Pub Date : 2025-09-03DOI: 10.1109/TLA.2025.11150629
Roberto Gonçalves de Magalhães Júnior;Rafael Nobre Orsi;Tatiany Marcondes Heiderich;Marina Carvalho de Moraes Barros;Ruth Guinsburg;Carlos Eduardo Thomaz
This paper introduces the application of novel eyetracking metrics to assess visual attention and cognitive load in neonatal pain assessment. Our goal is to evaluate pediatrician experts, non-experts, and parents using the relative Explore- Exploit Ratio, along with the Task-Evoked Pupillary Response, while analyzing the frontal faces of distinct newborns before and after painful procedures. All the experiments were based on a benchmark image dataset considering clinically relevant areas of interest. The Tobii TX300 system was used to record the eye-tracking data in a closed room with controlled lighting. Our results disclose that the visual attention described by the traditional metrics does not correspond directly to the respective fixation patterns and pupillary changes quantified for all the sample groups of participants investigated, highlighting statistically significant differences in the visual behavior between participants with or without clinical experience only when using the novel metrics proposed instead.
{"title":"Visual and Pupillary Behavior in Neonatal Pain Assessment using Eye-Tracking","authors":"Roberto Gonçalves de Magalhães Júnior;Rafael Nobre Orsi;Tatiany Marcondes Heiderich;Marina Carvalho de Moraes Barros;Ruth Guinsburg;Carlos Eduardo Thomaz","doi":"10.1109/TLA.2025.11150629","DOIUrl":"https://doi.org/10.1109/TLA.2025.11150629","url":null,"abstract":"This paper introduces the application of novel eyetracking metrics to assess visual attention and cognitive load in neonatal pain assessment. Our goal is to evaluate pediatrician experts, non-experts, and parents using the relative Explore- Exploit Ratio, along with the Task-Evoked Pupillary Response, while analyzing the frontal faces of distinct newborns before and after painful procedures. All the experiments were based on a benchmark image dataset considering clinically relevant areas of interest. The Tobii TX300 system was used to record the eye-tracking data in a closed room with controlled lighting. Our results disclose that the visual attention described by the traditional metrics does not correspond directly to the respective fixation patterns and pupillary changes quantified for all the sample groups of participants investigated, highlighting statistically significant differences in the visual behavior between participants with or without clinical experience only when using the novel metrics proposed instead.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 10","pages":"931-937"},"PeriodicalIF":1.3,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150629","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144934556","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-03DOI: 10.1109/TLA.2025.11150627
Enrique C. Quispe;Julio Rafael Gómez Sarduy;Zaid García Sánchez;Isidoro Fraga Hurtado;Roy Reyes Calvo;Yuri Ulianov López Castrillon
Forecasting electricity generation from renewable resources is crucial for the efficient planning and operation of power systems. The development of forecasting models based on local meteorological variables is common, however, sometimes this information is unavailable. This study explores the use of multivariate models that do not incorporate meteorological variables, but use historical power-generated data from eight PV plants located in the same region to predict the future value of a target plant. This allows for improved forecasting when meteorological variables are unavailable and the only information available is the generation of the PV plants. The performance of LSTM and BiLSTM networks is compared for different time horizons, considering various lags of the power series itself for estimating future values. The main contributions of this study include the introduction of power time series from other plants as model inputs, the use of spatial interpolation to fill in missing data and the application of causality tests between time series for the selection of predictor variables, and the uncertainty associated with the predictions is analyzed using quantile regression techniques.
