Enhancing the selection of a model-based clustering with external categorical variables

In cluster analysis, it can be useful to interpret the partition built from the data in the light of external categorical variables which are not directly involved to cluster the data. An approach is proposed in the model-based clustering context to select a number of clusters which both fits the data well and takes advantage of the potential illustrative ability of the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is noteworthy that each mixture model is fitted by the maximum likelihood methodology to the data, excluding the external variables which are used to select a relevant mixture model only. Numerical experiments illustrate the promising behaviour of the derived criterion. © 2014 Springer-Verlag Berlin Heidelberg.

Integração Intranet – Pro/Engineer Wildfire 3.0 Via Pro/Web.Link

Mestrado em Engenharia Informática, Área de Especialização em Arquiteturas, Sistemas e Redes

A comprehensive high-throughput FTIR spectroscopy-based method for evaluating the transfection event: estimating the transfection efficiency and extracting associated metabolic responses

Reporter genes are routinely used in every laboratory for molecular and cellular biology for studying heterologous gene expression and general cellular biological mechanisms, such as transfection processes. Although well characterized and broadly implemented, reporter genes present serious limitations, either by involving time-consuming procedures or by presenting possible side effects on the expression of the heterologous gene or even in the general cellular metabolism. Fourier transform mid-infrared (FT-MIR) spectroscopy was evaluated to simultaneously analyze in a rapid (minutes) and high-throughput mode (using 96-wells microplates), the transfection efficiency, and the effect of the transfection process on the host cell biochemical composition and metabolism. Semi-adherent HEK and adherent AGS cell lines, transfected with the plasmid pVAX-GFP using Lipofectamine, were used as model systems. Good partial least squares (PLS) models were built to estimate the transfection efficiency, either considering each cell line independently (R 2 ≥ 0.92; RMSECV ≤ 2 %) or simultaneously considering both cell lines (R 2 = 0.90; RMSECV = 2 %). Additionally, the effect of the transfection process on the HEK cell biochemical and metabolic features could be evaluated directly from the FT-IR spectra. Due to the high sensitivity of the technique, it was also possible to discriminate the effect of the transfection process from the transfection reagent on KEK cells, e.g., by the analysis of spectral biomarkers and biochemical and metabolic features. The present results are far beyond what any reporter gene assay or other specific probe can offer for these purposes.

Enhancing the selection of a model-based clustering with external categorical variables

In cluster analysis, it can be useful to interpret the partition built from the data in the light of external categorical variables which are not directly involved to cluster the data. An approach is proposed in the model-based clustering context to select a number of clusters which both fits the data well and takes advantage of the potential illustrative ability of the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is noteworthy that each mixture model is fitted by the maximum likelihood methodology to the data, excluding the external variables which are used to select a relevant mixture model only. Numerical experiments illustrate the promising behaviour of the derived criterion.

Multiobjective optimization of viscoelastic laminated sandwich structures using the Direct MultiSearch method

A multiobjective approach for optimization of passive damping for vibration reduction in sandwich structures is presented in this paper. Constrained optimization is conducted for maximization of modal loss factors and minimization of weight of sandwich beams and plates with elastic laminated constraining layers and a viscoelastic core, with layer thickness and material and laminate layer ply orientation angles as design variables. The problem is solved using the Direct MultiSearch (DMS) solver for derivative-free multiobjective optimization and solutions are compared with alternative ones obtained using genetic algorithms.

Hopelessness and Related Variables among Cancer Patients in the Southern European Psycho-Oncology Study (SEPOS).

BACKGROUND: Although hopelessness has been studied in cancer, no data are available in non-English-speaking countries. OBJECTIVE: The authors sought to amass data from Southern European countries (Italy, Portugal, Spain, and Switzerland) in order to fill this void. METHOD: A group of 312 cancer patients completed the Mini-MAC Hopelessness subscale, the Hospital Anxiety and Depression Scale (HADS), the Cancer Worry Inventory (CWI), and a six-item Visual Analog scale (VAS) to measure intensity of physical symptoms, general well-being, difficulty in coping with cancer, intensity of social support from close relationships, leisure activity, and support from religious beliefs. RESULTS: Regression analysis indicated that HADS-Depression, VAS Maladaptive Coping and Well-Being, and the CWI explained 42% of the variance. CONCLUSION: Hopelessness in cancer patients seems not exclusively to correspond to depression, but is related to various other psychosocial factors, such as maladaptive coping, as well.

