Home › Research › Brain scans and imaging
Predictive EEG Signatures of Alzheimer's Disease: A Cross-Validated, Cross-Method Analysis Using ANCOVA Filtering and Sparse Logistic Regression
Joshi G.. Research Square, 2026-10-20
Topic: Brain imaging, connectivity and assessment
Plain-language summary
Alzheimer’s Disease (AD) changes how the brain’s electrical activity behaves, and these changes can, in principle, be picked up directly from a resting-state EEG recording without any invasive testing. This study asks a much narrower question: does a small subset of four EEG-derived measurement families, namely signal complexity, spectral power, functional connectivity and network topology, reliably separate individuals with AD from Healthy Controls (HC), and does that separation hold up under strict, leakage-free testing? Using a public dataset of 36 AD and 29 HC subjects, candidate features were selected through permutation-based ANCOVA with false discovery rate correction, controlling for age and gender, and each selected feature was then re-checked using a complementary method (Mann-Whitney U with FDR correction, plus network-based statistics for connectivity), applied to the same subjects, to confirm the finding wasn’t an artifact of the first method’s particular assumptions. Agreement between the two methods was strong for signal complexity (Jaccard = 0.85) and spectral power (Jaccard = 0.87), essentially perfect for two of the four network measures (Jaccard = 1.00), and weaker for connectivity (Jaccard = 0.60, assessed at the level of frequency bands). A sparse logistic regression classifier was then trained using 5-fold cross-validation in which feature selection was independently repeated inside each fold, a design specifically intended to stop information from test subjects leaking into which features get chosen. This classifier reached 78.5% ± 5.8% accuracy at discriminating AD from HC (sensitivity 0.803 ± 0.116, specificity 0.753 ± 0.100, AUC 0.808 ± 0.134). A single spectral power feature was among the two features retained in three of the five folds, more than any other individual feature. Taken together, the results point to a real but modest EEG signature of AD in this dataset, identified using a validation design built specifically so that reported accuracy reflects genuine generalization rather than an artifact of how the features were selected.
Cite this study APA, Vancouver, BibTeX · download .bib or .ris
APA 7th
Joshi, G. (2026). Predictive EEG Signatures of Alzheimer's Disease: A Cross-Validated, Cross-Method Analysis Using ANCOVA Filtering and Sparse Logistic Regression. Research Square. https://doi.org/10.21203/rs.3.rs-11390599/v1
Vancouver
Joshi G. Predictive EEG Signatures of Alzheimer's Disease: A Cross-Validated, Cross-Method Analysis Using ANCOVA Filtering and Sparse Logistic Regression. Research Square. 2026. doi:10.21203/rs.3.rs-11390599/v1.
BibTeX
@article{joshi2026-495ba8c6,
author = {Joshi, G.},
title = {{Predictive EEG Signatures of Alzheimer's Disease: A Cross-Validated, Cross-Method Analysis Using ANCOVA Filtering and Sparse Logistic Regression}},
journal = {Research Square},
year = {2026},
doi = {10.21203/rs.3.rs-11390599/v1},
url = {https://doi.org/10.21203/rs.3.rs-11390599/v1}
}
Built from the record alzhub holds for this paper, not looked up. Volume, issue and page numbers are not held in the record. Check the citation against the original before you use it.
Key findings
- Taken together, the results point to a real but modest EEG signature of AD in this dataset, identified using a validation design built specifically so that reported accuracy reflects genuine generalization rather than an artifact of how the features were selected.
New to this subject? Start with our plain-English guide: Brain scans & tests →
Study details
- Effect size78.5%; 5.8%; AUC 0.808
- Evidence strengthLow
How this study scored 16/100
Preprint — not yet peer-reviewed. It scored 26 on the same rubric, but non-peer-reviewed sources are capped at 16, so it shows as 16. Scored zero for study design, sample size, studied in people and randomized / blinded because the record does not state it. We have no data on times cited, citations per year, impact for its field and year, cited by studies here, still being cited, senior author's track record, institution's track record and multicenter collaboration, so those were left out: it earned 16 of the 62 points we could judge it on, which is 26 out of 100.
| Component | Points | Why |
|---|---|---|
| Journal tier | 2 / 10 | Tier 4 — not a peer-reviewed journal (Research Square) |
| Journal's citation record | 0 / 4 | Not a journal article, so the journal's record does not count |
| Study design | 0 / 12 | No study design recorded — scores zero, not guessed (not stated in the record — scored 0) |
| Strength of evidence | 1 / 5 | Rated “low” in our summary |
| Peer-review validation | 4 / 10 | Preprint — not yet peer-reviewed. Hosted on a preprint server. |
| How recent | 5 / 5 | 2026 — within 1 year(s) of the newest study here |
| Sample size | 0 / 4 | Participant count not stated — scores zero (not stated in the record — scored 0) |
| Effect reported | 3 / 3 | Effect reported with a statistic (interval, p-value, ratio or similar) |
| Studied in people | 0 / 3 | Not stated whether this was studied in people — scores zero (not stated in the record — scored 0) |
| Randomized / blinded | 0 / 2 | No randomization or blinding stated — scores zero (not stated in the record — scored 0) |
| Times cited | — / 7 | No citation count (not found in OpenAlex) — left out of this score |
| Citations per year | — / 5 | No citation count or no year — left out of this score |
| Impact for its field and year | — / 6 | No field-weighted citation impact (OpenAlex has none yet) — left out of this score |
| Cited by studies here | — / 6 | Not yet checked against this library's citation graph — left out of this score |
| Still being cited | — / 4 | No yearly citation counts — left out of this score |
| Senior author's track record | — / 4 | Senior author's record not available — left out of this score |
| Institution's track record | — / 3 | No institution listed in OpenAlex — left out of this score |
| Multicenter collaboration | — / 3 | Institutions not listed — left out of this score |
| Indexed in PubMed | 1 / 2 | Not in PubMed, but has a DOI |
| Free to read | 0 / 2 | Behind a paywall — scores zero |
| Points | 16 / 62 | Scored on 12 of 20 parts. We have no data for the rest, so they are left out: 16 of the 62 points we could judge it on is 26 out of 100. |
| Total | 26 / 100 | Capped to 16 because this is not a peer-reviewed article. |
What each of these 20 components measures, and what this score is not →
Read the original
Free copy unknown: no free-copy check has an answer for this record yet. That does not mean it is paywalled.
Related
Studies in the library that share something with this one. The line above each title says what.
Related studies in Brain scans and imaging from nearby years
Also tagged “convolutional”
205 studies in the library carry the topic tag “convolutional”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 205 tagged studies.