Information

19 pages, 700 KiB

Open AccessArticle

Driving International Collaboration Beyond Boundaries Through Hackathons: A Comparative Analysis of Four Hackathon Setups

by Alice Barana, Vasiliki Eirini Chatzea, Kelly Henao, Ania Maria Hildebrandt, Ilias Logothetis, Marina Marchisio Conte, Alexandros Papadakis, Alberto Rueda, Daniel Samoilovich, Georgios Triantafyllidis and Nikolas Vidakis

Information 2025, 16(6), 488; https://doi.org/10.3390/info16060488 - 12 Jun 2025

Abstract

Hackathon events have become increasingly popular in recent years as a modern tool for innovation in the education sector as they offer important learning advantages. Within the “INVITE” Erasmus+ project, four distinct hackathons were organized to bring together academic institutions, teachers, and students [...] Read more.

Hackathon events have become increasingly popular in recent years as a modern tool for innovation in the education sector as they offer important learning advantages. Within the “INVITE” Erasmus+ project, four distinct hackathons were organized to bring together academic institutions, teachers, and students in the design of innovative international virtual and blended collaborations. In addition, as part of the “INVITE” project, an Open Interactive Digital Ecosystem (digital platform) has been developed to facilitate hackathons organization and was tested within two of the events. This platform can enhance hosting action-training programs providing a shared open resources space for educators to contact peers and design projects. All four hackathons were held during 2024 and their duration and type (onsite, blended, hybrid, and online) varied significantly. However, all hackathon topics were related to sustainability, SDGs, and Green Agenda. In total, more than 220 participants enrolled in the four events, including students, researchers, and professors from different disciplines, age groups, and countries. All participants were provided with qualitative surveys to explore their satisfaction and experiences. The results compare different hackathon setups to reveal valuable insights regarding the optimal design for higher education hackathons. Full article

(This article belongs to the Special Issue Accessibility and Inclusion in Education: Enabling Digital Technologies)

► Show Figures

Figure 1

31 pages, 550 KiB

Open AccessReview

Advances in Application of Federated Machine Learning for Oncology and Cancer Diagnosis

by Mohammad Nasajpour, Seyedamin Pouriyeh, Reza M. Parizi, Meng Han, Fatemeh Mosaiyebzadeh, Yixin Xie, Liyuan Liu and Daniel Macêdo Batista

Information 2025, 16(6), 487; https://doi.org/10.3390/info16060487 - 12 Jun 2025

Abstract

Machine learning has brought about a revolutionary transformation in healthcare. It has traditionally been employed to create predictive models through training on locally available data. However, privacy concerns can sometimes impede the collection and integration of data from diverse sources. Conversely, a lack [...] Read more.

Machine learning has brought about a revolutionary transformation in healthcare. It has traditionally been employed to create predictive models through training on locally available data. However, privacy concerns can sometimes impede the collection and integration of data from diverse sources. Conversely, a lack of sufficient data may hinder the construction of accurate models, thereby limiting the ability to produce meaningful outcomes. Especially in the field of healthcare, collecting datasets centrally is challenging due to privacy concerns. Indeed, federated learning (FL) emerges as a sophisticated distributed machine learning approach that comes to the rescue in such scenarios. It allows multiple devices hosted at different institutions, like hospitals, to collaboratively train a global model without sharing raw data. In addition, each device retains its data securely on locally, addressing the challenges of time-consuming annotation and privacy concerns. In this paper, we conducted a comprehensive literature review aimed at identifying the most advanced federated learning applications in cancer research and clinical oncology analysis. Our main goal was to present a comprehensive overview of the development of federated learning in the field of oncology. Additionally, we discuss the challenges and future research directions. Full article

(This article belongs to the Special Issue From Data to Diagnosis: Recent Advances of Machine Learning in Biomedical and Health Informatics)

► Show Figures

Figure 1

26 pages, 4278 KiB

Open AccessArticle

The Interpretative Effects of Normalization Techniques on Complex Regression Modeling: An Application to Real Estate Values Using Machine Learning

by Debora Anelli, Pierluigi Morano, Francesco Tajani and Maria Rosaria Guarini

Information 2025, 16(6), 486; https://doi.org/10.3390/info16060486 - 11 Jun 2025

Abstract

The performance of machine learning models depends on several factors, including data normalization, which can significantly improve its accuracy. There are many standardization techniques, and none is universally suitable; the choice depends on the characteristics of the problem, the predictive task, and the [...] Read more.

The performance of machine learning models depends on several factors, including data normalization, which can significantly improve its accuracy. There are many standardization techniques, and none is universally suitable; the choice depends on the characteristics of the problem, the predictive task, and the needs of the model used. This study analyzes how normalization techniques influence the outcomes of real estate price regression models using machine learning to uncover complex relationships between urban and economic factors. Six normalization techniques are employed to assess how they affect the estimation of relationships between property value and factors like social degradation, resident population, per capita income, green spaces, building conditions, and degraded neighborhood presence. The study’s findings underscore the pivotal role of normalization in shaping the perception of variables, accentuating critical thresholds, or distorting anticipated functional relationships. The work is the first application of a methodological approach to define the best technique on the basis of two criteria: statistical reliability and empirical evidence of the functional relationships obtainable with each standardization technique. Notably, the study underscores the potential of machine-learning-based regression to circumvent the limitations of conventional models, thereby yielding more robust and interpretable results. Full article

23 pages, 2407 KiB

Open AccessArticle

Enhancing Quantum Information Distribution Through Noisy Channels Using Quantum Communication Architectures

by Francisco Delgado

Information 2025, 16(6), 485; https://doi.org/10.3390/info16060485 - 11 Jun 2025

Abstract

Quantum information transmission is subject to imperfections in communication processes and systems. These phenomena alter the original content due to decoherence and noise. However, suitable communication architectures incorporating quantum and classical redundancy can selectively remove these errors, boosting destructive interference. In this work, [...] Read more.

