Wearable sensors are widely used to gather psychophysiological data in the laboratory and real-world applications. However, the accuracy of these devices should be carefully assessed. The study focused on testing the accuracy of the Empatica 4 (E4) wristband for the detection of heart rate variability (HRV) and electrodermal activity (EDA) metrics in stress-inducing conditions and growing-risk driving scenarios. Fourteen healthy subjects were recruited for the experimental campaign, where HRV and EDA were recorded over six experimental conditions (Baseline, Video Clip, Scream, No-Risk Driving, Low-Risk Driving, and High-Risk Driving) and by means of two measurement systems: the E4 device and a gold standard system. The overall quality of the E4 data was investigated; agreement and reliability were assessed by performing a Bland–Altman analysis and by computing the Spearman’s correlation coefficient. HRV time-domain parameters reported high reliability levels in Baseline (r > 0.72), Video Clip (r > 0.71), and No-Risk Driving (r > 0.67), while HRV frequency domain parameters were sufficient in Baseline (r > 0.58), Video Clip (r > 0.59), No-Risk (r > 0.51), and Low-Risk Driving (r > 0.52). As for the EDA parameters, no correlation was found. Further studies could enhance the HRV and EDA quality through further optimizations of the acquisition protocol and improvement of the processing algorithms.

Robot-Assisted Gait Rehabilitation (RAGR) is an established clinical practice to encourage neuroplasticity in patients with neuromotor disorders. Nevertheless, tasks repetition imposed by robots may induce boredom, affecting clinical outcomes. Thus, quantitative assessment of engagement towards rehabilitation using physiological data and subjective evaluations is increasingly becoming vital. This study aimed at methodologically exploring the performance of artificial intelligence (AI) algorithms applied to structured datasets made of heart rate variability (HRV) and electrodermal activity (EDA) features to predict the level of patient engagement during RAGR. The study recruited 46 subjects (38 underage, 10.3 ± 4.0 years old; 8 adults, 43.0 ± 19.0 years old) with neuromotor impairments, who underwent 15 to 20 RAGR sessions with Lokomat. During 2 or 3 of these sessions, ad hoc questionnaires were administered to both patients and therapists to investigate their perception of a patient’s engagement state. Their outcomes were used to build two engagement classification targets: self-perceived and therapist-perceived, both composed of three levels: “Underchallenged”, “Minimally Challenged”, and “Challenged”. Patient’s HRV and EDA physiological signals were processed from raw data collected with the Empatica E4 wristband, and 33 features were extracted from the conditioned signals. Performance outcomes of five different AI classifiers were compared for both classification targets. Nested k-fold cross-validation was used to deal with model selection and optimization. Finally, the effects on classifiers performance of three dataset preparation techniques, such as unimodal or bimodal approach, feature reduction, and data augmentation, were also tested. The study found that combining HRV and EDA features into a comprehensive dataset improved the synergistic representation of engagement compared to unimodal datasets. Additionally, feature reduction did not yield any advantages, while data augmentation consistently enhanced classifiers performance. Support Vector Machine and Extreme Gradient Boosting models were found to be the most effective architectures for predicting self-perceived engagement and therapist-perceived engagement, with a macro-averaged F1 score of 95.6% and 95.4%, respectively. The study displayed the effectiveness of psychophysiology-based AI models in predicting rehabilitation engagement, thus promoting their practical application for personalized care and improved clinical health outcomes.

