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

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wazir, Saad, Kim, Daeyoung
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915357256581120
author Wazir, Saad
Kim, Daeyoung
author_facet Wazir, Saad
Kim, Daeyoung
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2506_18335
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking Decoder Design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention
Wazir, Saad
Kim, Daeyoung
Image and Video Processing
Computer Vision and Pattern Recognition
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
title Rethinking Decoder Design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.18335