Maryam Ghahramani

Background: Early identification of balance impairments is critical for detecting fall risk in older adults. Force plates are the standard for measuring postural sway but are restricted in practice because they are cumbersome and expensive. The Balance Mat is a portable device that requires comprehensive validation against force plates and clinical benchmarks in older adult populations. Objective: To evaluate the technical validity and clinical discriminative ability of the Balance Mat against a laboratory-grade force plate, clinical tests, and falls history in an older adult cohort. Methods: Fifty-six community-dwelling older adults performed static balance assessments across six stance conditions. Postural sway data were recorded simultaneously using the Balance Mat and a force plate. Technical validity was assessed using Spearman’s rank correlation and Intraclass Correlation Coefficients. Linear regression models were applied to calibrate Balance Mat outputs against the force plate. Diagnostic accuracy for classifying fall risk against the Timed Up and Go test, Falls Efficacy Scale-International, and retrospective falls history was evaluated using Area Under the Curve analysis. Results: The Balance Mat demonstrated strong associations with force plate measurements, particularly for sway path and sway velocity (r=0.851). Following calibration, absolute agreement for these parameters reached excellent levels (ICC=0.93), whereas mean sway demonstrated poor agreement and was excluded. For fall-risk classification, the calibrated Balance Mat achieved fair accuracy for retrospective falls history and high Falls Efficacy Scale-International concern (Area Under the Curve 0.74-0.78) and moderate accuracy for Timed Up and Go thresholds (Area Under the Curve 0.70). Conclusion: The calibrated Balance Mat provides valid measurements of postural sway that align with force plate parameters, particularly for sway path and velocity. Given its fair diagnostic accuracy, the device is best utilised as a portable screening tool in combination with standard clinical assessments and falls history rather than as a standalone diagnostic test.

Objective pain assessment remains a significant challenge in clinical and research settings due to the subjective nature of self-reported measures. Physiological signals, particularly electrodermal activity (EDA), have emerged as promising indicators of autonomic responses associated with pain. Although recent advances in deep learning have improved the modelling of complex biosignals, many existing approaches remain computationally demanding, limiting their applicability for real-time monitoring in wearable and embedded systems. This paper proposes a fully convolutional network (FCN) for automated pain recognition using EDA signals. The proposed model is designed to efficiently capture temporal patterns in physiological data while maintaining low computational complexity. The approach is evaluated on the AI4Pain dataset for three-class pain classification (No Pain, Low Pain, High Pain). Experimental results show that the proposed FCN achieves an accuracy of 79.23% in offline evaluation. Furthermore, the model enables real-time inference with a latency of 0.47 ms, achieving 73.14% accuracy during real-time operation. These results demonstrate that convolutional architectures can provide an effective balance between predictive performance and computational efficiency, supporting the development of real-time physiological pain monitoring systems using wearable sensing technologies.

Accurate pain assessment remains challenging, because pain is subjective, and common clinical tools rely on self-report or observation. Functional near-infrared spectroscopy (fNIRS) provides a non-invasive window into cortical haemodynamics and shows promise for objective pain assessment. In this study, we propose a subject-independent framework that models pain states using functional and effective connectivity features derived from fNIRS signals. Using the AI4PAIN dataset (65 participants, 24 channels), we extracted correlation, partial correlation, coherence, and Granger causality matrices from haemodynamic signals (HbO₂, HHb, and HbT), yielding a comprehensive connectivity feature space. A three-class classification task (No Pain vs. Low Pain vs. High Pain) was evaluated using leave-one-subject-out (LOSO) cross-validation. Within each training fold, we standardised features and applied mutual-information-based feature selection; a systematic sweep showed that a reduced feature set of 700 connectivity features achieves performance comparable to the full 1380 features and constitutes the minimum configuration at which High Pain instances begin to be detected (High Pain recall ≈ 50%), whereas smaller feature sets produce near-zero High Pain recall. The proposed framework achieved best accuracy of 69.6%, with HHb achieving the highest accuracy (69.6%), followed by HbT and HbO₂. Analysis of frequently selected features revealed a predominance of directed Granger-causality connections, indicating stable and interpretable connectivity patterns associated with pain processing. These findings demonstrate the feasibility of connectivity-based subject-independent pain assessment and highlight the potential of fNIRS for objective clinical pain monitoring.

