Digital pathology of the living brain: a voxel-level spatio-temporal network for explainable ADHD diagnosis from raw rs-fMRI across multiple scanner sites - Takeaways - MDSpire

Digital pathology of the living brain: a voxel-level spatio-temporal network for explainable ADHD diagnosis from raw rs-fMRI across multiple scanner sites

  • By

  • Punna Rao Vuyyuru

  • Sathya Babu Korra

  • Srinivas Naik Nenavath

  • June 30, 2026

  • 0 min

Share

  • 1

    ADHD affects approximately 5–7% of children and adolescents globally, with current diagnoses relying on subjective behavioral assessments.

  • 2

    VoxSTNet is a novel framework that processes raw rs-fMRI BOLD volumes, preserving the complete signal while reducing computational burden.

  • 3

    The model achieved an accuracy of 98.7% in five-fold cross-validation and 78.4% under the Leave-One-Site-Out protocol.

  • 4

    HiResCAM saliency maps identified the right caudate nucleus as a key region in ADHD diagnosis, demonstrating voxel-level interpretability.

  • 5

    VoxSTNet aims to improve cross-site generalizability and reduce reliance on specialist preprocessing in resource-limited settings.

Original Source(s)

Digital pathology of the living brain: a voxel-level spatio-temporal network for explainable ADHD diagnosis from raw rs-fMRI across multiple scanner sites

  • by Punna Rao Vuyyuru, Sathya Babu Korra, Srinivas Naik Nenavath

  • June 30, 2026

Related Content

Empowering healthier lifestyles: patterns of digital tool use, self-efficacy, and affective balance in IBD-patients and a community sample

Case Report Describes Possible Betahistine–Suicidality Link

A single case report described suicidal ideation that resolved after betahistine discontinuation and recurred after rechallenge, but the findings cannot establish causality or quantify risk.

Association of MAP2 gene polymorphisms and altered expression with schizophrenia risk in a Chinese Han population