Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 18
Encontrada(s) 1 página(s)

RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Dual-encoder Multi-Scale Refinement Network for Robust Crack Segmentation Across Diverse Domains

  • Al-Sameai, Habeb
  • Saleh, Radhwan A.A.
  • Moura, Joaquim de
  • Akay, Rustu
The code and links to the datasets supporting the findings of this study are available at https://github.com/habeeb96/DMSRCrack., [Abstract]: Accurate crack segmentation is critical for infrastructure monitoring but remains challenging due to diverse crack morphologies, complex backgrounds, and domain shifts. This paper proposes DMSRCrack, a Dual-encoder Multi-Scale Refinement network for robust crack segmentation across diverse domains. The proposed architecture integrates a hybrid CNN–ViT encoder for local and global feature extraction, a Crack Detail Enhancement Module (CDEM) for preserving thin crack structures, a Boundary Refinement Head (BRH) for contour sharpening, and a Multi-Scale Fusion (MSF) module for scale-consistent representation. Across eight benchmark datasets, DMSRCrack achieves an average Dice of 0.7949 ± 0.0655 and IoU of 0.6639 ± 0.0917 in dataset-wise training. Under leave-one-dataset-out evaluation, it attains the highest IoU on DeepCrack (0.4056), Rissbilder (0.3540), and Crack500 (0.3008). Ablation and computational analyses further confirm the effectiveness and practical efficiency of the proposed contributions., This study was supported by the European Union, through the 3i ICT project in the H2020-MSCA-COFUND-2020 programme [grant agreement GA 101034261], and the Conselleria de Cultura, Educacion, Formación Profesional e Universidades of the regional government Xunta de Galicia through the 3i ICT COFUND agreement. It is also supported by the VARPA group funding (ED431C 2024/33)., Xunta de Galicia; ED431C 2024/33
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Classifying Rodent Behaviors One Frame at a Time: A CNN-based Method

  • Noshahri, Ehsan
  • Molares-Ulloa, Andrés
  • Fraga Seijas, Carlota
  • Puente-Castro, Alejandro
  • Rodríguez, Álvaro
[Abstract]: The classification of animal behavior is essential in disciplines such as pharmacology, neuroscience, and ecology. Traditional methods are labor-intensive, time-consuming, and dependent on human observation, while many existing automated approaches rely on pose estimation and temporal models that demand extensive data, annotations, and computational resources. This study proposes a non-temporal, frame-by-frame deep learning approach for classifying rodent behaviors directly from annotated video frames, without relying on pose estimation, body-part segmentation, or temporal context, and requiring only frame-level annotations. We constructed a large dataset of images representing distinct rodent behaviors in operant conditioning chambers to evaluate multiple convolutional neural network (CNN) architectures of various sizes. Our results show that deep CNNs achieve high classification performance in terms of accuracy and F1-score. The models maintain strong performance under challenging conditions, including restricted training data, noisy labels, and low subject diversity, demonstrating their data efficiency and resilience. Performance remains stable even when trained on data from a limited number of animals, supporting the method’s practicality for population-level behavior classification. Together, these results suggest that robust behavior classification can be achieved without temporal context, even under realistic experimental constraints such as limited data availability and label noise. The proposed framework also offers flexibility across different data formats, including static images and partial video recordings. Overall, our approach provides an accessible and scalable alternative for behavioral researchers, lowering barriers to automated animal behavior analysis while preserving strong classification performance., This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No. 101034261. The project is also part of the R&D grant PID2021-126289OA-I00, funded by the Spanish Ministry of Science, Innovation and Universities (MCIN)/the State Research Agency (AEI)/10.13039/501100011033/ and by the European Regional Development Fund (ERDF), A way of making Europe. Additional funding was provided by the Department of Education, Science, Universities, and Vocational Training of the Xunta de Galicia and by the European Union through the FEDER Galicia 2021–2027 program (Refs. ED431G 2023/01 and ED431C 2026/56), and by CITIC, a member of the CIGUS Network, Xunta de Galicia; ED431G 2023/01, Xunta de Galicia; ED431C 2026/56
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Pixel-Based CF-mMIMO: Addressing the AP Cooperation Cluster Formation in Fronthaul-Limited O-RAN Architectures

