Other biomarker modalities
Transcranial sonography shows that enlarged areas of the SN echogenicity is a characteristic and stabile feature in idiopathic PD [46]. Thus, increased echogenicity of the SN seems to serve as a convenient and inexpensive biomarker candidate for the early diagnosis of idiopathic PD and differentiating it from atypical parkinsonian syndromes.
Electroencephalography (EEG) can be used to identify biomarker candidates for PD. Elevated spectral beta power within basal ganglia is implicated in PD, and recent advances in EEG-based analytic approaches to quantify oscillatory beta band synchrony seem promising [59]. Lower background rhythm frequency and increased relative power in delta and theta bands in resting-state EEG hold potential as predictive biomarkers for cognitive deterioration in PD [60]. In a prospective study, marked EEG slowing during rapid eye movement (REM) sleep together with increased relative powers in delta and theta frequencies were predictive biomarkers for patients who later develop dementia [60, 61]. EEG readouts are very well suited for machine and deep learning based techniques that may assist in objective screening and staging of PD as discussed in a recent review [62]. Despite being an inexpensive and easily accessible approach, the predictive and translational value of EEG biomarkers still needs to be confirmed.
Digital biomarkers (or technology-based objective measures) form a rapidly emerging field of research to improve the longitudinal tracking of neurodegenerative diseases in clinical care and research [63]. Digital biomarkers refer to the use of inbuilt sensors in portable (e.g. smartphone), wearable (e.g. smartwatch or ring) or implantable devices allowing for active or passive data collection on biological (e.g. blood glucose), physiological (e.g. heart rate or body temperature) or functional (e.g. motor activities, speech or facial expressions) parameters [64]. By their nature, digital biomarkers are unbiased and offer an opportunity to collect dense datasets during everyday life. Real-world data together with sophisticated artificial intelligence (AI) -based algorithms can detect even slight changes in daily life which could hardly be detected in a clinical setting. This could help predicting PD before the onset of classical symptoms or estimating the course of the disease, thus allowing for more personalized care. The diagnostic utility of digital biomarkers, however, needs further validation.