EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Automatic assessment of pc-ASPECTS in MRIs of patients with posterior circulation infarcts

Shun Liu, Manuela Y F Sung, Noam Rotenberg, Argye E. Hillis, Andreia V. Faria

NeuroImage Stroke · 2026-11 · 100014

DOI: 10.1016/j.ynist.2026.100014 ↗

Kurz gefasst

Fachartikel in der Zeitschrift NeuroImage Stroke, 2026-11. Verfasst von Shun Liu und 4 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Institute of Biomedical Imaging and Bioengineering, National Institute on Deafness and Other Communication Disorders.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The Alberta Stroke Program Early CT Score (ASPECTS) is widely used to assess the involvement of key brain regions and, consequently, the severity of ischemic middle cerebral artery (MCA) strokes. Its reliability is lower in the posterior circulation, where small structures and imaging artifacts hinder consistent interpretation. To improve objectivity and efficiency, we extended our automated MCAASPECTS framework to posterior circulation strokes (pc-ASPECTS). We used diffusion-weighted MRI to quantify lesion burden within pc-ASPECTS regions and interpretable machine-learning models to classify regional involvement. The models were trained on 314 MRIs and independently tested on 105 patients with infarcts confined to posterior circulation. Regional classification performance, measured by balanced accuracy, exceeded 0.85 across all regions. The automated system calculated total pcASPECTS with 85.7% accuracy, and 93.3% of estimations were within ±1 point of the reference score. Performance was stable across demographic and lesion-related characteristics. Feature-importance analyses confirmed that model decisions were anatomically meaningful and consistent. The full system has been integrated into the public Open Acute Stroke Detection and Segmentation (OpenADS) platform, enabling objective, accessible, and real-time quantification of posterior circulation ischemic strokes.

Förderung

  • National Institute of Biomedical Imaging and Bioengineering
  • National Institute on Deafness and Other Communication Disorders

Zitieren

APA

Liu, S., Sung, M. Y. F., Rotenberg, N., Hillis, A. E., & Faria, A. V. (2026). Automatic assessment of pc-ASPECTS in MRIs of patients with posterior circulation infarcts. NeuroImage Stroke, 100014. https://doi.org/10.1016/j.ynist.2026.100014

Vancouver

Liu S, Sung MYF, Rotenberg N, Hillis AE, Faria AV. Automatic assessment of pc-ASPECTS in MRIs of patients with posterior circulation infarcts. NeuroImage Stroke. 2026:100014. doi:10.1016/j.ynist.2026.100014

BibTeX
@article{Liu2026,
  author = {Liu, Shun and Sung, Manuela Y F and Rotenberg, Noam and Hillis, Argye E. and Faria, Andreia V.},
  title = {Automatic assessment of pc-ASPECTS in MRIs of patients with posterior circulation infarcts},
  journal = {NeuroImage Stroke},
  year = {2026},
  pages = {100014},
  doi = {10.1016/j.ynist.2026.100014}
}
Kostenlos starten →