Quantum information transmission is subject to imperfections in communication processes and systems. These phenomena alter the original content due to decoherence and noise. However, suitable communication architectures incorporating quantum and classical redundancy can selectively remove these errors, boosting destructive interference. In this work, a selection of architectures based on path superposition or indefinite causal order were analyzed under appropriate configurations, alongside traditional methods such as classical redundancy, thus enhancing transmission. For that purpose, we examined a broad family of decoherent channels associated with the qubit chain transmission by passing through tailored arrangements or composite architectures of imperfect channels. The outcomes demonstrated that, when combined with traditional redundancy, these configurations could significantly improve the transmission across a substantial subset of the channels. For quantum key distribution purposes, two alternative bases were considered to encode the information chain. Because a control system must be introduced in the proposed architectures, two strategies for its disposal at the end of the communication process were compared: tracing and measurement. In addition, eavesdropping was also explored under a representative scenario, to quantify its impact on the most promising architecture analyzed. Thus, in terms of transmission quality and security, the analysis revealed significant advantages over direct transmission schemes. Full article

(This article belongs to the Section Information and Communications Technology)

► Show Figures

Figure 1

19 pages, 546 KiB

Open AccessArticle

Antecedents and Consequences of Flow Experience in Virtual Reality Tourism: A Path Analysis of Visit Intention

by Lei Zhou, Huaqing Zhou, Xiaotang Cui and Jing Zhao

Information 2025, 16(6), 484; https://doi.org/10.3390/info16060484 - 11 Jun 2025

Abstract

This study examines the psychological mechanisms underlying virtual reality (VR) tourism experiences through an integrated theoretical framework centered on flow experience and visit destination intention. Drawing upon flow theory, the research investigates how interactivity, perceived vividness, and telepresence influence flow experience and subsequently [...] Read more.

This study examines the psychological mechanisms underlying virtual reality (VR) tourism experiences through an integrated theoretical framework centered on flow experience and visit destination intention. Drawing upon flow theory, the research investigates how interactivity, perceived vividness, and telepresence influence flow experience and subsequently affect hedonic motivation and perceived visual appeal in VR tourism contexts. Using partial least squares structural equation modeling (PLS-SEM) analysis of data collected from 255 VR tourism users across major Chinese metropolitan centers, the study reveals that perceived vividness and telepresence significantly impact flow experience, while interactivity shows no significant effect. Flow experience demonstrates significant positive relationships with hedonic motivation and perceived visual appeal. Furthermore, hedonic motivation and perceived visual appeal significantly positively affect visit destination intention. The findings advance the theoretical understanding of VR tourism by illuminating the psychological pathways through which technological characteristics influence behavioral intentions. These results offer practical implications for destination marketers and VR tourism developers in designing more effective virtual experiences that enhance destination visit intentions. Full article

(This article belongs to the Special Issue Extended Reality and Its Applications)

► Show Figures

Figure 1

24 pages, 354 KiB

Open AccessArticle

Dynamic Mixture of Experts for Adaptive Computation in Character-Level Transformers

by Zhigao Huang, Musheng Chen and Shiyan Zheng

Information 2025, 16(6), 483; https://doi.org/10.3390/info16060483 - 11 Jun 2025

Abstract

This paper challenges the prevailing assumption that Mixture of Experts (MoE) consistently improves computational efficiency through a systematic evaluation of MoE variants in Transformer models. We implement and compare three approaches: basic MoE, top-k routing, and capacity-factored routing, each progressively addressing load-balancing [...] Read more.

This paper challenges the prevailing assumption that Mixture of Experts (MoE) consistently improves computational efficiency through a systematic evaluation of MoE variants in Transformer models. We implement and compare three approaches: basic MoE, top-k routing, and capacity-factored routing, each progressively addressing load-balancing challenges. Our experiments reveal critical trade-offs between performance and efficiency: while MoE models maintain validation performance comparable to baselines, they require significantly longer training times (a 50% increase) and demonstrate reduced inference speeds (up to 56% slower). Analysis of routing behavior shows that even with load-balancing techniques, expert utilization remains unevenly distributed. These findings provide empirical evidence that MoE’s computational benefits are highly dependent on model scale and task characteristics, challenging common assumptions about sparse architectures and offering crucial guidance for adaptive neural architecture design across different computational constraints. Full article

(This article belongs to the Section Information Processes)

► Show Figures

Figure 1

35 pages, 2584 KiB

Open AccessArticle

A Framework for Participatory Creation of Digital Futures: A Longitudinal Study on Enhancing Media Literacy and Inclusion in K-12 Through Virtual Reality

by Chrysoula Lazou and Avgoustos Tsinakos

Information 2025, 16(6), 482; https://doi.org/10.3390/info16060482 - 11 Jun 2025

Abstract

The present study explores the affordances of virtual reality (VR) technologies to enhance digital and media literacy skills within an interdisciplinary and inclusive K-12 English as a Foreign Language (EFL) learning context. Addressing gaps in research on the design and impact of VR [...] Read more.

The present study explores the affordances of virtual reality (VR) technologies to enhance digital and media literacy skills within an interdisciplinary and inclusive K-12 English as a Foreign Language (EFL) learning context. Addressing gaps in research on the design and impact of VR experiences in secondary education, the study investigates VR affordances not only as a learning tool, but also as a medium for knowledge co-creation through learning by doing, with students acting as the agents within digital social contexts. The study was conducted for two years, with 59 participants aged 13–14 years old, following a structured five-phase intervention model with the intent to comply with DigComp 2.2 guidelines for digital citizenship and the Universal Design for Learning (UDL) for inclusive educational practices. The phases involved (a) training on the technological level to leverage digital tools; (b) media and information literacy (MIL) instruction in VR; (c) collaborative VR artifact creation; (d) peer evaluation; and (e) dissemination with peers from other sociocultural contexts for an iterative process of continuous content improvement and social discourse. Mixed methods data collection included pre/post-course surveys, pre/post-tests, observation journals, and student-generated VR artifact evaluations. The findings indicate consistent learning gains across both years, with an average pre–post gain of 18 points (Cohen’s d = −2.25; t = −17.3, p < 0.001). The VR-supported intervention fostered complex skillset building within a VR-supported dynamic learning environment that caters to diverse needs. Students’ reflections informed a framework for designing inclusive media literacy in VR, structured around three main pillars: Narrative Structure, Strategic Design, and Representation Awareness. These themes encapsulate the practical, cognitive, and ethical dimensions of VR design. Sub-themes with examples contribute to understanding the key design elements of VR in promoting participatory engagement, digital and media literacy, critical discourse, and inclusive education. The sub-themes per pillar are signaling and multisensory cues, storyline, and artful thinking; schema formation, multimedia encoding, and optimal cognitive load; and bias-free, respect for emotional impact, and language and symbols. Complementary quantitative findings confirmed the themes of the proposed framework, revealing a positive correlation between the perceived ease of use (PEoU) with digital skills development and a negative correlation between perceived usefulness (PU) and cognitive load. The study concludes with recommendations for pedagogy, curriculum design, and future research to empower learners in shaping sustainable digital futures. Full article

(This article belongs to the Special Issue Innovation in Education, Training and Game Design with Immersive Technologies and Spatial Computing)

► Show Figures

Graphical abstract

20 pages, 669 KiB

Open AccessArticle

Interference Management in UAV-Assisted Multi-Cell Networks

by Muchen Jiang, Honglin Ren, Yongxing Qi and Ting Wu

Information 2025, 16(6), 481; https://doi.org/10.3390/info16060481 - 10 Jun 2025

Viewed by 67

Abstract

This article considers a multi-cell wireless network comprising of conventional user equipment (UE), sensor devices and unmanned aerial vehicles (UAVs) or drones. UAVs are used to assist a base station, e.g., improve coverage or collect data from sensor devices. The problem at hand [...] Read more.