  • Pereira Ruisánchez, Dariel
  • Joham, Michael
  • Fresnedo, Óscar
  • Pérez-Adán, Darian
  • Castedo, Luis
  • Utschick, Wolfgang
This version of the article has been accepted for publication, after peer review. The Version of Record is available online at: https://doi.org/10.1109/TCOMM.2025.3646853., [Abstract]: This paper investigates access point (AP) cooperation cluster formation in user-centric cell-free massive MIMO (CF-mMIMO) communication systems characterized by fronthaul links with capacity restrictions. Specifically, we consider an open radio access network (O-RAN) architecture that, although it favors the deployment of ultra-dense networks, is constrained in the number of APs that can be active simultaneously. In this context, we propose an innovative framework termed pixel-based CF-mMIMO, which enables efficient control of both the AP activation and cooperation cluster formation. Recognizing the parallels with pixel-based reconfigurable antennas, the proposed framework allows dynamic reconfiguration of the network coverage map with reasonably low computational cost. The high scalability and performance of the framework are mainly supported by a learning model based on graph neural networks (GNNs) that effectively exploits the existing graph-like structures in CF-mMIMO systems. Extensive simulation experiments demonstrate that the proposed approach achieves competitive spectral efficiency (SE) in challenging scenarios with dense AP deployments and numerous user equipments (UEs)., This work has been supported by grant ED431C 2024/18 funded by Xunta de Galicia and by grant PID2022-137099NB-C42 (MADDIE) funded by MCIN/AEI/10.13039/501100011033. This work also received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101034261 and by the postdoctoral Grant No. ED481B-2025/092 funded by Xunta de Galicia., Xunta de Galicia; ED481B-2025/092, Xunta de Galicia; ED431C 2024/18
RUC. Repositorio da Universidade da Coruña
Publication Contribución de congreso (ConferenceOutput)

A GNN-Based Approach to AP Cooperation Cluster Formation in Cell-Free Massive MIMO

  • Pereira Ruisánchez, Dariel
  • Joham, Michael
  • Fresnedo, Óscar
  • Pérez-Adán, Darian
  • Castedo, Luis
  • Utschick, Wolfgang
Traballo presentado en: 2025 IEEE 101st Vehicular Technology Conference (VTC2025-Spring), 17-20 de xuño de 2025, Oslo, Noruega. © 2025 IEEE. This version of the article has been accepted for publication, after peer review. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The Version of Record is available online at: https://doi.org/10.1109/VTC2025-Spring65109.2025.11174418, [Abstract]: Forming effective access point (AP) cooperation clusters is a key challenge in user-centric cell-free massive MIMO (CF-mMIMO). Existing approaches to this task are either computationally prohibitive or overlook the complex interrelationships within communication networks. In this context, we introduce an innovative approach based on graph neural networks (GNNs). By leveraging the inherent graph structure of CF-mMIMO networks, we transform the rate maximization problem into a node classification task, enabling a competitive and robust solution. Simulation results show that the proposed method significantly outperforms conventional baselines in terms of spectral efficiency, computational complexity, and scalability., This work has been supported by grant ED431C 2024/18 funded by Xunta de Galicia and by grant PID2022-137099NB-C42 (MADDIE) funded by MCIN/AEI/10.13039/501100011033. This work also received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101034261, Xunta de Galicia; ED431C 2024/18
RUC. Repositorio da Universidade da Coruña
Publication Contribución de congreso (ConferenceOutput)

Parsing the Switch: LLM-Based UD Annotation for Complex Code-Switched and Low-Resource Languages

