EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

ABSTRACT NUMBER: ESOC2026A1207 AUTOMATED CLASSIFICATION OF INTRACRANIAL HEMORRHAGE SUBTYPE FROM HEAD CT SCANS: A COMPARISON OF TWO DEEP LEARNING APPROACHES

Laurens Jan de Vries, Qinghui Liu, Jostein Gleditsch, Elakkyen Murugesu, Jon André Ottesen, Pål Rønning, Karoline Skogen, Atle Bjørnerud, Bradley MacIntosh

European Stroke Journal · 06.05.2026 · 11 (Supplement_1) · i344-i344

DOI: 10.1093/esj/aakag023.596 ↗

Kurz gefasst

Fachartikel in der Zeitschrift European Stroke Journal, online erschienen am 06.05.2026. Verfasst von Laurens Jan de Vries und 8 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and aims Spontaneous and trauma-related intracranial hemorrhage (ICH) are acute brain injuries that require urgent care. Non-contrast computed tomography (NCCT) is the primary type of imaging to support diagnosis and identify the ICH subtype. Due to the time-sensitivity, automated characterization of ICH subtypes may offer clinical insight and acute diagnostic support. Existing ICH deep learning classifiers focus on binary classification, while our method focuses on multi-label classification.

Methods

RSNA ICH Detection Challenge 2019 CT imaging data (n=22061) were reconstructed and split into n=17649 training, n=2206 validation, and n=2206 test sets. Ground truth ICH subtypes included: epidural (EDH), acute subdural (ASDH), subarachnoid (SAH), intraparenchymal (IPH) and intraventricular (IVH) hemorrhage. Figure 1 illustrates the ICH subtypes (arrows). Two deep learning approaches were trained and tested: 1) 3D-DenseNet, a parameter-efficient variant of a Convolutional Neural Network, 2) DINOv3GRU, which uses a pretrained vision transformer combined with a Gated Recurrent Unit (GRU).

Results

Figure 2 shows the area-under-the-curve (AUCs) classification results over 5 folds of the data. Validation and test results are provided for the mean across all subtypes, and individual ICH subtypes. Mann-Whitney U tests (DenseNet vs DINOv3GRU) revealed a significant AUC difference for the validation (W=0, p=0.008) and test set (W=0, p=0.008).

Conclusions

Two image-based ICH classification approaches were developed with performance that exceeded an AUC of 0.90 for every bleed subtype. Mann-Whitney U tests illustrated significantly higher AUCs for DINOv3GRU. These findings demonstrate the potential of deep learning models for ICH classification, in particular for time-sensitive clinical settings and further management. Conflict of interest All authors have nothing to disclose Figure 1 - belongs to Methods Figure 2 - belongs to Results

Zitieren

APA

Vries, L. J. D., Liu, Q., Gleditsch, J., Murugesu, E., Ottesen, J. A., Rønning, P., Skogen, K., Bjørnerud, A., & MacIntosh, B. (2026). ABSTRACT NUMBER: ESOC2026A1207 AUTOMATED CLASSIFICATION OF INTRACRANIAL HEMORRHAGE SUBTYPE FROM HEAD CT SCANS: A COMPARISON OF TWO DEEP LEARNING APPROACHES. European Stroke Journal, 11(Supplement_1), i344-i344. https://doi.org/10.1093/esj/aakag023.596

Vancouver

Vries LJD, Liu Q, Gleditsch J, Murugesu E, Ottesen JA, Rønning P, et al. ABSTRACT NUMBER: ESOC2026A1207 AUTOMATED CLASSIFICATION OF INTRACRANIAL HEMORRHAGE SUBTYPE FROM HEAD CT SCANS: A COMPARISON OF TWO DEEP LEARNING APPROACHES. European Stroke Journal. 2026;11(Supplement_1):i344-i344. doi:10.1093/esj/aakag023.596

BibTeX
@article{Vries2026,
  author = {Vries, Laurens Jan de and Liu, Qinghui and Gleditsch, Jostein and Murugesu, Elakkyen and Ottesen, Jon André and Rønning, Pål and Skogen, Karoline and Bjørnerud, Atle and MacIntosh, Bradley},
  title = {ABSTRACT NUMBER: ESOC2026A1207 AUTOMATED CLASSIFICATION OF INTRACRANIAL HEMORRHAGE SUBTYPE FROM HEAD CT SCANS: A COMPARISON OF TWO DEEP LEARNING APPROACHES},
  journal = {European Stroke Journal},
  year = {2026},
  volume = {11},
  number = {Supplement_1},
  pages = {i344--i344},
  doi = {10.1093/esj/aakag023.596}
}
Kostenlos starten →