Deconver: A Deconvolutional Network for Medical Image Segmentation

2025-04-01

AI Summary

To address the limited receptive field of CNNs and the high computational cost of Vision Transformers (ViTs) in medical image segmentation, this paper proposes a lightweight U-shaped network incorporating learnable non-negative deconvolution (NDC). The core contribution is the first design of a differentiable, monotonically convergent NDC layer, which integrates physics-driven deconvolution—originally from image restoration—into end-to-end learning. By eliminating costly attention mechanisms and synergistically combining a U-shaped encoder-decoder architecture with a lightweight feature reconstruction module, the method achieves efficient recovery of high-frequency details and effective artifact suppression. Evaluated on four major benchmarks—ISLES’22, BraTS’23, GlaS, and FIVES—the model achieves state-of-the-art performance in Dice score and Hausdorff distance while reducing FLOPs by up to 90%, significantly improving the accuracy-efficiency trade-off.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Non-convex Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

While convolutional neural networks (CNNs) and vision transformers (ViTs) have advanced medical image segmentation, they face inherent limitations such as local receptive fields in CNNs and high computational complexity in ViTs. This paper introduces Deconver, a novel network that integrates traditional deconvolution techniques from image restoration as a core learnable component within a U-shaped architecture. Deconver replaces computationally expensive attention mechanisms with efficient nonnegative deconvolution (NDC) operations, enabling the restoration of high-frequency details while suppressing artifacts. Key innovations include a backpropagation-friendly NDC layer based on a provably monotonic update rule and a parameter-efficient design. Evaluated across four datasets (ISLES'22, BraTS'23, GlaS, FIVES) covering both 2D and 3D segmentation tasks, Deconver achieves state-of-the-art performance in Dice scores and Hausdorff distance while reducing computational costs (FLOPs) by up to 90% compared to leading baselines. By bridging traditional image restoration with deep learning, this work offers a practical solution for high-precision segmentation in resource-constrained clinical workflows. The project is available at https://github.com/pashtari/deconver.

Problem

• Improves medical image segmentation accuracy and detail restoration
• Reduces computational complexity in segmentation networks
• Enables efficient high-precision segmentation in clinical workflows

Innovation

• Integrates deconvolution in U-shaped architecture
• Replaces attention with efficient nonnegative deconvolution
• Reduces computational costs by 90 percent