ReFuSeg: Regularized Multi-Modal Fusion for Precise Brain Tumour Segmentation

TL;DR

ReFuSeg introduces a robust multi-modal brain tumor segmentation method that effectively handles missing MRI modalities using a novel regularization approach, improving reliability in real-world clinical scenarios.

Contribution

The paper proposes a novel regularization module for multi-modal brain tumor segmentation that maintains accuracy despite missing imaging modalities, enhancing clinical robustness.

Findings

Effective segmentation with missing modalities

Improved robustness to artifacts and data errors

Maintains high accuracy in real-world scenarios

Abstract

Semantic segmentation of brain tumours is a fundamental task in medical image analysis that can help clinicians in diagnosing the patient and tracking the progression of any malignant entities. Accurate segmentation of brain lesions is essential for medical diagnosis and treatment planning. However, failure to acquire specific MRI imaging modalities can prevent applications from operating in critical situations, raising concerns about their reliability and overall trustworthiness. This paper presents a novel multi-modal approach for brain lesion segmentation that leverages information from four distinct imaging modalities while being robust to real-world scenarios of missing modalities, such as T1, T1c, T2, and FLAIR MRI of brains. Our proposed method can help address the challenges posed by artifacts in medical imagery due to data acquisition errors (such as patient motion) or a reconstruction algorithm's inability to represent the anatomy while ensuring a trade-off in accuracy. Our proposed regularization module makes it robust to these scenarios and ensures the reliability of lesion segmentation.