Glioblastoma Tumor Segmentation using an Ensemble of Vision Transformers

TL;DR

This paper introduces BRAINNET, a novel ensemble of vision transformer models for accurate glioblastoma tumor segmentation in MRI scans, achieving state-of-the-art results on a large public dataset.

Contribution

The study presents a new ensemble pipeline using MaskFormer vision transformers trained on orthogonal MRI slices for improved tumor segmentation accuracy.

Findings

Achieved high Dice coefficients for tumor core, whole tumor, and enhancing tumor.

Outperformed existing methods on the UPenn-GBM dataset.

Demonstrated robustness across different tumor regions.

Abstract

Glioblastoma is one of the most aggressive and deadliest types of brain cancer, with low survival rates compared to other types of cancer. Analysis of Magnetic Resonance Imaging (MRI) scans is one of the most effective methods for the diagnosis and treatment of brain cancers such as glioblastoma. Accurate tumor segmentation in MRI images is often required for treatment planning and risk assessment of treatment methods. Here, we propose a novel pipeline, Brain Radiology Aided by Intelligent Neural NETworks (BRAINNET), which leverages MaskFormer, a vision transformer model, and generates robust tumor segmentation maks. We use an ensemble of nine predictions from three models separately trained on each of the three orthogonal 2D slice directions (axial, sagittal, and coronal) of a 3D brain MRI volume. We train and test our models on the publicly available UPenn-GBM dataset, consisting of 3D multi-parametric MRI (mpMRI) scans from 611 subjects. Using Dice coefficient (DC) and 95% Hausdorff distance (HD) for evaluation, our models achieved state-of-the-art results in segmenting all three different tumor regions -- tumor core (DC = 0.894, HD = 2.308), whole tumor (DC = 0.891, HD = 3.552), and enhancing tumor (DC = 0.812, HD = 1.608).