  • Kellert, Olga
  • Tyagi, Nemika
  • Imran, Muhammad
  • Licona-Guevara, Nelvin
  • Gómez-Rodríguez, Carlos
Presented at: Conference on Empirical Methods in Natural Language Processing (EMNLP 2025), Suzhou, China, November 4th to November 9th, 2025., [Abstract]: Code-switching presents a complex challenge for syntactic analysis, especially in low-resource language settings where annotated data is scarce. While recent work has explored the use of large language models (LLMs) for sequence-level tagging, few approaches systematically investigate how well these models capture syntactic structure in code-switched contexts. Moreover, existing parsers trained on monolingual treebanks often fail to generalize to multilingual and mixed-language input. To address this gap, we introduce the BiLingua Pipeline, an LLM-based annotation pipeline designed to produce Universal Dependencies (UD) annotations for code-switched text. First, we develop a prompt-based framework for Spanish-English and Spanish-Guaraní data, combining few-shot LLM prompting with expert review. Second, we release two annotated datasets, including the first Spanish-Guaraní UD-parsed corpus. Third, we conduct a detailed syntactic analysis of switch points across language pairs and communicative contexts. Experimental results show that BiLingua Pipeline achieves up to 95.29% LAS after expert revision, significantly outperforming prior baselines and multilingual parsers. These results show that LLMs, when carefully guided, can serve as practical tools for bootstrapping syntactic resources in under-resourced, code-switched environments., We thank the anonymous annotators and reviewers for their constructive suggestions and help. We ex- tend our gratitude to the Research Computing (RC) and Enterprise Technology at ASU for providing computing resources and access to the ChatGPT en- terprise version for experiments. We acknowledge grants GAP (PID2022-139308OA-I00) funded by MICIU/AEI/10.13039/501100011033/ and ERDF, EU; LATCHING (PID2023-147129OB-C21) funded by MICIU/AEI/10.13039/501100011033 and ERDF, EU. CITIC, as a center accredited for excellence within the Galician University System and a member of the CIGUS Network, receives subsidies from the Department of Education, Sci- ence, Universities, and Vocational Training of the Xunta de Galicia. Additionally, it is co-financed by the EU through the FEDER Galicia 2021-27 operational program (Ref. ED431G 2023/01). Furthermore, this research was supported by the International, Interdisciplinary and Intersectoral Information and Communications Technology PhD programme (3-i ICT) granted to CITIC and supported by the European Union through the Horizon 2020 research and innovation programme under a Marie Skłodowska-Curie agreement (H2020-MSCA-COFUND), GA 101034261., Xunta de Galicia; ED431G 2023/01
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Segmentation-Guided Multi-Task Learning for Automated Analysis of Post-Covid-19 Retinal Alterations in Oct Images

  • Saleh, Radhwan A.A.
  • Al-Sameai, Habeb
  • Cordón Ciordia, Beatriz
  • García-Martín, Elena
  • Ortega Hortas, Marcos
  • Moura, Joaquim de
The source code for the experiments conducted in this research is available upon reasonable request to the corresponding author., [Abstract] The COVID-19 pandemic has revealed significant systemic effects beyond the respiratory system, including structural alterations in the retina. This study proposes explainable, segmentation-guided multi-task deep learning for detecting post-COVID retinal abnormalities using OCT. A newly compiled dataset of 225 participants (healthy controls, non-persistent COVID-19, and persistent post-COVID sequelae) is analyzed using a two-stage framework that jointly optimizes disease classification and pixel-wise retinal layer segmentation. Seven deep learning backbones are evaluated across five experimental setups. The proposed two-stage multi-task learning framework enhances classification accuracy by +1.08% and F1-score by +1.91% compared to the classification-only baseline, with the segmentation branch improving feature robustness by guiding the encoder toward anatomically meaningful retinal structures. These findings demonstrate the potential of segmentation-guided multi-task deep learning for automated post-COVID retinal image analysis while highlighting that retinal imaging alone may not sufficiently discriminate between persistent and non-persistent symptom profiles., This study was supported by the Carlos III Health Institute through grants PI23/00935 and RD21/0002/0050 (Inflammatory Disease Network - RICORS), as well as by the Government of Aragon through group 𝐵⁢23_23⁢𝑅. Additionally, it was funded by the Ministerio de Ciencia e Innovación, Government of Spain, through the research project [grant number PID2023-148913OB-I00, PID2024-161024OB-I00], and by the Consellería de Educación, Universidade e Formación Profesional, Xunta de Galicia, under the Grupos de Referencia Competitiva [grant number ED431C 2024/33]. This work also received support from the Carlos III Health Institute under the grant [FORT23/00010] as part of the FORTALECE Program of the Ministerio de Ciencia e Innovación. Finally, it was also funded by the European Union, through the 3i ICT project in the H2020-MSCA-COFUND-2020 programme [grant agreement GA 101034261], and the Conselleria de Cultura, Educación, Formación Profesional e Universidades of the regional government Xunta de Galicia through the 3i ICT COFUND agreement., Xunta de Galicia; ED431C 2024/33
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

