Redefining the Down-Sampling Scheme of U-Net for Precision Biomedical Image Segmentation

2026-02-22

AI Summary

This study addresses the loss of long-range contextual information in conventional U-Net architectures caused by aggressive downsampling, which compromises segmentation accuracy in biomedical images. To mitigate this issue, the authors propose Stair Pooling—a novel strategy that replaces standard 2D pooling operations with a cascade of small-sized, narrow pooling steps applied in multiple directions. This design gradually reduces spatial resolution from a factor of 1/4 to 1/2 per downsampling stage and naturally extends to 3D. By integrating transfer entropy to quantitatively optimize the downsampling pathway, the method effectively preserves critical information and enhances spatial detail reconstruction. Evaluated on three biomedical image segmentation benchmarks, the enhanced 2D and 3D U-Net variants achieve an average Dice score improvement of 3.8% over baseline models, demonstrating significant performance gains.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

U-Net architectures have been instrumental in advancing biomedical image segmentation (BIS) but often struggle with capturing long-range information. One reason is the conventional down-sampling techniques that prioritize computational efficiency at the expense of information retention. This paper introduces a simple but effective strategy, we call it Stair Pooling, which moderates the pace of down-sampling and reduces information loss by leveraging a sequence of concatenated small and narrow pooling operations in varied orientations. Specifically, our method modifies the reduction in dimensionality within each 2D pooling step from $\frac{1}{4}$ to $\frac{1}{2}$. This approach can also be adapted for 3D pooling to preserve even more information. Such preservation aids the U-Net in more effectively reconstructing spatial details during the up-sampling phase, thereby enhancing its ability to capture long-range information and improving segmentation accuracy. Extensive experiments on three BIS benchmarks demonstrate that the proposed Stair Pooling can increase both 2D and 3D U-Net performance by an average of 3.8\% in Dice scores. Moreover, we leverage the transfer entropy to select the optimal down-sampling paths and quantitatively show how the proposed Stair Pooling reduces the information loss.

Problem

• biomedical image segmentation
• U-Net
• down-sampling
• information loss
• long-range information

Innovation

• Stair Pooling
• U-Net
• biomedical image segmentation
• information retention
• transfer entropy