{
  "id": 9009923,
  "title": "Tracing high transductive cohort AUC to same-site supervision in a site-aware population GNN for multisite fMRI",
  "url": "https://urgent.news/2026/09/21/tracing-high-transductive-cohort-auc-to-same-site-supervision-in-a",
  "topic": "science",
  "section": "Science",
  "published": "2026-09-21T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.09.15.751719v1?rss=1"
  },
  "original_language": "en",
  "account": "A previously reported high transductive cohort accuracy in multisite autism functional MRI studies has been traced back to site-specific supervision within a site-aware population graph neural network. Researchers examined various information pathways that contributed to a previously observed high transductive cohort accuracy in a multisite autism fMRI study. The cohort out-of-fold accuracy was approximately 0.94, which depended on site-linked edges.\n\nA site-only graph achieved the same high accuracy (0.948 vs. 0.941) as the full model, indicating that the high performance was directly linked to supervision based on site-specific information. The gain in accuracy required same-site supervision, which was crucial for the model's performance. Masking this supervision reduced the accuracy to a significantly lower 0.480. This effect was not explained by the supervision budget itself.\n\nAmong architecture-matched heads, the high accuracy was only observed with the evaluated sex-heterogeneous dual-channel head. In contrast, a canonical topology-only Parisot-GCN did not exhibit the same pattern. Under leave-one-site-out evaluation, the performance fell to near-chance accuracy (0.522 for C-PAC and 0.532 for NIAK). This starkly contrasted with an imaging-only reference, which maintained higher accuracy levels (0.653 for C-PAC and 0.588 for NIAK).\n\nIn summary, the high cohort accuracy in this study was primarily dependent on same-site supervision through the transductive graph under the applied protocol. This relationship did not extend to discrimination across different sites, highlighting the site-specific nature of the observed performance.",
  "summary": "Population-graph models can exploit cohort-level context, which complicates the interpretation of high multisite neuroimaging performance. We asked which information pathways account for a previously reported high transductive cohort AUC in multisite autism fMRI. Under a frozen cohort (canonical ABIDE-I, 871 subjects, 20 sites) and a fixed evaluation protocol, we applied controlled graph,…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}