TotalSynth framework generates synthetic CT from MRI and CBCT scans

Researchers have developed TotalSynth, a framework capable of generating synthetic CT images from MRI and CBCT scans. The system was evaluated on a dataset comprising SynthRAD challenge data and multiple prostate cohorts, with additional testing on external BIC-MAC data. The MRI-to-CT model achieved an MAE of 67.49 HU, while the CBCT-to-CT model reached an MAE of 53.55 HU, demonstrating robust performance across various anatomical regions. The study highlights the importance of local validation and optional fine-tuning when dealing with domain shifts. AI

IMPACT This framework could improve medical imaging workflows by enabling the generation of synthetic CT scans from more readily available MRI or CBCT data.

RANK_REASON The cluster describes a research paper detailing a new framework for medical image synthesis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• BIC-MAC
• cone beam computed tomography
• Hugging Face
• magnetic resonance imaging
• SynthRAD
• Valentin Boussot

• paper
• other

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