Search Results (208)

Stroke is a leading cause of chronic disability, with heterogeneous sensorimotor impairments that are not well captured by standard clinical assessments. While advanced motion capture and robotic systems provide precise measurements, they are not scalable for widespread clinical use. We developed C-MORE (Computer Vision for Movement Observation and Recovery Enhancement), a smartphone-based framework that uses computer vision and machine learning to automatically score the Box and Blocks Test (BBT) and extract quantitative kinematic metrics. The system combines hand tracking with a custom machine learning (ML) architecture to identify valid block transfers and segment task phases. We evaluated C-MORE in 7 individuals with chronic stroke and a cohort of 10 healthy adults. The system achieved 99.0% agreement with ground-truth scoring, with errors below clinically meaningful thresholds. Kinematic measures derived from the system were sensitive to stroke-related impairments, including reduced movement velocity and increased task duration in affected limbs. Exploratory analyses indicated that grasp-related metrics, particularly the ratio of grasp-to-transfer duration, were correlated with independent measures of proprioception. These findings demonstrate that smartphone-based computer vision can provide accurate, scalable assessment of upper-extremity function. C-MORE offers a practical approach for enhancing clinical evaluation and enabling more precise, individualized rehabilitation strategies. Full article

The objective of this study was to evaluate the efficacy of a novel digital platform (Visitrain VG, Alicante, Spain) as a visual rehabilitation tool for patients with convergence insufficiency (CI), in comparison with conventional in-office vision therapy (VT) supplemented with home reinforcement exercises. A retrospective comparative study was conducted comprising 33 patients diagnosed with CI, allocated into two groups: a digital group (DG; n = 16) receiving treatment with the aforementioned digital platform and a conventional group (CG; n = 17) undergoing conventional vision therapy. Binocular vision clinical parameters were assessed at baseline, one month, and three months of follow-up, including near point of convergence (NPC), positive fusional vergence (PFV), and binocular accommodative facility (BAF). Both groups demonstrated significant improvements following three months (p < 0.050). At the one-month evaluation, the CG showed a more rapid clinical response, with statistically significant between-group differences being observed in the NPC (p = 0.004) and near PFV (p = 0.040) compared with the DG. Nevertheless, at the three-month follow-up, no significant differences were found between the groups (p ≥ 0.060). The digital platform under investigation appears to constitute an effective therapeutic alternative to conventional vision therapy, albeit with a comparatively slower initial clinical response rate. It may be particularly indicated for patients requiring greater scheduling flexibility or those with limited access to in-office clinical care. Prospective controlled clinical trials are warranted to corroborate these preliminary outcomes. Full article

Rationale and objectives: Parkinson’s disease (PD) significantly affects visual function, especially convergence and eye movements, impacting tasks such as reading. The objective was to investigate preliminarily the impact of the use of digital visual training in PD patients with associated convergence insufficiency (CI). Materials and methods: Pre–post pseudo-experimental pilot study to evaluate the impact of a novel digital therapy system (video game for use on a mobile phone or tablet) in 13 patients with PD and CI, with a mean age of 67 years. A comprehensive visual assessment was performed before and after a 6-week home-based visual rehabilitation, including measurement of near point of convergence (NPC), near positive fusional vergence (PFV), oculomotor tests (NSUCO and King-Devick tests), and symptom assessments with two validated questionnaires (CISS and SQVD). Results: Treatment adherence was variable, ranging from 0.8% to 124.7%. Despite this, significant improvements were found after therapy in break (p = 0.022) and recovery points of the NPC (p = 0.007), as well as break (p = 0.003) and recovery points in near PFV (p < 0.001). In the NSUCO test, the total score improved significantly from 23.9 ± 4.2 to 26.2 ± 3.7 after therapy (p = 0.003). Furthermore, a significant reduction in the total King-Devick test time was observed, decreasing from 79.4 ± 28.8 s to 69.0 ± 21.5 s with therapy (p = 0.034). Finally, symptom questionnaire scores also decreased significantly with therapy (CISS p = 0.037, SQVD p < 0.001). Conclusions: The digital vision therapy system evaluated seems to improve oculomotor control and reduce visual symptoms associated with CI in PD patients. Studies with larger sample sizes and a control group are needed to fully validate the therapeutic effectiveness of this tool. Full article

