Abstract

Abstract
In this paper we describe a comparative classification of Pneumonia using Convolution Neural Network. The database used was the dataset Labeled Optical Coherence Tomography (OCT) and Chest X-Ray Images for Classification made available by (Kermany, 2018) with a total of 5863 images, with 2 classes: normal and pneumonia. To evaluate the generalization capacity of the models, cross-validation of k-fold was used. The classification models proved to be efficient compared to the work of (Kermany et al., 2018) which obtained 92.8 % and the present work had an average accuracy of 95.30 %.

Abstract
The wavelength sensitivity and spectral resolution of Mach-Zehnder fiber interferometers obtained through a combination of two identical uncoated and titanium dioxide (TiO2) coated long period fiber gratings (LPFGs) is presented and compared with single LPFGs-based refractometric sensors. A set of LPFGs were fabricated in single mode fiber with the resonance band having an amplitude of 3 dB in order to split in half the optical power between the core and the specific cladding modes. The separation between the pair of LPFG written in the fiber was varied between 1 and 3 cm and the thickness of the TiO2 coating around the fiber ranged from 20 to 40 nm. A wavelength shift sensitivity of 216 nm/refractive index units (RIU) was achieved for the device with 3 cm and a 30-nm thick TiO2 coating, which presented a spectral resolution of 1.1 x 10(-4 )Rill Despite the lower wavelength shift sensitivity of 142 nm/RIU, attained for a 2-cm long device and 30-nm thick TiO2 coating, a spectral resolution of 1.8 x 10(-5) RIU was measured, which is one order of magnitude lower than a single LPFG.

Abstract
A multiplicity of innovative applications has been developed based on the mobile crowdsourcing (MCS) paradigm. One group of applications addresses the creation of accessibility maps in large cities. In this context, a conceptual model of a system for the detection and the timely notification of the existence of temporary obstacles and other dangers in the urban environment is proposed in "Pervasive Crowd Mapping for Dynamic Environments". This concept (PCM4DE) encompasses, among other technologies, the use of crowdsourcing. The system will be particularly useful to people with disabilities and elderly people. An exploratory literature review showed that data quality and the motivation strategies fur participating in the systems based on MCS remain two of the key challenges to the effectiveness of those systems. This paper aims to contribute to the implementation of the PCM4DE concept by proposing the development of a mechanism that should improve the data quality through motivation forms which will enable a positive personal user experience, i.e., an experience which meets the participant's objectives, needs and preferences.

Abstract
Automatic speaker identification typically relies on sophisticated statistical modeling and classification which requires large amounts of data for good performance. However, in actual audio forensics casework, frequently only a few seconds of speech material are available. In this paper, we favor diversity in feature extraction, simple modeling and classification, and constructive combination of congruent classification scores. We use phase, spectral magnitude and F0-related features in speaker identification experiments on a database of 35 speakers most of whom are twins. Using only 4.4 sec. of vowel-like sounds per speaker, we characterize the performance that is reached with individual features and we characterize simple and yet effective ways of classification score fusion. Insights for further research are also presented.

Abstract
Technology can help participants develop basic skills but also improve students learning, since it can be more accessible, giving the opportunity to experience new knowledge. Therefore, it seems relevant an in-depth study on teaching methods implemented with students with special education needs (SEN) to describe different methodologies. For that, in this paper it is presented the analysis on teaching strategies to help learning of specific content for students with intellectual disabilities (ID). The purpose is to describe strategies used in traditional and technology-based education to be the starting point in the development of an effective learning multimedia resource for this specific audience. Overall, were found advantages in the use of technologies as a teaching resource for students with and without special SEN, specifically for students with ID. Studies showed that attention and motivation increase and students feel more stimulated to learn when the syllabus is presented more interactively. Another important finding is that educational games as educational resource can indeed be a valid strategy to retain attention and stimulate memory of students with intellectual disabilities. Finally, with these findings it is proposed features to develop a multimedia application that can enhance learning of a new language for students with ID.