IXGS-Intraoperative 3D Reconstruction from Sparse, Arbitrarily Posed Real X-rays

2025-04-20

AI Summary

To address the reliance on radiation-intensive 3D imaging in spinal surgery, this paper proposes a real-time, unsupervised 3D anatomical reconstruction method from sparse, arbitrarily oriented intraoperative X-ray images. Methodologically, it introduces (1) the first adaptation of R²-Gaussian Splatting to real-world sparse X-ray reconstruction; (2) an anatomy-guided radiographic style normalization module to enhance multi-view consistency and anatomical fidelity; and (3) an end-to-end instantiation framework requiring neither pretraining nor annotated data. Validation on ex vivo datasets demonstrates that only 20–30 X-ray views suffice for surgical navigation accuracy. Quantitative evaluation via PSNR and SSIM confirms that style normalization significantly improves reconstruction quality. Expert assessment further verifies anatomical plausibility and clinical applicability.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Spine surgery is a high-risk intervention demanding precise execution, often supported by image-based navigation systems. Recently, supervised learning approaches have gained attention for reconstructing 3D spinal anatomy from sparse fluoroscopic data, significantly reducing reliance on radiation-intensive 3D imaging systems. However, these methods typically require large amounts of annotated training data and may struggle to generalize across varying patient anatomies or imaging conditions. Instance-learning approaches like Gaussian splatting could offer an alternative by avoiding extensive annotation requirements. While Gaussian splatting has shown promise for novel view synthesis, its application to sparse, arbitrarily posed real intraoperative X-rays has remained largely unexplored. This work addresses this limitation by extending the $R^2$-Gaussian splatting framework to reconstruct anatomically consistent 3D volumes under these challenging conditions. We introduce an anatomy-guided radiographic standardization step using style transfer, improving visual consistency across views, and enhancing reconstruction quality. Notably, our framework requires no pretraining, making it inherently adaptable to new patients and anatomies. We evaluated our approach using an ex-vivo dataset. Expert surgical evaluation confirmed the clinical utility of the 3D reconstructions for navigation, especially when using 20 to 30 views, and highlighted the standardization's benefit for anatomical clarity. Benchmarking via quantitative 2D metrics (PSNR/SSIM) confirmed performance trade-offs compared to idealized settings, but also validated the improvement gained from standardization over raw inputs. This work demonstrates the feasibility of instance-based volumetric reconstruction from arbitrary sparse-view X-rays, advancing intraoperative 3D imaging for surgical navigation.

Problem

• Reconstructing 3D spinal anatomy from sparse X-rays
• Reducing reliance on radiation-intensive 3D imaging systems
• Improving reconstruction quality without extensive annotated data

Innovation

• Extends R2-Gaussian splatting for 3D reconstruction
• Uses anatomy-guided style transfer standardization
• Requires no pretraining for patient adaptability