The burden of disability is rising rapidly in Africa, where a severe shortage of rehabilitation professionals and limited infrastructure create a major treatment gap. Immersive virtual reality and serious games have shown promise for upper limb rehabilitation, but current extended reality (XR) solutions lack personalization, cultural adaptability, real-time feedback, and scalability. This perspective paper proposes a conceptual AI-enhanced XR framework tailored to African low- and middle-income countries. We identify how generative AI, large language models, multiagent systems, and explainable AI can address specific rehabilitation barriers. The framework integrates these four pillars into a three-layer architecture covering content creation, interaction, and decision support. We analyze implementation considerations specific to African contexts—infrastructure, capacity building, cultural adaptation, ethics, and financing—and outline a detailed research agenda with near, medium, and longer term priorities. Realizing this vision requires co-design with African communities, investment in local capacity, adaptation to infrastructure constraints, and development of ethical frameworks. AI-enhanced XR has the potential to democratize access to quality rehabilitation across Africa, but this potential must be validated through rigorous, context-sensitive research. Full article

Accurate, rapid, and consistent evaluation of pavement condition across large-scale road networks is critical for sustainable maintenance and rehabilitation planning. However, conventional approaches largely rely on manual visual inspections, which are time-consuming, subjective, and difficult to implement at the network level. In this study, a semi-automated pavement distress evaluation framework that integrates field-based assessment with computer vision techniques is proposed. The study was conducted on a 3 km roadway network located within the Yıldız Technical University Davutpaşa Campus. Field-based distress observations were used as reference data, while street-level images obtained from the Mapillary platform were analyzed using a deep learning-based YOLOv8 model trained on the RDD2022 dataset, which was specifically developed for road distress detection. The analysis focuses on crack and pothole distress, which have a dominant influence on PCR and are highly distinguishable in image-based approaches. Correlation analyses between automated detection results and field-based data demonstrate a strong agreement, reaching values of approximately $\rho \approx 0.90$ in some routes. These findings indicate that these distress types are effective in representing variations in pavement condition. The results demonstrate that multi-source image data and deep learning-based detection methods can be reliably used for section-level pavement condition assessment. The proposed approach addresses a key gap in the literature by transforming image-level detections into engineering-based decision-support information. Furthermore, by leveraging publicly available data sources, the framework offers a low-cost and scalable solution that enables rapid preliminary assessment over large road networks, thereby providing significant potential for sustainable infrastructure management and the development of data-driven maintenance strategies. Several practical challenges encountered during the detection process—including sensitivity to contrast enhancement parameters, false positives from shadows and surface reflections, heterogeneous image resolution across crowdsourced imagery, and training distribution gaps for locally prevalent infrastructure features—are discussed, and directions for reducing human intervention through adaptive preprocessing and targeted model refinement are identified. Full article

Background: Artificial intelligence (AI) is rapidly transforming healthcare, with increasing applications in sports medicine. Advances in machine learning, deep learning, and computer vision enable the analysis of large, heterogeneous datasets derived from imaging, wearable sensors, performance-monitoring systems, and electronic health records. While these technologies offer opportunities to enhance injury prevention, diagnostic accuracy, rehabilitation monitoring, and clinical decision-making, their integration into athlete care remains complex and context-dependent. Methods: A structured narrative review of the PubMed/MEDLINE database was conducted to identify clinically relevant AI applications in sports medicine. The search focused on key domains including injury risk prediction, musculoskeletal imaging, rehabilitation monitoring, return-to-play assessment, performance management, and clinical workflow support. Evidence from original studies, reviews, methodological reports, and regulatory documents was qualitatively synthesized to provide an overview of current applications, methodological limitations, and decision-level implications. Results: AI demonstrates growing utility across multiple domains of sports medicine. Machine learning models can identify complex, non-linear relationships among training load, physiological responses, and injury risk, though their predictive performance varies widely and is often limited by dataset heterogeneity and a lack of external validation. In musculoskeletal imaging, AI-based algorithms support automated detection and quantification of abnormalities, with performance in selected tasks approaching that of expert readers, yet remaining task-specific and context-dependent. Emerging applications include movement analysis and rehabilitation monitoring through wearable sensors and computer vision systems, as well as data-driven support for return-to-play decisions and clinical workflow optimization. However, current evidence highlights important limitations, including algorithmic bias, limited generalizability, poor interpretability, and the risk of misapplication in complex clinical decision-making contexts. Conclusions: AI is likely to become an important decision-support layer in sports medicine by enabling data integration and longitudinal monitoring. However, model performance does not necessarily translate into improved clinical outcomes, and AI-generated predictions remain probabilistic and context-sensitive. Consequently, clinical decisions—particularly high-stakes processes such as return-to-play—require structured integration of AI outputs within a broader clinical framework. The sports physician remains central as a human-in-the-loop integrator, responsible for contextualizing AI-derived information, mitigating potential errors, and ensuring safe, individualized athlete management. Full article

