Realistic reproduction of human facial expressions is essential for realistic interactions between humans and humanoid robots. This work presents a data-driven framework for transferring human facial expressions to a humanoid robot and a virtual avatar, aiming to enhance emotional expressiveness and assess its applicability in psychiatric training scenarios. The proposed approach enables cross-domain facial expression mapping while accounting for mechanical constraints of robotic actuation. A user study (n = 40) evaluated emotion recognition across three stimulus categories: human faces (H), unconstrained virtual avatars (A) and humanoid robots with limited facial actuation (R). Participants identified emotions from static images and from dynamic expression sequences, presented with and without speech. Perceived realism and uncanny valley effects were assessed using an eight-item questionnaire rated on a 7-point Likert scale. Results indicate that human-to-robot facial expression transfer is feasible but constrained by mechanical expressivity. Highly expressive emotions such as surprise (H: 87.5%; A: 57.5%; R: 65%) and fear (H: 45%; A: 27.5%; R: 57.5%) achieved moderate recognition rates, whereas subtle emotions such as anger (H: 65%; A: 40%; R: 12.5%) and disgust (H: 60%; A: 10%; R: 22.5%) were poorly recognized on the robot. Dynamic expressions combined with speech significantly improved recognition. These findings demonstrate the feasibility of transferring human facial expressions to humanoid robots while highlighting current limitations of robotic facial actuation. The proposed framework provides a promising basis for emotionally realistic patient simulation and training applications in mental healthcare.

Falls are a leading cause of emergency department (ED) presentations among older adults and frequently signal the onset of functional decline, reduced mobility, and recurrent falls. While evidence-based falls prevention strategies, particularly strength and balance training, can substantially reduce fall risk, secondary prevention is rarely initiated in ED settings. Building on insights from the observational SeFallED study, the iSeFallED study aims to implement an individualized secondary falls prevention program directly within the ED. The intervention integrates comprehensive geriatric assessment, tailored exercise options, wearable sensor-based monitoring, and perturbation-based treadmill training, combined with participatory research methods to ensure patient-centered refinement. iSeFallED is a pragmatic mixed-methods implementation trial enrolling adults aged 60 years and older who present to the ED of the Klinikum Oldenburg or the Evangelisches Krankenhaus Oldenburg following a fall and are discharged without hospital admission. A risk stratification algorithm developed from SeFallED data assigns participants to one of three intervention arms. Individuals classified as having mild risk for functional decline receive educational materials on physical activity and falls prevention. Participants identified as having at least moderate risk may choose between a home-based, tablet-guided strength and balance program or supervised group-based training delivered by local sports partners or at the university center. Optional treadmill perturbation-based balance training is available to all intervention groups. Assessments occur at baseline and at 6, and 12 months, capturing activities of daily living, functional performance, fall risk factors, quality of life, physical activity, and fall incidence. Continuous activity and mobility data are collected through wearable sensors, while focus groups with participants, caregivers, and stakeholders capture qualitative insights. A target sample size of 350 participants will enable comparison with the historical SeFallED sample, with change in activities of daily living serving as the primary outcome. Secondary outcomes include recurrent falls, mobility, and adherence to intervention pathways. The iSeFallED study will provide evidence on the feasibility of initiating secondary falls prevention in the ED and will evaluate its efficacy relative to standard care using a historical control group. By identifying barriers and facilitators to implementation and incorporating machine learning based analysis of wearable sensor data, the study aims to refine secondary falls prevention strategies and offer a scalable model for integration into challenging clinical environments such as the ED. Prospectively registered on 5 March 2025 in the Deutsches Register für Klinische Studien, (DRKS00035322; Date of registration in DRKS: 2025–03 – 05).