This article considers a multi-cell wireless network comprising of conventional user equipment (UE), sensor devices and unmanned aerial vehicles (UAVs) or drones. UAVs are used to assist a base station, e.g., improve coverage or collect data from sensor devices. The problem at hand is to optimize the (i) sub-carrier assigned to a cell or base station, (ii) position of each UAV, and (iii) transmit power of the following nodes: base stations and UAVs. We outline a two-stage approach to maximize the fairness-aware sum-rate of UE and UAVs. In the first stage, a genetic algorithm (GA)-based approach is used to assign a sub-band to all cells and to determine the location of each UAV. Then, in the second stage, a linear program is used to determine the transmit power of UE and UAVs. The results demonstrate that our proposed two-stage approach achieves approximately 97.43% of the optimal fairness-aware sum-rate obtained via brute-force search. It also attains on average 98.78% of the performance of a computationally intensive benchmark that requires over 478% longer run-time. Furthermore, it outperforms a conventional GA-based sub-band allocation heuristic by 221.39%. Full article

► Show Figures

Figure 1

19 pages, 1587 KiB

Open AccessArticle

Uncovering Key Factors of Student Performance in Math: An Explainable Deep Learning Approach Using TIMSS 2019 Data

by Abdelamine Elouafi, Ilyas Tammouch, Souad Eddarouich and Raja Touahni

Information 2025, 16(6), 480; https://doi.org/10.3390/info16060480 - 10 Jun 2025

Viewed by 81

Abstract

In 2019, the TIMSS study offered a closer look at how Moroccan eighth-grade students were doing in mathematics. The data came from a sample of 8390 students; 37% performed well, while the remaining 63% struggled. The goal was to better understand which contextual [...] Read more.

In 2019, the TIMSS study offered a closer look at how Moroccan eighth-grade students were doing in mathematics. The data came from a sample of 8390 students; 37% performed well, while the remaining 63% struggled. The goal was to better understand which contextual factors truly influence academic success. The dataset was dense, with over 700 variables drawn from students, teachers, and school questionnaires. To make sense of it, advanced machine learning techniques were applied, including an autoencoder to reduce dimensionality. This process helped narrow things down to 20 key variables that strongly correlated with student performance. These factors covered a range of influences, from teaching strategies and student engagement to teacher training and school-level resources. The insights from the study offer practical guidance for educators and policymakers looking to design targeted, effective interventions. At its core, the study underscores a familiar truth: success in math does not hinge on a single element but on a web of interconnected conditions. Improving outcomes requires a holistic approach, one that supports both learners and the people guiding them. Full article

(This article belongs to the Special Issue Artificial Intelligence and Games Science in Education)

► Show Figures

Figure 1

16 pages, 1978 KiB

Open AccessArticle

Learning-Assisted Multi-IMU Proprioceptive State Estimation for Quadruped Robots

by Xuanning Liu, Yajie Bao, Peng Cheng, Dan Shen, Zhengyang Fan, Hao Xu and Genshe Chen

Information 2025, 16(6), 479; https://doi.org/10.3390/info16060479 - 9 Jun 2025

Viewed by 33

Abstract

This paper presents a learning-assisted approach for state estimation of quadruped robots using observations of proprioceptive sensors, including multiple inertial measurement units (IMUs). Specifically, one body IMU and four additional IMUs attached to each calf link of the robot are used for sensing [...] Read more.

This paper presents a learning-assisted approach for state estimation of quadruped robots using observations of proprioceptive sensors, including multiple inertial measurement units (IMUs). Specifically, one body IMU and four additional IMUs attached to each calf link of the robot are used for sensing the dynamics of the body and legs, in addition to joint encoders. The extended Kalman filter (KF) is employed to fuse sensor data to estimate the robot’s states in the world frame and enhance the convergence of the extended KF (EKF). To circumvent the requirements for the measurements from the motion capture (mocap) system or other vision systems, the right-invariant EKF (RI-EKF) is extended to employ the foot IMU measurements for enhanced state estimation, and a learning-based approach is presented to estimate the vision system measurements for the EKF. One-dimensional convolutional neural networks (CNN) are leveraged to estimate required measurements using only the available proprioception data. Experiments on real data from a quadruped robot demonstrate that proprioception can be sufficient for state estimation. The proposed learning-assisted approach, which does not rely on data from vision systems, achieves competitive accuracy compared to EKF using mocap measurements and lower estimation errors than RI-EKF using multi-IMU measurements. Full article

(This article belongs to the Special Issue Sensing and Wireless Communications)

► Show Figures

Figure 1

15 pages, 3156 KiB

Open AccessArticle

Adaptive AR Navigation: Real-Time Mapping for Indoor Environment Using Node Placement and Marker Localization

by Bagas Samuel Christiananta Putra, I. Kadek Dendy Senapartha, Jyun-Cheng Wang, Matahari Bhakti Nendya, Dan Daniel Pandapotan, Felix Nathanael Tjahjono and Halim Budi Santoso

Information 2025, 16(6), 478; https://doi.org/10.3390/info16060478 - 7 Jun 2025

Viewed by 170

Abstract

Indoor navigation remains a challenge due to the limitations of GPS-based systems in enclosed environments. Current approaches, such as marker-based ones, have been developed for indoor navigation. However, it requires extensive manual mapping and makes indoor navigation time-consuming and difficult to scale. To [...] Read more.

Indoor navigation remains a challenge due to the limitations of GPS-based systems in enclosed environments. Current approaches, such as marker-based ones, have been developed for indoor navigation. However, it requires extensive manual mapping and makes indoor navigation time-consuming and difficult to scale. To enhance current approaches to indoor navigation, this study proposes a node-based mapping for indoor navigation, allowing users to dynamically construct navigation paths using a mobile device. The system leverages NavMesh, the A* algorithm for pathfinding, and is integrated into the ARCore for real-time AR guidance. Nodes are placed within the environment to define walkable paths, which can be stored and reused without requiring a full system to rebuild. Once the prototype has been developed, usability testing is conducted using the Handheld Augmented Reality Usability Scale (HARUS) to evaluate manipulability, comprehensibility, and overall usability. This study finds that using node-based mapping for indoor navigation can help enhance flexibility in mapping new indoor spaces and offers an effective AR-guided navigation experience. However, there are some areas of improvement, including interface clarity and system scalability, that can be considered for future research. This study contributes practically to improving current practices in adaptive indoor navigation systems using AR-based dynamic mapping techniques. Full article

(This article belongs to the Special Issue Extended Reality: A New Way of Interacting with the World, 2nd Edition)

► Show Figures

Figure 1

24 pages, 4943 KiB

Open AccessArticle

Towards MR-Only Radiotherapy in Head and Neck: Generation of Synthetic CT from Zero-TE MRI Using Deep Learning

by Souha Aouadi, Mojtaba Barzegar, Alla Al-Sabahi, Tarraf Torfeh, Satheesh Paloor, Mohamed Riyas, Palmira Caparrotti, Rabih Hammoud and Noora Al-Hammadi

Information 2025, 16(6), 477; https://doi.org/10.3390/info16060477 - 6 Jun 2025

Viewed by 157

Abstract

This study investigates the generation of synthetic CT (sCT) images from zero echo time (ZTE) MRI to support MR-only radiotherapy, which can reduce image registration errors and lower treatment planning costs. Since MRI lacks the electron density data required for accurate dose calculations, [...] Read more.