Artificial intelligence (AI) is reshaping medical rehabilitation by advancing from isolated assistive technologies toward data-driven program management. Beyond established applications in robotics and virtual reality, AI enables multimodal data integration, predictive analytics, adaptive therapy optimization, and real-time monitoring across rehabilitation domains. This review synthesizes 61 peer-reviewed studies to examine how AI supports the management, planning, and evaluation of rehabilitation programs. The evidence indicates strong technical maturity at the device and session levels, particularly in robotic control and wearable monitoring, whereas longitudinal program orchestration and system-level coordination remain at an emerging stage. Machine learning, reinforcement learning, computer vision, and time-series models facilitate patient phenotyping, therapy personalization, and prognostic modeling. However, their scalability is constrained by limited interoperability, heterogeneous outcome measures, and insufficient multicenter validation. A structured six-layer management architecture is proposed to conceptualize AI as an integrated orchestration framework. Advancing toward scalable and trustworthy rehabilitation ecosystems will require interoperable infrastructures, longitudinal validation, and embedded ethical and explainability mechanisms. Full article

Remote exercise supervision and functional movement assessment require sensing pipelines that can capture body motion, interpret protocol progression, and provide meaningful feedback within the same runtime environment. This paper presents ORAMA, an integrated computer vision platform for the execution and remote monitoring of digital exercises and clinically oriented assessment protocols related to physical fitness, mobility, balance, and health. The system combines ZED 2i stereo capture and depth-aware body tracking with a protocol-driven software architecture that includes a computer-vision pipeline, an exercise and assessment engine, a real-time feedback layer, persistent session handling, structured output generation, and a chatbot-assisted interaction path. Unlike solutions that focus only on movement recognition, ORAMA organizes each task as an explicit executable protocol with calibration stages, state transitions, task-specific metrics, and live visual guidance. The paper analyzes the system architecture, reviews the surrounding literature on virtual coaching and rehabilitation-oriented computer vision, and demonstrates representative user-interface and runtime views for both assessment and exercise scenarios. The present work reports a prototype architecture and representative operational demonstrations, rather than a completed clinical validation or participant-based efficacy study. The resulting platform shows how markerless 3D body tracking can be embedded within a unified and interpretable environment for guided exercise, functional testing, and remote follow-up without requiring wearable sensors. Full article

Telerehabilitation—the remote delivery of rehabilitation services—is undergoing a paradigm shift with the convergence of immersive virtual reality (VR) and wearable biosensor technologies. This perspective article outlines a vision for home-based motor and cognitive rehabilitation that is engaging, personalized, and data-driven. We describe how immersive VR environments (for example, simulations of home settings or supermarkets) coupled with wearable sensors can address current challenges in rehabilitation by increasing patient motivation, enabling real-time biofeedback, and supporting remote clinician supervision. Gamification mechanisms and rich sensory feedback in VR are highlighted as key strategies to enhance user engagement and adherence to therapy. We discuss conceptual innovations such as multi-sensor data integration, dynamic difficulty adaptation, and AI-driven personalization of exercises, derived from recent research and our development experience, and consider their potential benefits for patients with neuro-cognitive-motor impairments (e.g., stroke, Parkinson’s disease, and multiple sclerosis). Implementation scenarios for home-based therapy are presented, emphasizing scalability, standardized digital metrics for monitoring progress, and seamless involvement of clinicians via telehealth platforms. We also critically examine the current limitations of VR and telehealth rehabilitation and how an integrative model could overcome these barriers. More specifically, this perspective defines the engineering requirements of a closed-loop VR-based telerehabilitation framework, including multimodal data synchronization, calibration, signal-quality management, interpretable adaptive control, digital biomarker validation, and practical strategies to improve accessibility, privacy, and scalability in home-based neurological rehabilitation. Full article

Human pose estimation is a crucial task in computer vision with widespread applications in healthcare, rehabilitation, sports, and remote monitoring. In this paper, we propose a deep learning-based method for automatic human pose estimation and joint angle computation, tailored specifically for physiotherapy and telemedicine scenarios. Beyond pose estimation, the proposed method is able to compute angles between joints, enabling analysis of body alignment and posture. The proposed approach is built upon a customized skeleton with 25 anatomical keypoints and a dataset composed of over 150,000 annotated and augmented images derived from multiple open-source datasets. Experimental results demonstrate the effectiveness of the proposed method, achieving a mAP@50 of 0.58 for keypoint localization and 0.98 for object detection. Moreover, we demonstrate several real-world practical use cases in evaluating exercise correctness and identifying postural deviations by exploiting the proposed method, confirming that the proposed method can represent a promising approach for automated motion analysis, with potential impact on digital health, rehabilitation support, and remote patient care. Full article