Transcranial temporal interference stimulation (tTIS) is a non-invasive method designed to target deep brain regions, such as the basal ganglia, without affecting overlying cortical areas. This study investigated intermittent theta-burst (iTBS) tTIS effects on symptom severity in Parkinson’s disease (PD) and motor learning behavior, a condition associated with – among others – basal ganglia dysfunction. We hypothesized that iTBS-tTIS applied to the right putamen would alleviate PD symptoms and improve motor learning expressed by the contra-lateral hand. This randomized, double-blinded, crossover trial included 19 PD patients (mean age 64 years, 14 males) and 19 age- and sex-matched healthy controls (mean age 68.6 years). Structural MRI data were obtained for each participant, and individualized electric field simulations were performed to predict field strength in the right putamen. The motor part of the Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale (MDS-UPDRS III) served as a primary outcome parameter, an alternating finger tapping task (aFTT) and Motor learning assessed through a sequential finger-tapping tasks (sFTT) were secondary outcome parameters. ITBS-tTIS significantly reduced MDS-UPDRS motor scores in PD patients and the stimulation induced changes in motor performance correlated with the electric field strength in the targeted putamen region. No effect was found for motor performance or motor learning in neither of the groups. These findings indicate that iTBS-tTIS in general holds potential as a non-invasive approach for deep brain stimulation in PD. ### Competing Interest Statement The authors have declared no competing interest. ### Clinical Trial DRKS00030841 ### Funding Statement Deutsche Forschungsgemeinschaft (RTG 2783) ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The Medical Ethics Committee of the University of Oldenburg gave ethical approval for this work. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Following lower limb trauma, performing orthopaedic rehabilitation exercises is a crucial factor in successful recovery. However, many patients find it difficult to execute these movements with optimal biomechanical form. To address this problem, an electromyography-based sensor system combined with machine learning is proposed. This system records and analyses exercises performed with the aim of providing real-time feedback on execution quality. To evaluate the potential of this concept, high-density electromyographic data (32 electrodes per sensor pad) were recorded from the vastus lateralis and vastus medialis muscles while performing rehabilitation exercises. Four different exercises (squat, hip abduction, leg raises and rocking) were performed, each in one optimal and three non-optimal variations, by n = 19 participants, resulting in 3,040 exercise executions and 194,560 electromyographic recordings. Analysis of this dataset demonstrated that a Support Vector Machine algorithm can be used to classify execution quality (four classes per exercise) with an average accuracy of 83.3% (± 8.8%). In addition, it was shown that, One Class Support Vector Machine trained with solely optimal executions, an unknown exercise could be identified as either optimal or non-optimal execution with an accuracy of 76.6% (± 5.9%). These results highlight the potential of this approach to evaluate exercise execution quality during rehabilitation. In the long term, this approach could provide personalised rehabilitation feedback and improve patient outcome.

Background The purpose of preoperative informed consent is to provide patients with comprehensive information about their treatment, including risks and alternatives, to enable informed decision-making. However, studies have shown that patients are often unable to understand or remember important information. Mobile health (mHealth) and augmented reality (AR) apps have been identified as promising solutions to improve patient education and knowledge retention. Objective This study aims to identify the essential requirements for an mHealth app to support informed decision-making for patients with colorectal cancer, with a specific focus on the potential of AR for visualization. This research explores the patient and physician perspectives on these requirements, particularly regarding information delivery and visualization to guide app design. Methods A qualitative focus group study was conducted with groups of mostly patients with colorectal cancer and a physician’s group. Topics related to patient education were discussed, guided by a semistructured interview guide covering personal experience; information content; context of use; and acceptance and presentation of content, which included presenting various visualizations in 2D, 3D, and AR. The interviews were transcribed and analyzed using qualitative content analysis. Results We conducted 4 focus groups with patients (n=23) and 1 focus group with physicians (n=7), for a total of 30 participants. Relevant informational content for the app and its presentation was identified. Patients consistently expressed a desire for personalized, detailed, and visual information about their condition and treatment tailored to their specific case throughout the treatment journey, so they could prepare for the informed consent discussion after diagnosis, prepare for treatment, access guidance and track progress during hospitalization, and access information and resources during recovery after treatment. Patients demonstrated a strong preference for interactive 3D visualizations, while physicians favored simpler 2D images that could be easily integrated into their existing workflow. AR visualizations were seen as a potential tool to provide a general overview of anatomy and surgical approaches but more as a novelty feature and a supplement to more traditional visualizations. Conclusions An ideal patient education app combines comprehensive content with interactive, customizable visualizations like 3D models and AR and should be accessible throughout a patient’s treatment journey. This study highlights the need for a patient-centered design that balances detailed information with ease of understanding and considering different preferences for visualization modalities and levels of detail.

As care for old adults increasingly shifts to the home, integrating everyday health data into clinical practice remains a major challenge. Current solutions based on passive sensors or wearable data often lack contextual understanding, leaving clinicians disconnected from the lived experience at home. To address this gap, we propose a user-centered, natural language-based, asynchronous platform enabling patients to communicate relevant health events and contextual insights in their own words. Developed through a participatory design process, our web-based prototype integrates a large language model with interfaces visualizing health and sensor data, and facilitates communication with hospital care teams. The system supports real-time data sharing, thereby contributing to bridging the hospital-home divide. Early feedback has informed iterative technical refinements, and a real-world user study is in progress. This approach represents an important step toward empowering older adults as active partners in their care and enabling more individualized, responsive clinical decision-making.

