A Statistical Approach for Functional Reach-to-Grasp Segmentation Using a Single Inertial Measurement Unit.

Area of Science:

• Biomechanics
• Rehabilitation Engineering
• Wearable Sensors

Background:

• Accurate identification of voluntary movements is crucial for monitoring rehabilitation progress.
• Existing methods for segmenting movement from inertial data often suffer from inaccuracies.
• Home monitoring applications require reliable and automated movement analysis.

Purpose of the Study:

• To present DynAMoS, a novel method for segmenting voluntary movements using a single wrist-worn inertial measurement unit.
• To evaluate the performance of DynAMoS against existing state-of-the-art methods.
• To assess the potential of DynAMoS for home-based rehabilitation monitoring.

Main Methods:

• Inertial data (angular velocity norm) were collected from 25 healthy subjects performing reach-to-grasp movements.
• DynAMoS employs adaptive thresholding and statistics-based post-processing for movement segmentation.
• Segmentation accuracy was validated against a stereophotogrammetric system (gold standard).

Main Results:

• DynAMoS demonstrated superior performance compared to two existing methods.
• The proposed method achieved a low percentage of erroneous movements (3%) and high accuracy (onset/offset mean absolute error < 0.08 s).
• Analysis of movement sub-phases, including drinking, was performed.

Conclusions:

• DynAMoS offers a significant improvement in voluntary movement segmentation from inertial data.
• The method's accuracy and reliability make it suitable for effective home monitoring in rehabilitation.
• DynAMoS can aid in assessing patient motion improvements during domicile rehabilitation protocols.