Evaluation of Spatial interpolation techniques for mapping climate variables with low sample density: a case study using a new gridded dataset of Bangladesh

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

The sum-of-the-parts method: An international application

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Finance from the NOVA – School of Business and Economics

Contributo da curva de débito-volume na detecção da obstrução da via aérea central

RESUMO: Introdução: A obstrução da via aérea central (OVAC) refere-se a um processo patológico que conduz a limitação do fluxo de ar ao nível do espaço glótico e subglótico, traqueia e brônquios principais. O seu correcto diagnóstico e tratamento constituem um território de interesse e preocupação para os profissionais de saúde, e requerem um profundo conhecimento da sua etiologia, fisiologia, diagnóstico e opções terapêuticas dado o potencial em originar significativa morbilidade e mortalidade. A avaliação da OVAC abrange múltiplas vertentes, entre as quais se salienta o componente clínico (sinais e sintomas), a repercussão fisiopatológica (função respiratória) e o estudo imagiológico (TC do tórax e broncoscopia). A compilação destes dados associada à etiologia, constituem factores importantes para estabelecer o prognóstico, determinar a necessidade de tratamento ou delinear uma futura intervenção terapêutica. A broncoscopia é o Gold Standard de avaliação desta condição, mas desde há cerca de 40 anos a curva de débito-volume constitui uma ferramenta não invasiva de detecção de OVAC. Apesar deste método ser utilizado até os nossos dias, poucos têm sido os estudos com o objectivo de verificar a sensibilidade e especificidade da curva de débito-volume na detecção de OVAC, bem como averiguar a relação entre as alterações morfológicas e quantitativas da mesma com a localização, o tipo e o grau da obstrução. Material e Métodos: Entre 1 de Novembro de 2009 e 30 de Abril de 2010, os doentes com indicação para a realização de broncoscopia diagnóstica ou terapêutica na Unidade de Técnicas Invasivas Pneumológicas (UTIP) do Centro Hospitalar Lisboa Norte – Hospital Pulido Valente (CHLN – HPV) foram seleccionados de forma consecutiva de acordo com os critérios de inclusão e exclusão. As avaliações (broncoscopia, curva de débito-volume e avaliação da dispneia) realizaram-se com um intervalo de tempo máximo de sete dias. A broncoscopia flexível foi realizada segundo as normas da British Thoracic Society e as curvas de débito-volume segundo as normas da ATS/ERS TaskForce 2005. Para a avaliação da dispneia recorreu-se à escala de dispneia MRC (Medical Research Council). Um painel de peritos realizou a avaliação da morfologia da curva de débito-volume (sugestiva ou não de OVAC) e um elemento independente a verificação dos critérios quantitativos e morfológicos (variáveis intra e extratorácica e fixa) da curva. O estudo foi aprovado pela Comissão de Ética para a Saúde do CHLN e todos os doentes assinaram um consentimento informado de participação. Resultados: Estudaram-se 82 doentes, 36 (44%) dos quais com OVAC. A predominância foi do género masculino, em relação ao feminino. A sensibilidade e especificidade dos critérios quantitativos da curva de débito-volume na detecção de OVAC foi de 91.3% e 88.9% respectivamente. Quando se utilizaram os critérios morfológicos da curva de débito-volume os valores foram de 93.5% e 30.6%. A agregação dos critérios morfológicos e quantitativos permitiu alcançar uma sensibilidade de 95.7% e especificidade de 86.1%. Nesta amostra, o critério quantitativo com maior ocorrência foi o FEF50/FIF50≥1 (83% dos doentes com OVAC). Este mostrou relacionar-se com todas as localizações de obstrução excepto o terço médio da traqueia. Mostrou, ainda, ter uma relação forte e positiva com o grau e tipo de obstrução (intra e extraluminal). O segundo foi o FEV1/PEF≥8, presente em 36% dos casos de OVAC. Relacionou-se com as obstruções no terço inferior da traqueia e brônquio principal direito (BPD). Também apresentou relação forte e positiva com o grau de obstrução e com os tipos de obstrução anteriormente descritos. Quanto à sintomatologia foi possível associar o grau de obstrução com o de dispneia e a presença de estridor com o grau e localização da obstrução na traqueia. Conclusões: Os resultados deste estudo demonstram que os critérios quantitativos da curva de débito-volume têm elevada sensibilidade e especificidade na detecção de OVAC. O critério FEV50/FIF50≥1 tem um bom poder discriminativo na detecção dessa condição, tendo sido relacionado com a localização, o grau e o tipo de obstrução. O critério FEV1/PEF≥8, embora com menor poder discriminativo, também se relaciona com o grau, a localização e o tipo de obstrução. A morfologia da curva tem uma boa sensibilidade mas baixa especificidade na detecção de OVAC, mas a agregação entre os critérios morfológicos e quantitativos aumenta a sensibilidade e especificidade. A dispneia e o estridor foram relacionados com o grau de obstrução e o último com a localização ao nível da traqueia.