{"title":"Multivariate Models for Photovoltaic Power Forecasting with Non-climatic Exogenous Variables","authors":"Enrique C. Quispe;Julio Rafael Gómez Sarduy;Zaid García Sánchez;Isidoro Fraga Hurtado;Roy Reyes Calvo;Yuri Ulianov López Castrillon","doi":"10.1109/TLA.2025.11150627","DOIUrl":"https://doi.org/10.1109/TLA.2025.11150627","url":null,"abstract":"Forecasting electricity generation from renewable resources is crucial for the efficient planning and operation of power systems. The development of forecasting models based on local meteorological variables is common, however, sometimes this information is unavailable. This study explores the use of multivariate models that do not incorporate meteorological variables, but use historical power-generated data from eight PV plants located in the same region to predict the future value of a target plant. This allows for improved forecasting when meteorological variables are unavailable and the only information available is the generation of the PV plants. The performance of LSTM and BiLSTM networks is compared for different time horizons, considering various lags of the power series itself for estimating future values. The main contributions of this study include the introduction of power time series from other plants as model inputs, the use of spatial interpolation to fill in missing data and the application of causality tests between time series for the selection of predictor variables, and the uncertainty associated with the predictions is analyzed using quantile regression techniques.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 10","pages":"877-887"},"PeriodicalIF":1.3,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150627","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144934535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-03DOI: 10.1109/TLA.2025.11150633
Fernando Aparicio Urbano-Molano;Jaime Velasco-Medina
FrodoKEM, a key encapsulation mechanism (KEM) based on the learning with errors (LWE) problem, would be included for standardization by the International Organization for Standardization (ISO) and recommended for PQC migration by the BSI (German Federal Office for Information Security) and the ANSSI (French Cybersecurity Agency). It is closely related to the challenging time-computational problem inherent to algebraically unstructured lattices. However, hardware implementations of this scheme are required to verify its effectiveness in real-world applications. To the best of our knowledge, this is the first hardware implementation of FrodoKEM using High-Level Synthesis (HLS), which meets all requirements of the version submitted for standardization to ISO. The proposed design started with the profiling of the reference C software implementation using Valgrind software tools, to identify the functions that are the most time-consuming. The advantages of the proposed implementation include a 34% improvement in the speed metric of the Key Generation module in comparison with the reference software implementation. The results show that the key generation, encapsulation, and decapsulation use 26%, 39%, and 32%, respectively, of the total area utilization on the Artix-7.
{"title":"FrodoKEM Hardware Implementation for Post-Quantum Cryptography","authors":"Fernando Aparicio Urbano-Molano;Jaime Velasco-Medina","doi":"10.1109/TLA.2025.11150633","DOIUrl":"https://doi.org/10.1109/TLA.2025.11150633","url":null,"abstract":"FrodoKEM, a key encapsulation mechanism (KEM) based on the learning with errors (LWE) problem, would be included for standardization by the International Organization for Standardization (ISO) and recommended for PQC migration by the BSI (German Federal Office for Information Security) and the ANSSI (French Cybersecurity Agency). It is closely related to the challenging time-computational problem inherent to algebraically unstructured lattices. However, hardware implementations of this scheme are required to verify its effectiveness in real-world applications. To the best of our knowledge, this is the first hardware implementation of FrodoKEM using High-Level Synthesis (HLS), which meets all requirements of the version submitted for standardization to ISO. The proposed design started with the profiling of the reference C software implementation using Valgrind software tools, to identify the functions that are the most time-consuming. The advantages of the proposed implementation include a 34% improvement in the speed metric of the Key Generation module in comparison with the reference software implementation. The results show that the key generation, encapsulation, and decapsulation use 26%, 39%, and 32%, respectively, of the total area utilization on the Artix-7.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 10","pages":"922-930"},"PeriodicalIF":1.3,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150633","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144934473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-03DOI: 10.1109/TLA.2025.11150630
Anikó Kopacz;Enol García González;Camelia Chira;José Ramón Villar Flecha
In the past few years, path planning and scheduling became a high-impact research topic due to their real-world applications such as transportation, manufacturing and robotics. This paper focuses on the Multi-robot Path Planning (MPP) problem, which consists of planning the route for a set of robots in a given static environment. The main goal is to navigate the robots from a starting point to a destination point without colliding with other robots or static obstacles. We propose a hybrid method -- H* -- that combines adaptive route planning based on A* and local search algorithm to optimize routes in the context of the MPP problem. The A* algorithm finds the optimal solution for the route search problem and a heuristic approach is applied to scale up to the multi-agent scenario.The overall length of determined paths and the number of robot collisions is minimized during the evaluations specific small-scale environments.Computational experiments are conducted for multi-robot scenarios and the performance of H* is compared to several path-searching algorithms including A* variations extended for the multi-agent scenario and coevolutionary algorithms.Experimental results demonstrate that H* outperforms the A* based heuristic approaches in terms of path length. H* shows similar performance as the coevolutionary method and performs better on smaller-scale maps.
