http://hdl.handle.net/10256/1515 2025-07-22T11:04:32Z 2025-07-22T11:04:32Z Martínez Álvarez, David Molero Grau, Aniol Calle Ortega, Eusebi Canals Ametller, Dolors Jové, Albert http://hdl.handle.net/10256/26902 2025-06-12T09:54:42Z 2025-10-01T00:00:00Z

Large-scale web tracking and cookie compliance: Evaluating one million websites under GDPR with AI categorization Martínez Álvarez, David; Molero Grau, Aniol; Calle Ortega, Eusebi; Canals Ametller, Dolors; Jové, Albert With the increasing prevalence of web-tracking technologies, including tracking cookies, pixel tracking, and browser fingerprinting techniques, there is a pressing need to analyze their impact on user privacy. Despite the growing interest in the scholarly literature, large-scale, fully automatic evaluations of website compliance with privacy regulations remain scarce. In this paper, we present new algorithms, methods, and an AI categorization model designed for massive, fully automatic analyses of web-tracking and cookie compliance and usage with and without valid user consent. Utilizing the recently published Website Evidence Collector (WEC) software from the European Data Protection Supervisor (EDPS), these algorithms are applied to assess over one million websites from Tranco's top list under European GDPR regulation. A novel 22-category multilabel AI model for website categorization provides content-based context to compliance results, achieving 96.56% accuracy and an F1 score of 0.963. Results reveal that nearly half of the websites utilize tracking cookies, while over half employ pixel tracking without user consent, thus highlighting significant differences between websites' content categories. Additionally, our analysis demonstrates how web-tracking power is concentrated among just a few companies, with the top 10 tracking firms being responsible for most compliance violations related to obtaining valid user consent. This paper serves as a foundation for ongoing large-scale web-tracking analyses, essential for understanding trends over time and evaluating the effectiveness of privacy regulations

2025-10-01T00:00:00Z González Vargas, Juan Camilo Fabregat Gesa, Ramon Carrillo-Ramos, Angela Jové Lagunas, Teodor http://hdl.handle.net/10256/26577 2025-03-14T08:48:31Z 2025-02-21T00:00:00Z

Evaluation of Motiv-ARCHE in the Santa Clara Museum González Vargas, Juan Camilo; Fabregat Gesa, Ramon; Carrillo-Ramos, Angela; Jové Lagunas, Teodor Currently, heritage sites, such as museums, have focused on the preservation and conservation of heritage elements for present and future generations. However, when displaying their content, they often do not consider different types of visitors, their preferences regarding the type and format of content, their interests, or their information needs (the same content is always presented). All of this can reduce the number of visits and the motivation of visitors. To address this issue, Motiv-ARCHE was developed as an application designed to enhance motivation in learning about cultural and natural heritage using augmented reality (AR). Motiv-ARCHE was implemented using the design-based research (DBR) methodology, an iterative approach that allows user feedback. In this article, we concentrate on presenting an experiment conducted at the Santa Clara Museum (Bogotá) in which a group of 44 participants used Motiv-ARCHE to access content associated with 10 cultural heritage elements that had been previously co-created with heritage experts from the museum itself. To evaluate the experiment, motivation and technology acceptance tests were applied, along with a demographic questionnaire, to statistically analyze whether the examined variables influence motivation for learning about cultural and natural heritage. Among the results, it is noteworthy that users with greater knowledge of AR, cultural and natural heritage, and a higher frequency of using this type of application felt more motivated to learn about heritage elements

2025-02-21T00:00:00Z López Barajas, Salvador Sanz, Pedro José Marín Prades, Raúl Echagüe, Juan Realpe, Sebastian http://hdl.handle.net/10256/26548 2025-03-10T08:22:42Z 2025-01-01T00:00:00Z

Network congestion control algorithm for image transmission hri and visual light communications of an autonomous underwater vehicle for intervention López Barajas, Salvador; Sanz, Pedro José; Marín Prades, Raúl; Echagüe, Juan; Realpe, Sebastian In this study, the challenge of teleoperating robots in harsh environments such as underwater or in tunnels is addressed. In these environments, wireless communication networks are prone to congestion, leading to potential mission failures. Our approach integrates a Human-Robot Interface (HRI) with a network congestion control algorithm at the application level for conservative transmission of images using the Robot Operating System (ROS) framework. The system was designed to avoid network congestion by adjusting the image compression parameters and the transmission rate depending on the real-time network conditions. To evaluate its performance, the algorithm was tested in two wireless underwater use cases: pipe inspection and an intervention task. An Autonomous Underwater Vehicle for Intervention (I-AUV) equipped with a Visual Light Communication (VLC) modem was used. Characterization of the VLC network was performed while the robot performed trajectories in the tank. The results demonstrate that our approach allows an operator to perform wireless missions where teleoperation requires images and the network conditions are variable. This solution provides a robust framework for image transmission and network control in the application layer, which allows for integration with any ROS-based system

2025-01-01T00:00:00Z Regalado, María Carreras Blesa, Nuria Mata Miquel, Christian Oliver i Malagelada, Arnau Lladó Bardera, Xavier Agut, Thais http://hdl.handle.net/10256/26383 2025-03-13T08:22:00Z 2025-03-01T00:00:00Z

Automatic Segmentation of Sylvian Fissure in Brain Ultrasound Images of Pre-Term Infants Using Deep Learning Models Regalado, María; Carreras Blesa, Nuria; Mata Miquel, Christian; Oliver i Malagelada, Arnau; Lladó Bardera, Xavier; Agut, Thais Objective: Segmentation of brain sulci in pre-term infants is crucial for monitoring their development. While magnetic resonance imaging has been used for this purpose, cranial ultrasound (cUS) is the primary imaging technique used in clinical practice. Here, we present the first study aiming to automate brain sulci segmentation in pre-term infants using ultrasound images. Methods: Our study focused on segmentation of the Sylvian fissure in a single cUS plane (C3), although this approach could be extended to other sulci and planes. We evaluated the performance of deep learning models, specifically U-Net and ResU-Net, in automating the segmentation process in two scenarios. First, we conducted cross-validation on images acquired from the same ultrasound machine. Second, we applied fine-tuning techniques to adapt the models to images acquired from different vendors. Results: The ResU-Net approach achieved Dice and Sensitivity scores of 0.777 and 0.784, respectively, in the cross-validation experiment. When applied to external datasets, results varied based on similarity to the training images. Similar images yielded comparable results, while different images showed a drop in performance. Additionally, this study highlighted the advantages of ResU-Net over U-Net, suggesting that residual connections enhance the model's ability to learn and represent complex anatomical structures. Conclusion: This study demonstrated the feasibility of using deep learning models to automatically segment the Sylvian fissure in cUS images. Accurate sonographic characterisation of cerebral sulci can improve the understanding of brain development and aid in identifying infants with different developmental trajectories, potentially impacting later functional outcomes

2025-03-01T00:00:00Z