Artificial Intelligence-Enhanced Couinaud Segmentation for Precision Liver Cancer Therapy

TL;DR

This paper presents LiverFormer, a novel 3D hybrid CNN-Transformer model with data augmentation for precise Couinaud liver segmentation, improving treatment planning and patient outcomes in liver cancer therapy.

Contribution

LiverFormer effectively combines global context and local features for Couinaud segmentation, with a registration-based data augmentation strategy to perform well with limited labeled data.

Findings

High accuracy in liver segmentation on CT images

Strong agreement with expert annotations

Potential to improve surgical and radiation therapy outcomes

Abstract

Precision therapy for liver cancer necessitates accurately delineating liver sub-regions to protect healthy tissue while targeting tumors, which is essential for reducing recurrence and improving survival rates. However, the segmentation of hepatic segments, known as Couinaud segmentation, is challenging due to indistinct sub-region boundaries and the need for extensive annotated datasets. This study introduces LiverFormer, a novel Couinaud segmentation model that effectively integrates global context with low-level local features based on a 3D hybrid CNN-Transformer architecture. Additionally, a registration-based data augmentation strategy is equipped to enhance the segmentation performance with limited labeled data. Evaluated on CT images from 123 patients, LiverFormer demonstrated high accuracy and strong concordance with expert annotations across various metrics, allowing for enhanced treatment planning for surgery and radiation therapy. It has great potential to reduces complications and minimizes potential damages to surrounding tissue, leading to improved outcomes for patients undergoing complex liver cancer treatments.