Background:Nutritional status is an influential factor for functional status and rehabilitation outcomes in patients undergoing geriatric rehabilitation. Although there is evidence for the potential of eHealth interventions in patients undergoing geriatric rehabilitation in general, the evidence for eHealth interventions with a focus on nutrition is scarce. In other target groups with older people, eHealth applications to support nutrition, such as computer-based food records, have been used successfully. Objective:Therefore, the aim of this study was to verify whether it is feasible for patients undergoing geriatric rehabilitation to independently use a tablet computer-based food record (e-food record) to document their food and beverage intake. The e-food record was developed in advance and tailored to the age- and disease-specific needs of patients undergoing geriatric rehabilitation. Methods:This prospective pilot study investigated the general feasibility of an e-food record in older adults (≥70 y) in a geriatric rehabilitation center in Germany. It was tested whether the e-food record could be independently used by the participants over 3 days. Furthermore, the usability of the e-food record was assessed by the System Usability Scale (0-100 points) after usage. To compare nutritional data, the participants recorded their consumption of food and beverages by the e-food record and by a 24-hour recall for the same time period, and the mean difference was calculated as follows: the value of the 24-hour recall minus the value of the e-food record. As the study was characterized as a pilot, the data analysis was descriptive. Results:Seventeen out of 25 patients (n=6, 35.3% female, mean age 79.5, SD 3.7 y) maintained the e-food record independently over the study period. The mean System Usability Scale score of the e-food record was 76.0 (SD 11.3) points. Datasets of 9 out of 17 participants (n=5, 55.6% female, mean age 78.2, SD 2.9 y) were analyzed in terms of nutritional data. Mean differences in energy, protein, and fluid intake by the 24-hour recall compared to the e-food record were 4.9 (SD 10.2) kcal/kg body weight (bw), 0.1 (SD 0.3) g/kg bw, and 4.9 (SD 9.4) g/kg bw, respectively. Conclusions:The use of an e-food record is generally feasible for patients undergoing geriatric rehabilitation characterized by low technical experience, high mean age, and a high rate of functional impairment. Lower intake levels were observed for the e-food record compared to the 24-hour recall with regard to energy, protein, and fluid intake. Aspects for further development of the e-food record were identified to enable evaluation on a larger sample. Following successful evaluation, the e-food record could be used within nutrition therapy in the future to increase the efficiency of the nutritional therapy process.

We present ClaRO, an unsupervised, cluster-based odometry pipeline for 4D imaging radar to improve robustness and reduce long-term drift for radar-only odometry. Our core idea is to utilize density-based clustering to improve ego-motion estimation using per-cluster Doppler-residuals and find a local best result to seed a robust least squares in addition to finding static points for pose estimation. The resulting global inlier mask is used to refine the pose via a weighted iterative closest point (ICP) approach that fuses Doppler and radar cross-section (RCS) cues while down-weighting the radar's weak elevation axis. The method is fully unsupervised and sensor-agnostic. We compare our method on multiple public datasets (View-of-Delft, HeRCULES, and NTU4DRadLM) and show that our approach provides results that match or exceed recent radar-only odometry baselines (Radar4Motion, EFEAR-4D) and pose graph based 4DRadarSLAM. Our cluster-wise approach significantly reduces long-term drift, achieving a 72.0% improvement in mean absolute trajectory error (ATE) and a 73.2% improvement in absolute rotation error (ARE) compared to state-of-the-art radar odometry methods on the HeRCULES dataset. While remaining competitive in relative pose estimation, our method improves global trajectory consistency and demonstrates robustness across different radar sensors and environments on the VoD and NTU4DRadLM benchmarks.

The assessment tool “READY?” supports care facilities and services in their institutional reflection on the use of robotics for nursing care. The tool contains a digitally supported question catalog completed in an accompanying workshop in the respective institution with the involvement of various stakeholders. The question catalog includes questions from the categories “Care”, “Privacy and legal issues”, “Ethical criteria”, “Technology and infrastructure”, “Institutional and social embeddedness”, and “Economic criteria” and thus pursues a multi-perspective approach. The assessment tool is empirically and theoretically based and was tested in four care facilities and services in the field. Four focus groups and a supplementary survey (n = 32) were conducted during the testing. The results indicate that the assessment tool can assist in initiating an institutional debate on the prerequisites for the possible use of robotics. The focus group participants report that the tool offers an opportunity for professional discussion and the possibility to strengthen collaboration within the institution. Suggestions for improving the assessment tool were collected, including the prospective provision of literature-based recommendations and optimizing the introduction of the workshops.

