Parkinson’s connectivity retrieval found similarly mixed neighbors for cases and controls
Source check: Source checked. Not independent scientific confirmation.
Finding: Among 40 participants in the TaoWu cohort, the ten closest matches in a learned connectivity representation were 79.5% Parkinson’s cases for diagnosed participants and 81.0% for controls, on average. The source pool included 139 people with Parkinson’s among 169 participants. Retrieval did not consistently improve classification beyond pretraining alone.
Technology: Graph convolutional autoencoder; functional-connectivity retrieval
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- A graph convolutional autoencoder learned functional-connectivity embeddings from 169 external individuals. Neighbor identities and a downstream classifier evaluated a separate 40-person sample.
- Test and result
- A frozen encoder was trained on external cohorts. The researchers summarized matched neighbor labels and compared prediction variants using repeated cross-validation in the separate TaoWu cohort.
- Limitation / still unknown
- The study is small and the source lookup is label-imbalanced. Same-fold checkpoint selection and cohort differences limit interpretation of classification metrics.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://arxiv.org/abs/2610.35107v1 · https://arxiv.org/html/2610.35107v1 · https://arxiv.org/pdf/2610.35107
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 21 Nov 2026. Review scope: v1 cohort Methods, Table 1 and Figure 2 controls; PDF author affiliation.
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.