Search

Objective To check if shoulders with acetabularization have better functional results in cases of rotator cuff arthropathy. Methods A clinical and radiological cross-sectional evaluation of 65 shoulders with rotator cuff arthropathy by measuring the range of motion (RoM) of the shoulder, the Constant-Murley score, and the radiological classifications of Hamada and Seebauer. The clinical findings were compared with the radiographic findings. Results According to the classification of Seebauer, we observed better results regarding the RoM in type-A shoulders. There was a statistically significant difference regarding anterior elevation and medial rotation between types A and B (p< 0.05). Lateral rotation did not show a statistically significant difference between types A and B. The Constant-Murley score presented better results in type A, and there was a statistically significant difference between groups A and B (p< 0.05). According to the classification of Hamada, we observed that the RoM had better results in types 3, 2 and 1, and there was a statistically significant difference regarding anterior elevation and medial rotation (p< 0.05) compared with groups 4A, 4B and 5. There was no statistically significant difference between the Hamada groups regarding lateral rotation. According to Hamada, the Constant-Murley score showed better results in types 3, 1 and 2, and there was a statistically significant difference between groups 3 and 5. Conclusion The RoM and shoulder function were better in patients with acetabularization (Seebauer 1A and Hamada 3), and worse in those with glenohumeral arthrosis (Seebauer 1B, 2B and Hamada 4A, 4B and 5). Resumo Objetivo Verificar se os ombros com acetabularização têm melhores resultados funcionais nos casos de artropatia do manguito rotador. Métodos Avaliação transversal clínica e radiológica de 65 ombros com artropatia do manguito rotador por meio da mensuração da amplitude de movimento (ADM) do ombro, do escore de Constant-Murley, e das classificações radiológicas de Hamada e Seebauer. Os achados clínicos foram comparados com os radiográficos. Resultados Segundo a classificação de Seebauer, com relação à ADM, observamos melhores resultados nos tipos A. Houve diferença estatística significativa na elevação anterior, e rotação medial entre os tipos A e B (p< 0.05). A rotação lateral não demonstrou diferença estatística significativa entre os tipos A e B. O escore de Constant-Murley apresentou melhores resultados nos tipos A, e houve diferença estatística significativa entre os grupos A e B (p< 0,05). Segundo a classificação de Hamada, observamos que a ADM teve melhores resultados nos tipos 3, 2 e 1, e houve diferença estatística significativa para a elevação anterior e a rotação medial (p< 0,05) quando comparadas com os grupos 4A, 4B e 5. Não houve diferença estatística significativa entre os grupos de Hamada em relação à rotação lateral. Ainda segundo Hamada, o escore de Constant-Murley apresentou melhores resultados nos tipos 3, 1 e 2, e houve diferença estatística significativa entre os grupos 3 e 5. Conclusão A ADM e a função do ombro apresentavam-se melhores nos pacientes com acetabularização (Seebauer 1A e Hamada 3), e piores naqueles com artrose glenoumeral (Seebauer 1B, 2B e Hamada 4A, 4B e 5).

Dissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e Tecnologia Os dados gerados por anualmente rondam os 40 trilhões de gigabytes. Este aumento significativo de dados todos os anos trás a necessidade de assegurar a proteção de informação sensível. A Inteligência Artificial tem vindo a melhorar cada vez mais os seus resultados, apresentando modelos capazes de responder rigorosamente em áreas de atuação críticas, por exemplo, medicina, veículos autónomos, robótica, etc. Estes algoritmos precisam de enormes quantidades de dados disponíveis para otimizarem ao máximo a sua resposta perante todos a sua área de operação.Surgiu a necessidade de continuar a melhorar estes algoritmos mantendo a privacidade e confidencialidade dos dados utilizados.Desta forma, foi criado o conceito de Federated Learning. O Federated Learning permite continuar a treinar algoritmos de Machine Learning sem partilhar os dados utilizados para a convergência do modelo. O Federated Learning apresenta apresenta algumas similaridades com o Distributed Learning. Em ambos os conceitos o treino é distribuido, no entanto o Federated Learning descentraliza também os dados de forma a manter a informação privada.O objetivo desta dissertação passa por explorar o conceito de Federated Learning, assim como comparar diretamente este conceito com o Machine Learning centralizado. Para tal, é mostrada a arquitetura necessária para a construção de uma solução federada. Este documento apresenta ainda resultados obtidos com soluções federadas tanto em ambiente de simulação como numa implementação em ambiente real. Finalmente, é também apresentado um ponto de vista dos resultados obtidos e opções de otimização de uma solução com Federated Learning são discutidas. The data generated annually is around 40 trillion gigabytes. This significant increase in data every year brings with it the need to ensure the protection of sensitive information. Artificial Intelligence has been improving its results more and more, presenting models capable of responding rigorously in critical areas, for example medicine, autonomous vehicles, robotics, etc. These algorithms need huge amounts of available data to optimize their response to all their area of operation.The urge to continue to improve these algorithms while maintaining the privacy and confidentiality of the data used emerged.Thus, the concept of Federated Learning was created. Federated Learning allows to continue training Machine Learning algorithms without sharing the data used for model convergence. Federated Learning has some similarities with Distributed Learning. In both concepts the training is distributed, however, Federated Learning also decentralizes the data in order to keep the information private.The objective of this dissertation is to explore the concept of Federated Learning, as well as to directly compare this concept with centralized Machine Learning. To this end, the architecture required to build a federated solution is analyzed in depth. This dissertation also presents results obtained with federated solutions in both simulation and real-world deployment. Finally, a viewpoint of the obtained results is also presented, and options for optimizing a solution with Federated Learning are discussed.