Human walking can be modeled using a springmass-damping (SMD) system. While most studies have focused on standard activities in healthy young to middle-aged populations, less attention has been given to participants' responses to unexpected gait perturbations. These responses may be valuable predictors of falls, particularly in older adults. Our previous study modeled walking in the vertical direction for a diverse group of participants. Since most gait perturbations occur in the medio-lateral and anteriorposterior directions, the next step involves extending the analysis to these directions and treadmill walking between perturbations.The study included 60 adults (aged 18-87 years), who walked on a perturbation treadmill while wearing an inertial measurement unit (IMU) at the lumbar region to capture body acceleration. Participants first walked at their preferred speed on the treadmill ("normal gait data"), followed by "perturbation trials" with gait perturbations. The gait data between the perturbations were analyzed as "inter-perturbation gait data". Force data was recorded using the treadmill's built-in force plates. The SMD model was applied to calculate damping and stiffness coefficients.The lowest median spring stiffness was found in medio-lateral direction and the highest in anterior-posterior direction. The lowest damping coefficient was found in medio-lateral direction and the highest in vertical direction. Compared to "normal gait data", "inter-perturbation gait data" showed higher stiffness and, for some participants, higher damping coefficients, while others exhibited decreased damping. Damping and stiffness coefficients were successfully extracted from treadmill walking data across all directions for a diverse group of participants and linked to gait dynamics. The analysis highlighted human gait adaptability under various conditions. This study provides groundwork for future research on individual responses to unexpected gait perturbations.Clinical relevance- Describing human walking with damping and stiffness coefficients in different directions could contribute to understand reactive dynamic balance, and thus give a sound estimation of a relevant risk factor for falls in older people.

The interdisciplinary AdaMeKoR project (An Adaptive Multi-Component Robot System for Nursing Care) focused on the development, evaluation, and reflection of a robotic system for nursing care. Conducted from 2020 to 2023, its goal was to create solutions that support both care recipients and professional caregivers. The project investigated the potential for physical relief of caregivers and patients’ increased autonomy, guided by the central question of how good care can be supported, while also considering implementation research and ethical implications. This article provides an overview of the project’s theoretical framework, a description of the developed system, and four exemplary empirical studies. The results include quantitative data on the physical relief potential and user experiences with different control formats, as well as qualitative findings from implementation research and ethical analysis. Further, these results are contextualized with respect to the support of good care. Finally, the article discusses broad implications for the research and implementation of robotic systems in nursing care and outlines key overarching insights. The project demonstrates the clear benefits of interdisciplinary collaboration that incorporates the perspectives of all stakeholders involved in care.

The five time chair rise test (5CRT) is commonly used in geriatric medicine and research to assess functional capacity and lower extremity strength to detect early age-related changes in older adults. Traditional stopwatch-based analyses may mask temporal variations in 5CRT transitions due to averaging. Temporal variations and dynamic characteristics are better assessed by motion variability analysis. This work employs k-means clustering using Dynamic Time Warping (DTW) as a metric for 5CRT to examine compensation mechanisms of older adults. The observational study included 172 healthy, community-dwelling adults aged 70+, yielding 860 chair rises recorded on a force plate and clustered using k-means. Descriptive statistics summarized performance distribution across clusters. Optimal clustering revealed two movement patterns, differing significantly (p $$<0.01$$ ) in 5CRT duration and forces during the stabilization phase. These patterns did not correlate directly with shorter or longer 5CRT durations, indicating overlap and highlighting the limitations of traditional stopwatch methods. This study demonstrates the potential of DTW and k-means clustering in geriatric medicine and research, enabling analysis of 5CRT performance independent of temporal variations, identifying potential health issues undetectable by conventional methods. The k-means model can be further trained to automate analysis, enhancing insights from 5CRT.

Changes in gait are associated with an increased risk of falling and may indicate the presence of movement disorders related to neurological diseases or age-related weakness. Continuous monitoring based on inertial measurement unit (IMU) sensor data can effectively estimate gait parameters that reflect changes in gait dynamics. Monitoring using a waist-level IMU sensor is particularly useful for assessing such data, as it can be conveniently worn as a sensor-integrated belt or observed through a smartphone application. Our work investigates the efficacy of estimating gait events and gait parameters based on data collected from a waist-worn IMU sensor. The results are compared to measurements obtained using a GAITRite® system as reference. We evaluate two machine learning (ML)-based methods. Both ML methods are structured as sequence to sequence (Seq2Seq). The efficacy of both approaches in accurately determining gait events and parameters is assessed using a dataset comprising 17,643 recorded steps from 69 subjects, who performed a total of 3588 walks, each covering approximately 4 m. Results indicate that the Convolutional Neural Network (CNN)-based algorithm outperforms the long short-term memory (LSTM) method, achieving a detection accuracy of 98.94% for heel strikes (HS) and 98.65% for toe-offs (TO), with a mean error (ME) of 0.09 ± 4.69 cm in estimating step lengths.