{
  "abstract": "Background\n\nFunctional brain time-series classification is constrained by small clinical cohorts. Handcrafted features are sample-efficient but discard temporal detail; end-to-end networks must estimate many temporal parameters from limited labels; and frozen time-series foundation models reduce that estimation burden but typically encode each channel independently, leaving coordinated activity across brain regions unmodeled.Methods\n\nWe introduce the Frozen Temporal-Hypergraph Network (FroHyNet), an asymmetric parallel framework that preserves channel-wise temporal representations from a frozen MantisV2 encoder while a trainable spatio-temporal hypergraph branch learns topology-aware spatial corrections. Bottleneck-mix Top-K (BM-TopK) then residual-injects these corrections only along selected channel and latent-feature axes, treating the spatial stream as a selective update rather than a dense overwrite of the temporal representation. We conduct five-fold cross-validation across two functional near-infrared spectroscopy (fNIRS) and three functional magnetic resonance imaging (fMRI) datasets.Results\n\nFroHyNet achieves the best average accuracy and area under the receiver-operating-characteristic curve (AUC) among methods evaluated on all five datasets, while keeping the trainable-parameter count well below those of from-scratch temporal networks. Gains are concentrated on fNIRS, with more modest improvements on the smaller fMRI cohorts. Branch ablations indicate dataset-dependent complementarity between the two streams, and fusion ablations on HC-D favor BM-TopK over the evaluated dense and recurrent alternatives.",
  "abstract_shown": true,
  "affiliations": [
    "College of Biomedical Engineering, Sichuan University, Chengdu, China.",
    "Mental Health Center, West China Hospital, Sichuan University and Sichuan Clinical Medical Research Center for Mental Disorders, Chengdu, China."
  ],
  "article_type": "research article",
  "arxiv": null,
  "author_total": 7,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Mingyao Gao"
    },
    {
      "affs": [
        0
      ],
      "name": "Ling He"
    },
    {
      "affs": [
        0
      ],
      "name": "Wen Li"
    },
    {
      "affs": [
        1
      ],
      "name": "Yajing Meng"
    },
    {
      "affs": [
        1
      ],
      "name": "Qing Li"
    },
    {
      "affs": [
        1
      ],
      "name": "Changjian Qiu"
    },
    {
      "affs": [
        0
      ],
      "name": "Jing Zhang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/c8qex99176662.bib",
  "cited_by": 0,
  "doi": "10.3389/fmed.2026.1966334",
  "fulltext_url": "https://europepmc.org/articles/PMC13683208",
  "id": "c8qex99176662",
  "issns": [
    "2296-858X"
  ],
  "landing_url": "https://europepmc.org/article/MED/42908832",
  "language": "eng",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "2805|Functional Brain Connectivity Studies"
    ]
  ],
  "openalex": "W7220840897",
  "pdf_url": "https://europepmc.org/articles/PMC13683208?pdf=render",
  "pmcid": "PMC13683208",
  "pmid": "42908832",
  "published": "2026-10-07",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/c8qex99176662.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:20:14Z",
      "native_id": "MED:42908832",
      "source": "europepmc",
      "url": "https://europepmc.org/article/MED/42908832"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "openalex:2805|Functional Brain Connectivity Studies",
      "subfield": "neuroscience"
    }
  ],
  "title": "FroHyNet: frozen temporal priors and trainable hypergraph fusion for brain time-series classification.",
  "title_text": "FroHyNet: frozen temporal priors and trainable hypergraph fusion for brain time-series classification.",
  "type": "peer-reviewed",
  "updated_at": "2026-10-30T12:20:14Z",
  "url": "https://research.draconitic.com/paper/c8qex99176662",
  "venue": "Frontiers in medicine",
  "versions": []
}