DRL-Based Sequential Scheduling for IRS-Assisted MIMO Communications

  • Pereira Ruisánchez, Dariel
  • Fresnedo, Óscar
  • Pérez-Adán, Darian
  • Castedo, Luis
[Abstract]: Efficient resource allocation strategies are pivotal in vehicular communications as connected devices steeply increase in scenarios with much more stringent requirements. In this work, we propose a deep reinforcement learning (DRL)-based sequential scheduling approach for sum-rate maximization in the uplink of intelligent reflecting surface (IRS)-assisted multi-user (MU) multiple-input multiple-output (MIMO) vehicular communications. We formulate the scheduling task as a partially observable Markov decision process (POMDP) and propose a novel stream-level sequential solution based on the proximal policy optimization (PPO) algorithm. We consider a realistic imperfect channel state information (ICSI) model and assess the proposal in several communication setups comprising both spatially uncorrelated and correlated links. Simulation results show that the proposed DRL-based sequential scheduling approach is a robust alternative to more computationally demanding benchmarks., This work was supported by MCIN/AEI/10.13039/501100011033 under Grants PID2019-104958RB-C42 (ADELE) and PID2022-137099NBC42 (MADDIE), in part by Marie Sklodowska-Curie through European Union’s Horizon 2020 Research and Innovation Programme underGrant 101034261, and in part by the Consellería de Cultura, Educación e Universidade of the Xunta de Galicia. The review of this article was coordinated by Dr. Shaowei Wang. (Corresponding author: Dariel Pereira-Ruisánchez.)
RUC. Repositorio da Universidade da Coruña
Publication Contribución de congreso (ConferenceOutput)

Machine Learning Decodes the Microbiome: Advancing Diagnostic Accuracy in Gastrointestinal Diseases

  • Wasim, Ayesha
  • Martin-De Arribas, E.
  • Ladra, Susana
  • Poza, Margarita
  • Fernández-Lozano, Carlos
Presentado en: 11th ISM World Congress on Targeting Microbiota, 14-15 October 2024, Malta Versión aceptada, [Abstract]: This study evaluates the diagnostic accuracy of machine learning (ML) models on gut microbiome data for gastrointestinal diseases. By assessing batch correction and profiling strategies, the results show that species-level taxonomic features excel for Crohn's disease, whereas ComBat-seq batch correction improves predictive accuracy in colorectal cancer. Algorithms like Random Forest and XGBoost proved most effective, offering guidance for reproducible non-invasive diagnostics., Spanish Ministry of Science and Innovation with funds from the European Union NextGenerationEU, Horizon 2020 research and innovation programme under a Marie Skłodowska-Curie agreement 101034261 (H2020-MSCA-COFUND), from GAIN and MRR funds, Recovery, Transformation and Resilience Plan (PRTRC17.I1), NextGenerationEU (DATAMICROCCR); MCIN/AEI/10.13039/501100011033, GAIN/Xunta de Galicia (GRC: ED431C 2021/53). The present study received funding from the Instituto de Salud Carlos III (ISCIII), Spain, through the projects PI20/00413 and PI23/00696, cofounded by the European Union (EU) to MP., Xunta de Galicia; ED431C 2021/53
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Low-Complexity K-Beams Clustering for Intra-Cell Pilot Reuse in Massive MIMO Communications

  • Pereira Ruisánchez, Dariel
  • Fresnedo, Óscar
  • Pérez-Adán, Darian
  • Castedo, Luis
Financiado para publicación en acceso aberto: Universidade da Coruña/CISUG, [Abstract]: Massive MIMO (mMIMO) communication systems are recognized as key enablers of next-generation wireless networks. However, the orthogonal pilot assignments typical of multiple-input multiple-output (MIMO) systems are not well suited to emerging use cases characterized by short channel coherence times and a high number of connected user equipments (UEs). In this work, we propose a novel approach for intra-cell pilot reuse that leverages the spatial features of correlated mMIMO channels to attain low pilot contamination while using a small number of pilot sequences. The first part of the proposed solution is a groundbreaking clustering algorithm termed K-beams, which splits the complex intra-cell pilot allocation into tractable problems without significant loss of optimality. Then, we introduce a heuristic approach called best-first pilot assignment (BFPA), designed to manage intra-cluster pilot assignments by minimizing interference among the most contaminating UEs. We evaluated the performance of our proposed solution (K-beams+BFPA) in terms of sum-normalized mean-squared error (NMSE) and sum-rate under various challenging network setups. The simulation results show that our approach is a robust alternative to more computationally demanding benchmarks., This work has been supported by grant ED431C 2024/18 funded by Xunta de Galicia and by grant PID2022-137099NB-C42 (MADDIE) funded by MCIN/AEI/10.13039/501100011033. This work also received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101034261 and the Ministry for Digital Transformation and Civil Service and “NextGenerationEU”/PRTR under grant TSI-100925-2023-1. Additionally, this work received funding for open-access charges from the agreement Universidade da Coruña/CISUG., Xunta de Galicia; ED431C 2024/18
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Improved distance correlation estimation

