Wavelet based automated postural event detection and activity classification with single imu - biomed 2013

Area of Science:

• Biomedical Engineering
• Gerontology
• Rehabilitation Technology

Background:

• Assessing mobility and activities of daily living (ADLs) is crucial for elderly independence.
• Current assessment methods are often costly, time-consuming, and subjective.
• Wearable sensors offer a potential for objective, low-cost, long-term monitoring.

Purpose of the Study:

• To develop and validate a low-cost, objective system for assessing functional mobility in the elderly.
• To utilize wearable Inertial Measurement Units (IMUs) for monitoring ADLs and postural transitions.
• To compare the accuracy of the IMU system with traditional motion capture systems.

Main Methods:

• Development of a portable wearable system (TEMPO) using MEMS-based IMUs.
• Acquisition of biomechanical data during various ADLs and mobility tasks.
• Implementation of a wavelet denoising algorithm for postural event detection and classification.
• Validation against a motion capture system in a laboratory setting.

Main Results:

• The IMU system achieved accuracy and recognition rates comparable to motion capture systems.
• Wavelet denoising effectively highlighted postural events and transition durations.
• The developed algorithm provided clinical insights into postural control and motor coordination.

Conclusions:

• A single wireless IMU placed at the sternum can accurately assess mobility characteristics.
• The developed system offers a cost-effective, objective method for long-term ambulatory monitoring.
• This technology has significant potential for assessing the condition of the elderly in real-life settings.