Search papers, labs, and topics across Lattice.
This study presents a hybrid framework that combines deep learning and iterative optimization to achieve high-fidelity 3D reconstructions of pelvic organs from MRI scans, specifically targeting the bladder, uterus, and rectum. By integrating a geometry-aware multi-level architecture with a two-stage optimization strategy, the framework not only enhances the accuracy of organ geometry but also improves computational efficiency compared to existing methods. The results show significant reductions in Chamfer Distance and improvements in Dice Similarity Coefficient scores, indicating superior geometric fidelity and mesh quality in the reconstructed models.
Achieving unprecedented geometric fidelity in 3D reconstructions of pelvic organs could revolutionize patient-specific modeling and analysis in medical applications.
Patient-specific 3D reconstruction of pelvic organ geometry from MRI is important for pelvic floor modeling and downstream patient-specific analysis. However, while previous studies have focused primarily on either image segmentation or downstream use of 3D models, the reconstruction of high-fidelity, high-quality geometries remains labor-intensive and poorly standardized. The study introduced a hybrid deformable shape modeling framework that integrates deep learning prediction with iterative optimization for the reconstruction of the bladder, uterus, and rectum. The framework consists of three core components: a geometry-aware multi-level deep learning architecture that preserves topological consistency of pelvic organs; a two-stage amortized optimization training strategy that balances global shape capture and local surface refinement; and a holistic synergy mechanism--where iterative optimization provides supervision for deep learning during the training phase, and during inference, deep learning rapidly predicts the global organ morphology, followed by iterative optimization to refine local surfaces and mesh quality. This framework demonstrated marked superiority in geometric fidelity than current mainstream deep learning-based organ reconstruction models. For individual anatomical structures, the reconstructed 3D geometries for the bladder, rectum, and uterus achieved significantly lower Chamfer Distance values and higher Dice Similarity Coefficient scores. In addition, while maintaining high computational efficiency, the proposed architecture yielded superior overall volumetric mesh quality. At the patient level, the framework achieved higher mean values for the 10 worst elements for both minSICN and minSIGE compared to traditional geometric post-processing algorithms.