  • Monroy-Castillo, Blanca E.
  • Jácome, M. A.
  • Cao, Ricardo
[Abstract]: Distance correlation is a novel class of multivariate dependence measure, taking positive values between 0 and 1, and applicable to random vectors of not necessarily equal arbitrary dimensions. It offers several advantages over the well-known Pearson correlation coefficient, the most important being that distance correlation equals zero if-and-only if- the random vectors are independent. There are two different estimators of the distance correlation available in the literature. The first estimator, proposed by Székely et al. (Ann Stat 35:2769–279 2007), is based on an asymptotically unbiased estimator of the distance covariance, which is a V-statistic. The second builds on an unbiased estimator of the distance covariance proposed in Székely and Rizzo (Stat 42:2382–2412 2014), shown to be a U-statistic by Huo and Székely (Technometrics 58:435–447 2016). This study evaluates their efficiency (mean squared error) and compares computational times for both methods under different dependence structures. Under conditions of independence or near-independence, the V-estimates are biased, while the U-estimator frequently cannot be computed due to negative values. To address this challenge, a convex linear combination of the former estimators is proposed and studied, yielding good results regardless of the level of dependence. Additionally, a medical database is studied and discussed., This research was supported by the International, Interdisciplinary and Intersectoral Information and Communications Technology PhD programme (3-i ICT) granted to CITIC and supported by the European Union through the Horizon 2020 research and innovation programme under a Marie Sklodowska-Curie agreement (H2020-MSCA-COFUND).
RUC. Repositorio da Universidade da Coruña
Publication Contribución de congreso (ConferenceOutput)

C-Footprints: A Statistic-Based Clustering for Pilot Allocation in Cell-Free Massive MIMO

  • Pereira Ruisánchez, Dariel
  • Fresnedo, Óscar
  • Pérez-Adán, Darian
  • Castedo, Luis
Traballo presentado en: ICC 2025 - IEEE International Conference on Communications, 8-12 de xuño de 2025, Montreal, Canada © 2025 IEEE. This version of the article has been accepted for publication, after peer review. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The Version of Record is available online at: https://doi.org/10.1109/ICC52391.2025.11160717, [Abstract]: User-centric cell-free massive MIMO (CF-mMIMO) communications rely on accurately knowing the channel state information (CSI) to perform coherent signal processing. However, achieving pilot contamination-free channel estimation is challenging due to the limited length of the coherence blocks. In this work, we combine a novel user equipment (UE) clustering method termed C-footprints and a sequential heuristic named cell-free-oriented best first pilot assignment (CF-BFPA) to find pilot allocations that effectively reduce contamination. First, the proposed clustering method groups potentially contaminating UEs based on the degree of orthogonality among the matrices of channel statistics. Subsequently, the CF-BFPA algorithm performs a greedy intra-cluster pilot assignment that ensures low interference between the most contaminating UEs. The performance of the proposed solution (C-footprints+CF-BFPA) is compared with several state-of-the-art algorithms across diverse CF-mMIMO network settings., This work has been supported by grant ED431C 2024/18 funded by Xunta de Galicia and by grant PID2022-137099NB-C42 (MADDIE) funded by MCIN/AEI/10.13039/501100011033. This work also received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101034261., Xunta de Galicia; ED431C 2024/18
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Novel Hybrid Machine Learning Framework for High-Fidelity Prediction of Fly Ash-Based Geopolymer Concrete Strength

  • Hassan, Amer
  • Saleh, Radhwan A.A.
  • Al-Sameai, Habeb
  • Moura, Joaquim de
  • Alomayri, Thamer
  • Zhang, Chunwei
[Abstract]: The complex, non-linear relationships between mix-design parameters and the mechanical properties of geopolymer concrete (GPC) are not fully understood, presenting a fundamental scientific challenge for accurate strength prediction and mix optimization. This challenge hinders the widespread adoption of GPC, a sustainable alternative to conventional concrete. Existing machine learning models for GPC often lack generalizability and interpretability due to the limited availability of datasets and basic architectures. This research introduces T-BoostNet, a novel hybrid machine learning (ML) framework combining Transformer architectures with XGBoost, designed for superior accuracyand interpretability in predicting GPC compressive strength. Leveraging an unprecedented dataset of 1117 unique GPC mixtures from 77 diverse studies, T-BoostNet effectively captures intricate local and global feature interactions. T-BoostNet consistently outperformed five benchmark ML algorithms, achieving the highest 𝑅2 = 0.848 ± 0.024 and MAE = 3.56 ± 0.30 MPa. SHAP analysis provided crucial interpretability, identifying curing period, water content in alkaline solution, specimen age, and curing temperature as the most influential factors. This framework advances sustainable construction by providing a reliable, interpretable tool that accelerates GPC adoption, reduces costly laboratory trials, and aligns with global low-carbon material goals., This study was supported by the European Union, through the 3i ICT project in the H2020-MSCA-COFUND-2020 programme [grant agreement GA 101034261], and the Conselleria de Cultura, Educacion, Formación Profesional e Universidades of the regional government Xunta de Galicia through the 3i ICT COFUND agreement. It is also supported by the VARPA group funding (ED431C 2024/33)., Xunta de Galicia; ED431C 2024/33
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

