Cyst-X: A Federated AI System Outperforms Clinical Guidelines to Detect Pancreatic Cancer Precursors and Reduce Unnecessary Surgery

2025/07/29 by Pan, Hongyi, Durak, Gorkem, Keles, Elif +28
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2507.22017

Abstract

Pancreatic cancer is projected to be the second-deadliest cancer by 2030, making early detection critical. Intraductal papillary mucinous neoplasms (IPMNs), key cancer precursors, present a clinical dilemma, as current guidelines struggle to stratify malignancy risk, leading to unnecessary surgeries or missed diagnoses. Here, we developed Cyst-X, an AI framework for IPMN risk prediction trained on a unique, multi-center dataset of 1,461 MRI scans from 764 patients. Cyst-X achieves significantly higher accuracy (AUC = 0.82) than both the established Kyoto guidelines (AUC = 0.75) and expert radiologists, particularly in correct identification of high-risk lesions. Clinically, this translates to a 20% increase in cancer detection sensitivity (87.8% vs. 64.1%) for high-risk lesions. We demonstrate that this performance is maintained in a federated learning setting, allowing for collaborative model training without compromising patient privacy. To accelerate research in early pancreatic cancer detection, we publicly release the Cyst-X dataset and models, providing the first large-scale, multi-center MRI resource for pancreatic cyst analysis.

Citations

• Large-Scale Multi-Center CT and MRI Segmentation of Pancreas with Deep Learning
• Frequency-Based Federated Domain Generalization for Polyp Segmentation
• Domain Generalization with Fourier Transform and Soft Thresholding
• Radiomics Boosts Deep Learning Model for IPMN Classification
• MONAI: An open-source framework for deep learning in healthcare
• Dynamic Linear Transformer for 3D Biomedical Image Segmentation
• Differentially Private Federated Learning with Local Regularization and Sparsification
• Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images
• Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images
• Federated Multi-Target Domain Adaptation
• Information Bottleneck Attribution for Visual Explanations of Diagnosis\n and Prognosis
• FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space
• nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation
• Federated Learning with Matched Averaging
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Federated Adversarial Domain Adaptation
• INN: Inflated Neural Networks for IPMN Diagnosis
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
• Federated Optimization in Heterogeneous Networks
• Efficient Attention: Attention with Linear Complexities
• European evidence-based guidelines on pancreatic cystic neoplasms
• Differentially Private Federated Learning: A Client Level Perspective
• Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?
• Deep Multi-Modal Classification of Intraductal Papillary Mucinous Neoplasms (IPMN) with Canonical Correlation Analysis
• Improving Deep Pancreas Segmentation in CT and MRI Images via Recurrent Neural Contextual Learning and Direct Loss Function
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
• Densely Connected Convolutional Networks
• Deep Residual Learning for Image Recognition
• Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy
• Restricting the Flow: Information Bottlenecks for Attribution

Related

• Radiomics Boosts Deep Learning Model for IPMN Classification
• Predicting the risk of pancreatic cancer with a CT-based ensemble AI algorithm
• IPMN Risk Assessment under Federated Learning Paradigm
• Early Detection and Localization of Pancreatic Cancer by Label-Free Tumor Synthesis
• Leveraging Fine-Tuned Large Language Models for Interpretable Pancreatic Cystic Lesion Feature Extraction and Risk Categorization