Stroke is a major cause of disability and an evident rehabilitation strategy is crucial. Mirror therapy (MT) is one of the popular rehabilitation methods that is known to be effective as the patients benefit from the mirror illusion. However, the patient’s attention to the mirror illusion during treatment is unclear. Therefore, the present study assesses the duration and frequency of the mirror gaze, distraction, and preparation of sixteen stroke patients during two MT methods using a behavioral coding software. During the 30 min treatment, the total mirror gaze duration during conventional bilateral MT (BMT) was 564.04 s, while it was 1482.45 s in unilateral MT using a screen (UMT). The total distracted time was 945.61 s in BMT, while it was only 162.03 s in UMT. The total preparatory duration was 290.35 s in BMT and 155.53 s in UMT. The total number of distracted bouts were 136.45 in BMT, while it was 73.38 in UMT. The total number of preparatory bouts were 18.42 in BMT and 9.56 in UMT. The average times of gaze duration per bout were 5.52 s in BMT and 21.81 s in UMT. The average times of distraction per bout were 9.22 s in BMT and 3.00 s in UMT. The total number of mirror gaze bouts and average time of preparation per bout did not present a statistical significance in the comparisons of the two methods. This study assesses two methods of MT using observational coding software to evaluate the duration and frequency of the mirror gaze during treatment. The results suggest that UMT may be an alternative option to provide MT for stroke patients to increase their attention towards the mirror.

No abstract

(1) Background: The present study investigated the agreement between the Azure Kinect and marker-based motion analysis during functional movements. (2) Methods: Twelve healthy adults participated in this study and performed a total of six different tasks including front view squat, side view squat, forward reach, lateral reach, front view lunge, and side view lunge. Movement data were collected using an Azure Kinect and 12 infrared cameras while the participants performed the movements. The comparability between marker-based motion analysis and Azure Kinect was visualized using Bland–Altman plots and scatter plots. (3) Results: During the front view of squat motions, hip and knee joint angles showed moderate and high level of concurrent validity, respectively. The side view of squat motions showed moderate to good in the visible hip joint angles, whereas hidden hip joint angle showed poor concurrent validity. The knee joint angles showed variation between excellent and moderate concurrent validity depending on the visibility. The forward reach motions showed moderate concurrent validity for both shoulder angles, whereas the lateral reach motions showed excellent concurrent validity. During the front view of lunge motions, both the hip and knee joint angles showed moderate concurrent validity. The side view of lunge motions showed variations in concurrent validity, while the right hip joint angle showed good concurrent validity; the left hip joint showed poor concurrent validity. (4) Conclusions: The overall agreement between the Azure Kinect and marker-based motion analysis system was moderate to good when the body segments were visible to the Azure Kinect, yet the accuracy of tracking hidden body parts is still a concern.

No abstract