Domain-Specialized Interactive Segmentation Framework for Meningioma Radiotherapy Planning

2025-09-30

AI Summary

Automatic segmentation of meningiomas in radiotherapy planning suffers from low accuracy, particularly in handling tumor heterogeneity and ill-defined boundaries. Method: We propose the first clinical radiotherapy-oriented interactive 3D segmentation framework, integrating radiologist prior knowledge with a lightweight deep learning architecture. It supports multiple efficient interaction modalities—including points, bounding boxes, scribbles, and lasso—to enhance robustness and interpretability for complex cases. Contribution/Results: Evaluated on 500 contrast-enhanced T1-weighted MRI scans, our method achieves a Dice score of 77.6% and an IoU of 64.8%, significantly outperforming general-purpose segmentation models (p < 0.01). Designed to align with clinical workflow, the system is fully open-sourced—including code and pre-trained models—providing a deployable AI-assisted tool for precision radiotherapy planning.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Precise delineation of meningiomas is crucial for effective radiotherapy (RT) planning, directly influencing treatment efficacy and preservation of adjacent healthy tissues. While automated deep learning approaches have demonstrated considerable potential, achieving consistently accurate clinical segmentation remains challenging due to tumor heterogeneity. Interactive Medical Image Segmentation (IMIS) addresses this challenge by integrating advanced AI techniques with clinical input. However, generic segmentation tools, despite widespread applicability, often lack the specificity required for clinically critical and disease-specific tasks like meningioma RT planning. To overcome these limitations, we introduce Interactive-MEN-RT, a dedicated IMIS tool specifically developed for clinician-assisted 3D meningioma segmentation in RT workflows. The system incorporates multiple clinically relevant interaction methods, including point annotations, bounding boxes, lasso tools, and scribbles, enhancing usability and clinical precision. In our evaluation involving 500 contrast-enhanced T1-weighted MRI scans from the BraTS 2025 Meningioma RT Segmentation Challenge, Interactive-MEN-RT demonstrated substantial improvement compared to other segmentation methods, achieving Dice similarity coefficients of up to 77.6% and Intersection over Union scores of 64.8%. These results emphasize the need for clinically tailored segmentation solutions in critical applications such as meningioma RT planning. The code is publicly available at: https://github.com/snuh-rad-aicon/Interactive-MEN-RT

Problem

• Develops specialized interactive segmentation for meningioma radiotherapy planning
• Addresses tumor heterogeneity challenges in clinical segmentation accuracy
• Integrates clinician input with AI for precise 3D meningioma delineation

Innovation

• Domain-specialized interactive segmentation for meningioma radiotherapy
• Integrates multiple clinical interaction methods for precision
• Achieves high Dice and IoU scores in evaluation