Search Results (1,229)

Background/Objectives: To describe the progression of axial and segmental ocular biometric lengths and refractive status in adolescents and study independent associations between these changes and baseline ocular biomechanics. Methods: Prospective cohort of 126 eyes from 63 individuals followed for 2.5 years. Data from general health and lifestyle were collected through a validated questionnaire. Data from ocular biometry (IOL MASTER 700^®), objective refraction, and ocular biomechanics (Corvis ST^®) were collected at baseline and the end of follow-up timepoints. Biomechanical parameters were correlated with the variation in axial length (d_AL), vitreous cavity length (d_VCL), and spherical equivalent (d_SE). Multivariable linear regression models (one eye randomly assigned) adjusted for age, SE, and AL were developed to identify independent associations between baseline biomechanics and d_AL, d_VCL, and d_SE. Results: The cohort of the present work had a mean age of 14.1 ± 2.6 years at baseline. Variations of 0.122 ± 0.17 mm, 0.092 ± 0.17 mm, and −0.32 ± 0.9 D were found in AL, VCL, and SE at follow-up, respectively. Within the multivariable regression models, the biomechanical parameters found to be independently associated with d_AL (model 1), d_VCL (model 2), and d_SE (model 3) were as follows: Model 1—Biomechanically corrected IOP (bIOP), Integrated Radius (IR), and A2 Deflection Area (A2DArea); Model 2—bIOP, IR, and A2DArea; and Model 3—IR and WholeEyeMovementMAxTime (MaxWEMT). Conclusions: The study of ocular biomechanical behavior may play a pivotal role in the risk assessment of ocular elongation and myopic progression. This work found independent associations between ocular biomechanical behavior at baseline and axial and segmental ocular elongation and refractive myopization, mainly including bIOP, IR, and MaxWEMT. Full article

Driver drowsiness detection has become an important application of sensor-based monitoring systems aimed at improving road safety. This review focuses on sensing technologies and physiological parameters used for real-time drowsiness detection in drivers. The surveyed approaches are categorized into physiological sensing methods, including electroencephalography (EEG), electrocardiography (ECG), galvanic skin response (GSR), and photoplethysmography (PPG), and mechanical sensing methods, including respiration rate, eye blinking, head movement, yawning, and steering wheel gripping force. Each method is analyzed from a sensor system perspective, considering signal acquisition principles, measurement location, and practical deployment constraints. In addition, the reviewed techniques are evaluated based on real-time capability, level of sensor attachment, cost, restriction of user movement, and suitability for standalone operation. The comparison highlights that mechanical sensing approaches provide non-invasive and cost-effective solutions; however, they are sensitive to environmental noise and behavioral variability. In contrast, physiological sensing methods offer more direct and earlier indicators of fatigue-related changes in biosignals, although they typically require wearable or contact-based sensors and more complex acquisition systems. The review further indicates that multimodal sensor fusion is increasingly being adopted to improve robustness and reliability in real-world driving conditions. Overall, this work provides a structured overview of sensing modalities and highlights key considerations for designing efficient, real-time driver monitoring systems. Full article

Background: This study investigated the clinical timeline, patient monitoring behaviors, and cumulative bilateral treatment burden in patients with bilateral neovascular age-related macular degeneration. Methods: We retrospectively analyzed follow-up patterns and treatment intensity from first-eye (FE) diagnosis to second-eye (SE) conversion. Results: SE conversion occurred within a mean of 2.0 years in the FE-active group (62.5%) while the FE remained exudative, contrasting with 6.2 years in the FE-inactive group (37.5%). Upon SE conversion, the total annual intravitreal injection burden escalated 3.4-fold (p = 0.002). Notably, the FE-inactive group exhibited numerically lower annual outpatient visit counts (4.40 ± 2.71 vs. 10.29 ± 5.02; p = 0.116), which potentially widened the monitoring window. Additionally, baseline SE retinal pigment epithelium (RPE) abnormalities independently predicted progression (aOR: 19.04; p = 0.032). Conclusions: While previous literature focuses on individual eyes, our findings highlight a vigilance gap in SE detection based on FE status. Clinicians must maintain proactive surveillance for patients with baseline SE RPE abnormalities, particularly when FE stability or next-generation long-acting therapies extend clinic intervals. Due to the limited sample size, these preliminary findings warrant validation in larger prospective cohorts. Full article

