major classes -- sensory, affective, evaluative, and miscellaneous. RESULTS: Patients with fibromyalgia reported, comparatively, more intense pain through their choice of pain descriptors, both sensory and affective; they also chose a higher number of words from these classes than patients in the other groups and were the only ones to choose specific affective descriptors such as "vicious", "wretched", "exhausting", "blinding". CONCLUSION: Assuming that each disease presents unique qualities of pain experience, and that these can be pointed out by means of this questionnaire by patients' choice of specific groups of words, the findings suggest that fibromyalgia include not only a physical component, but also a psycho-emotional component, indicating that they require both emotional/affective and physical care.

Hand gesture recognition for human computer interaction : a comparative study of different image features

Hand gesture recognition for human computer interaction, being a natural way of human computer interaction, is an area of active research in computer vision and machine learning. This is an area with many different possible applications, giving users a simpler and more natural way to communicate with robots/systems interfaces, without the need for extra devices. So, the primary goal of gesture recognition research is to create systems, which can identify specific human gestures and use them to convey information or for device control. For that, vision-based hand gesture interfaces require fast and extremely robust hand detection, and gesture recognition in real time. In this study we try to identify hand features that, isolated, respond better in various situations in human-computer interaction. The extracted features are used to train a set of classifiers with the help of RapidMiner in order to find the best learner. A dataset with our own gesture vocabulary consisted of 10 gestures, recorded from 20 users was created for later processing. Experimental results show that the radial signature and the centroid distance are the features that when used separately obtain better results, with an accuracy of 91% and 90,1% respectively obtained with a Neural Network classifier. These to methods have also the advantage of being simple in terms of computational complexity, which make them good candidates for real-time hand gesture recognition.

Construção e validação de uma escala de autoconceito para estudantes universitários de Moçambique

O presente artigo apresenta a construção e validação de um questionário de avaliação do autoconceito para a população de adolescentes e jovens-adultos que frequentam o ensino universitário em Moçambique. A construção do questionário teve como base a análise de conteúdo de descrições de si, fornecidas por 15 estudantes, e que informaram acerca das categorias centrais ao autoconceito e seus principais descritores. Com base nos resultados deste primeiro estudo, foram elaborados 77 itens que foram sucessivamente aplicados a pequenos grupos de alunos/as e a diferentes amostras, realizando-se análises qualitativas e quantitativas das respostas aos itens de modo a proceder à definição de dimensionalidade da escala e escolha dos melhores itens. Num último estudo, a versão ultimada da escala foi administrada a uma amostra de 250 estudantes (Midade=29.0, DP= 7.70). Os resultados da análise de componentes principais identificaram 24 itens que se organizam em cinco dimensões de autoconceito: autoconceito religioso, artístico, académico, social e físico. Estas cinco dimensões explicam 56.8% da variância total dos itens retidos na versão final da escala. As propriedades psicométricas são favoráveis à utilização deste instrumento de avaliação do autoconceito em estudantes universitários de Moçambique.

Women users of crack: systematic review of Brazilian literature

ABSTRACT Objective To present Brazilian’s empirical studies that address this issue between the period of 2004 to 2014. Methods It is a Brazilian literature Systematic Review using the descriptors “crack cocaine” AND “women”, in the database Scopus, Lilacs, Medline and SciELO. Results From the 785 articles found, 16 articles contemplated the inclusion criteria. It was evidenced that the use of crack by women is related to physical and sexual violence, provoking HIV risks in consequence of prostitution, and social prejudice. Conclusion Given this reality, studies evaluating treatments in the Brazilian context are essential, according the specificities of women crack users.

