CAMS: Convolution and Attention-Free Mamba-Based Cardiac Image Segmentation

Convolutional Neural Networks (CNNs) and Transformer-based self-attention models have become the standard for medical image segmentation. This paper demonstrates that convolution and self-attention, while widely used, are not the only effective methods for segmentation. Breaking with convention, we...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings / IEEE Workshop on Applications of Computer Vision S. 1893 - 1903
Hauptverfasser: Khan, Abbas, Asad, Muhammad, Benning, Martin, Roney, Caroline, Slabaugh, Gregory
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 26.02.2025
Schlagworte:
cardiac imaging
Computational modeling
Computer vision
Convolution
convolution-attention-free
Convolutional neural networks
Data mining
Feature extraction
Image segmentation
mamba based segmentation
semantic segmentation
Source coding
Transformers
ISSN:2642-9381
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Convolutional Neural Networks (CNNs) and Transformer-based self-attention models have become the standard for medical image segmentation. This paper demonstrates that convolution and self-attention, while widely used, are not the only effective methods for segmentation. Breaking with convention, we present a Convolution and self-Attention-free Mamba-based seman-tic Segmentation Network named CAMS-Net. Specifically, we design Mamba-based Channel Aggregator and Spatial Aggregator, which are applied independently in each encoder-decoder stage. The Channel Aggregator extracts information across different channels, and the Spatial Ag-gregator learns features across different spatial locations. We also propose a Linearly Interconnected Factorized Mamba (LIFM) block to reduce the computational complexity of a Mamba block and to enhance its decision function by introducing a non-linearity between two factor-ized Mamba blocks. Our model outperforms the existing state-of-the-art CNN, self-attention, and Mamba-based methods on CMR and M&Ms-2 Cardiac segmentation datasets, showing how this innovative, convolution, and self-attention-free method can inspire further research beyond CNN and Transformer paradigms, achieving linear complexity and reducing the number of parameters. Source code and pre-trained models are available at: https://github.com/kabbas570/CAMS-Net.
ISSN:2642-9381
DOI:10.1109/WACV61041.2025.00191