Unmanned aerial vehicles (UAVs) are widely used for documentation, surveying, and 3D modeling in the built environment, yet their outputs often remain difficult to reuse for immersive comparison of alternative construction scenarios. This study presents a low-cost UAV-to-3DGS-to-VR workflow for constructing scenario-comparable immersive environments for built-environment review. The workflow combines multi-angle UAV imagery, point-cloud-based geometric anchoring, 3D Gaussian Splatting (3DGS), and Unity-based virtual reality (VR) to transform drone-captured reality into a reusable scene for controlled scenario comparison. The workflow is demonstrated in Middenbeemster, the central town of the Beemster polder World Heritage property. One present-condition scene (M0) and three alternative construction scenarios (M1 to M3) were created within a shared spatial reference. Reconstruction quality was assessed using PSNR and SSIM, and the VR scenes were further evaluated through eye-tracking, head-motion recording, and subjective ranking. The results indicate that the workflow can generate visually reliable and directly comparable immersive scenes from UAV data in this case study. Behavioral and subjective findings showed a consistent pattern, with M1 appearing more compatible than M2 and M3 in this pilot evaluation. The study contributes a pilot UAV-based workflow that links reality capture, immersive scenario comparison, and supplementary behavioral evidence within one process. Full article

History museum exhibition spaces convey historical and cultural information through static artifacts, graphic–text narratives, and spatial atmosphere. The light environment affects not only exhibit visibility but also visitors’ visual comfort and perceptual preference. However, existing studies mainly focus on single lighting parameters, and perceptual differences across multiple lighting conditions remain insufficiently understood. This study investigated static exhibition spaces in history museums through a comparison of 12 virtual lighting conditions generated from different combinations of ambient illuminance, exhibit illuminance, and correlated color temperature. Visitors’ visual behavior and subjective perception were evaluated through eye-tracking experiments, heatmap analysis, and Likert-scale ratings. Different lighting combinations significantly affected visual attention allocation and subjective evaluation. Total duration of fixation, number of fixations, and average pupil diameter showed significant differences across conditions, whereas average fixation time did not. Overall, moderate ambient illuminance and higher exhibit illuminance were associated with more stable visual responses and more positive perceptual evaluations, while correlated color temperature showed a regulatory effect within the tested range of 3000–4000 K. These findings provide preliminary evidence for understanding perceptual responses to lighting combinations in static exhibition spaces and may inform future field-based validation of museum lighting design. Full article

Evaluations of medical artificial intelligence (AI) explanations often rely on self-reported trust, perceived usefulness, acceptance, or final decision outcomes, while less directly characterizing whether users check evidence around AI outputs during decision making. In AI-assisted chest radiograph interpretation, a critical process-level question is whether users return from the AI output to the original image evidence when further scrutiny is needed. To address this question, we examined whether verifiable explanations—explanations designed to make AI recommendations checkable against the original image evidence—are associated with process markers of visual evidence checking in AI-assisted chest radiograph interpretation using eye-tracking and human-factors process measures. A 2 × 2 between-subjects experiment manipulated verifiable explanations (present vs. absent) and risk context (high vs. low), with AI recommendation correctness embedded at the trial level. Fifty-six clinically trained participants each completed 24 interpretation trials. Analyses focused primarily on gaze transitions between the AI output and the original image and dwell time on the original image, with response time and exploratory verification-related behavioral states used as auxiliary process measures. Verifiable explanations did not simply increase acceptance of AI recommendations. Instead, when AI recommendations were incorrect, they were most clearly associated with more frequent AI–image transitions and longer absolute dwell time on the original image evidence. Exploratory state-based analyses further suggested a lower tendency toward no-verify adopt under incorrect AI recommendations, but these findings were treated as complementary rather than primary evidence. Overall, the value of verifiable explanations lies not only in final decisions but in whether they make AI recommendations more inspectable against the original evidence. These findings provide eye-tracking evidence consistent with visual evidence checking in AI-assisted diagnostic interfaces and underscore the value of process-sensitive human-factors measures in medical AI evaluation. Full article