{"title":"Hybrid Adaptive Greedy Algorithm Addressing the Multi-Robot Path Planning Problem","authors":"Anikó Kopacz;Enol García González;Camelia Chira;José Ramón Villar Flecha","doi":"10.1109/TLA.2025.11150630","DOIUrl":"https://doi.org/10.1109/TLA.2025.11150630","url":null,"abstract":"In the past few years, path planning and scheduling became a high-impact research topic due to their real-world applications such as transportation, manufacturing and robotics. This paper focuses on the Multi-robot Path Planning (MPP) problem, which consists of planning the route for a set of robots in a given static environment. The main goal is to navigate the robots from a starting point to a destination point without colliding with other robots or static obstacles. We propose a hybrid method -- H* -- that combines adaptive route planning based on A* and local search algorithm to optimize routes in the context of the MPP problem. The A* algorithm finds the optimal solution for the route search problem and a heuristic approach is applied to scale up to the multi-agent scenario.The overall length of determined paths and the number of robot collisions is minimized during the evaluations specific small-scale environments.Computational experiments are conducted for multi-robot scenarios and the performance of H* is compared to several path-searching algorithms including A* variations extended for the multi-agent scenario and coevolutionary algorithms.Experimental results demonstrate that H* outperforms the A* based heuristic approaches in terms of path length. H* shows similar performance as the coevolutionary method and performs better on smaller-scale maps.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 10","pages":"856-864"},"PeriodicalIF":1.3,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150630","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144934380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-08DOI: 10.1109/TLA.2025.11072500
Alfonso Vazquez Mendoza;Héctor Francisco Ruiz Paredes
Short-Term Hydrothermal Scheduling (STHS) is a very complex, multimodal, nonlinear optimization problem that has primarily been addressed by conventional and, more recently, metaheuristic optimization algorithms. The objective of conventional STHS is to optimize the hourly energy production of hydroelectric power plants and other generation sources over a specific period of time, allowing for the determination of the optimal economic operation of the Power Electrical System (PES). The conventional STHS formulation is widely used in the planning, analysis and operation of PES. However, nowadays PES incorporate variable renewable generation such as wind and solar photovoltaic power, as well as Energy Storage Systems (ESS), transmission grid models and load shedding scenarios in case of possible operational contingencies. This paper presents a STHS formulated and simulated using nonlinear programming for a day ahead, using artificial neural networks (ANN) for demand forecasting. The integration of wind and solar photovoltaic generation, ESS and cascaded hydroelectric power plants is considered, along with the transmission grid and load shedding models, all within a single optimization problem. The objective is to minimize generation costs and optimize power usage, dispatching the units in the most efficient manner. The efficient assignment of thermal, hydro, solar, wind units and ESS allows for optimal use of available water without exceeding reservoir limits. The formulation is validated using the IEEE 30-node system, obtaining optimal solutions in all scenarios, without the need to relax system constraints for convergence.
{"title":"Short-Term Day-Ahead Hydrothermal Scheduling with Energy Renewables Variable, Storage, Load Shedding using Artificial Intelligence Techniques for Demand Forecasting","authors":"Alfonso Vazquez Mendoza;Héctor Francisco Ruiz Paredes","doi":"10.1109/TLA.2025.11072500","DOIUrl":"https://doi.org/10.1109/TLA.2025.11072500","url":null,"abstract":"Short-Term Hydrothermal Scheduling (STHS) is a very complex, multimodal, nonlinear optimization problem that has primarily been addressed by conventional and, more recently, metaheuristic optimization algorithms. The objective of conventional STHS is to optimize the hourly energy production of hydroelectric power plants and other generation sources over a specific period of time, allowing for the determination of the optimal economic operation of the Power Electrical System (PES). The conventional STHS formulation is widely used in the planning, analysis and operation of PES. However, nowadays PES incorporate variable renewable generation such as wind and solar photovoltaic power, as well as Energy Storage Systems (ESS), transmission grid models and load shedding scenarios in case of possible operational contingencies. This paper presents a STHS formulated and simulated using nonlinear programming for a day ahead, using artificial neural networks (ANN) for demand forecasting. The integration of wind and solar photovoltaic generation, ESS and cascaded hydroelectric power plants is considered, along with the transmission grid and load shedding models, all within a single optimization problem. The objective is to minimize generation costs and optimize power usage, dispatching the units in the most efficient manner. The efficient assignment of thermal, hydro, solar, wind units and ESS allows for optimal use of available water without exceeding reservoir limits. The formulation is validated using the IEEE 30-node system, obtaining optimal solutions in all scenarios, without the need to relax system constraints for convergence.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 8","pages":"696-705"},"PeriodicalIF":1.3,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072500","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144581627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-08DOI: 10.1109/TLA.2025.11072497
