New LiftXR framework reconstructs CT volumes from X-rays

Researchers have developed LiftXR, a novel framework for reconstructing CT volumes from bi-planar X-ray images. This method addresses the inherent ambiguity in X-ray data by first generating a 3D anatomical layout, which then guides an intensity renderer to produce a CT volume. The reconstructed CT is further refined by an anatomical parser that uses boundary and intensity cues to improve the layout, enabling region-specific calibration. Experiments show LiftXR surpasses existing methods and yields improved anatomical fidelity for downstream tasks like segmentation. AI

IMPACT This research could lead to more accurate and detailed CT scans derived from simpler X-ray imaging, potentially improving diagnostic capabilities.

RANK_REASON The item is an academic paper detailing a new method for medical image reconstruction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• CT
• Connected Papers
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• LiftXR
• Litmaps
• ScienceCast
• scite Smart Citations
• X-ray

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →