MHITNet: a minimize network with a hierarchical context-attentional filter for segmenting medical ct images

TL;DR

MHITNet is a novel hierarchical transformer network that enhances medical CT image segmentation by capturing multi-scale features and global context, outperforming existing methods on multiple datasets.

Contribution

The paper introduces MHITNet, combining multi-scale, hierarchical context, and transformer modules with skip-connections for improved CT image segmentation, especially with limited data.

Findings

MHITNet outperforms current best practices on three datasets.

The multi-scale module captures deeper semantic information.

Hierarchical context attention effectively reweights pixels for better segmentation.

Abstract

In the field of medical CT image processing, convolutional neural networks (CNNs) have been the dominant technique.Encoder-decoder CNNs utilise locality for efficiency, but they cannot simulate distant pixel interactions properly.Recent research indicates that self-attention or transformer layers can be stacked to efficiently learn long-range dependencies.By constructing and processing picture patches as embeddings, transformers have been applied to computer vision applications. However, transformer-based architectures lack global semantic information interaction and require a large-scale training dataset, making it challenging to train with small data samples. In order to solve these challenges, we present a hierarchical contextattention transformer network (MHITNet) that combines the multi-scale, transformer, and hierarchical context extraction modules in skip-connections. The multi-scale module captures deeper CT semantic information, enabling transformers to encode feature maps of tokenized picture patches from various CNN stages as input attention sequences more effectively. The hierarchical context attention module augments global data and reweights pixels to capture semantic context.Extensive trials on three datasets show that the proposed MHITNet beats current best practises