![[2602.15067] Attention-gated U-Net model for semantic segmentation of brain tumors and feature extraction for survival prognosis](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.15067] Attention-gated U-Net model for semantic segmentation of brain tumors and feature extraction for survival prognosis
Summary
The article presents an Attention-Gated U-Net model for semantic segmentation of brain tumors, enhancing treatment planning through improved accuracy and feature extraction for survival prognosis.
Why It Matters
This research addresses the challenges in brain tumor treatment by improving segmentation accuracy, which can lead to better surgical planning and patient outcomes. The integration of advanced AI techniques in medical imaging highlights the potential of machine learning in healthcare.
Key Takeaways
- Introduces an advanced Attention-Gated U-Net model for brain tumor segmentation.
- Achieves a Dice Similarity Score of 0.900, indicating high segmentation accuracy.
- Integrates triplanar architecture for improved feature extraction and computational efficiency.
- Predicts survival prognosis with a 45.71% accuracy using extracted features.
- Demonstrates the potential of AI in enhancing medical imaging and treatment planning.
Computer Science > Artificial Intelligence arXiv:2602.15067 (cs) [Submitted on 14 Feb 2026] Title:Attention-gated U-Net model for semantic segmentation of brain tumors and feature extraction for survival prognosis Authors:Rut Pate, Snehal Rajput, Mehul S. Raval, Rupal A. Kapdi, Mohendra Roy View a PDF of the paper titled Attention-gated U-Net model for semantic segmentation of brain tumors and feature extraction for survival prognosis, by Rut Pate and 4 other authors View PDF HTML (experimental) 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 ...