# MGML: A Plug-and-Play Meta-Guided Multi-Modal Learning Framework for Incomplete Multimodal Brain Tumor Segmentation

Source: [MGML: A Plug-and-Play Meta-Guided Multi-Modal Learning Framework for Incomplete Multimodal Brain Tumor Segmentation](<https://www.sota2.com/research/paper/6955ff8253fdd5c36173dfc0>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-02T05:00:04.167Z

Authors: Yulong Zou, Bo Liu, Cun-Jing Zheng, Yuan-ming Geng, Siyue Li, Qiankun Zuo, Shuihua Wang, Yudong Zhang, Jin Hong

Published: 2025-12-30T01:37:04.000Z

## Abstract

Leveraging multimodal information from Magnetic Resonance Imaging (MRI) plays a vital role in lesion segmentation, especially for brain tumors. However, in clinical practice, multimodal MRI data are often incomplete, making it challenging to fully utilize the available information. Therefore, maximizing the utilization of this incomplete multimodal information presents a crucial research challenge. We present a novel meta-guided multi-modal learning (MGML) framework that comprises two components: meta-parameterized adaptive modality fusion and consistency regularization module. The meta-parameterized adaptive modality fusion (Meta-AMF) enables the model to effectively integrate information from multiple modalities under varying input conditions. By generating adaptive soft-label supervision signals based on the available modalities, Meta-AMF explicitly promotes more coherent multimodal fusion. In addition, the consistency regularization module enhances segmentation performance and implicitly reinforces the robustness and generalization of the overall framework. Notably, our approach does not alter the original model architecture and can be conveniently integrated into the training pipeline for end-to-end model optimization. We conducted extensive experiments on the public BraTS2020 and BraTS2023 datasets. Compared to multiple state-of-the-art methods from previous years, our method achieved superior performance. On BraTS2020, for the average Dice scores across fifteen missing modality combinations, building upon the baseline, our method obtained scores of 87.55, 79.36, and 62.67 for the whole tumor (WT), the tumor core (TC), and the enhancing tumor (ET), respectively. We have made our source code publicly available at https://github.com/worldlikerr/MGML.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Brain Tumor Segmentation (Whole Tumor) on BraTS 2020 (test)](<https://www.sota2.com/research/sota/brain-tumor-segmentation-whole-tumor-on-brats-2020-test.md>): Brain Tumor Segmentation; BraTS Whole Tumor 2020 (test); DSC: 91.21.

- [Enhancing Tumor Segmentation on BraTS ET 2023](<https://www.sota2.com/research/sota/enhancing-tumor-segmentation-on-brats-et-2023.md>): Enhancing Tumor (ET) Segmentation; BraTS ET 2023; DSC (FLAIR): 51.46.

- [Tumor Core (TC) Segmentation on BraTS TC 2023](<https://www.sota2.com/research/sota/tumor-core-tc-segmentation-on-brats-tc-2023.md>): Tumor Core (TC) Segmentation; BraTS TC 2023; DSC (FLAIR): 75.83.

- [Whole Tumor Segmentation on BraTS 2023](<https://www.sota2.com/research/sota/whole-tumor-segmentation-on-brats-2023.md>): Whole Tumor (WT) Segmentation; BraTS WT 2023; DSC (FLAIR): 90.95.

- [Brain Tumor Segmentation on BraTS Tumor Core 2020 (test)](<https://www.sota2.com/research/sota/brain-tumor-segmentation-on-brats-tumor-core-2020-test.md>): Brain Tumor Segmentation; BraTS Tumor Core 2020 (test); DSC: 85.63.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.23936>)

[PDF](<https://arxiv.org/pdf/2512.23936.pdf>)