This study investigates the generation of synthetic CT (sCT) images from zero echo time (ZTE) MRI to support MR-only radiotherapy, which can reduce image registration errors and lower treatment planning costs. Since MRI lacks the electron density data required for accurate dose calculations, generating reliable sCTs is essential. ZTE MRI, offering high bone contrast, was used with two deep learning models: attention deep residual U-Net (ADR-Unet) and derived conditional generative adversarial network (cGAN). Data from 17 head and neck cancer patients were used to train and evaluate the models. ADR-Unet was enhanced with deep residual blocks and attention mechanisms to improve learning and reconstruction quality. Both models were implemented in-house and compared to standard U-Net and Unet++ architectures using image quality metrics, visual inspection, and dosimetric analysis. Volumetric modulated arc therapy (VMAT) planning was performed on both planning CT and generated sCTs. ADR-Unet achieved a mean absolute error of 55.49 HU and a Dice score of 0.86 for bone structures. All the models demonstrated Gamma pass rates above 99.4% and dose deviations within 2–3%, confirming clinical acceptability. These results highlight ADR-Unet and cGAN as promising solutions for accurate sCT generation, enabling effective MR-only radiotherapy. Full article

(This article belongs to the Special Issue Deep Learning in Medical Image Analysis: Foundations, Techniques, and Applications)

► Show Figures

Figure 1

21 pages, 721 KiB

Open AccessArticle

Benchmarking Variants of Recursive Feature Elimination: Insights from Predictive Tasks in Education and Healthcare

by Okan Bulut, Bin Tan, Elisabetta Mazzullo and Ali Syed

Information 2025, 16(6), 476; https://doi.org/10.3390/info16060476 - 6 Jun 2025

Viewed by 172

Abstract

Originally developed as an effective feature selection method in healthcare predictive analytics, Recursive Feature Elimination (RFE) has gained increasing popularity in Educational Data Mining (EDM) due to its ability to handle high-dimensional data and support interpretable modeling. Over time, various RFE variants have [...] Read more.

Originally developed as an effective feature selection method in healthcare predictive analytics, Recursive Feature Elimination (RFE) has gained increasing popularity in Educational Data Mining (EDM) due to its ability to handle high-dimensional data and support interpretable modeling. Over time, various RFE variants have emerged, each introducing methodological enhancements. To help researchers better understand and apply RFE more effectively, this study organizes existing variants into four methodological categories: (1) integration with different machine learning models, (2) combinations of multiple feature importance metrics, (3) modifications to the original RFE process, and (4) hybridization with other feature selection or dimensionality reduction techniques. Rather than conducting a systematic review, we present a narrative synthesis supported by illustrative studies from EDM to demonstrate how different variants have been applied in practice. We also conduct an empirical evaluation of five representative RFE variants across two domains: a regression task using a large-scale educational dataset and a classification task using a clinical dataset on chronic heart failure. Our evaluation benchmarks predictive accuracy, feature selection stability, and runtime efficiency. Results show that the evaluation metrics vary significantly across RFE variants. For example, while RFE wrapped with tree-based models such as Random Forest and Extreme Gradient Boosting (XGBoost) yields strong predictive performance, these methods tend to retain large feature sets and incur high computational costs. In contrast, a variant known as Enhanced RFE achieves substantial feature reduction with only marginal accuracy loss, offering a favorable balance between efficiency and performance. These findings underscore the trade-offs among accuracy, interpretability, and computational cost across RFE variants, providing practical guidance for selecting the most appropriate algorithm based on domain-specific needs and constraints. Full article

► Show Figures

Figure 1

21 pages, 1126 KiB

Open AccessArticle

Spectral Adaptive Dropout: Frequency-Based Regularization for Improved Generalization

by Zhigao Huang, Musheng Chen and Shiyan Zheng

Information 2025, 16(6), 475; https://doi.org/10.3390/info16060475 - 6 Jun 2025

Viewed by 203

Abstract

Deep neural networks are often susceptible to overfitting, necessitating effective regularization techniques. This paper introduces Spectral Adaptive Dropout, a novel frequency-based regularization technique that dynamically adjusts dropout rates based on the spectral characteristics of network gradients. The proposed approach addresses the limitations of [...] Read more.

Deep neural networks are often susceptible to overfitting, necessitating effective regularization techniques. This paper introduces Spectral Adaptive Dropout, a novel frequency-based regularization technique that dynamically adjusts dropout rates based on the spectral characteristics of network gradients. The proposed approach addresses the limitations of traditional dropout methods by adaptively targeting high-frequency components that typically contribute to overfitting while preserving essential low-frequency information. Through extensive experimentation on character-level language modeling tasks, the study demonstrates that the method achieves a 1.10% improvement in validation loss while maintaining competitive inference speeds. Thise research explores several implementations including FFT-based analysis, wavelet decomposition, and per-attention-head adaptation, culminating in an optimized approach that balances computational efficiency with regularization effectiveness. Our results highlight the significant potential of incorporating frequency-domain information into regularization strategies for deep neural networks. Full article

(This article belongs to the Special Issue Intelligent Information Technology)

► Show Figures

Figure 1

20 pages, 3265 KiB

Open AccessArticle

Enhancing Rare Class Performance in HOI Detection with Re-Splitting and a Fair Test Dataset

by Gyubin Park and Afaque Manzoor Soomro

Information 2025, 16(6), 474; https://doi.org/10.3390/info16060474 - 6 Jun 2025

Viewed by 203

Abstract

In Human–Object Interaction (HOI) detection, class imbalance severely limits the performance of a model on infrequent interaction categories. To overcome this problem, a Re-Splitting algorithm has been developed. This algorithm implements DreamSim-based clustering and performs k-means-based partitioning to restructure the train–test splits. By [...] Read more.

