A Method for Automatic and Objective Scoring of Bradykinesia Using Orientation Sensors and Classification Algorithms

Area of Science:

• Neurology
• Biomedical Engineering
• Movement Disorders

Background:

• Accurate assessment of bradykinesia is crucial for diagnosing and monitoring Parkinson's disease (PD).
• Current evaluation relies on subjective rating scales, such as the Movement Disorders Society-Sponsored Revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS).
• Bradykinesia items in the MDS-UPDRS exhibit low inter-rater reliability, leading to inconsistencies in patient assessment.

Purpose of the Study:

• To develop an objective tool for quantifying bradykinesia in Parkinson's disease patients.
• To evaluate all characteristics of bradykinesia as defined in the MDS-UPDRS.
• To reduce inter-rater variability in bradykinesia assessment.

Main Methods:

• Twenty-five Parkinson's disease patients performed three MDS-UPDRS bradykinesia tasks.
• Patient movements were recorded using a nine degrees-of-freedom sensor and processed with sensor fusion for 3-D representation.
• Support vector machine classifiers were trained using selected features to automatically score movements.

Main Results:

• A set of features was defined and selected to objectively quantify bradykinesia.
• The developed system achieved classification errors significantly smaller than average inter-rater scoring errors for finger tapping, diadochokinesis, and toe tapping.
• Optimal classification performance was achieved using seven selected features.

Conclusions:

• Objective quantification of bradykinesia using sensor technology can significantly reduce assessment inconsistencies.
• This approach has the potential to improve the reliability of Parkinson's disease diagnosis and monitoring.
• Implementing objective scoring may overcome inter-rater disagreements and enhance the management of movement disorders.