PanGuide3D: Cohort-Robust Pancreas Tumor Segmentation via Probabilistic Pancreas Conditioning and a Transformer Bottleneck

2026-04-22

AI Summary

Pancreatic tumor segmentation in contrast-enhanced CT remains challenging, particularly for small lesions and anatomically complex regions, where generalization performance is often poor. To address this, this work proposes PanGuide3D, a novel architecture featuring a shared 3D encoder coupled with a decoder that generates a probabilistic pancreatic map to explicitly guide tumor segmentation via a multi-scale soft-gating mechanism. Additionally, a lightweight Transformer is embedded within the U-Net bottleneck to enhance long-range contextual modeling. Despite its structural simplicity, PanGuide3D significantly improves cross-cohort robustness and calibration, achieving state-of-the-art performance on both PanTS and MSD Task07 benchmarks. The method effectively reduces anatomically implausible false positives and notably enhances segmentation accuracy for small tumors and difficult anatomical regions.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Pancreatic tumor segmentation in contrast-enhanced computed tomography (CT) is clinically important yet technically challenging: lesions are often small, heterogeneous, and easily confused with surrounding soft tissue, and models that perform well on one cohort frequently degrade under cohort shift. Our goal is to improve cross-cohort generalization while keeping the model architecture simple, efficient, and practical for 3D CT segmentation. We introduce PanGuide3D, a cohort-robust architecture with a shared 3D encoder, a pancreas decoder that predicts a probabilistic pancreas map, and a tumor decoder that is explicitly conditioned on this pancreas probability at multiple scales via differentiable soft gating. To capture long-range context under distribution shift, we further add a lightweight Transformer bottleneck in the U-Net bottleneck representation. We evaluate cohort transfer by training on the PanTS (Pancreatic Tumor Segmentation) cohort and testing both in-cohort (PanTS) and out-of-cohort on MSD (Medical Segmentation Decathlon) Task07 Pancreas, using matched preprocessing and training protocols across strong baselines. We collect voxel-level segmentation metrics, patient-level tumor detection, subgroup analyses by tumor size and anatomical location, volume-conditioned performance analyses, and calibration measurements to assess reliability. Across the evaluated models, PanGuide3D achieves the best overall tumor performance and shows improved cross-cohort generalization, particularly for small tumors and challenging anatomical locations, while reducing anatomically implausible false positives. These findings support probabilistic anatomical conditioning as a practical strategy for improving cross-cohort robustness in an end-to-end model and suggest potential utility for contouring support, treatment planning, and multi-institutional studies.

Problem

• pancreatic tumor segmentation
• cohort shift
• cross-cohort generalization
• contrast-enhanced CT
• small lesion segmentation

Innovation

• probabilistic anatomical conditioning
• cohort-robust segmentation
• Transformer bottleneck
• multi-scale soft gating
• cross-cohort generalization