In Human–Object Interaction (HOI) detection, class imbalance severely limits the performance of a model on infrequent interaction categories. To overcome this problem, a Re-Splitting algorithm has been developed. This algorithm implements DreamSim-based clustering and performs k-means-based partitioning to restructure the train–test splits. By doing so, the approach balances the rarities and frequent classes of interaction equally, thereby increasing robustness. A Real-World test dataset has also been introduced. This dataset is comparable to a truly independent benchmark. It is designed to address class distribution bias, which is commonly present in traditional test sets. However, as shown in the Experiment and Evaluation subsection, a high level of performance can be achieved for the general case using different few-shot and rare-class training instances. Models trained solely on the re-split dataset show significant improvements in rare-class mAP, particularly for one-stage models. Evaluation on the test dataset from the real world further emphasizes previously overlooked model performance and supports fair structuring of dataset. The methods are validated with extensive experiments using five one-stage and two two-stage models. Our analysis shows that reshaping dataset distributions increases rare-class detection by as much as 8.0 mAP. This study paves the way for balanced training and evaluation leading to the formulation of a general framework for scalable, fair, and generalizable HOI detection. Full article

(This article belongs to the Special Issue Addressing Real-World Challenges in Recognition and Classification with Cutting-Edge AI Models and Methods)

► Show Figures

Figure 1

15 pages, 1000 KiB

Open AccessArticle

Integrating Large Language Models into Accessible and Inclusive Education: Access Democratization and Individualized Learning Enhancement Supported by Generative Artificial Intelligence

by Inigo Lopez-Gazpio

Information 2025, 16(6), 473; https://doi.org/10.3390/info16060473 - 3 Jun 2025

Viewed by 293

Abstract

This study explores the integration of large language models (LLMs) into educational environments, emphasizing enhanced accessibility, inclusivity, and individualized learning experiences. The study evaluates trends in the transformative potential of artificial intelligence (AI) technologies in their capacity to significantly mitigate traditional barriers related [...] Read more.

This study explores the integration of large language models (LLMs) into educational environments, emphasizing enhanced accessibility, inclusivity, and individualized learning experiences. The study evaluates trends in the transformative potential of artificial intelligence (AI) technologies in their capacity to significantly mitigate traditional barriers related to language diversity, learning disabilities, cultural differences, and socioeconomic inequalities. The result of the analysis highlights how LLMs personalize instructional content and dynamically respond to each learner’s educational and emotional needs. The work also advocates for an instructor-guided deployment of LLMs as pedagogical catalysts rather than replacements, emphasizing educators’ role in ethical oversight, cultural sensitivity, and emotional support within AI-enhanced classrooms. Finally, while recognizing concerns regarding data privacy, potential biases, and ethical implications, the study argues that the proactive and responsible integration of LLMs by educators is necessary for democratizing access to education and to foster inclusive learning practices, thereby advancing the effectiveness and equity of contemporary educational frameworks. Full article

(This article belongs to the Special Issue Accessibility and Inclusion in Education: Enabling Digital Technologies)

► Show Figures

Figure 1

22 pages, 1448 KiB

Open AccessArticle

A Framework for Generative AI-Driven Assessment in Higher Education

by Galina Ilieva, Tania Yankova, Margarita Ruseva and Stanimir Kabaivanov

Information 2025, 16(6), 472; https://doi.org/10.3390/info16060472 - 3 Jun 2025

Viewed by 333

Abstract

The rapid integration of generative artificial intelligence (AI) into educational environments raises both opportunities and concerns regarding assessment design, academic integrity, and quality assurance. While new generation AI tools offer new modes of interactivity, feedback, and content generation, their use in assessment remains [...] Read more.

The rapid integration of generative artificial intelligence (AI) into educational environments raises both opportunities and concerns regarding assessment design, academic integrity, and quality assurance. While new generation AI tools offer new modes of interactivity, feedback, and content generation, their use in assessment remains insufficiently pedagogically framed and regulated. In this study, we propose a new framework for generative AI-supported assessment in higher education, structured around the needs and responsibilities of three key stakeholders (branches): instructors, students, and control authorities. The framework outlines how teaching staff can design adaptive and AI-informed tasks and provide feedback, how learners can engage with these tools transparently, and how institutional bodies can ensure accountability through compliance standards, policies, and audits. This three-branch multi-level model contributes to the emerging discourse on responsible AI adoption in higher education by offering a holistic approach for integrating AI-based systems into assessment practices while safeguarding academic values and quality. Full article

(This article belongs to the Special Issue Generative AI Technologies: Shaping the Future of Higher Education)

► Show Figures

Figure 1

27 pages, 583 KiB

Open AccessSystematic Review

Public Engagement Through Programming in Archives: A Systematic Review of Activities and Resultant Outcomes

by Josiline Chigwada, Mthokozisi Masumbika Ncube and Patrick Ngulube

Information 2025, 16(6), 471; https://doi.org/10.3390/info16060471 - 3 Jun 2025

Viewed by 246

Abstract

Archives have the potential to contribute to national development by preserving historical records and providing access to information. However, their impact is constrained by ineffective outreach strategies, insufficient institutional investment, and low public visibility. Public programming has been used as a strategic approach [...] Read more.

Archives have the potential to contribute to national development by preserving historical records and providing access to information. However, their impact is constrained by ineffective outreach strategies, insufficient institutional investment, and low public visibility. Public programming has been used as a strategic approach to bridge the gap between archival institutions and their user communities through engagement initiatives. Therefore, the objective of this study was to systematically review and analyse the diverse public programming activities undertaken by archival institutions globally and to identify the resultant outcomes of these engagements. To achieve this, the study employed a systematic literature review methodology, examining scholarly publications to synthesise existing evidence on public engagement in archives, thereby providing a comprehensive overview of current practices and their demonstrated impacts. The systematic review was conducted in accordance with PRISMA guidelines, utilising a two-stage selection process involving a search of six databases and four specialised journals. This search yielded 39 publications that met the inclusion criteria. Methodological rigour was evaluated using the CASP checklist. The results from the study indicated that exhibitions, educational programmes, community outreach, and digital initiatives were the most common public programming strategies. These activities enhance public awareness, increase accessibility, and foster community engagement. Despite the availability of various public programming activities, challenges such as inadequate funding, lack of digital infrastructure, and bureaucratic constraints hinder their effectiveness. The need for structured outreach strategies, institutional support, and the integration of emerging technologies to optimise public programming in archives is emphasised. The findings contribute to improving archival accessibility and user engagement in a rapidly evolving digital landscape. Full article

► Show Figures

Figure 1

25 pages, 1352 KiB

Open AccessSystematic Review

Systematic Review of Graph Neural Network for Malicious Attack Detection

by Sarah Mohammed Alshehri, Sanaa Abdullah Sharaf and Rania Abdullrahman Molla

Information 2025, 16(6), 470; https://doi.org/10.3390/info16060470 - 2 Jun 2025

Viewed by 401

Abstract

As cyberattacks continue to rise alongside the rapid expansion of digital systems, effective threat detection remains a critical yet challenging task. While several machine learning approaches have been proposed, the use of graph neural networks (GNNs) for cyberattack detection has not yet been [...] Read more.