-------------ABSTRACT: Introduction: Central airway obstruction (CAO) refers to a pathological process that leads to restriction of airflow at the level of the glottis and subglottis, trachea and main bronchi. It’s proper diagnosis and treatment is an area of interest and concern to health professionals, and requires a deep knowledge of its etiology, physiology, diagnosis and treatment options, concerning the potential to cause significant morbidity and mortality. The evaluation of CAO covers multiple aspects: the clinical component (signs and symptoms), the pathophysiological effect (lung function) and the imaging study (bronchoscopy and chest CT). The compilation of this data associated with the etiology, are important for establishing prognosis, determine the need for treatment or outline a future therapeutic intervention. Bronchoscopy is the gold standard for evaluating this condition, but for the last 40 years the flow-volume loop has been used as a noninvasive tool for detecting CAO. Although this method is still in use, only few studies were made in order to verify its sensitivity and specificity in detecting CAO, and investigate the relation between morphological and quantitative changes of the curve to location, type and degree of obstruction. Methods: Between 1st November 2009 and 30th April 2010, patients with indication to perform diagnostic or therapeutic bronchoscopy in Interventional Pulmonology Unit - Hospital Pulido Valente (CHLN - HPV), were selected consecutively according to the inclusion and exclusion criteria. All assessments (bronchoscopy, flow-volume loop and dyspnea) were carried out within a period of seven days. The flexible bronchoscopy was performed according to the standards of the British Thoracic Society and the flow-volume loops in accordance with the standards of the ATS / ERS Taskforce 2005. For the evaluation of dyspnea was used to MRC dyspnea scale (Medical Research Council). A panel of experts evaluated the morphology of flow-volume loop (suggestive or non-suggestive of CAO) and an independent element established the quantitative criteria and morphological (intra and extrathoracic variables and fixed) of the curve. This study was approved by the Ethics Committee for Health CHLN and all the patients signed an informed consent to participate. Results: We’ve studied 82 patients, 36 (44%) of those with CAO. The majority of the patients were males, compared to females. The sensitivity and specificity of the quantitative criteria of the flow-volume curve in detecting CAO was 91.3% and 88.9% respectively. When we used the morphological criteria of flow-volume loop these values were 93.5% and 30.6%. The combination of quantitative and morphological criteria produced values of 95.7% sensitivity and 86.1% specificity. FEF50/FIF50≥1 was the most representative quantitative criterion (83% of patients with CAO) and it was correlated with all sites of obstruction except in the middle third of the trachea. It has shown a strong and positive association with the degree and type of obstruction (intra and extraluminal). The second was the FEV1/PEF ≥ 8, present in 36% of cases of CAO. It could be correlated with the obstruction in the lower third of the trachea and right main bronchus. It also showed a strong positive relation with the degree and types of obstruction described above. Regarding symptoms, we found a link between the degree of obstruction and dyspnea. The presence of stridor was correlated with the location and the degree of obstruction in the trachea. Conclusion: The results of this study demonstrate that the quantitative criteria of the flow-volume loop have a high sensitivity and specificity in detecting CAO. The criterion FEV50/FIF50 ≥ 1 has a good discriminative power to detect this condition and was related to the location, degree and type of obstruction. The criterion FEV1/PEF ≥ 8, although with a weaker discriminative power, also relates to the degree, location and type of obstruction. The morphology of the curve has a good sensitivity but low specificity in detecting CAO although the combination between the morphological and quantitative criteria increases sensitivity and specificity. Dyspnea and stridor were related to the degree of obstruction and the last one with its location in the trachea.