Using search tools for help-seeking during simulation-based video lectures has become increasingly popular and important. Intelligence search tools (such as GenAI) and traditional search engines, two of the most commonly used search methods, can aid students’ learning of scientific concepts. However, few studies have investigated how different search modes influence students’ understanding of these concepts. The present study aimed to identify the effects of GenAI search and traditional search on students’ understanding of scientific concepts in a simulation-based video lecture, focusing on visual attention, behavioral patterns, cognitive load, and learning performance. Our study adopted a between-subjects design with two experimental conditions: (a) Traditional search (n = 28) and (b) GenAI search (n = 29). We assigned participants to watch a video lecture, during which they could choose to operate a virtual simulation experiment or search for relevant information in a designated search area. Participants’ eye movements, behavior patterns, cognitive load and learning performance were measured and analyzed. The result revealed that traditional search prompted sustained attention to video learning materials, while GenAI search increased cross-area visual scanning and showed more frequent information search behavior. In addition, students’ retention performance improved, although their transfer performance did not show a significant increase in GenAI search. Furthermore, there was no significant difference in cognitive load between the two conditions. These results suggest that although GenAI encourages students to organize and process more information in the active exploration region, it has notable limitations in supporting deeper, context-dependent understanding and deductive reasoning regarding abstract concepts. Full article

Approximately one-third of the global population cooks over open fires in the home, which are responsible for millions of deaths per year globally. The aim of this longitudinal cohort study was to develop, implement, and evaluate an intervention to improve the cooking behaviors of women in a rural Nicaraguan community. The study employed the Community Health Action Model for Participatory Behavior Change (CHAMP-BC) framework to develop a novel approach to improving cooking behaviors and reducing symptoms associated with household air pollution. The study was conducted from October 2007 to March 2018. Community health needs assessments and community forums were conducted. Readiness to change surveys were administered. Pre- and post-test cookstove implementation surveys were conducted. One hundred sixty-seven primary cooks, primarily women (99%), consented to participate in the improved cookstove program and were followed from baseline for up to seven years post-intervention. There were significant improvements in cough (p < 0.0001), chest illness (p < 0.0001), shortness of breath (p < 0.0001), headaches (p < 0.0001), and eye irritation (p < 0.0001). The novel CHAMP-BC framework enhanced the research approach by empowering women to change their cooking behaviors while providing them with the autonomy and agency necessary for decision-making regarding evidence-based methods to improve their health. Full article

In equestrian circles, horse–rider harmony is understood intuitively, yet clear criteria are lacking. This study examined how different equestrians conceptualized and visually assessed harmony and how this influenced scoring. Qualitative interviews were combined with eye tracking technology. Thirty equestrians assessed five videos of horse–rider combinations performing in dressage, showjumping, eventing, working equitation and Icelandic riding, with eye movements being recorded using a mounted eye tracker. Verbal definitions were analyzed using thematic analysis, revealing three overarching themes: Horse Behavior, Horse–Rider Connection, and Rider Influence. Number of fixations and duration of fixation were reduced using principal component analysis (PCA), yielding five components each. These explained 70.9% and 64.5% of variance, respectively. Linear mixed-effects models showed significant effects for two PCA components: frequent fixation on the horse’s ears and eyes relative to the horse’s shoulder and rider leg predicted lower harmony scores (B = −0.34, SE = 0.13, z = −2.53, p < 0.05), whereas longer dwell time on the rider’s shoulder relative to the rider’s leg predicted higher scores (B = 0.25, SE = 0.12, z = 2.12, p < 0.05). Ears and eyes were also the most frequent first fixation. Harmony appears to be a shared construct at the conceptual level, but is personally enacted at the practical level. Equine facial expressions and rider posture serve as important perceptual indicators. Full article