As cyberattacks continue to rise alongside the rapid expansion of digital systems, effective threat detection remains a critical yet challenging task. While several machine learning approaches have been proposed, the use of graph neural networks (GNNs) for cyberattack detection has not yet been systematically explored in depth. This paper presents a systematic literature review (SLR) that analyzes 28 recent academic studies published between 2020 and 2025, retrieved from major databases including IEEE, ACM, Scopus, and Springer. The review focuses on evaluating how GNN models are applied in detecting various types of attacks, particularly those targeting IoT environments, web services, phishing, and network traffic. Studies were classified based on the type of dataset, GNN model architecture, and attack domain. Additionally, key limitations and future research directions were extracted and analyzed. The findings provide a structured comparison of current methodologies and highlight gaps that warrant further exploration. This review contributes a focused perspective on the potential of GNNs in cybersecurity and offers insights to guide future developments in the field. Full article

► Show Figures

Figure 1

30 pages, 1319 KiB

Open AccessArticle

AI-Powered Educational Agents: Opportunities, Innovations, and Ethical Challenges

by Diana-Margarita Córdova-Esparza

Information 2025, 16(6), 469; https://doi.org/10.3390/info16060469 - 31 May 2025

Viewed by 519

Abstract

Recent advances in large language models (LLMs) have triggered rapid growth in AI-powered educational agents, yet researchers and practitioners still lack a consolidated view of how these systems are engineered and validated. To address this gap, we conducted a systematic literature review of [...] Read more.

Recent advances in large language models (LLMs) have triggered rapid growth in AI-powered educational agents, yet researchers and practitioners still lack a consolidated view of how these systems are engineered and validated. To address this gap, we conducted a systematic literature review of 82 peer-reviewed and industry studies published from January 2023 to February 2025. Using a four-phase protocol, we extracted and coded them along six groups: technical and pedagogical frameworks, tutoring systems, assessment and feedback, curriculum design, personalization, and ethical considerations. Synthesizing these findings, we propose design principles that link technical choices to instructional goals and outline safeguards for privacy, fairness, and academic integrity. Across all domains, the evidence converges on a key insight: hybrid human–AI workflows, in which teachers curate and moderate LLM output, outperform fully autonomous tutors by combining scalable automation with pedagogical expertise. Limitations in the current literature, including short study horizons, small-sample experiments, and a bias toward positive findings, temper the generalizability of reported gains, highlighting the need for rigorous, long-term evaluations. Full article

(This article belongs to the Special Issue Advancing Educational Innovation with Artificial Intelligence)

► Show Figures

Figure 1

18 pages, 5468 KiB

Open AccessArticle

Symbolic Framework for Evaluation of NOMA Modulation Impairments Based on Irregular Constellation Diagrams

by Nenad Stefanovic, Vladimir Mladenovic, Borisa Jovanovic, Ron Dabora and Asutosh Kar

Information 2025, 16(6), 468; https://doi.org/10.3390/info16060468 - 31 May 2025

Viewed by 204

Abstract

Complexity of non-orthogonal multiple access (NOMA) digital signal processing schemes is particularly relevant in mobile environments because of the varying channel conditions of every single user. In contrast to legacy modulation and coding schemes (MCSs), NOMA MCSs typically have irregular symbol constellations with [...] Read more.

Complexity of non-orthogonal multiple access (NOMA) digital signal processing schemes is particularly relevant in mobile environments because of the varying channel conditions of every single user. In contrast to legacy modulation and coding schemes (MCSs), NOMA MCSs typically have irregular symbol constellations with asymmetric symbol decision regions affecting synchronization at the receiver. Research papers investigating signal processing in this emerging field usually lack sufficient details for facilitating software-defined radio (SDR) implementation. This work presents a new symbolic framework approach for simulating signal processing functions in SDR transmit–receive paths in a dynamic NOMA downlink use case. The proposed framework facilitates simple and intuitive implementation and testing of NOMA schemes and can be easily expanded and implemented on commercially available SDR hardware. We explicitly address several important design and measurement parameters and their relationship to different tasks, including variable constellation processing, carrier and symbol synchronization, and pulse shaping, focusing on quadrature amplitude modulation (QAM). The advantages of the proposed approach include intuitive symbolic modeling in a dynamic framework for NOMA signals; efficient, more accurate, and less time-consuming design flow; and generation of synthetic training data for machine-learning models that could be used for system optimization in real-world use cases. Full article

(This article belongs to the Special Issue Second Edition of Advances in Wireless Communications Systems)

► Show Figures

Figure 1

24 pages, 3152 KiB

Open AccessArticle

EHMQA-GPT: A Knowledge Augmented Large Language Model for Personalized Elderly Health Management

by Shaofu Lin, Yidan Duan, Tao Zhou, Xiliang Liu and Jiaojiao Wang

Information 2025, 16(6), 467; https://doi.org/10.3390/info16060467 - 30 May 2025

Viewed by 256

Abstract

Due to training limitations, general LLMs often lack sufficient accuracy and practicality in specialized domains such as elderly health management. To help alleviate this issue, this paper introduces EHMQA-GPT, the first domain-specific LLM tailored for non-specialist users (caregivers, elderly individuals, family members, and [...] Read more.

Due to training limitations, general LLMs often lack sufficient accuracy and practicality in specialized domains such as elderly health management. To help alleviate this issue, this paper introduces EHMQA-GPT, the first domain-specific LLM tailored for non-specialist users (caregivers, elderly individuals, family members, and community health workers) for low-risk, daily health consultations in real-world scenarios. EHMQA-GPT innovates in two aspects: (1) professional corpus construction: we established a multi-dimensional annotation system, integrating EHM-KB, EHM-SFT, and EHM-Eval, to achieve vector representation and hierarchical classification of domain knowledge; and (2) knowledge-enhanced large language model construction: based on ChatGLM3-6B, we integrated knowledge retrieval mechanisms and supervised fine-tuning strategies, enhanced the generation effect through knowledge base retrieval, and achieved deep alignment of domain knowledge through mixed supervised fine-tuning. The experimental verification part adopts testing in six fields. EHMQA-GPT has an accuracy rate of 78.1%, which is 22.3% higher than ChatGLM3-6B. Subjective assessment constructs a dual verification system (GPT-4 automatic scoring + gerontology expert blind review) and is significantly superior to the baseline model in three dimensions: knowledge accuracy (+38.9%), logical coherence (+39.4%), and practical guidance (+31.4%). The proposed framework and corpus provide a novel and scalable foundation for future research and deployment of LLMs in elderly health. Full article

► Show Figures

Figure 1

30 pages, 5729 KiB

Open AccessArticle

Signal-Induced Heap Transform-Based QR-Decomposition and Quantum Circuit for Implementing 3-Qubit Operations

by Artyom M. Grigoryan, Alexis Gomez, Isaac Espinoza and Sos S. Agaian

Information 2025, 16(6), 466; https://doi.org/10.3390/info16060466 - 30 May 2025

Viewed by 163

Abstract

This article presents a novel approach to the decomposition of unitary operations for 3-qubit systems by 28 controlled rotations and no permutations. The QR decomposition is described, which is based on the concept of the discrete signal-induced heap transform (DsiHT) and its quantum [...] Read more.