Using data mining for prediction of hospital length of stay: an application of the CRISP-DM Methodology

Hospitals are nowadays collecting vast amounts of data related with patient records. All this data hold valuable knowledge that can be used to improve hospital decision making. Data mining techniques aim precisely at the extraction of useful knowledge from raw data. This work describes an implementation of a medical data mining project approach based on the CRISP-DM methodology. Recent real-world data, from 2000 to 2013, were collected from a Portuguese hospital and related with inpatient hospitalization. The goal was to predict generic hospital Length Of Stay based on indicators that are commonly available at the hospitalization process (e.g., gender, age, episode type, medical specialty). At the data preparation stage, the data were cleaned and variables were selected and transformed, leading to 14 inputs. Next, at the modeling stage, a regression approach was adopted, where six learning methods were compared: Average Prediction, Multiple Regression, Decision Tree, Artificial Neural Network ensemble, Support Vector Machine and Random Forest. The best learning model was obtained by the Random Forest method, which presents a high quality coefficient of determination value (0.81). This model was then opened by using a sensitivity analysis procedure that revealed three influential input attributes: the hospital episode type, the physical service where the patient is hospitalized and the associated medical specialty. Such extracted knowledge confirmed that the obtained predictive model is credible and with potential value for supporting decisions of hospital managers.

Using customer lifetime value to improve the prediction of bank deposit subscription in telemarketing campaigns

Customer lifetime value (LTV) enables using client characteristics, such as recency, frequency and monetary (RFM) value, to describe the value of a client through time in terms of profitability. We present the concept of LTV applied to telemarketing for improving the return-on-investment, using a recent (from 2008 to 2013) and real case study of bank campaigns to sell long- term deposits. The goal was to benefit from past contacts history to extract additional knowledge. A total of twelve LTV input variables were tested, un- der a forward selection method and using a realistic rolling windows scheme, highlighting the validity of five new LTV features. The results achieved by our LTV data-driven approach using neural networks allowed an improvement up to 4 pp in the Lift cumulative curve for targeting the deposit subscribers when compared with a baseline model (with no history data). Explanatory knowledge was also extracted from the proposed model, revealing two highly relevant LTV features, the last result of the previous campaign to sell the same product and the frequency of past client successes. The obtained results are particularly valuable for contact center companies, which can improve pre- dictive performance without even having to ask for more information to the companies they serve.

Clustering barotrauma patients in ICU–A data mining based approach using ventilator variables

Lecture Notes in Computer Science, 9273

Predicting the distribution of forest tree species using topographic variables and vegetation index in eastern Acre, Brazil

Species distribution modeling has relevant implications for the studies of biodiversity, decision making about conservation and knowledge about ecological requirements of the species. The aim of this study was to evaluate if the use of forest inventories can improve the estimation of occurrence probability, identify the limits of the potential distribution and habitat preference of a group of timber tree species. The environmental predictor variables were: elevation, slope, aspect, normalized difference vegetation index (NDVI) and height above the nearest drainage (HAND). To estimate the distribution of species we used the maximum entropy method (Maxent). In comparison with a random distribution, using topographic variables and vegetation index as features, the Maxent method predicted with an average accuracy of 86% the geographical distribution of studied species. The altitude and NDVI were the most important variables. There were limitations to the interpolation of the models for non-sampled locations and that are outside of the elevation gradient associated with the occurrence data in approximately 7% of the basin area. Ceiba pentandra (samaúma), Castilla ulei (caucho) and Hura crepitans (assacu) is more likely to occur in nearby water course areas. Clarisia racemosa (guariúba), Amburana acreana (cerejeira), Aspidosperma macrocarpon (pereiro), Apuleia leiocarpa (cumaru cetim), Aspidosperma parvifolium (amarelão) and Astronium lecointei (aroeira) can also occur in upland forest and well drained soils. This modeling approach has potential for application on other tropical species still less studied, especially those that are under pressure from logging.

