Unveiling Incomplete Modality Brain Tumor Segmentation: Leveraging Masked Predicted Auto-Encoder and Divergence Learning
This addresses the problem of reduced segmentation accuracy due to missing modalities in clinical MRI for brain tumor analysis, representing an incremental advance.
The paper tackles brain tumor segmentation with incomplete MRI modalities by proposing masked predicted pre-training and knowledge distillation with Holder pseudo-divergence, achieving significant performance improvements on BRATS2018 and BRATS2020 datasets.
Brain tumor segmentation remains a significant challenge, particularly in the context of multi-modal magnetic resonance imaging (MRI) where missing modality images are common in clinical settings, leading to reduced segmentation accuracy. To address this issue, we propose a novel strategy, which is called masked predicted pre-training, enabling robust feature learning from incomplete modality data. Additionally, in the fine-tuning phase, we utilize a knowledge distillation technique to align features between complete and missing modality data, simultaneously enhancing model robustness. Notably, we leverage the Holder pseudo-divergence instead of the KLD for distillation loss, offering improve mathematical interpretability and properties. Extensive experiments on the BRATS2018 and BRATS2020 datasets demonstrate significant performance enhancements compared to existing state-of-the-art methods.