Abstract
Image inpainting is a crucial technique in computer vision, particularly for reconstructing corrupted images. In medical imaging, it addresses issues from instrumental errors, artifacts, or human factors. The development of deep learning techniques has revolutionized image inpainting, allowing for the generation of high-level semantic information to ensure structural and textural consistency in restored images. This article presents a comprehensive review of 53 studies on deep image inpainting in medical imaging, analyzing its evolution, impact, and limitations. The findings highlight the significance of deep image inpainting in artifact removal and enhancing the performance of multi-task approaches by localizing and inpainting regions of interest. Furthermore, the study identifies magnetic resonance imaging and computed tomography as the predominant modalities and highlights generative adversarial networks and U-Net as preferred architectures. Future research directions include the development of blind inpainting techniques, the exploration of techniques suitable for 3D/4D images, multiple artifacts, and multi-task applications, and the improvement of architectures.

Abstract
Background/Objectives: Type 2 diabetes mellitus (T2DM) is a chronic condition marked by hyperglycemia, which can affect metabolic, vascular, and hematological parameters. A low-carbohydrate, high-fat (LCHF) diet has been shown to improve glycemic control and blood pressure regulation. Exercise in hypoxia (EH) enhances insulin sensitivity, erythropoiesis, and angiogenesis. The combination of LCHF and EH may offer a promising strategy for managing T2DM and hypertension (HTN), although evidence remains limited. This study aimed to assess the effects of an eight-week normobaric EH intervention at 3000 m simulated altitude combined with an LCHF diet on hematological and lipid profiles, inflammation, and blood pressure in older patients with T2DM and HTN. Methods: Forty-two diabetic patients with HTN were randomly assigned to three groups: (1) control group (control diet + exercise in normoxia), (2) EH group (control diet + EH), and (3) intervention group (EH+LCHF) Baseline and eight-week measurements included systolic, diastolic, and mean blood pressure (SBP, DBP, MAP), hematological and lipid profiles, and inflammation biomarkers. Results: Blood pressure decreased after the intervention (p < 0.001), with no significant differences between groups (SBP: p = 0.151; DBP: p = 0.124; MAP: p = 0.18). No differences were observed in lipid profile or C-reactive protein levels (p > 0.05). Mean corpuscular hemoglobin (MCH) increased in the EH group (p = 0.027), while it decreased in the EH+LCHF group (p = 0.046). Conclusions: Adding hypoxia or restricting carbohydrates did not provide additional benefits on blood pressure in T2DM patients with HTN. Further elucidation of the mechanisms underlying hematological adaptations is imperative.

Abstract
Digestive disorders can be signs of long-term conditions such as cancer, and as such, they should be treated seriously. Endoscopic exams of the gastrointestinal tract allow for the early detection of these conditions and facilitate effective treatment; these procedures have their effectiveness limited by variations in operator performance, due to human error. Support systems are desired to help specialists detect and diagnose pathologies in this type of exam. This work used a seldom utilized dataset, the ERS dataset, which contains 121,399 labeled images, to evaluate eight models from the EfficientNet family of architectures, as well as three models from the EfficientNetV2 iteration of this architecture, for the task of binary classification of endoscopic images. This work also compared their performance to four other widely used CNN architectures for the same task, along with the baseline results published by the authors of the dataset. Each model was evaluated in a 5-fold cross-validation procedure, following the same training protocol. The experiments have shown that the best-performing architecture was EfficientNetV2M, followed closely by EfficientNetB7, with the former achieving average accuracy and F1-Score values of, respectively, 82.24% and 88.15%. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

Abstract
Food sustainability is a vital global challenge, as dietary choices affect both human health and the environment. This study evaluates Portuguese dietary patterns' environmental and nutritional sustainability dimensions using data from the National Food, Nutrition, and Physical Activity Survey (IAN-AF) 2015-2016. Environmental indicators (carbon footprint, water footprint, and land use) and a nutritional quality index (NRD9.3) were analysed. Sustainability scores were calculated based on deviations from population medians, with the environmental score estimated from a weighted mean of the three indicators. A quadrant analysis classified individuals into four sustainability segments: better environmental and better nutritional scores (reference group); worse environmental and worse nutritional scores; worse environmental and better nutritional scores; and better environmental and worse nutritional scores. The reference group, with higher plant-based food consumption, had the lowest environmental impacts, 33% lower carbon footprint, 36% lower water footprint, and 50% lower land use, while exhibiting 87% better nutritional quality. In contrast, the worse environmental and worse nutritional scores group, with a diet rich in red and processed meats, sweets, and alcohol, showed higher environmental impacts and poorer nutritional quality. The group with worse environmental and better nutritional scores favored dairy and seafood, whereas the group with better environmental and worse nutritional scores had higher intakes of white meat, sweets, and alcohol. Sociodemographic factors, including sex, age, and education, show to influence the sustainability dimensions. These findings highlight the need for tailored dietary strategies that consider differing environmental and nutritional profiles, supporting more effective and practical public health interventions.

Abstract
Optical fibers have been extensively applied in optical sensing platforms for their large bandwidth, stability, light weight and accessibility. This work presents a theoretical analysis of an optical fiber surface plasmon resonance system for refractometric sensing applications. The device consists of a multilayer hyperbolic metamaterial (HMM) composed of concentric Au/TiO2 alternate layers in optical fiber matrix. HMMs exhibit hyperbolic dispersion (HD) and the interaction of different plasmonic modes at each interface of the HMM is reported to enhance light-matter coupling, leading to an increased refractometric sensitivity. The HD and its effects on sensor performance are numerically investigated by effective medium theory (EMT) and backed by the exact transfer matrix method (TMM). The maximum sensor performance was attained for a configuration with 2 bilayers with 30 nm thickness for a metal fill fraction (rho) of 0.7, achieving a figure of merit (FOM) of 18.45. A direct comparison with a plasmonic Au optical fiber sensor returned an optimized FOM of 5.74, therefore achieving over a three-fold increase in sensor performance, assessing the potential of HMM as highly refractometric sensitive platforms.

Abstract