Falls are a multifactorial cause of injuries for older people. Subjects with osteoporosis are particularly vulnerable to falls. We study the performance of different computational methods to identify people with osteoporosis who experience a fall by analysing balance parameters. Balance parameters, from eyes open and closed posturographic studies, and prospective registration of falls were obtained from a sample of 126 community-dwelling older women with osteoporosis (age 74.3 ± 6.3) using World Health Organization Questionnaire for the study of falls during a follow-up of 2.5 years. We analyzed model performance to determine falls of every developed model and to validate the relevance of the selected parameter sets. The principal findings of this research were (1) models built using oversampling methods with either IBk (KNN) or Random Forest classifier can be considered good options for a predictive clinical test and (2) feature selection for minority class (FSMC) method selected previously unnoticed balance parameters, which implies that intelligent computing methods can extract useful information with attributes which otherwise are disregarded by experts. Finally, the results obtained suggest that Random Forest classifier using the oversampling method to balance the data independent of the set of variables used got the best overall performance in measures of sensitivity (>0.71), specificity (>0.18), positive predictive value (PPV >0.74), and negative predictive value (NPV >0.66) independent of the set of variables used. Although the IBk classifier was built with oversampling data considering information from both eyes opened and closed, using all variables got the best performance (sensitivity >0.81, specificity >0.19, PPV = 0.97, and NPV = 0.66).

Background: Despite advances in myoelectric control of hand prostheses, their dropout rate remains high. Methods: We analyzed 37 features extracted from surface electromyography (sEMG) recordings from 15 participants, distributed into three groups: non-impaired individuals, impaired individuals with limb loss due to trauma, and impaired individuals with limb loss due to electrical burn. Feature relationships were examined with correlation heatmaps and two feature-selection methods (ReliefF and Minimal Redundancy Maximum Relevance), and classification performance was evaluated using machine-learning models to characterize sEMG behavior across groups. Results: Individuals with electrical-burn injury exhibited increased forearm co-contraction on the affected side across normalized isometric contractions, indicating altered motor coordination and likely higher energetic cost for prosthetic control. Feature selection and model results revealed etiology-dependent differences in the most informative sEMG features, underscoring the need for personalized, etiology-aware myoelectric control strategies. Conclusions: These findings inform the design of adaptive prosthetic controllers and targeted rehabilitation protocols that account for injury-specific motor control adaptations.

Musculoskeletal disorders such as rheumatoid arthritis and osteoarthritis affect millions worldwide and are projected to rise sharply by 2050, highlighting the importance of scalable telerehabilitation. This paper introduces Rehabi, a mobile, user-friendly tele-rehabilitation platform that centrally integrates markerless motion for biomechanical assessment and monitoring. Rehabi development followed a user-centered methodology, combining questionnaires, interviews, and natural language processing to elicit requirements from patients and clinicians. The system architecture was implemented in accordance with Clean Architecture principles to ensure modularity and scalability. In a pilot clinical validation of the markerless motion algorithm integrated into Rehabi, 14 post-arthroplasty patients showed moderate agreement for hip flexion (ICC = 0.686) and good agreement for knee flexion (ICC = 0.801). Although the sample was small, the results show a promising trend suggesting that mobile markerless motion capture may be a viable option for remote assessment and monitoring.