No description available.

No description available.

The General Data Protection Regulation (GDPR) is a legal framework that sets guidelines for collecting, processing, and storing personal information from residents of the European Union (EU). The regulation applies to all companies, of any magnitude, that collect or handle information from European Citizens. This work aims to understand how the GDPR might negatively affect startups in Northern Portugal by either raising costs, hindering innovation, or placing barriers to the accelerated growth this type of company usually seeks. Using an online questionnaire, we inquired startup founders, c-level managers, and senior managers about the experience of their companies regarding the road to compliance with the GDPR and its impact on their business. We conclude, given the limitations of the study, that the regulation tends to create additional costs for startups, although not at a very significant level. In the innovation spectrum, the conclusion is that management often weighs the need to comply with the GDPR when ideating new products or services - the tendency, however, is not to discard any ideas exclusively because of the risk of non-compliance. Regarding growth, the GDPR is not seen as a blocker for growth for most companies, either because they tend

No description available.

The proposed research aims to design, develop and analyze a solution based on Federated Learning to support medical diagnosis of pneumonia using chest X-radiography (X-ray) images. / Medical image analysis is crucial for the efficient diagnosis of many diseases. Hospitals typically maintain vast repositories of images, which can be leveraged for various purposes, including research. However, access to such image collections is largely restricted to safeguard the privacy of the individuals whose images are being stored, as data protection concerns come into play. Recently, the development of Automated Medical Image Analysis has gained significant attention, with Deep Learning being one solution that has achieved remarkable results in medical image analysis. One promising approach for medical image analysis is Federated Learning (FL), which enables using a set of physically distributed data repositories (the nodes) for analysis, satisfying the restriction that data does not leave the repository. Under these conditions, FL can build high-quality accurate models using a lot of available data wherever it is. This approach can help researchers and clinicians to diagnose diseases and support medical decisions more efficiently and robustly. Detection of pneumonia on chest X-radiography (X-ray) images is proposed in a FL environment using Flower as framework, and FedAvg as strategy. This supervised learning approach uses pre-trained Convolutional Neural Network (CNN) models to leverage transfer learning: VGG-16, Resnet-18 and Resnet-50, and also data augmentation techniques are applied to fine-tune the models. Simulated a FL environment having 8 hospitals sharing their own images, Resnet-18 shows the best result with 98.46\% of accuracy, followed by Resnet-50 with 78.46\% of accuracy, then VGG-16 with 78.46\% of accuracy as well, all evaluated on the server-side after 5 rounds of training. The experiments suggested in the research work exhibited significant computational expense owing to an uneven dataset, a prevalent constraint encountered in the study as well, presented in the current state-of-the-art, which utilizes the identical dataset as this inquiry. Hence, it discusses applications, contributions, limitations, and challenges, and is suitable for those who want to understand how FL can contribute to the medical imaging domain. Furthermore, this solution is applicable as a baseline to solve other binary classification problems using medical images, such as Magnetic resonance imaging (MRI), Computed tomography (CT), X-radiography (X-ray), and histology images.

Neste projeto, temos como objetivo desenvolver um sistema de IA que recomende um conjunto de casos relativos (passados) para orientar a tomada de decisão do médico. Objetivo: A ambição é desenvolver um modelo de aprendizado baseado em IA para caracterização de câncer de pulmão, a fim de auxiliar na rotina clínica. Considerando a complexidade dos fenômenos biológicos que ocorrem durante o desenvolvimento do câncer, as relações entre eles e as manifestações visuais capturadas pela tomografia computadorizada (CT) têm sido exploradas nos últimos anos. No entanto, devido à falta de robustez dos métodos atuais de aprendizado profundo, essas correlações são frequentemente consideradas espúrias e se perdem quando confrontadas com dados coletados a partir de distribuições alteradas: diferentes instituições, características demográficas ou até mesmo estágios de desenvolvimento do câncer. / In this project, we aim to develop an AI system that recommends a set of relative (past) cases to guide the decision-making of the clinician. Objective: The ambition is to develop an AI-based learning model for lung cancer characterization in order to assist in clinical routine. Considering the complexity of the biological phenomenat hat occur during cancer development, relationships between these and visual manifestations captured by CT have been explored in recent years; however, given the lack of robustness of current deep learning methods, these correlations are often found spurious and get lost when facing data collected from shifted distributions: different institutions, demographics or even stages of cancer development.

No description available.

No description available.

Microsserviços tornaram-se o estilo de arquitetura mais utilizado entre todas as estratégias de desenvolvimento de software disponíveis. No entanto, as pesquisas sobre esse tema estão no início, o que dificulta a localização de aplicações de microsserviços em escala para análise. Portanto, há uma grande necessidade de novas investigações, bem como ferramentas para apoiar novos desenvolvimentos no campo de microsserviços. O primeiro objetivo deste trabalho é coletar características de microsserviços encontradas na literatura e traduzi-las em características mensuráveis no código. Com isso, fornecemos um conjunto abrangente de características, bem como métricas para identificá-las no código. Um segundo objetivo é usar essas métricas para identificar a base do código seguindo um estilo de arquitetura de microsserviço. Essa solução é disponibilizada por meio de uma ferramenta que permite aos usuários encontrar microsserviços em escalas e filtrá-los de acordo com suas necessidades. Isso pode ser usado para encontrar exemplos de microsserviços em uma linguagem de programação específica ou para criar corpora para estudos de pesquisa. Nossa avaliação mostra que nosso algoritmo pode identificar microsserviços com uma precisão de 85%. / Microservices have become the most used architectural style among all available software development strategies. However, it is difficult to find microservice applications at scale for analysis. Therefore, there is a great need for new investigations as well as tools to support new developments in the field of microservices. The first goal of this work is to collect microservices characteristics found in the literature and translate them into measurable features in the code. With this, we provide a comprehensive set of characteristics as well as metrics to identify them in the code. A second goal is to design an algorithm to use such metrics to identify code basis following a microservice architectural style. This solution is made available through a tool that allows users to find microservices at scales and filter them according to their needs. This can be used to find examples of microservices in a specific programming language or to create corpora for research studies. Our evaluation shows our algorithm can identify microservices with a precision of 85%.

No description available.