Back

Classifying Alzheimers Disease and Dementia Patients Using Non-invasive EEG Biomarkers

Hassan, W.; Khan, S.; Sohrabpour, A.

2024-10-04 neurology
10.1101/2024.10.03.24314841 medRxiv
Show abstract

Researchers are currently exploring methods to detect early-stage Alzheimers Disease (AD) and other forms of dementia such as frontotemporal dementia (FTD), especially through non-invasive biomarkers, i.e., measurements that reflect biological processes. This paper utilizes a dataset of electroencephalogram (EEG) recordings, a noninvasive biomarker, to distinguish individuals with Alzheimers or frontotemporal dementia from healthy control subjects. This paper explores the usage of machine learning methods to more accurately predict the cognitive status of patients from these non-invasive EEG Biomarkers. We found that AD patients could be easily separated from healthy controls based on their EEG features using simple linear classifiers with an accuracy of 77% and the AD, FTD, and healthy controls with an accuracy of around 57% (randomly selecting the right class is about one third or 33% in the 3-way classification).

Matching journals

The top 7 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.