EPRA U-Net: an efficient pyramid residual attention framework for accurate infarct segmentation in diffusion-weighted MRI


Ulutaş H., Sahin M. E., Erkoç M. F., Yüce E., Tuncer T., Dogan S., ...Daha Fazla

BMC Medical Imaging, cilt.26, sa.1, 2026 (SCI-Expanded, Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 26 Sayı: 1
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1186/s12880-026-02569-9
  • Dergi Adı: BMC Medical Imaging
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, EMBASE, MEDLINE, Directory of Open Access Journals, Natural Science Collection (ProQuest), Biological Science Database (ProQuest), Biomedical Reference Collection: Corporate Edition (EBSCO), Health Research Premium Collection (ProQuest), Technology Collection (ProQuest)
  • Anahtar Kelimeler: Deep learning, Infarct segmentation, Medical image analysis, Tversky loss, U-Net, UN SDG 3
  • Yozgat Bozok Üniversitesi Adresli: Evet

Özet

Objective: Accurate identification of acute ischemic infarcts on diffusion-weighted magnetic resonance imaging (DWI) is a critical prerequisite for reliable lesion quantification and effective clinical decision support in the management of cerebrovascular events. Methods: This study presents EPRA U-Net (Efficient Pyramid Residual Attention U-Net), a task-specific integrated architecture for efficient and accurate infarct segmentation of DWI images. In the proposed architecture, an EfficientNet-based encoder was used as a hierarchical feature extractor with a minimized parameterization. In addition, a Residual-Recurrent (R2) block (recurrent unrolling step t = 2, following the original formulation) and Atrous Spatial Pyramid Pooling (ASPP) were integrated to enhance the performance of spatial dependency modeling. Additionally, a dual attention mechanism was incorporated to highlight lesion-related activations while concurrently enabling the suppression of extraneous background responses. To prioritize lesion detection consistent with clinical imperative, a Tversky loss function (α = 0.4, β = 0.6) was adopted, thereby emphasizing the sensitivity of detection over its specificity during the optimization process. Results: Experimental evaluations were conducted utilizing an in-house dataset comprising 167 patients with 4,895 DWI slices; subsequently, the performance of the proposed EPRA U-Net was assessed in comparison with state-of-the-art models, specifically UNet++, DeepLabV3+, and TransUNet. The experimental results suggest that EPRA U-Net attained superior performance, evidenced by a pixel-aggregated Dice of 0.8984, a per-sample Dice of 0.9469, an IoU of 0.8155, a Recall of 0.8887, a Lesion F1 of 0.9378, and an HD95 of 11.62 px (median: 2.00 px). Furthermore, a clear reduction in the rate of missed lesions, specifically by 16%, 25%, and 29%, was observed when compared with UNet++, DeepLabV3+, and TransUNet, respectively. Additionally, a comprehensive ablation study was systematically conducted. Conclusion: These findings suggest that the proposed architecture affords a more robust and accurate infarct segmentation, consequently facilitating a more reliable estimation of treatment eligibility.