Clinical and echocardiographic predictors of mortality in chagasic cardiomyopathy - systematic review

Diagnosis, prognosis and evaluation of death risk in Chagas cardiomyopathy still constitute a challenge due to the diversity of manifestations, which determine the importance of using echocardiography, tissue Doppler and biomarkers. To evaluate, within a systematic review, clinical and echocardiographic profiles of patients with chronic chagasic cardiomyopathy, which may be related to worse prognosis and major mortality risk. To perform the systematic review, we used Medline (via PubMed), LILACS and SciELO databases to identify 82 articles published from 1991 to 2012, with the following descriptors: echocardiography, mortality and Chagas disease. We selected 31 original articles, involving diagnostic and prognostic methods. The importance of Chagas disease has increased due to its emergence in Europe and United States, but most evidence came from Brazil. Among the predictors of worse prognosis and higher mortality risk are morphological and functional alterations in the left and right ventricles, evaluated by conventional echocardiography and tissue Doppler, as well as the increase in brain natriuretic peptide and troponin I concentrations. Recently, the evaluations of dyssynchrony, dysautonomia, as well as strain, strain rate and myocardial twisting were added to the diagnostic arsenal for the early differentiation of Chagas cardiomyopathy. Developments in imaging and biochemical diagnostic procedures have enabled more detailed cardiac evaluations, which demonstrate the early involvement of both ventricles, allowing a more accurate assessment of the mortality risk in Chagas disease.

Nest site selection by Hypsiboas faber(Anura, Hylidae) in Southern Brazil

ABSTRACT Male gladiator frogs of Hypsiboas Wagler, 1830 build nests on available substrate surrounding ponds and streams where female spawn eggs during the breeding period. Although gladiator frogs seem to show plasticity in the way they construct their nests, there is no study reporting if these species present preferences about microhabitat conditions for nest-building (mainly under subtropical climate). Predation pressure and environmental conditions have been considered major processes shaping the great diversity of reproductive strategies performed by amphibians, but microhabitat conditions should explain where to build a nest as well as how nest looks. This study aimed to test nest site selection for nest-building by Hypsiboas faber(Wied-Neuwied, 1821), determining which factors are related to nest site selection and nest features. The survey was conducted at margins of two permanent ponds in Southern Brazil. Habitat factors were evaluated in 18 plots with nest and 18 plots in the surrounding without nest (control), describing vegetation structure and heterogeneity, and substrate characteristics. Water temperature was measured inside the nest and in its adjacency. Nest features assessed were area, depth and temperature. Habitat characteristics differed between plots with and without nest. Microhabitat selected for nest-building was characterized by great vegetation cover and height, as well as shallower water and lower cover of organic matter in suspension than in plots without nest. Differences between temperature inside nest and in its adjacency were not observed. No relationship between nest features and habitat descriptors was evidenced. Results revealed that Hypsiboas faber does not build nests anywhere. Males seem to prefer more protected habitats, probably avoiding predation, invasion of conspecific males and inclement weather. Lack of differences between temperature inside- and outside-nest suggest that nest do not improve this condition for eggs and tadpole development. Nest architecture was not related to habitat characteristics, which may be determined by other factors, as nest checking by females before amplexus. Nest site selection should increase offspring survival as well the breeding success of Hypsiboas faber.

Estudi de casos pel control de síntesi d’instruments musicals virtuals mitjançant la veu cantada

Report for the scientific sojourn carried out at the Music Technology Area (Sound Processing and Control Lab), Faculty of Music, McGill University, Montreal, Canada, from October to December 2005.The aim of this research is to study the singing voice for controlling virtual musical instrument synthesis. It includes analysis and synthesis algorithms based on spectral audio processing. After digitalising the acoustic voice signal in the computer, a number of expressive descriptors of the singer are extracted. This process is achieved synchronously, thus all the nuance of the singer performance have been tracked. In a second stage, the extracted parameters are mapped to a sound synthesizer, the so-called digital musical instruments. In order achieve it, several tests with music students of the Faculty of Music, McGill University have been developed. These experiments have contributed to evaluate the system and to derive new control strategies to integrate: clarinet synthesis, bass guitar, visual representation of voice signals.

