Abstract
Lung cancer late diagnosis has a large impact on the mortality rate numbers, leading to a very low five-year survival rate of 5%. This issue emphasises the importance of developing systems to support a diagnostic at earlier stages. Clinicians use Computed Tomography (CT) scans to assess the nodules and the likelihood of malignancy. Automatic solutions can help to make a faster and more accurate diagnosis, which is crucial for the early detection of lung cancer. Convolutional neural networks (CNN) based approaches have shown to provide a reliable feature extraction ability to detect the malignancy risk associated with pulmonary nodules. This type of approach requires a massive amount of data to model training, which usually represents a limitation in the biomedical field due to medical data privacy and security issues. Transfer learning (TL) methods have been widely explored in medical imaging applications, offering a solution to overcome problems related to the lack of training data publicly available. For the clinical annotations experts with a deep understanding of the complex physiological phenomena represented in the data are required, which represents a huge investment. In this direction, this work explored a TL method based on unsupervised learning achieved when training a Convolutional Autoencoder (CAE) using images in the same domain. For this, lung nodules from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) were extracted and used to train a CAE. Then, the encoder part was transferred, and the malignancy risk was assessed in a binary classification-benign and malignant lung nodules, achieving an Area Under the Curve (AUC) value of 0.936. To evaluate the reliability of this TL approach, the same architecture was trained from scratch and achieved an AUC value of 0.928. The results reported in this comparison suggested that the feature learning achieved when reconstructing the input with an encoder-decoder based architecture can be considered an useful knowledge that might allow overcoming labelling constraints.

Abstract

Abstract
Conventionally, vital sign monitoring for neonatal infants inside the Neonatal Intensive Care Unit is performed via probes affixed to their skin. However, such instruments may cause damage to the epidermis and increase the risk of infection as well as promote discomfort to the infant. As an alternative to traditional means of monitoring heart rate, remote Photoplethysmography techniques have been surging among the scientific community. These techniques have been vastly explored for adult subjects but not for neonatal infants, who would greatly benefit from such applications. This study aims at developing a regular consumer camera-based framework for continuous and contactless extraction of the heart rate in adult subjects in challenging conditions and investigating the tool's ability to adapt to a new domain which consists of newborn subjects and the real-world conditions of a Neonatal Intensive Care Unit.

Abstract
We study the electromagnetically induced transparency (EIT) phenomenon in a hollow-core fibre filled with rubidium gas. We analyse the impact of the guiding effect and of the temperature on the properties of the EIT phenomenon. The refractive index felt by the probe laser is found to vary due to the radial dependence of the fibre mode field at the pump frequency. Several results are presented for the transmission, dispersion, and group velocity of the probe field, considering both the free propagation regime and the guided propagation along the hollow-core fibre. We note that the EIT occurring in a waveguide has a great potential for practical applications since it can be controlled by adjusting the gas and the fibre properties.

Abstract
Telecommunication networks evolution is driving the development of new applications for mobile devices. Some of these applications are resource-intensive and push computational and energy demands of mobile devices beyond the mobile hardware capabilities. In this context, Mobile Cloud Computing (MCC) architecture emerges as a solution for offloading mobile devices that allows to execute these applications in cloud datacenters thus reducing the processing demand in mobile devices. However, more demanding applications, e.g. interactive and realtime applications, are sensitive to processing and communications delay. For these applications, Mobile Edge Computing (MEC) can be used as an intermediary technology, providing computing and storage resources in the network edge. This paper presents a study carried out to evaluate the performance of MEC and MCC architectures when executing two applications, Fluid and FaceSwap, representative of real time and computing intensive applications. A set of scenarios were designed to quantify the performance of these architectures in different settings.

Abstract
Prompting students' interest and engagement in learning environments is crucial to achieve the best results. Academia and educators in general are constantly adapting materials and methodologies in order to maximise the acquisition of contents by their students. In this case-study, a new teaching/learning methodology is presented and evaluated through a final questionnaire survey. This case-study aims to understand students' efficiency and motivation levels regarding a new teaching/learning methodology adopted in the second module of a Computer Systems and Architectures course attended by first-year Computer Sciences undergraduates. The new teaching/learning methodology relies on a specific programming language-ARMv6 assembly-to improve students' efficiency levels, and an innovative always-visible in-class mobile test scenario, implemented through a low-cost computing platform-Raspberry Pi 1 B+- A s a server, mimicking as much as possible a real-life environment, so that students believe they are working on real hardware, thus enhancing their motivation levels. The results of the questionnaire survey allowed to infer that the use of a specific programming language, such as ARMv6 assembly, coupled with a new always-visible in-class mobile test scenario were in fact efficient in raising the levels of motivation among Computer Sciences students and, consequently, improved their skills in Computer Architecture. 2012 ACM Subject Classification Computer systems organization.