Abstract

Accurate segmentation of brain tumors in MRI scans is crucial for effective treatment. Multimodal MRI, including FLAIR, T1ce, T1, and T2 images, provides valuable information for tumor delineation. However, challenges such data corruption and varying protocols can lead to missing modalities, affecting segmentation accuracy. Recent methods attempt to transfer knowledge from multimodal teacher models to unimodal student models, but often struggle to capture complex structural relationships within the data. This paper introduces a novel approach called Graph Neural Networkbased Knowledge Distillation (GKD), which uses Graph Neural Networks to create a comprehensive graph connecting intermediate layers of both teacher and student networks. This enables the student network to better learn essential features and their interactions. Evaluations on the BraTS2018 and BraTS2020 datasets demonstrate that GKD significantly enhances brain tumor segmentation accuracy by deepening the connection between teacher and student insights. The code is publicly available at https://github.com/T-Y-Liu/GKD.
Original languageEnglish
Journal2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI 2025)
Publication statusPublished - 2025

Fingerprint

Dive into the research topics of 'Leveraging Graph Neural Networks in Transferring Multimodal Knowledge for Unimodal Segmentation'. Together they form a unique fingerprint.

Cite this