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
During a pandemic, public discussion and decision-making may be required in face of limited evidence. Data-grounded analysis can support decision-makers in such contexts, contributing to inform public policies. We present an empirical analysis method based on regression modelling and hypotheses testing to assess events for the possibility of occurrence of superspreading contagion with geographically heterogeneous impacts. We demonstrate the method by evaluating the case of the May 1st, 2020 Demonstration in Lisbon, Portugal, on regional growth patterns of COVID-19 cases. The methodology enabled concluding that the counties associated with the change in the growth pattern were those where likely means of travel to the demonstration were chartered buses or private cars, rather than subway or trains. Consequently, superspreading was likely due to travelling to/from the event, not from participating in it. The method is straightforward, prescribing systematic steps. Its application to events subject to media controversy enables extracting well founded conclusions, contributing to informed public discussion and decision-making, within a short time frame of the event occurring.

Abstract
Process Analytical Technology (PAT) has been increasingly used in the pharmaceutical industry to monitor essential parameters in real-time during pharmaceutical processes. The concentration of Active Pharmaceutical Ingredients (APIs), such as paracetamol, is one of these parameters, and controlling its variations allows for optimization of the production process. In this study, a refractometric sensor, implemented by an interrogation system based on an Erbium-Doped Fiber Ring Cavity (EDFRC), was presented and experimentally demonstrated. The Cavity Ring proposed included a 1 x 3 coupler. One port of the coupler was used to increase the optical power of the system through a Fiber Bragg Grating (FBG), and the other two ports were used as sensing head and reference. The sensor detected variations of paracetamol concentration with a sensitivity of [(-1.00 +/- 0.05) x 10(-3)] nW/(g/kg) and a resolution of 5.53 g/kg. The results demonstrate the potential of this technology as a possible non-invasive PAT tool.

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
The knee is one of the most stressed joints of the human body, being susceptible to ligament injuries and degenerative diseases. Due to the rising incidence of knee pathologies, the number of knee X-rays acquired is also increasing. Such X-rays are obtained for the diagnosis of knee injuries, the evaluation of the knee before and after surgery, and the monitoring of the knee joint's stability. These types of diagnosis and monitoring of the knee usually involve radiography under physical stress. This widely used medical tool provides a more objective analysis of the measurement of the knee laxity than a physical examination does, involving knee stress tests, such as valgus, varus, and Lachman. Despite being an improvement to physical examination regarding the physician's bias, stress radiography is still performed manually in a lot of healthcare facilities. To avoid exposing the physician to radiation and to decrease the number of X-ray images rejected due to inadequate positioning of the patient or the presence of artefacts, positioning systems for stress radiography of the knee have been developed. This review analyses knee positioning systems for X-ray environment, concluding that they have improved the objectivity and reproducibility during stress radiographs, but have failed to either be radiolucent or versatile with a simple ergonomic set-up.

Abstract
In a clinical context, physicians usually take into account information from more than one data modality when making decisions regarding cancer diagnosis and treatment planning. Artificial intelligence-based methods should mimic the clinical method and take into consideration different sources of data that allow a more comprehensive analysis of the patient and, as a consequence, a more accurate diagnosis. Lung cancer evaluation, in particular, can benefit from this approach since this pathology presents high mortality rates due to its late diagnosis. However, many related works make use of a single data source, namely imaging data. Therefore, this work aims to study the prediction of lung cancer when using more than one data modality. The National Lung Screening Trial dataset that contains data from different sources, specifically, computed tomography (CT) scans and clinical data, was used for the study, the development and comparison of single-modality and multimodality models, that may explore the predictive capability of these two types of data to their full potential. A ResNet18 network was trained to classify 3D CT nodule regions of interest (ROI), whereas a random forest algorithm was used to classify the clinical data, with the former achieving an area under the ROC curve (AUC) of 0.7897 and the latter 0.5241. Regarding the multimodality approaches, three strategies, based on intermediate and late fusion, were implemented to combine the information from the 3D CT nodule ROIs and the clinical data. From those, the best model-a fully connected layer that receives as input a combination of clinical data and deep imaging features, given by a ResNet18 inference model-presented an AUC of 0.8021. Lung cancer is a complex disease, characterized by a multitude of biological and physiological phenomena and influenced by multiple factors. It is thus imperative that the models are capable of responding to that need. The results obtained showed that the combination of different types may have the potential to produce more comprehensive analyses of the disease by the models.