(1) Background: Gait analysis provides quantitative information on walking patterns and has proven invaluable for assessing motor function in rehabilitation programmes. A markerless motion capture system combining computer vision techniques provides low-cost, real-time, portable gait analysis. (2) Methods: The kinematics of the knee and ankle of twenty-seven healthy volunteers were assessed using a single smartphone camera combined with the MediaPipe human pose estimation framework. The system was validated using the OPAL wearable sensor system by APDM Wearable Technologies. (3) Results: Findings showed close correspondence between the two systems for knee kinematics showing a mean absolute error of 4.10° ± 2.32° and 3.15° ± 3.10° for right and left knee flexion, respectively, and a mean absolute error of 2.30° ± 2.01° and 3.12° ± 2.63° for right and left knee extension, respectively. The mean absolute error for right and left knee range of motion was found to be 4.55° ± 3.12° and 4.15° ± 3.01°, respectively. Moreover, Bland–Altman plots indicated minimal bias (average 0.6 for flexion, average 0.47 for the extension, and 0.30 for the range of motion) and excellent correlation for knee flexion bilaterally (0.916 and 0.845 for the right and left side, respectively), with slightly lower but still satisfactory agreement for knee extension (0.862 and 0.845 for the right and left side, respectively). Conversely, ankle measurements revealed poor concordance: dorsiflexion and range of motion presented significant differences and systematic errors, while plantarflexion showed no statistical difference but weak correlation. (4) Conclusions: This study demonstrated that combining a smartphone camera with a human pose estimation framework allows for low-cost, real-time, portable gait analysis, particularly of the knee joint. Full article

Background: Age-related macular degeneration (AMD) is the primary cause of severe visual acuity loss in individuals over 60 with increasing prevalence. Currently, no effective treatments exist for geographic atrophy and macular scarring, highlighting the need for visual rehabilitation in these patients. Microperimetry offers functional assessment at any AMD stage and employs fixation training to help patients utilize the most effective retinal areas for vision. Methods: A prospective study involving 25 patients (50 eyes) aged 67 to 90. The MAIA II microperimeter assessed scotoma size and location, retinal sensitivity, macular integrity, fixation parameters (P1, P2, 63%BCEA, 95%BCEA), fixation stability, and preferred retinal locus. Quality of life was evaluated using the National Eye Institute Visual Function Questionnaire (NEI-VFQ-25). A subgroup with inactive AMD-related macular changes, either bilateral geographic atrophy (13 patients, 26 eyes) or bilateral scarring (12 patients, 24 eyes), was identified, all exhibiting bilateral absolute central scotomas of at least 2 degrees. Each patient completed 10 fixation training sessions with a microperimeter, training the eye with better acuity weekly. One-week post-training, a functional assessment was performed on both trained and untrained eyes. Results: Fixation training significantly improved best corrected visual acuity (BCVA) in trained eyes (mean change −0.14 logMAR, p < 0.001, large effect size) and also in fellow untrained eyes (−0.16 logMAR, p < 0.001). BNVA improved from 2.25 to 1.86 in trained eyes (p < 0.001) and from 2.96 to 2.76 in untrained eyes (p = 0.004). Fixation stability parameters improved significantly, including increases in P1 and P2 and reductions in Bivariate Contour Ellipse Area (BCEA). Quality of life measured using the NEI-VFQ-25 questionnaire improved significantly in 9 of 11 domains. Conclusions: Microperimetry may be a valuable tool for assessing visual function in AMD patients. Fixation training with the MAIA II microperimeter is both safe and effective for vision rehabilitation in those with geographic atrophy and macular scarring. Full article

