Attention-gated U-Net model for semantic segmentation of brain tumors and feature extraction for survival prognosis

TL;DR

This paper introduces an advanced Attention-Gated R2U-Net model for brain tumor segmentation that improves accuracy and feature extraction for survival prognosis, demonstrating competitive performance on the BraTS2021 dataset.

Contribution

The study proposes a novel triplanar R2U-Net architecture with attention gating for enhanced brain tumor segmentation and survival feature extraction, combining residual, recurrent, and triplanar elements.

Findings

Achieved DSC of 0.900 for whole tumor segmentation.

Extracted 64 features per planar model for survival prediction.

Attained 45.71% accuracy in survival prognosis.

Abstract

Gliomas, among the most common primary brain tumors, vary widely in aggressiveness, prognosis, and histology, making treatment challenging due to complex and time-intensive surgical interventions. This study presents an Attention-Gated Recurrent Residual U-Net (R2U-Net) based Triplanar (2.5D) model for improved brain tumor segmentation. The proposed model enhances feature representation and segmentation accuracy by integrating residual, recurrent, and triplanar architectures while maintaining computational efficiency, potentially aiding in better treatment planning. The proposed method achieves a Dice Similarity Score (DSC) of 0.900 for Whole Tumor (WT) segmentation on the BraTS2021 validation set, demonstrating performance comparable to leading models. Additionally, the triplanar network extracts 64 features per planar model for survival days prediction, which are reduced to 28 using an Artificial Neural Network (ANN). This approach achieves an accuracy of 45.71%, a Mean Squared Error (MSE) of 108,318.128, and a Spearman Rank Correlation Coefficient (SRC) of 0.338 on the test dataset.