Brain tumor segmentation with missing modalities via latent multi-source correlation representation
This addresses a practical problem in clinical settings where missing imaging modalities hinder accurate brain tumor segmentation, offering an incremental improvement over existing methods.
The paper tackles brain tumor segmentation with missing MRI modalities by proposing a novel correlation representation block to discover latent multi-source correlations, resulting in robust segmentation that outperforms state-of-the-art methods on BraTS 2018 datasets.
Multimodal MR images can provide complementary information for accurate brain tumor segmentation. However, it's common to have missing imaging modalities in clinical practice. Since there exists a strong correlation between multi modalities, a novel correlation representation block is proposed to specially discover the latent multi-source correlation. Thanks to the obtained correlation representation, the segmentation becomes more robust in the case of missing modalities. The model parameter estimation module first maps the individual representation produced by each encoder to obtain independent parameters, then, under these parameters, the correlation expression module transforms all the individual representations to form a latent multi-source correlation representation. Finally, the correlation representations across modalities are fused via the attention mechanism into a shared representation to emphasize the most important features for segmentation. We evaluate our model on BraTS 2018 datasets, it outperforms the current state-of-the-art method and produces robust results when one or more modalities are missing.