Background/Objectives: Visuospatial impairments are among the earliest cognitive symptoms in Alzheimer’s disease (AD) and related dementias (ADRD), yet standard assessments often lack ecological validity and focus on isolated domains. This study examines whether integrating linguistic, behavioral, and eye-tracking measures provides a more comprehensive characterization of cognitive deficits within a multimodal, exploratory framework. Methods: Twenty older adults (10 with mild to moderate dementia, including AD/ADRD, and 10 age-matched controls) completed three tasks: (1) oral narrative production, (2) visuospatial behavioral tasks (manipulation, recognition, reproduction), and (3) free-viewing eye-tracking. Linguistic, behavioral (time, errors), and fixation-based measures were analyzed using non-parametric statistics, with emphasis on effect sizes and cross-domain patterns. Results: The clinical group differed consistently from controls across domains. Linguistic measures showed increased output but reduced quality, including lower syntactic complexity, more grammatical errors, greater pragmatic deviations, and reduced gist comprehension. Behavioral tasks revealed slower performance and more frequent failures. Eye-tracking differences were less pronounced, showing a tendency toward longer fixations and less efficient visual exploration. A composite multimodal index showed clear separation between groups, indicating a consistent pattern of impairment across measures. Conclusions: Cognitive differences in dementia are expressed across multiple domains, with the strongest effects in linguistic and behavioral measures. These findings highlight the value of multimodal profiles for capturing multidimensional impairment. Results should be interpreted as exploratory and require confirmation in larger, confirmatory studies. Full article

Animate attention bias reflects the visual system’s tendency to prioritize animate over inanimate stimuli. This bias is reduced in autism spectrum disorder (ASD), suggesting that similar patterns may also be observed in individuals with high autistic traits (AT). Although previous research has reported reduced animate attention bias during early attentional orienting in individuals with high AT, how this bias unfolds across processing stages remains unclear. Using a dot-probe task combined with eye-tracking, the present study examined this stage-related pattern in individuals with high and low AT. Response time results showed that the low AT group had a significant animate-probe advantage, whereas the high AT group showed no significant advantage, broadly replicating prior findings. In stage-wise analyses, the low AT group showed a significant animate-probe advantage at the late stage, whereas the high AT group showed no significant advantage at either stage. However, this group difference was not reflected in most fixation-based measures. This RT–fixation dissociation suggests that reduced animate attention bias in high AT should not be interpreted simply as reduced overt fixation allocation to animate stimuli, but may reflect differences in using animacy-related cue-location information to facilitate subsequent probe detection and response selection. Full article

This study examines the association between Integrated Reporting (IR) quality and tax avoidance among South African listed firms from 2012 to 2021, and whether this relationship differs across the highest and lowest levels of IR quality. The extent to which the adoption of a Combined Assurance (CA) model strengthens the IR monitoring role in reducing tax avoidance, as well as the IR quality link with ESG-related implications of tax avoidance, are also explored. IR quality is directly derived from the EY Excellence in Integrated Reporting Awards ranking. This ranking evaluates firms’ adherence to the IR framework and is thus employed as a comprehensive proxy for IR quality. Tax avoidance is captured through multiple proxies. The main findings reveal no significant overall association between IR quality and tax avoidance, suggesting a decoupling between IR and tax behavior. However, when examining firms at the highest and lowest levels of IR quality, a significant negative relationship emerges only for the top performers (highest IR quality), indicating that IR constrains tax avoidance only when supported by a strong ethical corporate culture. Firms adopting CA exhibit higher tax avoidance, suggesting that IR and CA may be constrained by underlying corporate culture and used symbolically. Higher IR quality is also associated with lower tax avoidance relative to GDP and reduced potential revenue losses relative to government expenditures on education, health, and environmental protection. These findings contribute to the literature on IR, corporate governance, and tax avoidance, while also informing policymakers and regulators on the need to strengthen IR and CA frameworks through enhanced tax transparency requirements, thereby supporting equitable resource mobilization, institutional trust, and long-term sustainable development. Full article

