The identification of cognitive impairment in Parkinson's disease using biofluids, neuroimaging, and artificial...

Parkinson's disease (PD) is a neurodegenerative movement disorder causing severe disability and cognitive impairment as the disease progresses. It is necessary to develop biomarkers for cognitive decline in PD for earlier detection and prediction of disease progression.

We reviewed literature which used artificial intelligence-based techniques, which can be more sensitive than other analyses, to determine potential biomarkers for cognitive impairment in PD.

We found that combining biomarker types, including those from neuroimaging and biofluids, resulted in higher accuracy. Focused analysis on each biomarker type revealed that using structural and functional magnetic resonance imaging (MRI) resulted in accuracy and area under the curve (AUC) values above 80%/0.80, and that beta-amyloid-42 and tau were able to classify PD subjects by cognitive function with accuracy and AUC values above 90%/0.90.

We can conclude that applying both blood-based and imaging-based biomarkers may improve diagnostic accuracy and prediction of cognitive impairment in PD.