Background: Tuberculosis (TB) remains a major health problem within the Community of Portuguese Language Speaking Countries (CPLP). Despite the marked variation in TB incidence across its member-states and continued human migratory flux between countries, a considerable gap in the knowledge on the Mycobacterium tuberculosis population structure and strain circulation between the countries still exists. Materials and Methods: To address this, we have assembled and analyzed the largest CPLP M. tuberculosis molecular and drug susceptibility dataset, comprised by a total of 1447 clinical isolates, including 423 multidrug-resistant isolates, from five CPLP countries. Genotyping analysis was carried out by 15/24 Mycobacterial Interspersed Repetitive Unit – Variable Number of Tandem Repeat (MIRU-VNTR) and Spoligotyping. Drug Susceptibility testing was performed using standardized BACTEC 960 MGIT methodology or through the resazurin microtiter assay (REMA). Results: The data herein presented reinforces Latin American and Mediterranean (LAM) strains as the hallmark of M. tuberculosis populational structure in the CPLP coupled with country-specific differential prevalence of minor clades. Moreover, using high-resolution typing by 24-loci MIRU-VNTR, six cross-border genetic clusters were detected, thus supporting recent clonal expansion across the Lusophone space. To make this data available to the scientific community and public health authorities we developed CPLP-TB (available at http://cplp-tb.ff.ulisboa.pt), an online database coupled with web-based tools for exploratory data analysis. Conclusions: As a public health tool, CPLP-TB is expected to contribute to improved knowledge on the M. tuberculosis population structure and strain circulation within the CPLP, thus supporting risk assessment of strain-specific trends. N/A

Rifampicin is one of the most important chemotherapeutic agents used in the treatment of tuberculosis. M. tuberculosis clinical strains resistant to rifampicin harbor mainly mutation in an 81-base pair region of rpoB. These mutations mainly consist of single amino acid substitutions. However insertions also can be related with rifampicin resistance strains. Herein, we described an insertion of 12 nucleotides in clinical isolates of M. tuberculosis resistant to rifampicin, all obtained from inmates. To evaluate the importance this insertion in surviving and drug resistance, it were carried out fitness experimental assays as well as in silico studies of 3D structural models, molecular docking simulations and virtual screening. The medical records of the seven patients showed all were previously treated to tuberculosis. Growth curves shown that the insertion determines a biological cost when compared to wild type rpoB and katG; or the double mutated rpoB S531L and katG S315T. From docking and molecular dynamics simulations it can be inferred that the insertion does not affect the process of synthesis of RNA transcripts. On the other hand, in the mutant RNAP-RIF complex rifampicin confirmed a low affinity interaction for the mutant form. Interesting, virtual screening for potential inhibitors for wtRNAP and mRNAP using a library of 1446 compounds approved by the FDA showed that the best ligands were mainly compounds with antibiotic activity, although the targets involved in the pharmacological action are other than RNAP. In conclusion, seven strains of M. tuberculosis RIF resistant that present an insertion of four amino acids in RNA polymerase showed by growth curve assays, a biological cost. Further, bioinformatics tools had characterized the putative drug resistance dynamic as well as the maintenance of RNA polymerase activity.

Anatomical variations at the origin of the biceps tendon have been described by several authors, but occurrences of an origin in the supraspinatus are rare. It is unclear whether this variation might contribute toward pathological conditions of the shoulder. Our objective here was to describe a case of an anatomical variation in the origin of the tendon of the long head of the biceps. The clinical information, preoperative images and arthroscopic images relating to a patient with an aberrant origin of the long head of the biceps, which was observed during shoulder arthroscopy, were reviewed. In this case study, the origin of the biceps was found in the rotator cuff, without any origin from the supraglenoid tubercle or upper labrum. This variant did not seem to contribute toward the pathological condition of the shoulder, and standard treatment for the concomitant condition was sufficient for treating it. RESUMO As variações anatômicas na origem do tendão do bíceps foram descritas por vários autores, mas a ocorrência de sua origem no supraespinhal é rara. Não está claro se essa variação pode contribuir para condições patológicas do ombro. Nosso objetivo é descrever um caso de uma variação anatômica da origem da cabeça longa do tendão do bíceps. Informações clínicas, imagens pré-operatórias e imagens artroscópicas foram revisadas a partir de um paciente que teve uma origem aberrante da cabeça longa do bíceps observada durante a artroscopia do ombro. Neste estudo de caso, a origem do bíceps foi encontrada no manguito rotador, sem origem do tubérculo supraglenoidal ou labrum superior. Essa variante não parece contribuir para a patologia ombro e o tratamento padrão de patologia concomitante foi suficiente.