Rethinking Decoder Design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention

2025-06-23

AI Summary

To address insufficient multi-scale feature propagation between encoder and decoder and low decoding efficiency in biomedical image biomarker segmentation, this paper proposes a novel decoder architecture compatible with both Transformer- and CNN-based encoders. Key contributions include: (1) a Depth-to-Space upsampling mechanism that enhances feature reconstruction fidelity; (2) a residual linear attention module that jointly strengthens discriminative feature representation across channel and spatial dimensions; and (3) a multi-scale local–global context fusion strategy to improve robustness under limited training samples. Evaluated on four benchmark datasets—MoNuSeg, DSB, electron microscopy, and TNBC—our method consistently outperforms existing state-of-the-art approaches, achieving average Dice score improvements of 2.76–4.03 percentage points.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Deep Generative Models & Autoencoders
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Segmenting biomarkers in medical images is crucial for various biotech applications. Despite advances, Transformer and CNN based methods often struggle with variations in staining and morphology, limiting feature extraction. In medical image segmentation, where datasets often have limited sample availability, recent state-of-the-art (SOTA) methods achieve higher accuracy by leveraging pre-trained encoders, whereas end-to-end methods tend to underperform. This is due to challenges in effectively transferring rich multiscale features from encoders to decoders, as well as limitations in decoder efficiency. To address these issues, we propose an architecture that captures multi-scale local and global contextual information and a novel decoder design, which effectively integrates features from the encoder, emphasizes important channels and regions, and reconstructs spatial dimensions to enhance segmentation accuracy. Our method, compatible with various encoders, outperforms SOTA methods, as demonstrated by experiments on four datasets and ablation studies. Specifically, our method achieves absolute performance gains of 2.76% on MoNuSeg, 3.12% on DSB, 2.87% on Electron Microscopy, and 4.03% on TNBC datasets compared to existing SOTA methods. Code: https://github.com/saadwazir/MCADS-Decoder

Problem

• Improving biomarker segmentation in medical images
• Addressing limitations in decoder feature transfer efficiency
• Enhancing accuracy with multi-scale contextual information integration

Innovation

• Depth-to-space restoration enhances spatial reconstruction
• Residual linear attention improves feature integration
• Multi-scale context capture boosts segmentation accuracy