EUVIMEDEUVIMED entdecken
Open Access

Scanning the Horizon of Replicability in Neuroscience: A Recipe of Developing Replicable Deep Models for Functional Neuroimages

Jiaqi Ding, Tingting Dan, Ziquan Wei, Paul J Laurienti, Guorong Wu

IEEE Transactions on Biomedical Engineering · 2026-01 · 73 (1) · 281-292

DOI: 10.1109/tbme.2025.3581167 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Biomedical Engineering, 2026-01. Verfasst von Jiaqi Ding und 4 weiteren Autor:innen. Hauptthemen laut MEDLINE: Brain, Deep Learning, Neurosciences, Image Processing, Computer-Assisted. Die Arbeit nennt 48 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIA NIH HHS, National Institutes of Health, Foundation of Hope for Research and Treatment of Mental Illness.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neuroimaging techniques have revolutionized our capacity to understand the neurobiological underpinnings of behavior in-vivo. Leveraging an unprecedented wealth of public neuroimaging data, there is a surging interest to answer novel neuroscience questions using machine learning techniques. Despite the remarkable successes in existing deep models, current state-of-arts have not yet recognized the potential issues of experimental replicability arising from ubiquitous cognitive state changes, which might lead to spurious conclusions and impede generalizability across neuroscience studies. In this work, we first dissect the critical (but often missed) challenge of ensuring prediction replicability in spite of task-irrelevant functional fluctuations. Then, we formulate the solution as a domain adaptation where we devise a cross-attention mechanism with discrepancy loss in a Transformer backbone. We have evaluated the cognitive task recognition accuracy and consistency on multi-run functional neuroimages (successive imaging measurements of the same cognitive task in a short period of time) from Human Connectome Project, where the significantly enhanced replicability and accuracy by our proposed deep model indicate the great potential of addressing real-world neuroscience questions through the lens of reliable deep models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

BrainDeep LearningNeurosciencesImage Processing, Computer-AssistedModels, NeurologicalHumansConnectomeMagnetic Resonance Imaging

Förderung

  • NIA NIH HHS
  • National Institutes of Health
  • Foundation of Hope for Research and Treatment of Mental Illness

Zitieren

APA

Ding, J., Dan, T., Wei, Z., Laurienti, P. J., & Wu, G. (2026). Scanning the Horizon of Replicability in Neuroscience: A Recipe of Developing Replicable Deep Models for Functional Neuroimages. IEEE Transactions on Biomedical Engineering, 73(1), 281-292. https://doi.org/10.1109/tbme.2025.3581167

Vancouver

Ding J, Dan T, Wei Z, Laurienti PJ, Wu G. Scanning the Horizon of Replicability in Neuroscience: A Recipe of Developing Replicable Deep Models for Functional Neuroimages. IEEE Transactions on Biomedical Engineering. 2026;73(1):281-292. doi:10.1109/tbme.2025.3581167

BibTeX
@article{Ding2026,
  author = {Ding, Jiaqi and Dan, Tingting and Wei, Ziquan and Laurienti, Paul J and Wu, Guorong},
  title = {Scanning the Horizon of Replicability in Neuroscience: A Recipe of Developing Replicable Deep Models for Functional Neuroimages},
  journal = {IEEE Transactions on Biomedical Engineering},
  year = {2026},
  volume = {73},
  number = {1},
  pages = {281--292},
  doi = {10.1109/tbme.2025.3581167}
}
Kostenlos starten →