Artificial intelligence in gut microbiome analysis for dementia diagnosis: a scoping review
AI-based predictive models identify intestinal microbial patterns associated with Alzheimer's disease, but evidence remains preliminary.
| Population | — |
|---|---|
| Intervention | — |
| Comparator | — |
| Outcome | — |
What the study showed
Twenty-eight studies applying predictive models (random forests, neural networks) to the gut microbiome in dementia were reviewed. Alzheimer's patients showed reduced butyrate-producing bacteria (Butyrivibrio, Eubacterium, Faecalibacterium) and increased Odoribacter splanchnicus, Klebsiella pneumoniae, Bacteroides, and Prevotella. Findings suggest intestinal dysbiosis potentially affecting the gut-brain axis.
How it was done
PRISMA-ScR scoping review searching PubMed, Web of Science, Scopus, and ScienceDirect through July 2025; 896 articles screened by title and abstract, yielding 28 for critical appraisal.
Risk of bias
Scoping reviews do not allow quantitative synthesis or effect size estimation; the 28 included primary studies are heterogeneous in methods, populations, and AI platforms. The abstract does not detail methodological quality or risk of bias of primary studies.
What this study does NOT prove
The review does not prove causality between gut dysbiosis and dementia, nor does it validate any AI model for diagnostic use.
In clinical practice
There is insufficient basis for clinical application of microbial biomarkers in dementia diagnosis. The field remains exploratory.
Limitations
Scoping reviews do not allow quantitative synthesis or effect size estimation; the 28 included primary studies are heterogeneous in methods, populations, and AI platforms. The abstract does not detail methodological quality or risk of bias of primary studies.
Technical appendix
Version history
- 1.0 · 2026-07-05 — Auto-generated under Evidence Standard v1.0
Paid access: structured summary from public metadata; consult the original study at the source.