CLET: Computation of Latencies in Event-related potential Triggers using photodiode on virtual reality apparatuses.

To investigate event-related activity in human brain dynamics as measured with EEG, triggers must be incorporated to indicate the onset of events in the experimental protocol. Such triggers allow for the extraction of ERP, i.e., systematic electrophysiological responses to internal or external stimuli that must be extracted from the ongoing oscillatory activity by averaging several trials containing similar events. Due to the technical setup with separate hardware sending and recording triggers, the recorded data commonly involves latency differences between the transmitted and received triggers. The computation of these latencies is critical for shifting the epochs with respect to the triggers sent. Otherwise, timing differences can lead to a misinterpretation of the resulting ERPs. This study presents a methodical approach for the CLET using a photodiode on a non-immersive VR (i.e., LED screen) and an immersive VR (i.e., HMD). Two sets of algorithms are proposed to analyze the photodiode data. The experiment designed for this study involved the synchronization of EEG, EMG, PPG, photodiode sensors, and ten 3D MoCap cameras with a VR presentation platform (Unity). The average latency computed for LED screen data for a set of white and black stimuli was 121.98 ± 8.71 ms and 121.66 ± 8.80 ms, respectively. In contrast, the average latency computed for HMD data for the white and black stimuli sets was 82.80 ± 7.63 ms and 69.82 ± 5.52 ms. The codes for CLET and analysis, along with datasets, tables, and a tutorial video for using the codes, have been made publicly available.

Performance of machine learning models in estimation of ground reaction forces during balance exergaming.

BACKGROUND: Balance training exercise games (exergames) are a promising tool for reducing fall risk in elderly. Exergames can be used for in-home guided exercise, which greatly increases availability and facilitates independence. Providing biofeedback on weight-shifting during in-home balance exercise improves exercise efficiency, but suitable equipment for measuring weight-shifting is lacking. Exergames often use kinematic data as input for game control. Being able to useg such data to estimate weight-shifting would be a great advantage. Machine learning (ML) models have been shown to perform well in weight-shifting estimation in other settings. Therefore, the aim of this study was to investigate the performance of ML models in estimation of weight-shifting during exergaming using kinematic data. METHODS: Twelve healthy older adults (mean age 72 (± 4.2), 10 F) played a custom exergame that required repeated weight-shifts. Full-body 3D motion capture (3DMoCap) data and standard 2D digital video (2D-DV) was recorded. Weight shifting was directly measured by 3D ground reaction forces (GRF) from force plates, and estimated using a linear regression model, a long-short term memory (LSTM) model and a decision tree model (XGBoost). Performance was evaluated using coefficient of determination ([Formula: see text]) and root mean square error (RMSE). RESULTS: Results from estimation of GRF components using 3DMoCap data show a mean (± 1SD) RMSE (% total body weight, BW) of the vertical GRF component ([Formula: see text]) of 4.3 (2.5), 11.1 (4.5), and 11.0 (4.7) for LSTM, XGBoost and LinReg, respectively. Using 2D-DV data, LSTM and XGBoost achieve mean RMSE (± 1SD) in [Formula: see text] estimation of 10.7 (9.0) %BW and 19.8 (6.4) %BW, respectively. [Formula: see text] was [Formula: see text] for the LSTM in the [Formula: see text] component using 3DMoCap data, and [Formula: see text] using 2D-DV data. For XGBoost, [Formula: see text] [Formula: see text] was [Formula: see text] using 3DMoCap data, and [Formula: see text] using 2D-DV data. CONCLUSION: This study demonstrates that an LSTM model can estimate 3-dimensional GRF components using 2D kinematic data extracted from standard 2D digital video cameras. The [Formula: see text] component is estimated more accurately than [Formula: see text] and [Formula: see text] components, especially when using 2D-DV data. Weight-shifting performance during exergaming can thus be extracted using kinematic data only, which can enable effective independent in-home balance exergaming.

Experiences of Stroke Survivors and Clinicians With a Fully Immersive Virtual Reality Treadmill Exergame for Stroke Rehabilitation: A Qualitative Pilot Study.

Use of VR-games is considered a promising treatment approach in stroke rehabilitation. However, there is little knowledge on the use and expectations of patients and health professionals regarding the use of treadmill walking in a fully immersive virtual environment as a rehabilitation tool for gait training for stroke survivors. The objectives of the current study were to determine whether stroke survivors can use fully immersive VR utilizing modern HMDs while walking on a treadmill without adverse effects, and to investigate the experiences of stroke survivors and clinicians after testing with focus on acceptability and potential utilization in rehabilitation. A qualitative research design with semi-structured interviews was used to collect data. Five stroke survivors and five clinicians participated in the study and tested a custom-made VR-game on the treadmill before participating in individual semi-structured interview. Data were analyzed through thematic analysis. The analysis of the interview data identified two main categories: (1) experiencing acceptability through safety and motivation, and (2) implementing fully immersive VR in rehabilitation. Both stroke survivors' and clinicians enjoyed the treadmill-based VR-game and felt safe when using it. The stroke survivors experienced motivation for exercising and achievement by fulfilling tasks during the gaming session as the VR-game was engaging. The clinicians found additional motivation by competing in the game. Both groups saw a potential for use in gait rehabilitation after stroke, on the premise of individual adaptation to each patient's needs, and the technology being easy to use. The findings from this qualitative study suggest that a fully immersive treadmill-based VR-game is acceptable and potentially useful as part of gait rehabilitation after stroke, as it was positively received by both stroke survivors and clinicians working within stroke rehabilitation. The participants reported that they experienced motivation in the game through safety, engagement and achievement. They also saw the potential of implementing such a setup in their own rehabilitation setting. Elements that enable safety and engaging experience are important to maintain when using a fully immersive VR-game in stroke rehabilitation.

Comparison of a Deep Learning-Based Pose Estimation System to Marker-Based and Kinect Systems in Exergaming for Balance Training.

Using standard digital cameras in combination with deep learning (DL) for pose estimation is promising for the in-home and independent use of exercise games (exergames). We need to investigate to what extent such DL-based systems can provide satisfying accuracy on exergame relevant measures. Our study assesses temporal variation (i.e., variability) in body segment lengths, while using a Deep Learning image processing tool (DeepLabCut, DLC) on two-dimensional (2D) video. This variability is then compared with a gold-standard, marker-based three-dimensional Motion Capturing system (3DMoCap, Qualisys AB), and a 3D RGB-depth camera system (Kinect V2, Microsoft Inc). Simultaneous data were collected from all three systems, while participants (N = 12) played a custom balance training exergame. The pose estimation DLC-model is pre-trained on a large-scale dataset (ImageNet) and optimized with context-specific pose annotated images. Wilcoxon's signed-rank test was performed in order to assess the statistical significance of the differences in variability between systems. The results showed that the DLC method performs comparably to the Kinect and, in some segments, even to the 3DMoCap gold standard system with regard to variability. These results are promising for making exergames more accessible and easier to use, thereby increasing their availability for in-home exercise.