Spinal orthoses are the most viable conservative treatment for scoliosis, and additive manufacturing techniques have shown huge perspective in producing patient-specific braces, reducing material waste, and production times. This pilot study aimed at determining whether 3D-printed braces could induce advantages or disadvantages compared to conventional braces in terms of mobility and gait, and at quantitatively evaluating the effects of braces on mobility and gait. Ten participants were included in the study, eight with adolescent idiopathic scoliosis and two with osteogenesis imperfecta. Participants were asked to perform Timed-Up and Go (TUG) tests wearing a triaxial accelerometer under three conditions: unbraced, wearing a conventional (i.e., thermoformed) brace, and wearing a 3D-printed brace. After segmenting each TUG test in sub-phases, metrics quantifying gait and mobility were computed, and Friedman tests among all conditions were performed. No significant differences in scoliotic patients mobility and gait between conventional and 3D-printed brace conditions were found, potentially suggesting that 3D-printed braces are as effective as conventional ones. Conversely, Stand flexion amplitude and Sit extension amplitude were lower in both conventional and 3D-printed brace conditions compared to the unbraced, meaning that braces limited the trunk range of motion. As for gait parameters, no significant differences in Walk Cadence and Walk Velocity among the three conditions were found, indicating that braces did not affect gait, at least during TUG tests. The study was registered at Clinicaltrials.gov (Study ID NCT04282408, Date of Registration February 11th, 2020). In summary, these preliminary results may help to depict 3D-printed braces as a viable alternative to conventional braces for scoliotic patients.

Objectives: To examine the emotional, cognitive and dispositional experience of children and adolescents undergoing Lokomat rehabilitation by integrating self-evaluation, therapist observations and physiological metrics across repeated sessions, with the aim of characterising how patient experience evolves throughout paediatric robot-assisted gait training. Design: Prospective observational study using a multidimensional assessment approach combining self-report, therapist ratings and physiological measures. Setting: Inpatients undergoing robot-assisted gait training (RAGT) with the Lokomat at the Scientific Institute Eugenio Medea in Bosisio Parini (Italy). Participants: 42 children and adolescents (N=42; mean age 11.66±5.59 years) undergoing RAGT. Interventions: Robot-assisted gait therapy with the Lokomat. Participants underwent 30-minute therapy sessions as per routine rehabilitation protocols, with treatment durations ranging from 15 to 20 sessions, as prescribed by their referring clinician. Primary and secondary outcome measures: Participants completed ad-hoc questionnaires about emotional, cognitive and dispositional factors before and after therapy; therapists provided structured assessments of patient engagement and psychological states. Physiological data, such as heart rate variability (HRV) and electrodermal activity (EDA), were recorded using wearable sensors to capture physiological correlates of emotional and cognitive engagement. Results: The results showed that by the end of Lokomat therapy, patients displayed increased cognitive engagement and better emotional regulation, along with higher vagal activity (normalised high-frequency) and increased phasic EDA responses. According to the therapists, patients appeared more confident, calm and cooperative. Sympathetic activation observed during satisfaction ratings reflected the involvement of the autonomic nervous system in positive emotional experiences. Conclusions: This study, therefore, emphasises a multidimensional approach to rehabilitation, which involves subjective patient self-assessments, therapist observations and physiological signals in an effort to capture a more comprehensive patient experience. The findings highlight the importance of personalised, patient-centred approaches and contribute new evidence on the psychological and physiological effects of RAGT in paediatric populations. Further research is warranted to confirm these results and explore their clinical implications. Trial registration number: NCT05767268 .

Driving with assistive devices creates complex cognitive and emotional demands that require systematic investigation. This study uses a multivariate approach based on subjective and objective measures to evaluate mental workload (MWL), stress and emotional state during simulated driving with an assistive device. Thirty healthy adults ($42\pm 13$ years of age, 7 females) completed four driving tasks combining two levels of difficulty (Easy vs. Hard) and two steering tools (wheel vs. single-pin aid). Subjective measures from NASA Task Load Index and Self-Assessment Manikin were collected, as well as physiological parameters from electroencephalographic, electrocardiographic, and electrodermal activity signals. The results revealed that the assistive device significantly induced increases in perceived physical demand, frustration, loss of emotional control and stress, yet reducing intrinsic sympathetic response represented by electrodermal activity parameters. Multivariate analyses highlighted that combining different physiological predictors improved MWL estimation. This study marks an initial step towards understanding the impact of assistive devices on MWL and stress in post-acute individuals returning to driving.