Machine learning on Parkinson's disease? Let's translate into clinical practice.

Area of Science:

• Neuroimaging
• Machine Learning
• Neurological Disorders

Background:

• Clinical neuroimaging traditionally focuses on anatomical changes in neurological disorders.
• Machine learning (ML) represents a shift towards predictive analysis in this field.

Purpose of the Study:

• To evaluate the potential of multivariate machine learning approaches in classifying neurological disorders.
• To assess the predictive capabilities of ML using neuroimaging data for unseen patient cohorts.

Main Methods:

• Application of machine learning techniques to clinical neuroimaging data.
• Multivariate analysis of neuroimaging abnormalities for classification.

Main Results:

• Machine learning enables a predictive approach in neuroimaging, moving beyond anatomical descriptions.
• The study explores the potential of ML in classifying previously unseen clinical cohorts.

Conclusions:

• Machine learning is poised to significantly impact the clinical practice of Parkinson's disease.
• Important barriers must be addressed before widespread adoption of ML in clinical neuroimaging for Parkinson's disease.