Fuzzy inference model evaluating turn for Parkinson's disease patients

Area of Science:

• Biomedical Engineering
• Neurology
• Rehabilitation Science

Background:

• Parkinson's disease (PD) significantly impacts motor function, particularly gait and turning.
• The Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) assesses PD severity, but turning assessment lacks clear guidelines and relies on subjective visual observation.
• Current subjective rating methods for turning in PD are prone to variability and uncertainty.

Purpose of the Study:

• To develop an objective and reliable method for quantifying turning impairment in Parkinson's disease patients.
• To propose a fuzzy inference model utilizing biomechanical data from lower limb sensors for turning assessment.
• To address the limitations of subjective visual assessment in the MDS-UPDRS gait evaluation.

Main Methods:

• Extraction of four biomechanical features from inertial sensors worn on the lower limbs (ankles) of 46 PD patients.
• Development of a fuzzy inference model integrating examiner knowledge to compute a turning assessment score.
• Comparison of sensor-derived scores with clinician-based MDS-UPDRS motor examination ratings.

Main Results:

• The computed turning assessment scores demonstrated reasonable consistency with expert clinician opinions.
• The proposed fuzzy model offers objective and reproducible scoring, mitigating the variability inherent in subjective examiner observations.
• The implemented continuous scale avoids the floor/ceiling effects often associated with discrete rating scales.

Conclusions:

• The sensor-based fuzzy inference model provides a more objective and consistent approach to assessing turning impairment in Parkinson's disease.
• This method enhances the quantitative evaluation of PD motor symptoms, offering a valuable complement to existing clinical scales.
• The continuous nature of the scale improves sensitivity and reduces bias in PD motor assessment.