Automated Versus Traditional Scoring Agreeability During the Balance Error Scoring System

Area of Science:

• Biomechanics
• Clinical assessment
• Human movement analysis

Background:

• The Balance Error Scoring System (BESS) traditionally relies on subjective visual assessment for evaluating postural control.
• Objective and automated methods are needed to enhance the reliability of BESS scoring.

Purpose of the Study:

• To compare the accuracy and reliability of an automated, computer-based BESS scoring system against traditional manual scoring methods.
• To evaluate an instrumented pressure mat system for BESS assessment in healthy individuals.

Main Methods:

• A descriptive cross-sectional study involved 51 healthy participants performing BESS trials on an instrumented pressure mat.
• Balance errors were scored manually from video recordings and compared with automated scoring using center of force data.
• Statistical analyses, including linear mixed models and Bland-Altman analyses, were used to determine discrepancies and limits of agreement.

Main Results:

• Significant differences were found between automated and manual BESS scoring across most conditions (P < .05).
• The largest scoring discrepancies occurred during tandem foam stance, while tandem firm stance showed the smallest differences.
• Bland-Altman analysis revealed wide limits of agreement between the two scoring methods.

Conclusions:

• Automated and manual BESS scoring methods yield significantly different results with substantial variability.
• The choice between automated and manual scoring should consider the specific clinical or research application.
• Further research is warranted to understand the clinical implications of these discrepancies in postural control assessment.