Mario Luiz Tronco;Ingrid Lorena Argote Pedrazza;Emerson Carlos Pedrino;Carlos Roberto Valencio
Computer vision systems are essential for automating agricultural tasks such as disease detection and fruit defect identification. However, their application in coffee farming faces significant challenges due to environmental variability and the complex structure of coffee trees, which complicate image acquisition. Thus, this study addresses two key questions: 1) Can low-cost, user-friendly equipment adapt to crop conditions while ensuring image quality 2) Can a computer vision algorithm accurately count and classify coffee beans with over 80% accuracy using data from low-cost cameras To answer these questions, an image acquisition system was developed based on the phenological characteristics of coffee plants, ensuring focused and consistent image capture. Additionally, a novel algorithm was created, utilizing statistical analysis of color spaces to effectively separate fruits from the background, segment images, and count fruits. The algorithm achieved accuracy rates, when compared with a traditional approach, within the desired range for each coffee fruit class: green (83%), green-olive (79%), cherry (86%), and raisin (80%). These results demonstrate the potential of this approach for accurate and efficient fruit processing in coffee farming, particularly when images are captured directly from tree branches.
{"title":"Estimation of fruit number in coffee trees by maturity level, based on color space weighting, using a new segmentation algorithm","authors":"Mario Luiz Tronco;Ingrid Lorena Argote Pedrazza;Emerson Carlos Pedrino;Carlos Roberto Valencio","doi":"10.1109/TLA.2025.11072497","DOIUrl":"https://doi.org/10.1109/TLA.2025.11072497","url":null,"abstract":"Computer vision systems are essential for automating agricultural tasks such as disease detection and fruit defect identification. However, their application in coffee farming faces significant challenges due to environmental variability and the complex structure of coffee trees, which complicate image acquisition. Thus, this study addresses two key questions: 1) Can low-cost, user-friendly equipment adapt to crop conditions while ensuring image quality 2) Can a computer vision algorithm accurately count and classify coffee beans with over 80% accuracy using data from low-cost cameras To answer these questions, an image acquisition system was developed based on the phenological characteristics of coffee plants, ensuring focused and consistent image capture. Additionally, a novel algorithm was created, utilizing statistical analysis of color spaces to effectively separate fruits from the background, segment images, and count fruits. The algorithm achieved accuracy rates, when compared with a traditional approach, within the desired range for each coffee fruit class: green (83%), green-olive (79%), cherry (86%), and raisin (80%). These results demonstrate the potential of this approach for accurate and efficient fruit processing in coffee farming, particularly when images are captured directly from tree branches.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 8","pages":"736-742"},"PeriodicalIF":1.3,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072497","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144581629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-08DOI: 10.1109/TLA.2025.11072504
Mudadla Dhananjaya;Devendra Potnuru;Ramesh Devarapalli;Malleswara Rao K Durga;Thanikanti Sudhakar Babu
High-gain converters are well-established circuit designs that find practical use in industrial and commercial settings, particularly in applications demanding high power ratings, such as Fuel Cell Electric Vehicles (FCEV) and grid-connected Renewable Energy Sources (RES). High-gain topologies from a single source pose reliability issue in RES applications due to increased device count and stress. In the event of source failure, these topologies may lead to an energy supply gap for the loads. Addressing this challenge, integrating diverse energy sources with step-up voltage capability stands as a promising solution for both DC microgrid and Electric Vehicle (EV) applications. In this study, a Dual-Input Single-Output (DISO) converter is introduced to integrate various sources and achieve an increased output voltage gain by charging the inductors in parallel and discharging them in series. Moreover, if any sources fail, the converter can supply the energy to the load from the available source and it can be operated in bidirectional mode. This paper also extensively discusses theoretical analysis, considerations related to design and circuit modeling. Furthermore, include a comparison of this converter with several other topologies. It examined to validated with a 250 W laboratory prototype.
