Abstract
Small to medium-sized shipyards play a crucial role in the European naval industry. However, the globalization of technology has increased competition, posing significant challenges to shipyards, particularly in domestic markets for short sea, work, and inland vessels. Many shipyard operations still rely on manual, labor-intensive tasks performed by highly skilled operators. In response, the adoption of new tools is essential to enhance efficiency and competitiveness. This paper presents a methodology for developing a human-centric portfolio of advanced technologies tailored for shipyard environments, covering processes such as shipbuilding, retrofitting, outfitting, and maintenance. The proposed technological solutions, which have achieved high technology readiness levels, include 3D modeling and digitalization, robotics, augmented and virtual reality, and occupational exoskeletons. Key findings from real-scale demonstrations are discussed, along with major development and implementation challenges. Finally, best practices and recommendations are provided to support both technology developers seeking fully tested tools and end users aiming for seamless adoption. © 2025 by the authors.

Abstract
The public transportation sector generates large volumes of sensor data that, if analyzed adequately, can help anticipate failures and initiate maintenance actions, thereby enhancing quality and productivity. This work contributes to a real-time data-driven predictive maintenance solution for Intelligent Transportation Systems. The proposed method implements a processing pipeline comprised of sample pre-processing, incremental classification with Machine Learning models, and outcome explanation. This novel online processing pipeline has two main highlights: (i) a dedicated sample pre-processing module, which builds statistical and frequency-related features on the fly, and (ii) an explainability module. This work is the first to perform online fault prediction with natural language and visual explainability. The experiments were performed with the Metropt data set from the metro operator of Porto, Portugal. The results are above 98 % for f-measure and 99 % for accuracy. In the context of railway predictive maintenance, achieving these high values is crucial due to the practical and operational implications of accurate failure prediction. In the specific case of a high f-measure, this ensures that the system maintains an optimal balance between detecting the highest possible number of real faults and minimizing false alarms, which is crucial for maximizing service availability. Furthermore, the accuracy obtained enables reliability, directly impacting cost reduction and increased safety. The analysis demonstrates that the pipeline maintains high performance even in the presence of class imbalance and noise, and its explanations effectively reflect the decision-making process. These findings validate the methodological soundness of the approach and confirm its practical applicability for supporting proactive maintenance decisions in real-world railway operations. Therefore, by identifying the early signs of failure, this pipeline enables decision-makers to understand the underlying problems and act accordingly swiftly.

Abstract
Accurate retrieval of post-surgical images plays a critical role in surgical planning for breast cancer patients. However, current content-based image retrieval methods face challenges related to limited interpretability, poor robustness to image noise, and reduced generalization across clinical settings. To address these limitations, we propose a multistage retrieval pipeline integrating saliency-based explainability, noise-reducing image pre-processing, and ensemble learning. Evaluated on a dataset of post-operative breast cancer patient images, our approach achieves contrastive accuracy of 77.67% for Excellent/Good and 84.98% for Fair/Poor outcomes, surpassing prior studies by 8.37% and 11.80%, respectively. Explainability analysis provided essential insight by showing that feature extractors often attend to irrelevant regions, thereby motivating targeted input refinement. Ablations show that expanded bounding box inputs improve performance over original images, with gains of 0.78% and 0.65% contrastive accuracy for Excellent/Good and Fair/Poor, respectively. In contrast, the use of segmented images leads to a performance drop (1.33% and 1.65%) due to the loss of contextual cues. Furthermore, ensemble learning yielded additional gains of 0.89% and 3.60% over the best-performing single-model baselines. These findings underscore the importance of targeted input refinement and ensemble integration for robust and generalizable image retrieval systems. © 2025 Elsevier B.V., All rights reserved.

Abstract
Quality inspection inspection systems are critical for maintaining product integrity. Being a repetitive task, when performed by operators only, it can be slow and error-prone. This paper introduces an automated inspection system for quality assessment in casting aluminum parts resorting to a robotic system. The method comprises two processes: filing detection and hole inspection. For filing detection, five deep learning modes were trained. These models include an object detector and four instance segmentation models: YOLOv8, YOLOv8n-seg, YOLOv8s-seg, YOLOv8m-seg, and Mask R-CNN, respectively. Among these, YOLOv8s-seg exhibited the best overall performance, achieving a recall rate of 98.10%, critical for minimizing false negatives and yielding the best overall results. Alongside, the system inspects holes, utilizing image processing techniques like template-matching and blob detection, achieving a 97.30% accuracy and a 2.67% Percentage of Wrong Classifications. The system improves inspection precision and efficiency while supporting sustainability and ergonomic standards, reducing material waste and reducing operator fatigue. © The Author(s) 2025.

Abstract
Background: During their training, nurses must develop interprofessional collaboration skills, which are essential in clinical settings. Aim: This study aims to describe the development and testing stages of a virtual escape room, named Lock-down Treatment, to enhance interprofessional collaboration. Methods: The User-Centered Design methodology was used, involving users from requirement gathering to iterative prototyping. Requirements were established through interviews with 6 healthcare professionals, and a prototype was developed and tested for final assessment. Results: The results identified key areas for improvement, particularly in terms of timing and support during the game and demonstrated the effectiveness of the escape room in promoting interdisciplinary collaboration. This study proves that tools like escape rooms can significantly enrich nursing education. Conclusion: It is essential to integrate innovative methods into interprofessional training, making it more engaging and interactive. However, it is crucial that such tools are meticulously planned and validated to ensure their suitability through a rigorous validation process. Future research should evaluate the 'Lockdown Treatment' to assess its long-term effectiveness and applicability in clinical practice and patient outcomes. (c) 2025 The Authors. Published by Elsevier Inc. on behalf of Organization for Associate Degree Nursing. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

Abstract
Background Insufficient physical activity is one of the leading risk factors for death worldwide. Regular exercise can improve physical performance and quality of life, reduce the risks of falls and depressive symptoms, and reduce the likelihood of cognitive decline in older adults. Virtual reality (VR) and serious games (SG) are promising tools to improve physical and cognitive functioning. As part of the VR2Care project activities, four pilot sites explored the capabilities of the VR environment in a remote psychomotor training with SG and a hybrid approach with local groups of older adults performing physical activity.Objective The present study aimed to explore and measure the impact on older adults' quality of life and physical activity of using VR2Care solution and the level of usability, satisfaction and acceptance.Methods The study is a mixed method study, using qualitative and quantitative surveys to evaluate quality of life and physical activity of older users, and usability, satisfaction and acceptance of the solution. The data collection is a mix of investigator site data entry and users' self-reported data through the solutions or through online and paper-based means. Data were collected at baseline and after a follow-up of 6 weeks. Data are expressed as mean +/- standard deviation (SD) unless otherwise stated. Within the group, baseline to end of observation differences were assessed by paired sample t-test. A p = 0.05 was considered significant.Results No significant improvements in quality of life and physical activity were found. Little improvement, although not significant, in physical activity was found, comparing the Total MET average value of users who participated in phase I and II, therefore using SmartAL and Rehability. Little improvement, although not significant, in physical activity applies in >= 76 population. Users' feedback on usability, satisfaction and acceptance of VR2Care is generally positive. VR2Care was appreciated mostly for its usefulness in managing physical activity and the capacity to influence the consistency of attending physical activity sessions as prescribed by doctors.Conclusion Our results suggest that randomized controlled trial will be needed to assess correlations between specific features of the solution and health outcomes.