Abstract
Artificial Intelligence (AI) is one of the big hopes for the future of a positive revolution in the use of medical data to improve clinical routine and personalized medicine [...]

Abstract
Bone marrow edema (BME) is the term given to the abnormal fluid signal seen within the bone marrow on magnetic resonance imaging (MRI). It usually indicates the presence of underlying pathology and is associated with a myriad of conditions/causes. However, it can be misleading, as in some cases, it may be associated with normal changes in the bone, especially during the growth period of childhood, and objective methods for assessment are lacking. In this work, learning models for BME detection were developed. Transfer learning was used to overcome the size limitations of the dataset, and two different regions of interest (ROI) were defined and compared to evaluate their impact on the performance of the model: bone segmention and intensity mask. The best model was obtained for the high intensity masking technique, which achieved a balanced accuracy of 0.792 +/- 0.034. This study represents a comparison of different models and data regularization techniques for BME detection and showed promising results, even in the most difficult range of ages: children and adolescents. The application of machine learning methods will help to decrease the dependence on the clinicians, providing an initial stratification of the patients based on the probability of edema presence and supporting their decisions on the diagnosis.

Abstract
Biomedical engineering has been targeted as a potential research candidate for machine learning applications, with the purpose of detecting or diagnosing pathologies. However, acquiring relevant, high-quality, and heterogeneous medical datasets is challenging due to privacy and security issues and the effort required to annotate the data. Generative models have recently gained a growing interest in the computer vision field due to their ability to increase dataset size by generating new high-quality samples from the initial set, which can be used as data augmentation of a training dataset. This study aimed to synthesize artificial lung images from corresponding positional and semantic annotations using two generative adversarial networks and databases of real computed tomography scans: the Pix2Pix approach that generates lung images from the lung segmentation maps; and the conditional generative adversarial network (cCGAN) approach that was implemented with additional semantic labels in the generation process. To evaluate the quality of the generated images, two quantitative measures were used: the domain-specific Frechet Inception Distance and Structural Similarity Index. Additionally, an expert assessment was performed to measure the capability to distinguish between real and generated images. The assessment performed shows the high quality of synthesized images, which was confirmed by the expert evaluation. This work represents an innovative application of GAN approaches for medical application taking into consideration the pathological findings in the CT images and the clinical evaluation to assess the realism of these features in the generated images.

Abstract
Dissipative quartic solitons have gained interest in the field of mode-locked lasers for their energy-width scaling which allows the generation of ultrashort pulses with high energies. Pursuing the characterization of such pulses, here we found soliton solutions of a distributed model for mode locked lasers in the presence of either positive or negative fourth-order dispersion (4OD). We studied the impact the laser parameters may have on the profiles, range of existence, and energy-width relation of the output pulses. The most energetic and narrowest solutions occur for negative 4OD, with the energy having an inverse cubic dependence with the width in most cases. Our simulations showed that the spectral filtering has the biggest contribution in the generation of short (widths as low as 39 fs) and very energetic (391 nJ) optical pulses.(c) 2023 Optica Publishing Group

Abstract
We investigate the properties of time-dependent dissipative solitons for a cubic complex Ginzburg-Landau equation stabilized by nonlinear gradient terms. The separation of initially nearby trajectories in the asymptotic limit is predominantly used to distinguish qualitatively between time-periodic behavior and chaotic localized states. These results are further corroborated by Fourier transforms and time series. Quasiperiodic behavior is obtained as well, but typically over a fairly narrow range of parameter values. For illustration, two examples of nonlinear gradient terms are examined: the Raman term and combinations of the Raman term with dispersion of the nonlinear gain. For small quintic perturbations, it turns out that the chaotic localized states are showing a transition to periodic states, stationary states, or collapse already for a small magnitude of the quintic perturbations. This result indicates that the basin of attraction for chaotic localized states is rather shallow.

Abstract
These are the short papers proceedings of the 3rd Symposium of Applied Science for Young Researchers – SASYR. This scientific event welcomed works by junior researchers on any research topic covered by the following three research centers: ADiT-lab (from IPVC, Instituto Politécnico de Viana do Castelo), 2Ai (from IPCA, Instituto Politécnico do Cávado e do Ave) and CeDRI (from IPB, Instituto Politécnico de Bragança). The main objective of SASYR is to provide a friendly and relaxed environment for young researchers to present their work, discuss recent results, and develop new ideas. In this way, this event offered an opportunity for the ADiT-lab, 2Ai, and CeDRI research communities to gather synergies and promote collaborations, thus improving the quality of their research. The SASYR 2023 took place at Instituto Politécnico do Cávado e do Ave, Barcelos, Portugal, on the 11th of July, 2023.