This article presents a novel approach to the decomposition of unitary operations for 3-qubit systems by 28 controlled rotations and no permutations. The QR decomposition is described, which is based on the concept of the discrete signal-induced heap transform (DsiHT) and its quantum analogue. This transform is generated by a given signal and may use different paths, or orders, of processing the data, and, among them, one can find paths that allow one to construct efficient quantum circuits for implementing multi-qubit unitary gates. The case of real unitary matrices is considered. The proposed approach is described in detail, and quantum circuits are presented for computing 3-qubit operations. This approach allowed us to write simple Qiskit codes to implement the decomposition of 3-qubit operations. Examples with quantum circuits for the quantum 3-qubit quantum cosine and Hartley transforms are described. Full article

(This article belongs to the Section Information Theory and Methodology)

► Show Figures

Figure 1

14 pages, 469 KiB

Open AccessArticle

SEO in Rural Tourism: A Case Study of Terras de Trás-os-Montes—Portugal

by Elisabete Paulo Morais, Elsa Tavares Esteves and Carlos R. Cunha

Information 2025, 16(6), 465; https://doi.org/10.3390/info16060465 - 30 May 2025

Viewed by 234

Abstract

This research investigates the application of search engine optimization (SEO) in developing the digital image of rural tourism businesses in the Terras de Trás-os-Montes region of Portugal. With digital marketing becoming increasingly important for businesses to stay competitive, SEO has become a vital [...] Read more.

This research investigates the application of search engine optimization (SEO) in developing the digital image of rural tourism businesses in the Terras de Trás-os-Montes region of Portugal. With digital marketing becoming increasingly important for businesses to stay competitive, SEO has become a vital tool for developing online recognition, qualified traffic acquisition, and enhancement of conversion rates. The research performs an SEO analysis of 21 rural tourism websites by applying the Ubersuggest tool, analyzing such key indicators as on-page SEO scores, organic traffic, keyword ranking, backlinks, and technical performance. The results identify wide SEO performance discrepancies, with some sites registering excellent practices and others with critical errors that impair the sites’ online recognizability. In particular, low word count, absent meta description, and loading speed issues are very much present. The research emphasizes the need for effective SEO methods, such as on-page maintenance, content creation, and link building, to advance search engine ranking and end-user experience. Moreover, the study emphasizes the necessity for rural tourism businesses to evolve and adapt to current SEO trends, i.e., voice search optimization and local SEO, in the changing digital business environment. The results provide recommendations for rural tourism businesses to develop their digital marketing activities and make progress online. Full article

► Show Figures

Figure 1

29 pages, 577 KiB

Open AccessArticle

Offloaded Computation for QoS Routing in Wireless Sensor Networks

by Basma Mostafa and Miklos Molnar

Information 2025, 16(6), 464; https://doi.org/10.3390/info16060464 - 30 May 2025

Viewed by 270

Abstract

In Wireless Sensor Networks (WSNs) used for real-time applications, ensuring Quality of Service (QoS) is essential for maintaining end-to-end performance guarantees. QoS requirements are typically defined by a set of end-to-end constraints, including delay, jitter, and packet loss. In multi-hop scenarios, this requires [...] Read more.

In Wireless Sensor Networks (WSNs) used for real-time applications, ensuring Quality of Service (QoS) is essential for maintaining end-to-end performance guarantees. QoS requirements are typically defined by a set of end-to-end constraints, including delay, jitter, and packet loss. In multi-hop scenarios, this requires multi-constrained path computation. This research examines the standard Routing Protocol for Low-Power and Lossy Networks (RPL), which employs a Destination-Oriented Directed Acyclic Graph (DODAG) for data transmission. Nonetheless, there are several challenges related to multi-constrained route computation in the RPL: (1) The DODAG originates from an objective function that cannot manage multiple constraints. (2) The process of computing multi-constrained routes is resource-intensive, even for a single path. (3) The collection of QoS-compatible paths does not necessarily form a DODAG. To address these challenges, this paper suggests modifications to the existing protocols that shift computationally demanding tasks to edge servers. Such a strategic adjustment allows for the implementation of QoS-compatible route computation in WSNs using the RPL. It enhances their ability to meet increasingly stringent demands for QoS in numerous application environments. Full article

(This article belongs to the Special Issue Internet of Things and Cloud-Fog-Edge Computing, 2nd Edition)

► Show Figures

Figure 1

26 pages, 2222 KiB

Open AccessArticle

Investigating Service Robot Acceptance Factors: The Role of Emotional Design, Communication Style, and Gender Groups

by Gang Ren, Xuezhen Wu, Zhihuang Huang and Baoyi Zhang

Information 2025, 16(6), 463; https://doi.org/10.3390/info16060463 - 30 May 2025

Viewed by 287

Abstract

Service robots (SRs) are increasingly deployed in commercial settings, yet the factors influencing their acceptance, particularly emotional design elements, remain understudied. This research investigates SR acceptance factors by integrating the technology acceptance model, the Computers Are Social Actors (CASA) framework, Kansei engineering (KE), [...] Read more.

