Abstract

Abstract
Glaucoma is a silent disease that leads to vision loss or irreversible blindness. Current deep learning methods can help glaucoma screening by extending it to larger populations using retinal images. Low-cost lenses attached to mobile devices can increase the frequency of screening and alert patients earlier for a more thorough evaluation. This work explored and compared the performance of classification and segmentation methods for glaucoma screening with retinal images acquired by both retinography and mobile devices. The goal was to verify the results of these methods and see if similar results could be achieved using images captured by mobile devices. The used classification methods were the Xception, ResNet152 V2 and the Inception ResNet V2 models. The models' activation maps were produced and analysed to support glaucoma classifier predictions. In clinical practice, glaucoma assessment is commonly based on the cup-to-disc ratio (CDR) criterion, a frequent indicator used by specialists. For this reason, additionally, the U-Net architecture was used with the Inception ResNet V2 and Inception V3 models as the backbone to segment and estimate CDR. For both tasks, the performance of the models reached close to that of state-of-the-art methods, and the classification method applied to a low-quality private dataset illustrates the advantage of using cheaper lenses.

Abstract
Search Engine Optimization, or SEO for short, helps to improve the inbound user traffic of a website. Optimizing a website for search engines is now crucial to its success and ultimately to the company's ability to increase the business. Many companies websites on the Internet fail to use any SEO technique or strategy to improve their positioning in search engine results. Therefore, if other digital marketing strategies are not used to promote the website, its traffic will be very restricted. In this study several different SEO techniques were used to improve a website's overall indexing on the Google search engine. Through the implemented strategy it was possible to improve the positioning in search results of the company's website used in the case study, allowing several searches with different keywords to show the website on the first page of results. With the results obtained, it was possible to conclude that it is very important to find the perfect balance between SEO keyword competition and monthly search volume. To help select the best keywords for this particular website, the use of SEO tools are of the utmost importance.

Abstract
This research paper presents and discusses the main results generated and obtained with the proprietary computer platform CIDIUS (R), developed by the authors of this work, which aims to support the decision-making process of Portuguese mayors. Thus, keeping in mind the theoretical models and based on the data collected through the questionnaire given to the population, we tried to understand the influence that the dimensions Notoriety, Image, and Reputation (NIR), Citizen and Voter Expectations (CVE), Contestation and Complaint of the Municipal Executive (CCME), Perceived Value (PV), and Organizational Performance and Perceived Quality (OPPQ) has a positive effect on Municipe Satisfaction (MS). The parishes of the municipality of Valongo were selected and analyzed, namely the parishes of Alfena, Campo e Sobrado, Valongo, and Ermesinde, and a total of 998 valid questionnaires were collected. It was concluded that all studied dimensions except the Organizational Performance and Perceived Quality (OPPQ) dimension had a positive and statistically significant impact on Municipe Satisfaction (MS). The results of this research suggest the need for the use of these opinion-gathering techniques to encourage active citizen involvement in the daily life of their municipality, as well as the need for valid information that gives executives the ability to take political action that is appropriate to the interests and expectations of citizens.

Abstract
We present MONARCH, a framework-agnostic storage middleware that transparently employs storage tiering to accelerate Deep Learning (DL) training. It leverages existing storage tiers of modern supercomputers (i.e., compute node's local storage and shared parallel file system (PFS)), while considering the I/O patterns of DL frameworks to improve data placement across tiers. MONARCH aims at accelerating DL training and decreasing the I/O pressure imposed over the PFS. We apply MONARCH to TensorFlow and PyTorch, while validating its performance and applicability under different models and dataset sizes. Results show that, even when the training dataset can only be partially stored at local storage, MONARCH reduces TensorFlow's and PyTorch's training time by up to 28% and 37% for I/O-intensive models, respectively. Furthermore, MONARCH decreases the number of I/O operations submitted to the PFS by up to 56%.

Abstract
Background: Regional Development (RD) allows countries to balance regional differences by providing economic and social benefits to communities. This research highlights the importance of logistics activities to regional social development, and a framework to assess these connections is proposed. Objectives: How to boost regional socioeconomic development through logistics. Methods/Approach: The contributions of logistics to socioeconomic development are analysed based on the previous research, and the case of the Alto Minho (AM) region in Portugal was used to illustrate the connection between logistics and regional development. Results showed that logistics had created jobs, increased company turnover and exports, and increased GDP growth in several regions. For the AM region, results indicate that many companies are operating in this area, contributing to supporting municipalities to reduce regional disparities. Conclusions: A framework for assessing regional logistics performance is proposed together with several logistics performance indicators. This approach is essential for future developments integrating logistics into socioeconomic development.