Effective workplace collaboration is essential for productivity and creativity, yet achieving the necessary mutual understanding can be challenging, particularly involving individuals from different neurotypes. This work evaluates VIRTUES, a Virtual Reality (VR) platform designed to foster mutual understanding and collaborative behaviors between autistic and neurotypical individuals. VIRTUES integrates multimodal sensing (eye tracking, interaction logs, and transcribed speech) to objectively quantify five defined dimensions of collaboration while providing real-time, context-aware support through an embedded rule-based feedback mechanism. A user study involving 12 autistic–neurotypical pairs demonstrates that VIRTUES can assess and support collaborative efforts across different neurotypes. Through synchronized sensing data, we identified that Information Pooling serves as a critical driving factor for successful collaborative performance. These preliminary findings suggest that VIRTUES provides a foundation for exploring inclusive teamwork and may inform the design of future interventions to support neurodiverse social-technical skill acquisition. Full article

This study explores the use of transfer learning in gaze estimation, focusing on the development of personalized models tailored to individual users. Our approach involves collecting gaze data using standard laptop webcams, designed to operate effectively within resource-limited settings, thereby enhancing accessibility and affordability. We conducted a comparative analysis of models using transfer learning against models trained without pre-trained weights, examining their convergence behavior and sensitivity to different dataset sizes. The analysis includes both eye and face images, providing a comprehensive view of model adaptability. Our findings show that while both methods produce comparable results overall, transfer learning offers notable advantages—particularly faster convergence, reduced computational cost, and enhanced stability when data are limited. However, the results also reveal that transfer learning is not universally superior; for face images, models trained from scratch achieved lower mean errors but exhibited higher variability, whereas transfer learning ensured more consistent performance. These insights highlight that the benefits of transfer learning depend on the data characteristics and task complexity. In the most data-constrained setting (100 images), transfer learning reduced the mean error by 20.99 px for the Left Eye model and 35.56 px for the Right Eye model, whereas for face images the models trained from scratch consistently achieved lower mean error across all evaluated dataset sizes. Overall, this study underscores the potential of transfer learning for efficient and scalable gaze estimation, particularly in small-data environments, while providing practical guidance on when and how transfer learning yields the greatest benefit for real-time applications such as human–computer interaction (HCI), assistive technologies, and personalized user experiences. Full article

Background/Objectives: Uncorrected refractive errors (UREs) are a primary cause of preventable visual impairment in children globally, impacting education and quality of life. In Jakarta, prevalence has surged to 40% post-pandemic, categorizing it as a serious public health problem. This study aimed to develop and validate the CIPSEL questionnaire as a rapid, culturally adapted screening tool for identifying visual impairment consistent with possible UREs among Indonesian school-aged children. Methods: A cross-sectional study was conducted in South Jakarta with 131 students aged 8–12 years. The 10-item CIPSEL questionnaire, exploring visual behaviors and symptoms, was administered via face-to-face interviews. Visual acuity was assessed using a standard Snellen chart by medical personnel blinded to the questionnaire results. Diagnostic accuracy was assessed using receiver operating characteristic (ROC) curve analysis, with optimal thresholds determined via the Youden Index and the shortest distance to (0, 1). Results: Visual impairment was identified in 26 students (19.8%). Mean CIPSEL scores were significantly higher in students with visual impairment (4.73) compared to those with normal vision (1.95). ROC analysis showed considerable diagnostic accuracy with an AUC of 0.887 (95% CI: 0.829–0.946). A safety-first cutoff of 2.5 prioritized sensitivity (96.2%), while a balanced cutoff of 3.5 provided 80.8% sensitivity and 79.0% specificity. A tiered risk system (Low, Medium, and High) demonstrated a robust statistical association with actual clinical findings (Cramer’s V = 0.534, p < 0.001). Conclusions: CIPSEL is a reliable and scalable screening tool for the early detection of visual impairment in Indonesian children. Its tiered risk stratification framework facilitates nuanced clinical decision-making and efficient resource allocation in school-based settings. Its accessibility for non-medical personnel and potential for digital integration support national efforts toward universal eye health. Full article