SynNER: Syntax-infused Named Entity Recognition in the Biomedical Domain

  • Imran, Muhammad
  • Zamaraeva, Olga
  • Gómez-Rodríguez, Carlos
Financiado para publicación en acceso aberto: Universidade da Coruña/CISUG The code and other resources developed for this work are available in our GitHub repository at: https://github.com/chimran135/SynNER. The datasets used in this study are publicly available from third-party sources. The MTSamples and VAERS datasets can be downloaded from: https://github.com/BIDS-Xu-Lab/Clinical_Entity_Recognition_Using_GPT_models. The NCBI-Disease, BC2GM, and JNLPBA datasets can be downloaded from: https://github.com/cambridgeltl/MTL-Bioinformatics-2016. Supplementary material is available at JAMIA Open online, [Abstract]: Named Entity Recognition (NER) is a technology that helps computers automatically find and classify important terms in text, such as names of diseases, drugs, or medical procedures. This is especially valuable in the biomedical field, where researchers and clinicians need to process large volumes of text from scientific articles, clinical notes, or patient records. In this work, we present SynNER, a system that improves the accuracy of NER by teaching computers to pay attention not only to the words themselves, but also to syntax, ie, the internal structure of sentences. For example, recognizing how words are connected in a sentence (which parts of the sentence are subjects, objects or modifiers) can make it easier to correctly identify medical terms, even when they appear in complex contexts. We tested our method on five different collections of biomedical texts. The results showed that incorporating grammatical knowledge significantly boosted accuracy. Our SynNER system outperformed previous state-of-the-art methods on three of the five datasets. Our results show that using syntax can help researchers and healthcare professionals more reliably and quickly extract vital information from a vast amount of text, which could ultimately help improve biomedical research and clinical decision support tools., We acknowledge the European Research Council (ERC), which has funded this research under the Horizon Europe research and innovation programme (SALSA, grant agreement No 101100615), SCANNER-UDC (PID2020-113230RB-C21) funded by MICIU/AEI/10.13039/501100011033, LATCHING (PID2023-147129OB-C21) funded by MICIU/AEI/10.13039/501100011033 and ERDF (EU), Ministry for Digital Transformation and Civil Service and “NextGenerationEU” PRTR under grant TSI-100925-2023-1, Xunta de Galicia (ED431C 2024/02), and Galician Research Center “CITIC”, funded by Xunta de Galicia through the collaboration agreement between the Consellería de Cultura, Educación, Formación Profesional e Universidades and the Galician universities for the reinforcement of the research centres of the Galician University System (CIGUS). Furthermore, this research was supported by the International, Interdisciplinary and Intersectoral Information and Communications Technology PhD programme (3-i ICT) granted to CITIC and supported by the European Union through the Horizon 2020 research and innovation programme under a Marie Skłodowska-Curie agreement (H2020-MSCA-COFUND), GA 101034261. Funding for open access charge: Universidade da Coruña/CISUG., Xunta de Galicia; ED431C 2024/02
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

