Background and aim Regular physical activity preserves functional independence in older adults, yet care-home residents often miss out because personalized supervision is scarce. Autonomous, technology-supported exercise platforms could deliver such guidance without additional staff time—but only if sessions are automatically monitored for safety and quality. We therefore designed a deep learning (DL) system that (a) recognizes individual exercise types and (b) estimates joint angle trajectories from a standard video recording. These outputs are used to compute objective exercise performance metrics (EPMs) such as duration, repetition count, motion variability, and range of motion.

Methods Seven care-home residents (aged between 65–94 years) performed six common rehabilitation exercises in front of a single camera while wearing 17 inertial sensors (Xsens MVN Awinda) that provided ground-truth joint angles. Two-dimensional skeleton poses estimated from the video were fed into a temporal convolutional neural network to recognize the exercises and estimate three-dimensional joint angles. We evaluated exercise segmentation with F1@50 and angle regression with mean per-joint angular error (MPJAE) across nine trunk and lower-limb joints, using leave-one-subject-out cross-validation. Pearson correlations assessed agreement between estimated and ground-truth EPMs.

Results The DL model achieved an F1@50 of 0.92 (± 0.04) for exercise recognition and an MPJAE of 7.7° (± 0.91) for joint angle estimation. The estimated EPMs aligned closely with ground truth, achieving correlation scores of 0.93 (95% CI [0.90, 0.95]) for duration, 0.86 (95% CI [0.80, 0.90]) for repetition count, and between 0.3 and 0.9 for motion variability and range of motion across exercises.

Conclusion The DL algorithm reliably estimates key exercise outcomes from a single video stream. This video-based monitoring pipeline could enable unsupervised, technology-supported exercise assessment in residential care homes while safeguarding session quality and safety. Future work will validate the approach in larger and more diverse cohorts.

Introduction: Frailty, characterized by a reduction in intrinsic capacity across multiple physiological systems, is a key concern in healthy aging. Insight in the trajectory of an individual’s functional ability and intrinsic reserve capacity in a relatively younger population of older adults is lacking. This study aims to investigate the early stages of frailty by tracking trajectories of physical indicators of intrinsic capacity before frailty becomes clinically evident. Methods: The AMersfoort COhort study on functional decline, Healthy aging and Frailty (AMCOHF) is a unique 10-year prospective cohort study evaluating the predictive value of longitudinal trajectories of physical parameters for frailty onset or robustness maintenance. An a-select community-dwelling robust population of Amersfoort (55–75 years) in the Netherlands will undergo baseline assessments for inclusion criteria and will be followed longitudinally every 2.5 years. Frailty status is assessed using the Fried phenotype, Rockwood frailty index, and Groningen frailty indicator. Testing procedures and questionnaire completion include physical performance tests in the domains: (1) musculoskeletal system, (2) articular system, (3) cardiorespiratory system, (4) sensory system, (5) immune system, and 6) uro-gynecological system. Study outcomes focus on intrinsic capacity, functional ability, explanatory data, and frailty. Statistical analyses evaluating the predictive capacity include logistic regression, confirmatory factor or latent class analysis, and structural equation modeling. Nonprobability convenience sampling recruits 2,078 robust participants, estimating a 1-year frailty incidence of 1.5%–6.0%. Ethical approval was obtained, and the trial is prospectively registered on Open Science Framework (DOI: 10.17605/OSF.IO/RMBQV). Conclusion: The AMCOHF study will contribute to knowledge about markers to predict an accelerated decline in intrinsic capacity in an early stage. This knowledge is important to deploy prevention strategies at an earlier stage in life then those currently undertaken, ultimately reducing healthcare costs and contributing to a healthy aging population.

Assistive lower-limb exoskeletons (LLEs) have been recognized as promising tools for enhancing physical capacity in stroke survivors. Involving end-users in the early development stages is essential to ensure these technologies meet user needs. Co-design approaches, which actively engage end-users, support this goal. This study aims to (1) evaluate the impact of fatigue on daily living, (2) identify activities that could benefit from LLE assistance, (3) outline design and usability requirements for home-based LLEs, and (4) define physical parameters LLEs should monitor and assess. Discussions were structured using the PERCEPT co-design methodology and thematically analyzed. Four chronic stroke survivors participated in three focus group sessions, each lasting approximately 2 hours. Fatigue was identified as a significant factor in daily life, underscoring the importance of assistive technologies, such as LLEs, to help mitigate exhaustion. Participants recognized LLEs as valuable tools for enhancing physical performance, with benefits for muscle strength, balance, fatigue management, coordination, and general mobility. Design considerations included system modularity, battery efficiency, ease of donning and doffing, and practical needs for daily use. Our findings offer valuable insights into stroke survivors’ design and usability concerns regarding LLEs and provide a foundation for advancing the development and adoption of new assistive technologies.