MFA-Net: Multi-Scale feature fusion attention network for liver tumor segmentation

TL;DR

MFA-Net is a novel attention-based multi-scale feature fusion network that improves liver tumor segmentation accuracy in CT images by effectively integrating features across different scales.

Contribution

The paper introduces MFA-Net, a new multi-scale feature fusion attention network that enhances segmentation accuracy by better integrating multi-scale features.

Findings

MFA-Net outperforms state-of-the-art methods on liver CT datasets.

MFA-Net achieves more precise segmentation across different image scales.

Experimental results validate the effectiveness of the proposed attention mechanism.

Abstract

Segmentation of organs of interest in medical CT images is beneficial for diagnosis of diseases. Though recent methods based on Fully Convolutional Neural Networks (F-CNNs) have shown success in many segmentation tasks, fusing features from images with different scales is still a challenge: (1) Due to the lack of spatial awareness, F-CNNs share the same weights at different spatial locations. (2) F-CNNs can only obtain surrounding information through local receptive fields. To address the above challenge, we propose a new segmentation framework based on attention mechanisms, named MFA-Net (Multi-Scale Feature Fusion Attention Network). The proposed framework can learn more meaningful feature maps among multiple scales and result in more accurate automatic segmentation. We compare our proposed MFA-Net with SOTA methods on two 2D liver CT datasets. The experimental results show that our MFA-Net produces more precise segmentation on images with different scales.