Tractography-Guided Dual-Label Collaborative Learning for Multi-Modal Cranial Nerves Parcellation

2025-08-03

AI Summary

Existing cranial nerve segmentation methods underutilize diffusion MRI (dMRI) information, limiting the effectiveness of multimodal fusion. To address this, we propose a dual-label collaborative learning framework: (1) it incorporates coarse anatomical prior labels generated via atlas-based tractography alongside expert-annotated ground-truth labels to establish complementary supervisory signals; and (2) it introduces a modality-adaptive encoding module (MEM) that dynamically weights structural MRI and dMRI features to mitigate boundary ambiguity. Experiments on the Human Connectome Project (HCP) dataset demonstrate statistically significant improvements in segmentation accuracy over single-label baselines (p < 0.01). These results validate the efficacy and robustness of our dual-label collaboration mechanism and modality-adaptive fusion strategy for challenging cranial nerve segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

The parcellation of Cranial Nerves (CNs) serves as a crucial quantitative methodology for evaluating the morphological characteristics and anatomical pathways of specific CNs. Multi-modal CNs parcellation networks have achieved promising segmentation performance, which combine structural Magnetic Resonance Imaging (MRI) and diffusion MRI. However, insufficient exploration of diffusion MRI information has led to low performance of existing multi-modal fusion. In this work, we propose a tractography-guided Dual-label Collaborative Learning Network (DCLNet) for multi-modal CNs parcellation. The key contribution of our DCLNet is the introduction of coarse labels of CNs obtained from fiber tractography through CN atlas, and collaborative learning with precise labels annotated by experts. Meanwhile, we introduce a Modality-adaptive Encoder Module (MEM) to achieve soft information swapping between structural MRI and diffusion MRI. Extensive experiments conducted on the publicly available Human Connectome Project (HCP) dataset demonstrate performance improvements compared to single-label network. This systematic validation underscores the effectiveness of dual-label strategies in addressing inherent ambiguities in CNs parcellation tasks.

Problem

• Improves multi-modal cranial nerves parcellation using dual-label learning
• Addresses insufficient diffusion MRI information in existing fusion methods
• Reduces ambiguities in CNs parcellation via tractography-guided collaboration

Innovation

• Tractography-guided dual-label collaborative learning
• Modality-adaptive encoder for soft information swapping
• Coarse and precise label fusion for CN parcellation