Large-kernel Attention for Efficient and Robust Brain Lesion Segmentation

Liam Chalcroft; Ruben Lourenço Pereira; Mikael Brudfors; Andrew S. Kayser; Mark D'Esposito; Cathy J. Price; Ioannis Pappas; John Ashburner

Large-kernel Attention for Efficient and Robust Brain Lesion Segmentation

Image and Video Processing 2023-08-15 v1 Computer Vision and Pattern Recognition

Authors: Liam Chalcroft , Ruben Lourenço Pereira , Mikael Brudfors , Andrew S. Kayser , Mark D'Esposito , Cathy J. Price , Ioannis Pappas , John Ashburner

View on arXiv ↗ PDF ↗

Abstract

Vision transformers are effective deep learning models for vision tasks, including medical image segmentation. However, they lack efficiency and translational invariance, unlike convolutional neural networks (CNNs). To model long-range interactions in 3D brain lesion segmentation, we propose an all-convolutional transformer block variant of the U-Net architecture. We demonstrate that our model provides the greatest compromise in three factors: performance competitive with the state-of-the-art; parameter efficiency of a CNN; and the favourable inductive biases of a transformer. Our public implementation is available at https://github.com/liamchalcroft/MDUNet .

Keywords

image segmentation deep learning segmentation neural networks for signal processing

Cite

@article{arxiv.2308.07251,
  title  = {Large-kernel Attention for Efficient and Robust Brain Lesion Segmentation},
  author = {Liam Chalcroft and Ruben Lourenço Pereira and Mikael Brudfors and Andrew S. Kayser and Mark D'Esposito and Cathy J. Price and Ioannis Pappas and John Ashburner},
  journal= {arXiv preprint arXiv:2308.07251},
  year   = {2023}
}

Large-kernel Attention for Efficient and Robust Brain Lesion Segmentation

Abstract

Keywords

Cite

Related papers