A Syntax-Injected Approach for Faster and More Accurate Sentiment Analysis

  • Imran, Muhammad
  • Kellert, Olga
  • Gómez-Rodríguez, Carlos
The following information was supplied regarding data availability: The OpeNERen, OpeNERes, UD_English-EWT and UD_Spanish-AnCora datasets are available at Zenodo: Muhammad Imran. (2025). chimran135/syntax-injected-sentimentanalysis: Syntax-injected Sentiment Analysis (1.0.0). Zenodo. https://doi.org/10.5281/zenodo.15323755. The Rest-Mex 2023 shared task website provides access to the training data for registered participants. The data are held privately for assessment purposes and can be obtained from Rest-Mex 2023 organizers https://sites.google.com/cimat.mx/rest-mex2023 by contacting Miguel Ángel Álvarez Carmona at miguel.alvarez@cimat.mx. The Rest-Mex 2023 dataset was used under license for this study., [Abstract]: Sentiment Analysis (SA) is a crucial aspect of Natural Language Processing (NLP), focusing on identifying and interpreting subjective assessments in textual content. Syntactic parsing is useful in SA as it improves accuracy and provides explainability; however, it often becomes a computational bottleneck due to slow parsing algorithms. This article proposes a solution to this bottleneck by using a Sequence Labeling Syntactic Parser (SELSP) to integrate syntactic information into SA via a rule-based sentiment analysis pipeline. By reformulating dependency parsing as a sequence labeling task, we significantly improve the efficiency of syntax-based SA. SELSP is trained and evaluated on a ternary polarity classification task, demonstrating greater speed and accuracy compared to conventional parsers like Stanza and heuristic approaches such as Valence Aware Dictionary and sEntiment Reasoner (VADER). The combination of speed and accuracy makes SELSP especially attractive for sentiment analysis applications in both academic and industrial contexts. Moreover, we compare SELSP with Transformer-based models trained on a 5-label classification task. In addition, we evaluate multiple sentiment dictionaries with SELSP to determine which yields the best performance in polarity prediction. The results show that dictionaries accounting for polarity judgment variation outperform those that ignore it. Furthermore, we show that SELSP outperforms Transformer-based models in terms of speed for polarity prediction., The European Research Council (ERC) funded this research under the Horizon Europe research and innovation programme (SALSA, grant agreement No. 101100615), SCANNER-UDC (PID2020-113230RB-C21) funded by MICIU/AEI/10.13039/501100011033, LATCHING (PID2023-147129OB-C21) funded by MICIU/AEI/10.13039/501100011033 and ERDF (EU), project GAP (PID2022-139308OA-I00) funded by MICIU/AEI/10.13039/501100011033/ and ERDF (EU), Ministry for Digital Transformation and Civil Service and “NextGenerationEU” PRTR under grant TSI-100925-2023-1, Xunta de Galicia (ED431C 2024/02), and Galician Research Center “CITIC”, funded by Xunta de Galicia through the collaboration agreement between the Consellería de Cultura, Educación, Formación Profesional e Universidades and the Galician universities for the reinforcement of the research centres of the Galician University System (CIGUS). Furthermore, this research was supported by the International, Interdisciplinary and Intersectoral Information and Communications Technology PhD programme (3-i ICT) granted to CITIC and supported by the European Union through the Horizon 2020 research and innovation programme under a Marie Skłodowska-Curie agreement (H2020-MSCA-COFUND), GA 101034261. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript., Xunta de Galicia; ED431C 2024/02
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Beyond Accuracy: Cross-session Stability in Representation-based Animal Identification

  • Rodríguez, Álvaro
  • Noshahri, Ehsan
  • Puente-Castro, Alejandro
  • Molares-Ulloa, Andrés
All datasets used in this study are publicly available and can be accessed via the links provided in the key resources table. Zenodo: https://doi.org/10.5281/zenodo.15112879, [Abstract]: Identification of unmarked animals is central to behavioral and ecological research. Representation learning currently achieves near-perfect performance in single-session experiments. However, stability across sessions remains poorly characterized. We evaluate identity consistency, accuracy, and learned representation using synthetic and real cross-session experiments. The results show that identification degrades even under minimal visual changes. Although exposure to multiple sessions improves generalization, substantial degradation remains. This indicates that the reported accuracy alone is insufficient to guarantee stable identities across sessions., This work was partially funded by the EU through the FEDER Galicia 2021-27 operational program ref. ED431G 2023/01; by the Consellería de Educación, Ciencia, Universidades e Formación Profesional of Xunta de Galicia under the program “Axudas para a consolidación e estruturación de unidades de investigación competitivas (Grupos de Referencia Competitiva)”, ref. ED431C 2026/56; by grant PID2021-126289OA-I00 funded by MCIN/AEI/https://doi.org/10.13039/501100011033 and ERDF A way of making Europe; and by the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement no. 101034261., Xunta de Galicia; ED431G 2023/01, Xunta de Galicia; ED431C 2026/56
RUC. Repositorio da Universidade da Coruña
Publication Contribución de congreso (ConferenceOutput)

