Topology-Driven Fusion of nnU-Net and MedNeXt for Accurate Brain Tumor Segmentation on Sub-Saharan Africa Dataset

2026-04-17

AI Summary

This work addresses the challenge of low segmentation accuracy for brain tumors in sub-Saharan Africa, where poor image quality and inconsistent imaging protocols in low-field MRI scans lead to geometric and topological distortions. To tackle this issue, the authors propose a novel architecture that integrates nnU-Net and MedNeXt, augmented with a topology optimization module—introduced here for the first time—to correct topological errors in predictions. The model is pretrained on the BraTS 2025 dataset and subsequently fine-tuned on BraTS-Africa. Experimental results demonstrate substantial improvements in segmentation accuracy, achieving normalized surface distance (NSD) scores of 0.810, 0.829, and 0.895 for the SNFH, NETC, and ET tumor subregions, respectively, thereby effectively mitigating the adverse effects of low-quality MRI data.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Non-convex Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate automatic brain tumor segmentation in Low and Middle-Income (LMIC) countries is challenging due to the lack of defined national imaging protocols, diverse imaging data, extensive use of low-field Magnetic Resonance Imaging (MRI) scanners and limited health-care resources. As part of the Brain Tumor Segmentation (BraTS) Africa 2025 Challenge, we applied topology refinement to the state-of-the-art segmentation models like nnU-Net, MedNeXt, and a combination of both. Since the BraTS-Africa dataset has low MRI image quality, we incorporated the BraTS 2025 challenge data of pre-treatment adult glioma (Task 1) to pre-train the segmentation model and use it to fine-tune on the BraTS-Africa dataset. We added an extra topology refinement module to address the issue of deformation in prediction that arose due to topological error. With the introduction of this module, we achieved a better Normalized Surface Distance (NSD) of 0.810, 0.829, and 0.895 on Surrounding Non-Enhancing FLAIR Hyperintensity (SNFH) , Non-Enhancing Tumor Core (NETC) and Enhancing tumor (ET).

Problem

• brain tumor segmentation
• low-field MRI
• LMIC
• topological error
• BraTS-Africa

Innovation

• topology refinement
• nnU-Net
• MedNeXt
• brain tumor segmentation
• low-field MRI