Evaluation de la douleur chez les personnes cérébrolésées non-communicantes aux soins intensifs

La douleur est fréquente en milieu de soins intensifs et sa gestion est l'une des missions des infirmières. Son évaluation est une prémisse indispensable à son soulagement. Cependant lorsque le patient est incapable de signaler sa douleur, les infirmières doivent se baser sur des signes externes pour l'évaluer. Les guides de bonne pratique recommandent chez les personnes non communicantes l'usage d'un instrument validé pour la population donnée et basé sur l'observation des comportements. A l'heure actuelle, les instruments d'évaluation de la douleur disponibles ne sont que partiellement adaptés aux personnes cérébrolésées dans la mesure où ces personnes présentent des comportements qui leur sont spécifiques. C'est pourquoi, cette étude vise à identifier, décrire et valider des indicateurs, et des descripteurs, de la douleur chez les personnes cérébrolésées. Un devis d'étude mixte multiphase avec une dominante quantitative a été choisi pour cette étude. Une première phase consistait à identifier des indicateurs et des descripteurs de la douleur chez les personnes cérébrolésées non communicantes aux soins intensifs en combinant trois sources de données : une revue intégrative des écrits, une démarche consultative utilisant la technique du groupe nominal auprès de 18 cliniciens expérimentés (6 médecins et 12 infirmières) et les résultats d'une étude pilote observationnelle réalisée auprès de 10 traumatisés crâniens. Les résultats ont permis d'identifier 6 indicateurs et 47 descripteurs comportementaux, vocaux et physiologiques susceptibles d'être inclus dans un instrument d'évaluation de la douleur destiné aux personnes cérébrolésées non- communicantes aux soins intensifs. Une deuxième phase séquentielle vérifiait les propriétés psychométriques des indicateurs et des descripteurs préalablement identifiés. La validation de contenu a été testée auprès de 10 experts cliniques et 4 experts scientifiques à l'aide d'un questionnaire structuré qui cherchait à évaluer la pertinence et la clarté/compréhensibilité de chaque descripteur. Cette démarche a permis de sélectionner 33 des 47 descripteurs et valider 6 indicateurs. Dans un deuxième temps, les propriétés psychométriques de ces indicateurs et descripteurs ont été étudiés au repos, lors de stimulation non nociceptive et lors d'une stimulation nociceptive (la latéralisation du patient) auprès de 116 personnes cérébrolésées aux soins intensifs hospitalisées dans deux centres hospitaliers universitaires. Les résultats montrent d'importantes variations dans les descripteurs observés lors de stimulation nociceptive probablement dues à l'hétérogénéité des patients au niveau de leur état de conscience. Dix descripteurs ont été éliminés, car leur fréquence lors de la stimulation nociceptive était inférieure à 5% ou leur fiabilité insuffisante. Les descripteurs physiologiques ont tous été supprimés en raison de leur faible variabilité et d'une fiabilité inter juge problématique. Les résultats montrent que la validité concomitante, c'est-à-dire la corrélation entre l'auto- évaluation du patient et les mesures réalisées avec les descripteurs, est satisfaisante lors de stimulation nociceptive {rs=0,527, p=0,003, n=30). Par contre la validité convergente, qui vérifiait l'association entre l'évaluation de la douleur par l'infirmière en charge du patient et les mesures réalisés avec les descripteurs, ainsi que la validité divergente, qui vérifiait si les indicateurs discriminent entre la stimulation nociceptive et le repos, mettent en évidence des résultats variables en fonction de l'état de conscience des patients. Ces résultats soulignent la nécessité d'étudier les descripteurs de la douleur chez des patients cérébrolésés en fonction du niveau de conscience et de considérer l'hétérogénéité de cette population dans la conception d'un instrument d'évaluation de la douleur pour les personnes cérébrolésées non communicantes aux soins intensifs. - Pain is frequent in the intensive care unit (ICU) and its management is a major issue for nurses. The assessment of pain is a prerequisite for appropriate pain management. However, pain assessment is difficult when patients are unable to communicate about their experience and nurses have to base their evaluation on external signs. Clinical practice guidelines highlight the need to use behavioral scales that have been validated for nonverbal patients. Current behavioral pain tools for ICU patients unable to communicate may not be appropriate for nonverbal brain-injured ICU patients, as they demonstrate specific responses to pain. This study aimed to identify, describe and validate pain indicators and descriptors in brain-injured ICU patients. A mixed multiphase method design with a quantitative dominant was chosen for this study. The first phase aimed to identify indicators and descriptors of pain for nonverbal brain- injured ICU patients using data from three sources: an integrative literature review, a consultation using the nominal group technique with 18 experienced clinicians (12 nurses and 6 physicians) and the results of an observational pilot study with 10 traumatic brain injured patients. The results of this first phase identified 6 indicators and 47 behavioral, vocal and physiological descriptors of pain that could be included in a pain assessment tool for this population. The sequential phase two tested the psychometric properties of the list of previously identified indicators and descriptors. Content validity was tested with 10 clinical and 4 scientific experts for pertinence and comprehensibility using a structured questionnaire. This process resulted in 33 descriptors to be selected out of 47 previously identified, and six validated indicators. Then, the psychometric properties of the descriptors and indicators were tested at rest, during non nociceptive stimulation and nociceptive stimulation (turning) in a sample of 116 brain-injured ICLI patients who were hospitalized in two university centers. Results showed important variations in the descriptors observed during the nociceptive stimulation, probably due to the heterogeneity of patients' level of consciousness. Ten descriptors were excluded, as they were observed less than 5% of the time or their reliability was insufficient. All physiologic descriptors were deleted as they showed little variability and inter observer reliability was lacking. Concomitant validity, testing the association between patients' self report of pain and measures performed using the descriptors, was acceptable during nociceptive stimulation (rs=0,527, p=0,003, n=30). However, convergent validity ( testing for an association between the nurses' pain assessment and measures done with descriptors) and divergent validity (testing for the ability of the indicators to discriminate between rest and a nociceptive stimulation) varied according to the level of consciousness These results highlight the need to study pain descriptors in brain-injured patients with different level of consciousness and to take into account the heterogeneity of this population forthe conception of a pain assessment tool for nonverbal brain-injured ICU patients.

