Meta-D: Metadata-Aware Architectures for Brain Tumor Analysis and Missing-Modality Segmentation

TL;DR

Meta-D introduces a metadata-aware architecture that leverages scanner metadata to enhance brain tumor analysis and missing-modality segmentation, improving accuracy and robustness in medical imaging tasks.

Contribution

The paper presents Meta-D, a novel architecture that explicitly incorporates scanner metadata to improve feature extraction and handle missing modalities in brain tumor segmentation.

Findings

Up to 2.62% F1-score improvement in 2D tumor detection.

Up to 5.12% Dice score increase in missing-modality segmentation.

Reduces model parameters by 24.1%.

Abstract

We present Meta-D, an architecture that explicitly leverages categorical scanner metadata such as MRI sequence and plane orientation to guide feature extraction for brain tumor analysis. We aim to improve the performance of medical image deep learning pipelines by integrating explicit metadata to stabilize feature representations. We first evaluate this in 2D tumor detection, where injecting sequence (e.g., T1, T2) and plane (e.g., axial) metadata dynamically modulates convolutional features, yielding an absolute increase of up to 2.62% in F1-score over image-only baselines. Because metadata grounds feature extraction when data are available, we hypothesize it can serve as a robust anchor when data are missing. We apply this to 3D missing-modality tumor segmentation. Our Transformer Maximizer utilizes metadata-based cross-attention to isolate and route available modalities, ensuring the network focuses on valid slices. This targeted attention improves brain tumor segmentation Dice scores by up to 5.12% under extreme modality scarcity while reducing model parameters by 24.1%.