# DynTransNet: Dynamic Transformer Network with multi-scale attention for liver cancer segmentation

## TL;DR

This paper introduces a new AI model for accurately identifying liver cancer regions in medical scans, which could improve treatment planning and reduce manual work.

## Contribution

A novel U-shaped segmentation framework with dynamic multi-head self-attention and feature refinement modules for liver tumor segmentation.

## Key findings

- The model achieved a mean Dice score of 86.12 on the ATLAS dataset.
- It achieved a mean Dice score of 93.12 on the LiTS dataset.

## Abstract

Hepatocellular carcinoma (HCC), a predominant subtype of liver cancer, remains Q7 a major contributor to global cancer mortality. Accurate delineation of liver tumors in CT and MRI scans is critical for treatment planning and clinical decision-making. However, manual segmentation is time-consuming, errorprone, and inconsistent, necessitating reliable automated approaches.

This study presents a novel U-shaped segmentation framework inspired by U-Net, designed to enhance accuracy and robustness. The encoder incorporates Dynamic Multi-Head Self-Attention (D-MSA) to capture both global and local spatial dependencies, while the decoder uses skip connections to preserve spatial detail. Additionally, a Feature Mix Module (FM-M) blends multiscale features, and a Residual Module (RM) refines feature representations and stabilizes training. The proposed framework addresses key challenges such as boundary precision, complex structural relationships, and dataset imbalance.

Experimental results demonstrate superior segmentation performance, achieving a mean Dice score of 86.12 on the ATLAS dataset and 93.12 on the LiTS dataset.

The proposed method offers a robust, efficient tool for liver tumor segmentation and holds strong potential to streamline diagnostic workflows and improve automated medical image analysis in clinical practice.