Accelerometric signals in automatic balance assessment.

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Wearable Sensor Technology

Background:

• Patients with limited mobility require continuous monitoring for effective diagnosis and rehabilitation.
• Existing balance assessment methods can be subjective and difficult to implement in home environments.
• Objective, automated systems are needed to support remote patient care and rehabilitation progress tracking.

Purpose of the Study:

• To develop and evaluate an automatic computer-aided balance assessment system for patients with limited mobility.
• To utilize accelerometric data from inertial sensors for objective balance evaluation.
• To enhance the diagnosis and rehabilitation process for disabled individuals in a home setting.

Main Methods:

• The system integrates seven Berg Balance Scale activities.
• Accelerometric signals from inertial sensors are analyzed using specific medium-frequency features.
• Feature vector reduction is performed using multilevel Fisher's linear discriminant.
• Classification of balance activities is achieved using a multilayer perceptron artificial neural network.

Main Results:

• The developed system demonstrates direct assessment effectiveness ranging from 75% to 94% across different activities.
• The computer-aided approach provides objective and quantifiable balance metrics.
• The system successfully supports and monitors the rehabilitation process for patients with limited mobility.

Conclusions:

• The automatic balance assessment system offers a viable solution for remote patient monitoring and rehabilitation.
• The use of wearable inertial sensors and advanced signal processing provides accurate balance assessment.
• This technology has the potential to significantly improve care for individuals with mobility impairments in home environments.