NeuroTS-Net improves brain tumor segmentation in pediatric MRI scans
NeuroTS-Net: Multi-Class Semantic Segmentation of Pediatric Brain Tumors in Multi-Modal MRI
Computer Vision and Pattern RecognitionMachine Learning
Summary
Brain tumors in children are hard to outline accurately because their affected areas are small and not easy to see clearly in MRI images. The authors developed NeuroTS-Net, a special AI system that can automatically find and label different parts of brain tumors in 3D MRI scans from children. NeuroTS-Net keeps important details while understanding the bigger picture, helping it perform better than other popular methods. This tool could help doctors diagnose and plan treatments more accurately and quickly.
What this means in practice
- •For medical imaging teams: Automate detailed segmentation of pediatric brain tumors in MRI to support clinical diagnosis and treatment planning.
- •For healthcare ai developers: Integrate NeuroTS-Net as a component in tools for assessing tumor response and guiding pediatric brain cancer therapies.
Authors
Darius Peteleaza, Razvan-Gabriel Dumitru, Bogdan Neamtu, Arpad Gellert, Mariana Sandu, Claudiu Matei
Abstract
Pediatric brain tumors are a leading cause of cancer-related mortality in children, and their small, rare, and often low-contrast subregions make accurate manual delineation challenging. Reliable automated segmentation is therefore needed to support diagnosis, treatment planning, and response assessment. Accordingly, we introduce NeuroTS-Net, a three-dimensional encoder-decoder convolutional neural network architecture for multi-class semantic segmentation that incorporates a dual-scale raw-detail stream, adaptive low-resolution context selection, and detail-preserving multipath downsampling. These components preserve fine intensity and boundary information while efficiently modeling broader tumor context. NeuroTS-Net was trained on the BraTS 2026 pediatric dataset without external data or pretrained weights and evaluated against nnU-Net and MedNeXt under the same experimental protocol. NeuroTS-Net outperformed the baseline methods, achieving whole-tumor and tumor-core Dice scores of 0.938 and 0.937 on the internal validation set and 0.927 and 0.926 on the official challenge validation set. The code is open-sourced at: https://github.com/maenstru56/NeuroTS.