The AHEPA EEG benchmark: setting the standard for machine learning in dementia diagnosis, a scoping review
Andreas Miltiadous, Aimilia Ntetska, Vasileios Aspiotis, Efthalia Moustakli, Markos G Tsipouras, Alexandros T Tzallas, Nikolaos Giannakeas, Euripidis Glavas, Pantelis Angelidis, Katerina D Tzimourta · 2026
Read The AHEPA EEG benchmark: setting the standard for machine learning in dementia diagnosis, a scoping review by Andreas Miltiadous, Aimilia Ntetska, Vasileios Aspiotis, Efthalia Moustakli, Markos G Tsipouras, Alexandros T Tzallas, Nikolaos Giannakeas, Euripidis Glavas, Pantelis Angelidis, Katerina D Tzimourta (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13184051.txt.
Description
Andreas Miltiadous and nine co-authors audited every machine learning study that used one popular brain-wave dataset for dementia diagnosis, and found the impressive scores mostly rest on shaky testing. The AHEPA recordings are the field's standard benchmark for separating Alzheimer's and frontotemporal dementia from healthy controls. The review sorted forty-six studies into three tiers by how honestly they validated their models.
The pattern was blunt. The stricter the testing, the lower the accuracy: for Alzheimer's, the average fell from about ninety-one percent overall to eighty-two percent in the top tier, and the frontotemporal task fell further. Weak validation alone added seven to ten points of phantom accuracy, explaining over half the spread between papers. Fancy deep models lost their edge once testing was done properly. The authors set baselines and rules they say every future study on this dataset should be judged against.
Catalog record
| CREATOR | Andreas Miltiadous, Aimilia Ntetska, Vasileios Aspiotis, Efthalia Moustakli, Markos G Tsipouras, Alexandros T Tzallas, Nikolaos Giannakeas, Euripidis Glavas, Pantelis Angelidis, Katerina D Tzimourta |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 119 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |