EPRA U-Net: An Efficient Pyramid Residual Attention Framework for Accurate Infarct Segmentation in Diffusion-Weighted MRI

2026-07-03

AI Summary

This study addresses the challenge of inaccurate segmentation of acute ischemic cerebral infarction in diffusion-weighted MRI (DWI) by proposing the EPRA U-Net architecture. The model integrates an EfficientNet encoder, residual recurrent blocks (R2 with t=2), atrous spatial pyramid pooling (ASPP), and a dual attention mechanism, optimized with the Tversky loss function. By leveraging an efficient pyramid residual attention mechanism, the approach significantly enhances feature responses for small lesions and models spatial dependencies while reducing parameter count. Evaluated on 4,895 DWI slices from 167 patients, the method achieves Dice scores of 0.8984 at the pixel level and 0.9469 at the sample level, reducing false-negative rates by 16%, 25%, and 29% compared to UNet++, DeepLabV3+, and TransUNet, respectively, thereby substantially improving segmentation accuracy and clinical reliability.

Abstract

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 was adopted, 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. 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.

Problem

• infarct segmentation
• diffusion-weighted MRI
• acute ischemic stroke
• lesion quantification
• clinical decision support

Innovation

• Efficient Pyramid Residual Attention
• Dual Attention Mechanism
• Residual-Recurrent Block
• Atrous Spatial Pyramid Pooling
• Tversky Loss