Purpose: The purpose of this preliminary study was to investigate the prevalence of Charles Bonnet Syndrome (CBS) among patients attending the National Centre of Service and Research for the Prevention of Blindness and Vision Rehabilitation of the Visually Impaired, Rome, Italy. Furthermore, the research aimed to delineate the psychological profile of these individuals to determine whether significant differences exist compared with visually impaired patients who do not experience hallucinatory phenomena and to identify likely predictors. Methods: A preliminary cross-sectional analysis was conducted on a convenience sample of patients recruited between January 2025 and December 2025. Prevalence was calculated based on structured clinical interviews, while the psychological profile was assessed by comparing the CBS group with a control group (non-CBS) matched for visual acuity. Participants underwent comprehensive ophthalmological and psychological assessments, including best-corrected visual acuity (BCVA), reading acuity (RA), contrast sensitivity (CS), fixation stability, and retinal sensitivity (RS). Psychological status was evaluated using the Symptom Check List-90-Revised (SCL-90-R), the Patient Health Questionnaire (PHQ-9), and the Generalized Anxiety Disorder Questionnaire (GAD-7). Patients experiencing CBS were further interviewed regarding the specific characteristics and patterns of their hallucinations. The association between CBS and both psychological profiles and visual function parameters was evaluated using regression analysis. Results: Out of 385 individuals screened, 120 participants (58% women; mean age 55.4 ± 18.8 years) were included; CBS was detected in 19%. No significant differences were observed between participants with and without CBS in demographic variables or psychological questionnaire scores (p > 0.05). Mean SCL-90-R, PHQ-9, and GAD-7 scores indicated mild psychological distress, depression, and anxiety, with no significant group differences (p > 0.05). Using standard cut-off values, depressive and anxiety symptoms were prevalent in 65% and 88% of participants, respectively, but were not significantly associated with CBS in chi-square or logistic regression analyses (p > 0.05). Logistic regression analysis of SCL-90 scores showed that only anxiety was significantly associated with hallucination occurrence among the visually impaired participants (OR = 0.27; 95% CI = 0.08–0.87; p < 0.05). Among the visual function parameters, poorer RA in the worse eye was significantly associated with CBS (p < 0.05). Conclusions: This study confirms that CBS is a prevalent, yet frequently under-reported, condition within rehabilitation settings. While overall visual function did not differ significantly between patients with and without CBS, reduced reading acuity (RA) in the worse eye emerged as a potential specific risk factor. Characterizing the psychological profile of these patients is essential to differentiate the syndrome from psychiatric disorders and to develop tailored support pathways. Despite its preliminary nature, this research underscores the necessity of systematic screening to enhance clinical management and the emotional well-being of visually impaired individuals. Consequently, integrating psychological support into visual rehabilitation programs is vital to addressing the high prevalence of comorbid anxiety and depression. Full article

Background/Objectives: Due to their positions, professional baseball catchers are at elevated risk of concussion, which can impair visual processing. There is a need for sensitive sensorimotor monitoring tools to track concussion-related neurophysiological changes more accurately. We investigated whether oculometrics can address this need. Methods: Four Major League Baseball catchers completed an oculometric assessment shortly after suffering a concussion (Time 1) and again after completing vision rehabilitation (Time 2). The assessment produces 10 z-scored measures, including a summary score. Results: Players’ Time 1 summary score tended to be typical of a normal healthy adult (Mean = 0.07 z-scored units). On average, players improved by 1.3 z-score units from their Time 1 summary score (SD = 1.07). Exploratory analyses revealed that sensorimotor recovery was driven by smooth pursuit latency, proportion of tracking comprising smooth pursuit, and the amplitude of catch-up saccades. Conclusions: Our analysis was based on a very small sample of concussion cases, each of which was unique. Despite this limitation, our data show how oculometrics can measure improvements in visual processing following a concussion among baseball players with exceptional perceptual-motor skills. Our data highlight the risk that brain injuries in high-performing individuals go undetected due to standard-of-care tools normed to behavior from healthy control populations; for these athletes, “normal” scores cannot be interpreted as neurologically “healthy”. Full article

Dry eye disease (DED) is a complex, multifactorial, progressive disease that has consequences both for individuals and society. Symptoms reported by patients include discomfort in the eye and periodic blurred vision, while in the broader perspective, the disease is associated with economic burdens and challenges for healthcare systems. Globally, dry eye disease remains a growing problem observed in many countries. It is estimated that symptoms of dry eye syndrome occur in approximately 10 to 20 per cent of people over the age of 40. This prevalence is on the rise, which is associated with both the aging population and increased incidence among younger adults. In this group, factors such as contact lens wear and prolonged use of digital devices are considered to be contributing factors. Further epidemiological studies, conducted in different regions of the world, covering diverse populations and a wide range of age groups, with a particular focus on younger cohorts, may contribute to a more accurate understanding of the prevalence of dry eye disease. There are more and more methods of diagnosing DED. In addition to well-known procedures like the Schirmer test or tear break-up time, there are also methods that focus on the evaluation of the tear film or imaging of the ocular surface. Moreover, usage of artificial intelligence is also playing a significant role in it. However, the key issue in individual cases is introducing the most effective treatment based on combining available substances, including corticosteroids, antibiotics and supplements, which leads to a reduction in inflammation and improvement in visual comfort. Full article