A Methodology to measure shareholder value orientation and shareholder value creation aimed at providing a research basis to investigate the link between both magnitudes

Does shareholder value orientation lead to shareholder value creation? This article proposes methods to quantify both, shareholder value orientation and shareholder value creation. Through the application of these models it is possible to quantify both dimensions and examine statistically in how far shareholder value orientation explains shareholder value creation. The scoring model developed in this paper allows quantifying the orientation of managers towards the objective to maximize wealth of shareholders. The method evaluates information that comes from the companies and scores the value orientation in a scale from 0 to 10 points. Analytically the variable value orientation is operationalized expressing it as the general attitude of managers toward the objective of value creation, investment policy and behavior, flexibility and further eight value drivers. The value creation model works with market data such as stock prices and dividend payments. Both methods where applied to a sample of 38 blue chip companies: 32 firms belonged to the share index IBEX 35 on July 1st, 1999, one company represents the “new economy” listed in the Spanish New Market as per July 1st, 2001, and 5 European multinational groups formed part of the EuroStoxx 50 index also on July 1st, 2001. The research period comprised the financial years 1998, 1999, and 2000. A regression analysis showed that between 15.9% and 23.4% of shareholder value creation can be explained by shareholder value orientation.

Insertion dans les soins après une première hospitalisation dans un secteur pour psychose [Linkage to care after first hospitalisation for psychosis]

BACKGROUND: First hospitalisation for a psychotic episode causes intense distress to patients and families, but offers an opportunity to make a diagnosis and start treatment. However, linkage to outpatient psychiatric care remains a notoriously difficult step for young psychotic patients, who frequently interrupt treatment after hospitalisation. Persistence of symptoms, and untreated psychosis may therefore remain a problem despite hospitalisation and proper diagnosis. With persisting psychotic symptoms, numerous complications may arise: breakdown in relationships, loss of family and social support, loss of employment or study interruption, denial of disease, depression, suicide, substance abuse and violence. Understanding mechanisms that might promote linkage to outpatient psychiatric care is therefore a critical issue, especially in early intervention in psychotic disorders. OBJECTIVE: To study which factors hinder or promote linkage of young psychotic patients to outpatient psychiatric care after a first hospitalisation, in the absence of a vertically integrated program for early psychosis. Method. File audit study of all patients aged 18 to 30 who were admitted for the first time to the psychiatric University Hospital of Lausanne in the year 2000. For statistical analysis, chi2 tests were used for categorical variables and t-test for dimensional variables; p<0.05 was considered as statistically significant. RESULTS: 230 patients aged 18 to 30 were admitted to the Lausanne University psychiatric hospital for the first time during the year 2000, 52 of them with a diagnosis of psychosis (23%). Patients with psychosis were mostly male (83%) when compared with non-psychosis patients (49%). Furthermore, they had (1) 10 days longer mean duration of stay (24 vs 14 days), (2) a higher rate of compulsory admissions (53% vs 22%) and (3) were more often hospitalised by a psychiatrist rather than by a general practitioner (83% vs 53%). Other socio-demographic and clinical features at admission were similar in the two groups. Among the 52 psychotic patients, 10 did not stay in the catchment area for subsequent treatment. Among the 42 psychotic patients who remained in the catchment area after discharge, 20 (48%) did not attend the scheduled or rescheduled outpatient appointment. None of the socio demographic characteristics were associated with attendance to outpatient appointments. On the other hand, voluntary admission and suicidal ideation before admission were significantly related to attending the initial appointment. Moreover, some elements of treatment seemed to be associated with higher likelihood to attend outpatient treatment: (1) provision of information to the patient regarding diagnosis, (2) discussion about the treatment plan between in- and outpatient staff, (3) involvement of outpatient team during hospitalisation, and (4) elaboration of concrete strategies to face basic needs, organise daily activities or education and reach for help in case of need. CONCLUSION: As in other studies, half of the patients admitted for a first psychotic episode failed to link to outpatient psychiatric care. Our study suggests that treatment rather than patient's characteristics play a critical role in this phenomenon. Development of a partnership and involvement of patients in the decision process, provision of good information regarding the illness, clear definition of the treatment plan, development of concrete strategies to cope with the illness and its potential complications, and involvement of the outpatient treating team already during hospitalisation, all came out as critical strategies to facilitate adherence to outpatient care. While the current rate of disengagement after admission is highly concerning, our finding are encouraging since they constitute strategies that can easily be implemented. An open approach to psychosis, the development of partnership with patients and a better coordination between inpatient and outpatient teams should therefore be among the targets of early intervention programs. These observations might help setting up priorities when conceptualising new programs and facilitate the implementation of services that facilitate engagement of patients in treatment during the critical initial phase of psychotic disorders.