Automated segmenta-on of pediatric neuroblastoma on multi-modal MRI: Results of the SPPIN challenge at MICCAI 2023

2025-05-01

AI Summary

Manual preoperative MRI segmentation for pediatric neuroblastoma is time-consuming and operator-dependent, with no publicly available benchmark for extracranial pediatric tumor segmentation. Method: We establish the first multimodal MRI automatic segmentation benchmark and challenge specifically for pediatric retroperitoneal tumors, incorporating T1-weighted, T2-weighted, and DWI 3D volumes. We systematically evaluate the generalizability of large-scale pretrained models—particularly STU-Net—on small-scale, highly heterogeneous pediatric MRI data, using comprehensive metrics: Dice coefficient, Hausdorff distance at 95% (HD95), and volumetric similarity (VS). Contribution/Results: The winning model achieves a median Dice score of 0.82 (HD95 = 7.69 mm; VS = 0.91), demonstrating feasibility of automated segmentation. However, performance degrades markedly for post-chemotherapy small tumors (Dice = 0.59), revealing a critical robustness bottleneck for clinical deployment. This benchmark fills a key gap in pediatric oncologic imaging and provides actionable insights into model limitations under real-world clinical variability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Surgery plays an important role within the treatment for neuroblastoma, a common pediatric cancer. This requires careful planning, often via magnetic resonance imaging (MRI)-based anatomical 3D models. However, creating these models is often time-consuming and user dependent. We organized the Surgical Planning in Pediatric Neuroblastoma (SPPIN) challenge, to stimulate developments on this topic, and set a benchmark for fully automatic segmentation of neuroblastoma on multi-model MRI. The challenge started with a training phase, where teams received 78 sets of MRI scans from 34 patients, consisting of both diagnostic and post-chemotherapy MRI scans. The final test phase, consisting of 18 MRI sets from 9 patients, determined the ranking of the teams. Ranking was based on the Dice similarity coefficient (Dice score), the 95th percentile of the Hausdorff distance (HD95) and the volumetric similarity (VS). The SPPIN challenge was hosted at MICCAI 2023. The final leaderboard consisted of 9 teams. The highest-ranking team achieved a median Dice score 0.82, a median HD95 of 7.69 mm and a VS of 0.91, utilizing a large, pretrained network called STU-Net. A significant difference for the segmentation results between diagnostic and post-chemotherapy MRI scans was observed (Dice = 0.89 vs Dice = 0.59, P = 0.01) for the highest-ranking team. SPPIN is the first medical segmentation challenge in extracranial pediatric oncology. The highest-ranking team used a large pre-trained network, suggesting that pretraining can be of use in small, heterogenous datasets. Although the results of the highest-ranking team were high for most patients, segmentation especially in small, pre-treated tumors were insufficient. Therefore, more reliable segmentation methods are needed to create clinically applicable models to aid surgical planning in pediatric neuroblastoma.

Problem

• Automated segmentation of pediatric neuroblastoma on multi-modal MRI
• Reducing time-consuming and user-dependent 3D model creation
• Improving reliability for small, pre-treated tumor segmentation

Innovation

• Used large pre-trained STU-Net network
• Multi-modal MRI for tumor segmentation
• Benchmarked with Dice and Hausdorff metrics