Instrumented finger-to-nose test classification in children with ataxia or developmental coordination disorder and...

Area of Science:

• Neurology
• Biomedical Engineering
• Pediatrics

Background:

• Childhood coordination is affected by physiological development and pathological conditions like early onset ataxia (EOA) and developmental coordination disorder (DCD).
• Current diagnoses rely on clinical specialists but lack objective reproducibility due to the absence of a gold phenotypic standard.

Purpose of the Study:

• To determine if quantitative analysis of the finger-to-nose (FNT) test can differentiate between physiological and pathological coordination issues.
• To assess the utility of inertial measurement units (IMUs) and random forest classification for analyzing FNT performance.

Main Methods:

• Participants performed the FNT, while IMUs captured upper limb movement trajectories.
• Random forests were used to classify participants into control, EOA, or DCD groups based on movement data.

Main Results:

• The model achieved 87.4% accuracy for controls, 74.4% for EOA, and 24.8% for DCD.
• Classification accuracy was good for controls and EOA, but significantly lower for DCD patients.

Conclusions:

• Quantitative FNT analysis shows potential as a reliable tool for supporting the assessment of early onset ataxia.
• The FNT's effectiveness in reflecting distinct DCD symptoms remains uncertain due to classification limitations.