Human guided glioma segmentation improves surgical accuracy and safety
Uncertainty-Guided Handshake: Efficient Human-in-the-Loop Refinement for Surgical-Grade Glioma Segmentation
Computer Vision and Pattern Recognition
Summary
Medical imaging models can miss important details in brain tumor scans, which is risky for surgery. The authors developed a system where a computer works with humans to find and fix these errors more efficiently. Their method uses uncertainty measurements to highlight areas needing attention, reducing the workload on clinicians. Tests show it improves the accuracy of tumor boundaries, helping to make brain surgery safer.
What this means in practice
- •For radiology departments: Support clinicians in refining automated brain tumor segmentations to improve surgical planning accuracy while reducing manual workload.
- •For medical ai developers: Incorporate uncertainty-driven feedback to build safer, more reliable interactive segmentation tools for neuro-oncology imaging.
Authors
Samuel Hart, Ahmad Yahya, Ahmed Karam Eldaly
Abstract
While state-of-the-art automated models for medical image segmentation achieve high mean performance, they frequently suffer from localized, catastrophic failures that preclude safe clinical deployment, particularly in neuro-oncology. Interactive segmentation frameworks mitigate this by incorporating human oversight, but traditionally impose prohibitive cognitive and temporal workloads by requiring clinicians to manually search for errors. In this project, we present an efficient, Hybrid Structural-Aleatoric Human-in-the-Loop framework for glioma segmentation that bridges the gap between automated baseline performance and surgical-grade precision, achieving sub-2.0 mm HD95 on curated benchmarks while providing safety-net routing for structural failures across real-world clinical data. By extracting voxel-wise Test-Time Augmentation (TTA) uncertainty and applying hierarchical topological filtering, our method proactively isolates high-risk structural anomalies. We comprehensively evaluated our approach on a challenging out-of-distribution clinical stress-test cohort (N = 362). Operating under a simulated Human Oracle, the framework improved the Whole Tumor (WT) Dice score from 0.891 to 0.914 and reduced the 95th percentile Hausdorff Distance (HD95) from 5.82 mm to 4.76 mm. Critically for surgical safety, the system rescued severe boundary failures in the Tumor Core, reducing mean HD95 from 17.96 mm to 14.83 mm (improving absolute TC Dice to 0.356). These spatial rescues were achieved while demanding a median interactive workload of just 11.3% of the target volume. Acknowledging this as a simulated upper bound lacking real-world cognitive friction, the framework nevertheless demonstrates a highly Pareto-efficient pathway for safely deploying clinical AI.