High-Fidelity 3D Geometric Reconstruction of Pelvic Organs from MRI: A Hybrid Deep Learning and Iterative Optimization Approach

The study presents a hybrid framework for high-fidelity 3D geometric reconstruction of pelvic organs from MRI, combining deep learning with iterative optimization. Key components include a multi-level deep learning architecture that ensures topological consistency, a two-stage optimization strategy for shape capture and surface refinement, and a synergy mechanism that enhances training and inference processes. This approach outperforms existing models in geometric fidelity, achieving lower Chamfer Distance and higher Dice Similarity Coefficients for reconstructed organs, while maintaining computational efficiency, making it significant for practitioners involved in patient-specific anatomical modeling.