Reliability of 3D Depth Motion Sensors for Capturing Upper Body Motions and Assessing the Quality of Wheelchair Transfers

Abstract

Wheelchair users must use proper technique when performing sitting-pivot-transfers (SPTs) to prevent upper extremity pain and discomfort. Current methods to analyze the quality of SPTs include the TransKinect, a combination of machine learning (ML) models, and the Transfer Assessment Instrument (TAI), to automatically score the quality of a transfer using Microsoft Kinect V2. With the discontinuation of the V2, there is a necessity to determine the compatibility of other commercial sensors. The Intel RealSense D435 and the Microsoft Kinect Azure were compared against the V2 for inter- and intra-sensor reliability. A secondary analysis with the Azure was also performed to analyze its performance with the existing ML models used to predict transfer quality. The intra- and inter-sensor reliability was higher for the Azure and V2 (n = 7; ICC = 0.63 to 0.92) than the RealSense and V2 (n = 30; ICC = 0.13 to 0.7) for four key features. Additionally, the V2 and the Azure both showed high agreement with each other on the ML outcomes but not against a ground truth. Therefore, the ML models may need to be retrained ideally with the Azure, as it was found to be a more reliable and robust sensor for tracking wheelchair transfers in comparison to the V2.

Effect of an interactive Pilates-based telerehabilitation intervention in people with multiple sclerosis: Protocol for a Randomized Controlled Trial. (Preprint)

“…Thus, because specific prescriptions should be tailored on the clinical condition, a dedicated trial should be considered for patients with higher level of disability. Secondly, owing to the discontinuation of the Kinect Sensor V2 adopted in MS-FIT, for the successful translation of the exergame into clinical practice, a new compatible commercial sensor will be identified and the app will be adapted; the Microsoft AzureKinect and MentorAge are valid and comparable systems already taken into account by other researchers (Koontz et al, 2022;Petsani et al, 2022;Tacchino et al, 2023). Thirdly, although the MS-FIT tool has not been developed to pursue rehabilitative purposes, it would be helpful to analyse the MS-FIT intervention as a complement to the work of healthcare professionals and assess if it is able to potentiate the effect of a rehabilitative intervention, eventually also in PwMS with higher level of disability.…”

An Internet- and Kinect-Based Multiple Sclerosis Fitness Intervention Training With Pilates Exercises: Development and Usability Study

“…Moreover, a full-body skeletal tracking library in Python has recently become available. These aspects suggest that the Microsoft AzureKinect may be an adequate replacement within the MS-FIT system [84]. Another alternative solution to the Kinect Sensor V2 is MentorAge, a depth RGB (Red, Green, Blue) image sensor tracking system that operates on the Android system and uses infrared 3D capturing technology.…”