The prefrontal cortex (PFC) of the brain is involved in processing visual, vestibular, and somatosensory inputs to stabilise postural balance. However, the PFC's activation map for a standing person and different sensory inputs remains unclear. This study aimed to explore the PFC activity map and distinct haemodynamic responses during postural control when sensory inputs change. To this end, functional near-infrared spectroscopy (fNIRS) was employed to capture the haemodynamic responses throughout the PFC from a group of young adults standing in four sensory conditions. The results revealed distinct PFC activation patterns supporting sensory processing, motor planning, and cognitive control to maintain balance under different degraded sensory conditions. Furthermore, by applying machine learning classifiers and multivariate feature selection, the PFC locations and haemodynamic responses indicative of different sensory conditions were identified. The findings of this study offer valuable insights for optimising rehabilitation approaches, enhancing the design of fNIRS studies, and advancing brain-computer interface technologies for balance assessment and training.

Parkinson’s disease is a progressive neurodegenerative disorder where diagnosis is essential for effective management. However, current diagnostic approaches often lack sensitivity during the early clinical stages of the disease. This study investigates the potential of wearable inertial measurement units combined with machine learning techniques for detecting Parkinson’s disease during mobility tasks. Data were collected from 28 individuals diagnosed with Parkinson’s disease and 34 age-matched healthy controls while performing the Timed Up and Go test and its dual-task variants. A set of gait and mobility features was extracted and refined using two feature selection techniques. Four machine learning models, including Support Vector Machine, Logistic Regression, Linear Discriminant Analysis, and Gradient Boosting, were trained to classify participants. Support Vector Machine and Logistic Regression achieved the highest classification accuracy of 95%. Key discriminative features included turn duration, sit-to-stand time, and stand-to-sit time, which demonstrated potential as objective digital biomarkers for early-stage Parkinson’s disease detection. Additionally, higher motor impairment scores were associated with increased shank range of motion and prolonged turn duration, particularly during dual-task conditions. These findings highlight the potential of wearable sensor-based mobility assessments combined with machine learning as a non-invasive, accessible, and reliable tool for early-stage detection and monitoring of Parkinson’s disease.

Parkinson’s disease (PD) leads to impairments in cortical structures, resulting in motor and cognitive symptoms. Given the connection between brain structure deficits and physical symptoms in PD, assessing objective brain activity and body motion could provide valuable insights for PD assessment and understanding its underlying mechanisms. This study aimed to explore the connection between brain activity and body movement metrics in a group of individuals with PD and an age-matched healthy control (HC) group. The goal was to evaluate the feasibility of using brain and body motion measures for assessing PD. Participants from both groups underwent the Timed Up and Go (TUG) test under three conditions: simple TUG, cognitive dual-task TUG (CDTUG), and motor dual-task TUG (MDTUG). Key findings include: Both groups exhibited similar activation patterns in the prefrontal cortex (PFC) during the simple TUG, with motor performance differences observed in cadence. During CDTUG, both groups showed the highest PFC activation with more pronounced motor impairments, such as higher stride and step time. During MDTUG, the HC group exhibited significantly higher PFC activity compared to the PD group. While both groups had similar patterns of activation in PFC area while TUG and CDTUG, they showed distinct behaviour during MDTUG. These results suggest that motor and cognitive impairments in PD are more pronounced during complex activities. While MDTUG effectively differentiated between PD and HC groups, the findings indicate that both cognitive and motor dual-tasks are essential for comprehensive PD assessment.

Locomotive syndrome (LS) refers to a condition where individuals face challenges in performing activities of daily living. Early detection of such deterioration is crucial to reduce the need for nursing care. The Geriatric Locomotive Function Scale (GLFS-25), a 25-question assessment, has been proposed for categorizing individuals into different stages of LS. However, its subjectivity has prompted interest in technology-based quantitative assessments. In this study, we utilized machine learning and an instrumented five-time sit-to-stand test (FTSTS) to assess LS stages. Younger and older participants were recruited, with older individuals classified into LS stages 0–2 based on their GLFS-25 scores. Equipped with a single inertial measurement unit at the pelvis level, participants performed the FTSTS. Using acceleration data, 144 features were extracted, and seven distinct machine learning models were developed using the features. Remarkably, the multilayer perceptron (MLP) model demonstrated superior performance. Following data augmentation and principal component analysis (PCA), the MLP+PCA model achieved an accuracy of 0.9, a precision of 0.92, a recall of 0.9, and an F1 score of 0.91. This underscores the efficacy of the approach for LS assessment. This study lays the foundation for the future development of a remote LS assessment system using commonplace devices like smartphones.