Multi-modal Liver Segmentation and Fibrosis Staging Using Real-world MRI Images

2025-09-30

AI Summary

This study addresses the clinical need for automatic liver segmentation (LiSeg) and non-invasive liver fibrosis staging (LiFS) in real-world, multicenter, multimodal, multiphase MRI. Method: We propose an integrated framework combining co-registered pseudo-label guidance, a deep segmentation network, and STAD (Shape, Texture, Appearance, Direction) radiomic features extracted from segmentation masks. To mitigate annotation scarcity, we employ multimodal image co-registration and weakly supervised pseudo-label generation. The segmentation model is designed for high robustness, enabling end-to-end joint LiSeg-LiFS modeling. Contribution/Results: Our method achieves state-of-the-art performance across all subtasks in the CARE 2025 Challenge. It generalizes effectively to diverse MRI sequences using only limited annotated data, demonstrating strong reproducibility and translational potential for clinical deployment.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Liver fibrosis represents the accumulation of excessive extracellular matrix caused by sustained hepatic injury. It disrupts normal lobular architecture and function, increasing the chances of cirrhosis and liver failure. Precise staging of fibrosis for early diagnosis and intervention is often invasive, which carries risks and complications. To address this challenge, recent advances in artificial intelligence-based liver segmentation and fibrosis staging offer a non-invasive alternative. As a result, the CARE 2025 Challenge aimed for automated methods to quantify and analyse liver fibrosis in real-world scenarios, using multi-centre, multi-modal, and multi-phase MRI data. This challenge included tasks of precise liver segmentation (LiSeg) and fibrosis staging (LiFS). In this study, we developed an automated pipeline for both tasks across all the provided MRI modalities. This pipeline integrates pseudo-labelling based on multi-modal co-registration, liver segmentation using deep neural networks, and liver fibrosis staging based on shape, textural, appearance, and directional (STAD) features derived from segmentation masks and MRI images. By solely using the released data with limited annotations, our proposed pipeline demonstrated excellent generalisability for all MRI modalities, achieving top-tier performance across all competition subtasks. This approach provides a rapid and reproducible framework for quantitative MRI-based liver fibrosis assessment, supporting early diagnosis and clinical decision-making. Code is available at https://github.com/YangForever/care2025_liver_biodreamer.

Problem

• Developing non-invasive liver fibrosis staging using multi-modal MRI images
• Creating automated liver segmentation from limited annotated MRI data
• Providing reproducible quantitative assessment for early fibrosis diagnosis

Innovation

• Multi-modal co-registration for pseudo-labelling generation
• Deep neural networks for automated liver segmentation
• STAD features from masks and MRI for fibrosis staging