Back

Diagnostic Utility of Electrophysiological Markers for Early and Differential Diagnosis of Alzheimers, Frontotemporal, and Lewy Body Dementias: A Systematic Review

De Keulenaer, S.; Van Mossevelde, S.; Van den Bossche, T.; Crossiers, D.; Cras, P.; Ellender, T.; Bruffaerts, R.

2024-09-23 neurology
10.1101/2024.09.20.24314042 medRxiv
Show abstract

BackgroundAn early and accurate diagnosis is crucial to provide optimal patient care in neurodegenerative diseases. Although an EEG shows advantages in availability and cost compared to the current diagnostic tools, it is not routinely used in clinical practice. Previous reviews have either focused on single disease populations and/or solely on resting state EEG. To evaluate the utility of EEG for early diagnosis and differential diagnosis, we conducted a systematic review across Alzheimers disease (AD), Frontotemporal Dementia (FTD) and Lewy Body Dementia (DLB). MethodsWe searched databases Pubmed, Cochrane, Web of Science, and Scopus for articles published from 2000 to 2023 investigating resting-state and task-based EEG-markers in biomarker-proven AD, FTD and DLB. ResultsOur search yielded a total of 12010 studies, of which 71 papers were eligible: 34 on AD, 18 on DLB and 9 on FTD. Slowing of the frequency spectrum was a common observation across diseases, achieving excellent sensitivity in AD and DLB. Research on FTD was limited and with varying results in the discrimination from healthy controls, although connectivity analysis and microstates are promising avenues. In differential diagnosis, both spectral and connectivity metrics show encouraging results. Task-based EEG emerges as a promising tool in early AD. ConclusionEEG shows promise as a cost-effective, non-invasive tool for early detection and differential diagnosis. Future research should aim to collect standardized data from multicentric cohorts, across multiple diseases and stages, and explore the neural underpinnings of these diseases, to improve interpretability of the findings.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.