A Foundation Model for Brain Lesion Segmentation with Mixture of Modality Experts

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Xinru, Ou, Ni, Basaran, Berke Doga, Visentin, Marco, Qiao, Mengyun, Gu, Renyang, Ouyang, Cheng, Liu, Yaou, Matthew, Paul M., Ye, Chuyang, Bai, Wenjia
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909256346763264
author Zhang, Xinru
Ou, Ni
Basaran, Berke Doga
Visentin, Marco
Qiao, Mengyun
Gu, Renyang
Ouyang, Cheng
Liu, Yaou
Matthew, Paul M.
Ye, Chuyang
Bai, Wenjia
author_facet Zhang, Xinru
Ou, Ni
Basaran, Berke Doga
Visentin, Marco
Qiao, Mengyun
Gu, Renyang
Ouyang, Cheng
Liu, Yaou
Matthew, Paul M.
Ye, Chuyang
Bai, Wenjia
contents Brain lesion segmentation plays an essential role in neurological research and diagnosis. As brain lesions can be caused by various pathological alterations, different types of brain lesions tend to manifest with different characteristics on different imaging modalities. Due to this complexity, brain lesion segmentation methods are often developed in a task-specific manner. A specific segmentation model is developed for a particular lesion type and imaging modality. However, the use of task-specific models requires predetermination of the lesion type and imaging modality, which complicates their deployment in real-world scenarios. In this work, we propose a universal foundation model for 3D brain lesion segmentation, which can automatically segment different types of brain lesions for input data of various imaging modalities. We formulate a novel Mixture of Modality Experts (MoME) framework with multiple expert networks attending to different imaging modalities. A hierarchical gating network combines the expert predictions and fosters expertise collaboration. Furthermore, we introduce a curriculum learning strategy during training to avoid the degeneration of each expert network and preserve their specialization. We evaluated the proposed method on nine brain lesion datasets, encompassing five imaging modalities and eight lesion types. The results show that our model outperforms state-of-the-art universal models and provides promising generalization to unseen datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2405_10246
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Foundation Model for Brain Lesion Segmentation with Mixture of Modality Experts
Zhang, Xinru
Ou, Ni
Basaran, Berke Doga
Visentin, Marco
Qiao, Mengyun
Gu, Renyang
Ouyang, Cheng
Liu, Yaou
Matthew, Paul M.
Ye, Chuyang
Bai, Wenjia
Image and Video Processing
Computer Vision and Pattern Recognition
Brain lesion segmentation plays an essential role in neurological research and diagnosis. As brain lesions can be caused by various pathological alterations, different types of brain lesions tend to manifest with different characteristics on different imaging modalities. Due to this complexity, brain lesion segmentation methods are often developed in a task-specific manner. A specific segmentation model is developed for a particular lesion type and imaging modality. However, the use of task-specific models requires predetermination of the lesion type and imaging modality, which complicates their deployment in real-world scenarios. In this work, we propose a universal foundation model for 3D brain lesion segmentation, which can automatically segment different types of brain lesions for input data of various imaging modalities. We formulate a novel Mixture of Modality Experts (MoME) framework with multiple expert networks attending to different imaging modalities. A hierarchical gating network combines the expert predictions and fosters expertise collaboration. Furthermore, we introduce a curriculum learning strategy during training to avoid the degeneration of each expert network and preserve their specialization. We evaluated the proposed method on nine brain lesion datasets, encompassing five imaging modalities and eight lesion types. The results show that our model outperforms state-of-the-art universal models and provides promising generalization to unseen datasets.
title A Foundation Model for Brain Lesion Segmentation with Mixture of Modality Experts
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.10246