EEG may serve as a biomarker in Huntington's disease using machine learning automatic classification.

Area of Science:

• Neuroscience
• Biomarker Discovery
• Medical Technology

Background:

• Huntington's disease (HD) lacks reliable biomarkers for tracking progression, hindering therapeutic development.
• Quantitative electroencephalography (qEEG) offers a potential method to detect subcortical dysfunction in HD.
• Early detection and monitoring of HD progression are crucial for guiding effective therapies.

Purpose of the Study:

• To develop an automatic classifier using qEEG to differentiate HD gene carriers from healthy controls.
• To identify qEEG features that correlate with established clinical markers of HD progression.
• To explore the potential of qEEG as a non-invasive biomarker for HD.

Main Methods:

• A pilot study involving 26 HD gene carriers and 25 healthy controls.
• Resting-state electroencephalography (EEG) recordings analyzed using quantitative electroencephalography (qEEG).
• Development of a classification index based on statistical pattern recognition of EEG features, validated with 10-fold cross-validation.

Main Results:

• The qEEG-based classifier achieved 83% specificity, 83% sensitivity, and 83% accuracy in distinguishing HD gene carriers from controls.
• The area under the receiver operating characteristic curve was 0.9, indicating strong classification performance.
• Specific qEEG features showed highly significant correlations with key clinical markers of HD progression.

Conclusions:

• qEEG demonstrates potential as a reliable biomarker for Huntington's disease.
• qEEG-derived indices correlating with disease progression could facilitate the development of tools for monitoring therapeutic efficacy.
• Further research is warranted to validate qEEG as a clinical tool for HD management.