Real-time prediction of breast cancer sites using deformation-aware graph neural network

2025-11-17

AI Summary

To address tumor target displacement caused by tissue deformation in indirect MRI-guided breast biopsy, this paper proposes a patient-specific deformation-aware model based on graph neural networks (GNNs). The method replaces computationally intensive finite element simulation with a geometric prior that jointly encodes surface displacement and distance maps, enabling millisecond-scale inference of the full-tissue displacement field. A patient-specific deformable model is constructed from MRI structural information. Validation on phantom and clinical datasets demonstrates a tumor displacement prediction RMSE of 0.2 mm and a Dice similarity coefficient of 0.977. Inference is over 4,000× faster than finite element methods, enabling real-time navigation. This work significantly improves spatial accuracy and clinical feasibility of ex vivo MRI-guided biopsy.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Early diagnosis of breast cancer is crucial, enabling the establishment of appropriate treatment plans and markedly enhancing patient prognosis. While direct magnetic resonance imaging-guided biopsy demonstrates promising performance in detecting cancer lesions, its practical application is limited by prolonged procedure times and high costs. To overcome these issues, an indirect MRI-guided biopsy that allows the procedure to be performed outside of the MRI room has been proposed, but it still faces challenges in creating an accurate real-time deformable breast model. In our study, we tackled this issue by developing a graph neural network (GNN)-based model capable of accurately predicting deformed breast cancer sites in real time during biopsy procedures. An individual-specific finite element (FE) model was developed by incorporating magnetic resonance (MR) image-derived structural information of the breast and tumor to simulate deformation behaviors. A GNN model was then employed, designed to process surface displacement and distance-based graph data, enabling accurate prediction of overall tissue displacement, including the deformation of the tumor region. The model was validated using phantom and real patient datasets, achieving an accuracy within 0.2 millimeters (mm) for cancer node displacement (RMSE) and a dice similarity coefficient (DSC) of 0.977 for spatial overlap with actual cancerous regions. Additionally, the model enabled real-time inference and achieved a speed-up of over 4,000 times in computational cost compared to conventional FE simulations. The proposed deformation-aware GNN model offers a promising solution for real-time tumor displacement prediction in breast biopsy, with high accuracy and real-time capability. Its integration with clinical procedures could significantly enhance the precision and efficiency of breast cancer diagnosis.

Problem

• Predicts breast cancer lesion locations during biopsy procedures in real time
• Creates accurate deformable breast models using graph neural networks
• Overcomes limitations of MRI-guided biopsies through computational efficiency

Innovation

• Graph neural network predicts breast cancer site deformation
• Finite element model simulates tissue behavior from MRI
• Real-time inference with 4000x speedup over simulations