Indexació de continguts televisius: anàlisis de vídeo

Aquest és un projecte que tracta sobre la indexació automàtica de continguts televisius. És una tasca que guanyarà importància amb els imminents canvis que hi haurà en la televisió que coneixem. L'entrada de la nova televisió digital farà que hi hagi una interacció molt més fluida entre l'espectador i la cadena, a més de grans quantitats de canals, cada un amb programes de tipus totalment diferents. Tot això farà que tenir mètodes de cerca basats en els continguts d'aquests programes sigui del tot imprescindible. Així doncs, el nostre projecte està basat plenament en poder extreure alguns d'aquests descriptors que faran possible la categorització dels diferents programes televisius.

Establiment de bases científiques per al desenvolupament d'una estratègia integrada per avaluar toxicitat en reproducció

Report for the scientific sojourn at the National Institute for Public Health and the Environment (RIVM), Netherlands, from 2006 to 2008. This project aimed at creating scientific elements that could help deriving an integrated testing strategy for reproductive toxicity. Part of the project focused on the use of alternative tests for regulatory purposes. An in vitro-in vivo extrapolation method for embryotoxicity was proposed and evaluated. In vitro and in vivo dose descriptors were correlated; however, the scatter in the correlation was too large to allow an accurate estimation of an in vivo dose from an in vitro dose. The in vitro-in vivo extrapolation method together with toxicokinetic data was also applied to a category of substances (phthalates). Although based on a limited number of substances, the results suggested that in vitro-in vivo extrapolation for embryotoxicity is possible within a category of compounds if adequate toxicokinetic data is available. Because of the limitations that still remain in the use of alternative tests for reproductive toxicicity, other approaches to reduce animal testing were explored. Thus, a database of reproductive toxicity studies was created to retrospectively evaluate the comparative value of some studies or elements in a particular study. When compared to the subchronic toxicity study, the rat two-generation reproductive toxicity study had a considerable impact on the identification of hazard for reproductive toxicity, but not on the overall NOAEL. Among the two-generation studies included in our database, the second generation affected neither the overall NOAEL nor the critical effect. The rat and the rabbit developmental toxicity studies were, on average, similarly sensitive. However, for certain substances the developmental study in either one of the two species appeared to be more sensitive than in the other species.