Self-Supervised Deep Learning Design for MU-SIMO Beyond-Diagonal RIS

  • Pérez-Adán, Darian
  • Pereira Ruisánchez, Dariel
  • Fresnedo, Óscar
  • Santamaría, Ignacio
  • Castedo, Luis
  • Thompson, John S.
Presented at: ICC 2026 - IEEE International Conference on Communications, 24-28 May 2026, Glasgow, United Kingdom © 2026 IEEE. This version of the article has been accepted for publication, after peer review. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The Version of Record is available online at: https://doi.org/10.1109/ICC59461.2026.11587048, [Abstract]: Beyond-diagonal reconfigurable intelligent surfaces (BD-RISs) have emerged as a promising technology for wireless communications. BD-RIS enhances the capabilities of conventional RIS architectures by enabling full coupling among surface elements connected by adjustable reactances. However, leveraging such extended flexibility requires effectively addressing the added complexity in optimization tasks. In this paper, we propose a self-supervised (SS) deep learning (DL) approach for BD-RIS optimization in multi-user (MU) single-input multiple-output (SIMO) uplink systems. Unlike supervised methods that require expensive optimization solvers for training labels, the learning process of our approach is guided by a loss function derived from the optimization objective function. The proposed neural network (NN) model learns to map the channel state information (CSI) to the BD-RIS scattering matrices that maximize the system sum-rate, while guaranteeing the matrices’ structural constraints. Simulation results show that our approach outperforms traditional optimization-based methods in realistic 3GPP channels, while exhibiting lower computational complexity and maintaining comparable user fairness to state-of-the-art baselines., This work has been supported by grant ED431C 2024/18 funded by Xunta de Galicia and by grant PID2022-137099NB-C42/C43 (MADDIE) funded by MCIN/AEI/10.13039/501100011033 and FEDER UE. This work also received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101034261 and by the postdoctoral Grant No. ED481B-2025/092 funded by Xunta de Galicia., Xunta de Galicia; ED431C 2024/18, Xunta de Galicia; ED481B-2025/092
RUC. Repositorio da Universidade da Coruña
Publication Artículo científico (JournalArticle)

Data driven modeling and experimental validation of alkali activated materials for construction applications

  • Saleh, Radhwan A.A.
  • Hassan, Amer
  • Al-Sameai, Habeb
  • Moura, Joaquim de
  • Ortega Hortas, Marcos
  • Saleh, Zead
  • Alomayri, Thamer
  • Zhang, Chunwei
[Abstract]: This study presents a robust machine learning framework for predicting the compressive strength of fly ash geopolymer concrete (FAGPC) using a comprehensive database of over 1000 mix designs. The core innovation is a two-stage residual learning architecture that combines a LightGBM base model with error-correcting layers (KNN, SVR, or Extra Trees) to capture complex, non-linear relationships between chemical ratios, aggregate proportions, and curing conditions. To ensure model reliability and prevent data leakage, a nested cross-validation strategy was employed, resulting in an average R2 of 0.837 and RMSE of 6.05 MPa. Experimental validation on unseen specimens confirmed the framework’s predictive power, with the LightGBM-KNN hybrid achieving an R2 of 0.866 and an MAE of 2.39 MPa. By significantly reducing the reliance on resource-intensive trial-and-error methods, this computationally efficient tool provides a validated pathway to accelerate the adoption of low-carbon materials in sustainable construction., This study was supported by the European Union, through the 3i ICT project in the H2020-MSCA-COFUND-2020 programme [grant agreement GA 101034261], and the Conselleria de Cultura, Educacion, Formación Profesional e Universidades of the regional government Xunta de Galicia through the 3i ICT COFUND agreement. Additional support was provided by the National Natural Science Foundation of China (Grant No. 52361135807) and by the VARPA group funding (ED431C 2024/33), Xunta de Galicia; ED431C 2024/33, China. National Natural Science Foundation; 52361135807
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Segmentation-guided multi-task learning for automated analysis of post-COVID-19 retinal alterations in OCT images

  • Saleh, Radhwan A. A.
  • Al-Sameai, Habeb
  • Cordón, Beatriz
  • Garcia-Martin, Elena
  • Ortega, Marcos
  • de Moura, Joaquim
The COVID-19 pandemic has revealed significant systemic effects beyond the respiratory system, including structural alterations in the retina. This study proposes explainable, segmentation-guided multi-task deep learning for detecting post-COVID retinal abnormalities using OCT. A newly compiled dataset of 225 participants (healthy controls, non-persistent COVID-19, and persistent post-COVID sequelae) is analyzed using a two-stage framework that jointly optimizes disease classification and pixel-wise retinal layer segmentation. Seven deep learning backbones are evaluated across five experimental setups. The proposed two-stage multi-task learning framework enhances classification accuracy by +1.08% and F1-score by +1.91% compared to the classification-only baseline, with the segmentation branch improving feature robustness by guiding the encoder toward anatomically meaningful retinal structures. These findings demonstrate the potential of segmentation-guided multi-task deep learning for automated post-COVID retinal image analysis while highlighting that retinal imaging alone may not sufficiently discriminate between persistent and non-persistent symptom profiles.