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

TL;DR

This paper introduces a novel decoder architecture for biomarker segmentation in medical images, combining depth-to-space restoration and residual linear attention to improve feature integration and spatial reconstruction, outperforming current state-of-the-art methods.

Contribution

The paper presents a new decoder design that effectively captures multi-scale context and enhances feature transfer from encoders, leading to superior segmentation accuracy across multiple datasets.

Findings

Achieved up to 4.03% performance improvement over SOTA methods.

Demonstrated effectiveness across four diverse medical imaging datasets.

Validated the decoder's compatibility with various encoder architectures.

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