{"title":"Bidirectional Step-Up Multi-Input Converter with Improved Voltage Gain for DC Microgrids","authors":"Mudadla Dhananjaya;Devendra Potnuru;Ramesh Devarapalli;Malleswara Rao K Durga;Thanikanti Sudhakar Babu","doi":"10.1109/TLA.2025.11072504","DOIUrl":"https://doi.org/10.1109/TLA.2025.11072504","url":null,"abstract":"High-gain converters are well-established circuit designs that find practical use in industrial and commercial settings, particularly in applications demanding high power ratings, such as Fuel Cell Electric Vehicles (FCEV) and grid-connected Renewable Energy Sources (RES). High-gain topologies from a single source pose reliability issue in RES applications due to increased device count and stress. In the event of source failure, these topologies may lead to an energy supply gap for the loads. Addressing this challenge, integrating diverse energy sources with step-up voltage capability stands as a promising solution for both DC microgrid and Electric Vehicle (EV) applications. In this study, a Dual-Input Single-Output (DISO) converter is introduced to integrate various sources and achieve an increased output voltage gain by charging the inductors in parallel and discharging them in series. Moreover, if any sources fail, the converter can supply the energy to the load from the available source and it can be operated in bidirectional mode. This paper also extensively discusses theoretical analysis, considerations related to design and circuit modeling. Furthermore, include a comparison of this converter with several other topologies. It examined to validated with a 250 W laboratory prototype.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 8","pages":"675-686"},"PeriodicalIF":1.3,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072504","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144581478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-08DOI: 10.1109/TLA.2025.11072932
Bruno Waideman;Plínio Thomaz Aquino
The production, storage, and dissemination of information have evolved from ancient communication methods to modern digital technologies, with digital media playing a key role in connecting individuals. While keyboards are common tools for interaction, they present challenges for individuals with motor impairments. Augmentative and Alternative Communication (AAC) techniques, including gesture input, voice commands, and sensor-based systems, have emerged to address these limitations. Eye tracking, used in accessibility systems, offers both opportunities and challenges, such as visual fatigue and inaccuracies that lead to slower typing. To address these challenges, this study proposes an interaction approach integrating eye movement tracking with a virtual keyboard, utilizing an artificial neural network to interpret gaze data and translate intentions within the interface at a low cost for the user. Additionally, a Language Model (LM) aids in predicting next-word suggestions. This research will assess the impact of these technologies on typing speed, error rate, and linguistic predictability, contributing both scientifically and societally to the advancement of accessible communication systems.
{"title":"Augmentative and Alternative Communication Using Eye Tracking and Word Recommendation Using Language Models","authors":"Bruno Waideman;Plínio Thomaz Aquino","doi":"10.1109/TLA.2025.11072932","DOIUrl":"https://doi.org/10.1109/TLA.2025.11072932","url":null,"abstract":"The production, storage, and dissemination of information have evolved from ancient communication methods to modern digital technologies, with digital media playing a key role in connecting individuals. While keyboards are common tools for interaction, they present challenges for individuals with motor impairments. Augmentative and Alternative Communication (AAC) techniques, including gesture input, voice commands, and sensor-based systems, have emerged to address these limitations. Eye tracking, used in accessibility systems, offers both opportunities and challenges, such as visual fatigue and inaccuracies that lead to slower typing. To address these challenges, this study proposes an interaction approach integrating eye movement tracking with a virtual keyboard, utilizing an artificial neural network to interpret gaze data and translate intentions within the interface at a low cost for the user. Additionally, a Language Model (LM) aids in predicting next-word suggestions. This research will assess the impact of these technologies on typing speed, error rate, and linguistic predictability, contributing both scientifically and societally to the advancement of accessible communication systems.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 8","pages":"637-645"},"PeriodicalIF":1.3,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072932","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144581651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-08DOI: 10.1109/TLA.2025.11072502
{"title":"Table of Contents August 2025","authors":"","doi":"10.1109/TLA.2025.11072502","DOIUrl":"https://doi.org/10.1109/TLA.2025.11072502","url":null,"abstract":"","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 8","pages":"636-636"},"PeriodicalIF":1.3,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072502","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144581652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-08DOI: 10.1109/TLA.2025.11072499
Kauê Tartarotti Nepomuceno Duarte;Murilo Costa de Barros;Abhijot Singh Sidhu;David Gobbi;Cheryl McCreary;Feryal Saad;Richard Camicioli;Eric Smith;Marco Carvalho;Richard Frayne
White matter hyperintensities (WMHs) are a common finding on magnetic resonance (MR) images in older individuals, appearing as high-signal intensity regions on fluid-attenuated inversion recovery (FLAIR) imaging. People with high WMH volume are at increased risk for dementia and stroke, controlling for vascular risk factors, but WMH burden is not reliably assessed in clinical practice. Manual segmentation of WMHs is accepted as the gold standard (or ground truth), however, it is a laborious and time-consuming method. Newer machine learning (ML)-based approaches are being proposed as alternatives to manual segmentation. Among these approaches, U-Net convolutional neural networks have demonstrated good WMH segmentation performance. However, even state-of-the-art ML models sometimes fail to correctly identify WMHs and their boundaries with sufficient accuracy. Attention blocks have emerged as a potential solution for improving the performance of U-Net models by enhancing the ability of the model to focus on relevant features in the data. We investigated the effectiveness of attention blocks in U-Net models for WMH segmentation compared to three other models (U-Net++, U-Net3+, and a standard U-Net). Attention blocks significantly improved the F-measure score for WMH segmentation (0.811 vs 0.789 for next best model, p=0.04) in a diverse brain imaging dataset. This study demonstrates that attention blocks enhance U-Net models used for WMH identification and classification.
