PARF-Net: integrating pixel-wise adaptive receptive fields into hybrid Transformer-CNN network for medical image segmentation

TL;DR

PARF-Net introduces pixel-wise adaptive receptive fields into hybrid Transformer-CNN architectures, enhancing local feature learning and global dependency integration for improved medical image segmentation performance.

Contribution

The paper proposes Conv-PARF, a novel adaptive convolution method, and integrates it into a hybrid Transformer-CNN network to better capture local and global features in medical images.

Findings

PARF-Net achieves 84.27% mean Dice on Synapse dataset.

Outperforms state-of-the-art methods on four medical image datasets.

Effectively captures lesions of varying shapes and scales.

Abstract

Convolutional neural networks (CNNs) excel in local feature extraction while Transformers are superior in processing global semantic information. By leveraging the strengths of both, hybrid Transformer-CNN networks have become the major architectures in medical image segmentation tasks. However, existing hybrid methods still suffer deficient learning of local semantic features due to the fixed receptive fields of convolutions, and also fall short in effectively integrating local and long-range dependencies. To address these issues, we develop a new method PARF-Net to integrate convolutions of Pixel-wise Adaptive Receptive Fields (Conv-PARF) into hybrid Network for medical image segmentation. The Conv-PARF is introduced to cope with inter-pixel semantic differences and dynamically adjust convolutional receptive fields for each pixel, thus providing distinguishable features to disentangle the lesions with varying shapes and scales from the background. The features derived from the Conv-PARF layers are further processed using hybrid Transformer-CNN blocks under a lightweight manner, to effectively capture local and long-range dependencies, thus boosting the segmentation performance. By assessing PARF-Net on four widely used medical image datasets including MoNuSeg, GlaS, DSB2018 and multi-organ Synapse, we showcase the advantages of our method over the state-of-the-arts. For instance, PARF-Net achieves 84.27% mean Dice on the Synapse dataset, surpassing existing methods by a large margin.