Stroke commonly leads to long-term gait impairments, underscoring the need for objective and continuous functional assessment during rehabilitation. This study employs machine learning methods to assess a smart insole-based system in stroke gait recognition. Data were collected from stroke survivors and healthy control participants during Walk and Timed-Up-and-Go tasks. After preprocessing, group differences were quantified using Hedges' g, and multiple machine learning models were applied to classify two participant groups. Support Vector Machine and KNN achieved the best performance, with accuracies of 0.88. The results demonstrate that sensor-based gait features can be used to distinguish stroke gait patterns from control gait patterns, highlighting the potential of this approach for future homebased monitoring and personalised rehabilitation.

Patients falling from their beds is still one of the major complications of the hospital’s treatments. To address this issue, this research investigates the use of social robots for the identification of potential bed-related falls in hospitals. Using a robot camera and human pose estimation techniques, the patient’s position in the bed is extracted, and a threshold-based algorithm is used to identify any anomalies that could indicate a fall. Due to the absence of publicly available datasets, a synthetic dataset was created using a simulation environment to develop and tune the detection algorithm. A user study was conducted to validate the proposed approach and evaluate people’s perception of the robot. The system achieved an accuracy of 90.9% in the controlled setting using real data. Participants rated the robot as significantly more trustworthy and behaviorally aware when it detects a possible fall, suggesting that timely and meaningful assistance improved the perceived social competence of the robot. These findings highlight the feasibility of deploying social robots as monitoring systems in sensitive clinical settings, offering a cost-effective and socially acceptable solution.

Social robots have gained significant importance in Human-Robot Interaction, particularly in domains requiring personalized support, such as education, healthcare, and entertainment. The rise of e-sports has created a demand for effective coaching systems that can provide tailored guidance to players, paving the way for the integration of social robots as e-coaches. This pilot study explores the role of Furhat, a social robot, as an e-coach in a football video game. Thanks to computer vision techniques, Furhat analyzes the human player’s performance in real-time and provides adaptive feedback tailored to individual gameplay styles. The study investigates the effectiveness of the robot in providing both technical guidance and emotional support. Forty participants, divided into casual and hardcore gamers, engaged with Furhat in a controlled experimental setting. Results revealed that casual gamers sought general guidance and linguistic clarity, while hardcore gamers prioritized context-relevant, well-timed feedback. Although robot gender had a minimal overall impact, a statistically significant interaction was observed between robot gender and gamer type on adaptability perception (p = 0.046). Performance data showed that 85% of participants either maintained or improved their gameplay, with a majority reporting positive comfort and engagement levels during robotic interaction.

Recent years have witnessed the increasing literature on using smart insoles in health and well-being, and yet, their capability of daily living activity recognition has not been reviewed. This paper addressed this need and provided a systematic review of smart insole-based systems in the recognition of Activities of Daily Living (ADLs). The review followed the PRISMA guidelines, assessing the sensing elements used, the participants involved, the activities recognised, and the algorithms employed. The findings demonstrate the feasibility of using smart insoles for recognising ADLs, showing their high performance in recognising ambulation and physical activities involving the lower body, ranging from 70% to 99.8% of Accuracy, with 13 studies over 95%. The preferred solutions have been those including machine learning. A lack of existing publicly available datasets has been identified, and the majority of the studies were conducted in controlled environments. Furthermore, no studies assessed the impact of different sampling frequencies during data collection, and a trade-off between comfort and performance has been identified between the solutions. In conclusion, real-life applications were investigated showing the benefits of smart insoles over other solutions and placing more emphasis on the capabilities of smart insoles.

Abstract Health monitoring, rehabilitation, and fitness are just a few domains where human activity recognition can be applied. In this study, a deep learning approach has been proposed to recognise ambulation and fitness activities from data collected by five participants using smart insoles. Smart insoles, consisting of pressure and inertial sensors, allowed for seamless data collection while minimising user discomfort, laying the baseline for the development of a monitoring and/or rehabilitation system for everyday life. The key objective has been to enhance the deep learning model performance through several techniques, including data segmentation with overlapping technique (2 s with 50% overlap), signal down-sampling by averaging contiguous samples, and a cost-sensitive re-weighting strategy for the loss function for handling the imbalanced dataset. The proposed solution achieved an Accuracy and F1-Score of 98.56% and 98.57%, respectively. The Sitting activities obtained the highest degree of recognition, closely followed by the Spinning Bike class, but fitness activities were recognised at a higher rate than ambulation activities. A comparative analysis was carried out both to determine the impact that pre-processing had on the proposed core architecture and to compare the proposed solution with existing state-of-the-art solutions. The results, in addition to demonstrating how deep learning solutions outperformed those of shallow machine learning, showed that in our solution the use of data pre-processing increased performance by about 2%, optimising the handling of the imbalanced dataset and allowing a relatively simple network to outperform more complex networks, reducing the computational impact required for such applications.

Determining whether a subject has a gait impairment due to a disease or to the loss of muscularity due to advancing age is fundamental for an early diagnosis of musculoskeletal diseases. Parkinson’s is the second most common neurodegenerative disease. The disease’s most prevalent symptom is slow movement or sluggish gait, which can adversely impact the individual’s quality of life. Generally, the gait analysis is carried out on long test sessions, which include for example long periods of walking, that cause inconvenience when the subjects under test have marked gait impairments. To help the diagnosis of Parkinson’s disease, in this study we investigated the classification of Parkinson’s disease by analysing only a few seconds of walking data using smart insoles, statistical analysis and machine learning techniques. The data from the smart insoles was assessed using correlation analysis. By creating pressure groups and analysing their values, it was found that the number of sensors could be reduced from 16 to 7. Furthermore, a feature vector representing the subject’s gait was created by applying on the data a time windowing segmentation of 5 seconds and extracting six statistical features (mean, variance, skewness, kurtosis, energy and entropy). Four different models have been compared in terms of classification performance, reaching an F1-Score in the classification of patients with Parkinson’s against healthy subjects, considering adult and elderly subjects as two separate classes, of 97.04% using the Random Forest. Such metric increased to 98.89%, using the K-Nearest Neighbours when healthy subjects were considered as a single class. The models’ performance for each experiment was determined to be statistically equivalent, demonstrating the potential of this approach to provide the groundwork for the rapid detection of Parkinson’s disease. Although the performance obtained is promising the number of subjects included in the study was fairly low, with a high bias towards the number of healthy subjects. Hence, in future work, the proposed solution will be tested on a larger cohort to ascertain its robustness.