Adaptable Segmentation Pipeline for Diverse Brain Tumors with Radiomic-guided Subtyping and Lesion-Wise Model Ensemble

2025-12-16

AI Summary

This study addresses the challenge of generalizable segmentation of multicenter, heterogeneous brain tumors—including pediatric/adult primary gliomas, meningiomas, post-radiation meningiomas, and pre-/post-treatment brain metastases—on multiparametric MRI. Methodologically, we propose a radiomics-driven, subtype-aware segmentation framework: (1) introducing radiomics-guided tumor subtype identification and adaptive training; (2) constructing a lesion-level dynamic weighted ensemble of heterogeneous architectures (e.g., nnUNet, TransBTS); and (3) designing a lesion-specific post-processing mechanism informed by localized performance metrics. Our approach breaks reliance on fixed network architectures, substantially improving cross-disease and cross-cohort robustness. Evaluated on the BraTS 2025 Lighthouse benchmark across four clinical challenges—PED (pediatric), MEN (meningioma), MEN-RT (post-radiation meningioma), and MET (metastasis)—our method achieves state-of-the-art performance, enabling reliable clinical quantitative analysis and prognostic assessment.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Segmentation
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Robust and generalizable segmentation of brain tumors on multi-parametric magnetic resonance imaging (MRI) remains difficult because tumor types differ widely. The BraTS 2025 Lighthouse Challenge benchmarks segmentation methods on diverse high-quality datasets of adult and pediatric tumors: multi-consortium international pediatric brain tumor segmentation (PED), preoperative meningioma tumor segmentation (MEN), meningioma radiotherapy segmentation (MEN-RT), and segmentation of pre- and post-treatment brain metastases (MET). We present a flexible, modular, and adaptable pipeline that improves segmentation performance by selecting and combining state-of-the-art models and applying tumor- and lesion-specific processing before and after training. Radiomic features extracted from MRI help detect tumor subtype, ensuring a more balanced training. Custom lesion-level performance metrics determine the influence of each model in the ensemble and optimize post-processing that further refines the predictions, enabling the workflow to tailor every step to each case. On the BraTS testing sets, our pipeline achieved performance comparable to top-ranked algorithms across multiple challenges. These findings confirm that custom lesion-aware processing and model selection yield robust segmentations yet without locking the method to a specific network architecture. Our method has the potential for quantitative tumor measurement in clinical practice, supporting diagnosis and prognosis.

Problem

• Adaptable pipeline for diverse brain tumor segmentation
• Radiomic-guided subtyping to ensure balanced training
• Lesion-wise model ensemble for robust predictions

Innovation

• Modular pipeline combining state-of-the-art segmentation models
• Radiomic features guide tumor subtype detection for balanced training
• Lesion-wise ensemble and post-processing tailored to each case