Multimodal Conditional MeshGAN for Personalized Aneurysm Growth Prediction

2025-08-27

AI Summary

This study addresses the challenge of jointly modeling local geometric deformations and global anatomical changes in predicting three-dimensional aortic aneurysm evolution. We propose Mesh2Mesh-GAN—the first multimodal conditional mesh-to-mesh generative adversarial network tailored for aneurysm growth prediction. Methodologically, it introduces a dual-branch architecture: KNN convolution for local geometric modeling and graph convolution for global topological modeling, conditioned on clinical features and temporal variables to support both retrospective and prospective longitudinal modeling. Evaluated on our newly established TAAMesh dataset, the model achieves statistically significant improvements over state-of-the-art methods in geometric reconstruction fidelity and key diameter prediction accuracy. The codebase and baselines are publicly released, establishing a novel paradigm for clinically deployable, patient-specific disease trajectory modeling.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• 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
• Web Mining and Content Analysis: Models for Web evolution
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Personalized, accurate prediction of aortic aneurysm progression is essential for timely intervention but remains challenging due to the need to model both subtle local deformations and global anatomical changes within complex 3D geometries. We propose MCMeshGAN, the first multimodal conditional mesh-to-mesh generative adversarial network for 3D aneurysm growth prediction. MCMeshGAN introduces a dual-branch architecture combining a novel local KNN-based convolutional network (KCN) to preserve fine-grained geometric details and a global graph convolutional network (GCN) to capture long-range structural context, overcoming the over-smoothing limitations of deep GCNs. A dedicated condition branch encodes clinical attributes (age, sex) and the target time interval to generate anatomically plausible, temporally controlled predictions, enabling retrospective and prospective modeling. We curated TAAMesh, a new longitudinal thoracic aortic aneurysm mesh dataset consisting of 590 multimodal records (CT scans, 3D meshes, and clinical data) from 208 patients. Extensive experiments demonstrate that MCMeshGAN consistently outperforms state-of-the-art baselines in both geometric accuracy and clinically important diameter estimation. This framework offers a robust step toward clinically deployable, personalized 3D disease trajectory modeling. The source code for MCMeshGAN and the baseline methods is publicly available at https://github.com/ImperialCollegeLondon/MCMeshGAN.

Problem

• Predicting personalized aortic aneurysm growth progression accurately
• Modeling both local deformations and global anatomical changes
• Overcoming over-smoothing limitations in 3D geometry predictions

Innovation

• Multimodal conditional GAN for 3D aneurysm prediction
• Dual-branch architecture with KCN and GCN networks
• Clinical attribute encoding for personalized temporal modeling