白质高强度（WMHs）是老年人磁共振（MR）图像上的常见发现，在流体衰减反转恢复（FLAIR）成像上表现为高信号强度区域。在控制血管危险因素的情况下，WMH体积高的人患痴呆和中风的风险增加，但临床实践中对WMH负担的评估并不可靠。人工分割wmh是公认的金标准（或基础真理），但它是一种费时费力的方法。新的基于机器学习（ML）的方法被提出作为人工分割的替代方案。在这些方法中，U-Net卷积神经网络表现出了良好的WMH分割性能。然而，即使是最先进的ML模型有时也无法以足够的精度正确识别wmh及其边界。通过增强模型关注数据中相关特征的能力，注意力块已经成为改善U-Net模型性能的一种潜在解决方案。与其他三种模型（U-Net++、U-Net3+和标准U-Net）相比，我们研究了U-Net模型中注意力块对WMH分割的有效性。在不同的脑成像数据集中，注意力块显著提高了WMH分割的f测量得分（0.811 vs 0.789，次优模型，p=0.04）。本研究表明，注意块增强了用于WMH识别和分类的U-Net模型。
{"title":"Attention Blocks Improve White Matter Hyperintensity Semantic Segmentation using U-Nets","authors":"Kauê Tartarotti Nepomuceno Duarte;Murilo Costa de Barros;Abhijot Singh Sidhu;David Gobbi;Cheryl McCreary;Feryal Saad;Richard Camicioli;Eric Smith;Marco Carvalho;Richard Frayne","doi":"10.1109/TLA.2025.11072499","DOIUrl":"https://doi.org/10.1109/TLA.2025.11072499","url":null,"abstract":"White matter hyperintensities (WMHs) are a common finding on magnetic resonance (MR) images in older individuals, appearing as high-signal intensity regions on fluid-attenuated inversion recovery (FLAIR) imaging. People with high WMH volume are at increased risk for dementia and stroke, controlling for vascular risk factors, but WMH burden is not reliably assessed in clinical practice. Manual segmentation of WMHs is accepted as the gold standard (or ground truth), however, it is a laborious and time-consuming method. Newer machine learning (ML)-based approaches are being proposed as alternatives to manual segmentation. Among these approaches, U-Net convolutional neural networks have demonstrated good WMH segmentation performance. However, even state-of-the-art ML models sometimes fail to correctly identify WMHs and their boundaries with sufficient accuracy. Attention blocks have emerged as a potential solution for improving the performance of U-Net models by enhancing the ability of the model to focus on relevant features in the data. We investigated the effectiveness of attention blocks in U-Net models for WMH segmentation compared to three other models (U-Net++, U-Net3+, and a standard U-Net). Attention blocks significantly improved the F-measure score for WMH segmentation (0.811 vs 0.789 for next best model, p=0.04) in a diverse brain imaging dataset. This study demonstrates that attention blocks enhance U-Net models used for WMH identification and classification.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":"23 8","pages":"646-661"},"PeriodicalIF":1.3,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072499","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144581748","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：