Service robots (SRs) are increasingly deployed in commercial settings, yet the factors influencing their acceptance, particularly emotional design elements, remain understudied. This research investigates SR acceptance factors by integrating the technology acceptance model, the Computers Are Social Actors (CASA) framework, Kansei engineering (KE), and social presence theory (SPT) to examine how design elements influence user responses to SRs. Using structural equation modeling of survey data from 318 shoppers and hotel guests in China, we tested relationships between CASA attributes, emotional perceptions, social presence, and usage intention. The results revealed that communication style produced the strongest effects across all emotional dimensions, with cuteness and coolness directly influencing usage intention, while warmth and novelty operate through social presence mediation. Multi-group analysis identified significant gender differences in response patterns: male users prioritized communication-driven perceptions while female users responded more strongly to coolness attributes. These findings extend our understanding of acceptance factors in service robot adoption, highlighting the critical roles of emotional design, communication style, and gender differences, while suggesting differentiated design approaches for diverse user segments. Full article

(This article belongs to the Special Issue Artificial Intelligence Methods for Human-Computer Interaction)

► Show Figures

Figure 1

19 pages, 569 KiB

Open AccessArticle

Artificial Intelligence from Google Environment for Effective Learning Assessment

by Sergio Miranda

Information 2025, 16(6), 462; https://doi.org/10.3390/info16060462 - 30 May 2025

Viewed by 263

Abstract

This study investigates the use of Google NotebookLM for the automatic generation of educational assessment items. A mixed-methods approach was adopted, combining quantitative psychometric evaluation with qualitative student feedback. Six tests, each composed of 15 multiple-choice questions generated from diverse sources such as [...] Read more.

This study investigates the use of Google NotebookLM for the automatic generation of educational assessment items. A mixed-methods approach was adopted, combining quantitative psychometric evaluation with qualitative student feedback. Six tests, each composed of 15 multiple-choice questions generated from diverse sources such as PDFs, web slides, and YouTube videos, were administered to undergraduate students. Quantitative analysis involved calculating key indices which confirmed that many AI-generated items met acceptable psychometric criteria, though some items revealed reliability concerns and potential bias. Concurrently, a structured questionnaire assessed the clarity, relevance, and fairness of the test items. Students generally rated the AI-generated questions positively in terms of clarity and pedagogical alignment, while also noting areas for improvement. In conclusion, the findings suggest that generative AI can offer a scalable and efficient solution for test item creation; however, further methodological refinements are needed to ensure consistent validity, reliability, and ethical fairness in learning assessments. Full article

► Show Figures

Figure 1

16 pages, 334 KiB

Open AccessArticle

Exploring the Merchera Ethnic Group Through ChatGPT: The Risks of Epistemic Exclusion

by Soraya Oronoz, Albert Miró Pérez and Juan Peña-Martínez

Information 2025, 16(6), 461; https://doi.org/10.3390/info16060461 - 30 May 2025

Viewed by 610

Abstract

The study of underrepresented ethnic groups in the social sciences is often hindered by structural and epistemic barriers that limit access to culturally embedded knowledge. This article examines the potential of the GPT-4 version of the ChatGPT language model as a complementary research [...] Read more.

The study of underrepresented ethnic groups in the social sciences is often hindered by structural and epistemic barriers that limit access to culturally embedded knowledge. This article examines the potential of the GPT-4 version of the ChatGPT language model as a complementary research tool used to generate insights into the Merchera ethnic group, whose presence in the academic literature remains minimal and often characterised by misrepresentation. Through a comparative analysis considering ChatGPT responses and the scarce number of existing sources, this study explores the model’s reliability, depth, and limitations. The findings reveal that while ChatGPT offers a coherent synthesis of available knowledge, it tends to reproduce the prevailing biases and informational gaps found in the existing academic discourse. The paper concludes that generative AI may serve as a provisional support for research on marginalised communities, but its outputs must be interpreted with caution and situated within a framework of critical inquiry and ethical responsibility. Full article

(This article belongs to the Special Issue Feature Papers in Information in 2024–2025)

► Show Figures

Graphical abstract

15 pages, 7136 KiB

Open AccessArticle

Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges

by Xiyu Zhang, Xu Chen, Yang Wang, Dongliang Liu and Yifeng Hong

Information 2025, 16(6), 460; https://doi.org/10.3390/info16060460 - 29 May 2025

Viewed by 247

Abstract

Abdominal organ segmentation in CT images is crucial for accurate diagnosis, treatment planning, and condition monitoring. However, the annotation process is often hindered by challenges such as low contrast, artifacts, and complex organ structures. While unsupervised domain adaptation (UDA) has shown promise in [...] Read more.

Abdominal organ segmentation in CT images is crucial for accurate diagnosis, treatment planning, and condition monitoring. However, the annotation process is often hindered by challenges such as low contrast, artifacts, and complex organ structures. While unsupervised domain adaptation (UDA) has shown promise in addressing these issues by transferring knowledge from a different modality (source domain), its reliance on both source and target data during training presents a practical challenge in many clinical settings due to data privacy concerns. This study aims to develop a cross-modality abdominal multi-organ segmentation model for label-free CT (target domain) data, leveraging knowledge solely from a pre-trained source domain (MRI) model without accessing the source data. To achieve this, we generate source-like images from target-domain images using a one-way image translation approach with the pre-trained model. These synthesized images preserve the anatomical structure of the target, enabling segmentation predictions from the pre-trained model. To further enhance segmentation accuracy, particularly for organ boundaries and small contours, we introduce an auxiliary translation module with an image decoder and multi-level discriminator. The results demonstrate significant improvements across several performance metrics, including the Dice similarity coefficient (DSC) and average symmetric surface distance (ASSD), highlighting the effectiveness of the proposed method. Full article

► Show Figures

Figure 1

21 pages, 1755 KiB

Open AccessArticle

Wi-Fi Sensing and Passenger Counting: A Statistical Analysis of Local Factors and Error Patterns

by Cristina Pronello, Deepan Anbarasan and Alessandra Boggio Marzet

Information 2025, 16(6), 459; https://doi.org/10.3390/info16060459 - 29 May 2025

Viewed by 238

Abstract

Automatic passenger counting (APC) systems are an important asset for public transport operators, allowing them to optimise networks by monitoring lines’ utilisation. However, the cost of these systems is high and the development of alternative devices, cheaper than the most widely used optical [...] Read more.

Automatic passenger counting (APC) systems are an important asset for public transport operators, allowing them to optimise networks by monitoring lines’ utilisation. However, the cost of these systems is high and the development of alternative devices, cheaper than the most widely used optical systems, seems promising. This paper aims at understanding the influence of local factors on the accuracy of a Wi-Fi APC system by analysing error patterns in a real-world scenario. The APC system was installed on a bus operating regularly within the public transport network and, in the meantime, ground truth data were collected through manual counting. The collected data were then analysed to calculate accuracy and, finally, multilevel modelling was used to identify error patterns due to local factors. This study challenges traditional assumptions, revealing that factors like high pedestrian traffic or intense vehicular movement around the bus have minimal impact on accuracy, if effective received signal strength indicator filters are used. Instead, the number of passengers within the bus affects Wi-Fi systems significantly, especially when the bus is carrying more than 10 passengers, which leads to undercounting due to signal obstruction. This research lays the foundation for strategic error correction to improve accuracy in real-world scenarios. Full article

(This article belongs to the Special Issue Sensing and Wireless Communications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Information, Volume 16, Issue 6 (June 2025) – 66 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI