←Back to feed
🧠 AI🟢 Bullish
3D Wavelet-Based Structural Priors for Controlled Diffusion in Whole-Body Low-Dose PET Denoising
arXiv – CS AI|Peiyuan Jing, Yue Yang, Chun-Wun Cheng, Zhenxuan Zhang, Liutao Yang, Thiago V. Lima, Klaus Strobel, Antoine Leimgruber, Angelica Aviles-Rivero, Guang Yang, Javier A. Montoya-Zegarra|
🤖AI Summary
Researchers developed WCC-Net, a 3D wavelet-based diffusion model that significantly improves low-dose PET imaging denoising while reducing patient radiation exposure. The AI framework uses frequency-domain structural priors to maintain anatomical accuracy and outperforms existing CNN, GAN, and diffusion baselines across multiple dose levels.
Key Takeaways
- →WCC-Net combines 3D diffusion models with wavelet-based structural guidance to denoise low-dose PET scans more effectively than existing methods.
- →The framework improves PSNR by +1.21 dB and SSIM by +0.008 over strong diffusion baselines on 1/20-dose test sets.
- →The model generalizes robustly to unseen dose levels including 1/50 and 1/4 dose scenarios while maintaining volumetric anatomical consistency.
- →By decoupling anatomical structure from noise, WCC-Net preserves diagnostic reliability while reducing patient radiation exposure.
- →The research demonstrates superior performance in whole-body imaging applications with reduced structural distortion and intensity errors.
#medical-ai#diffusion-models#pet-imaging#denoising#3d-processing#wavelets#healthcare-ai#computer-vision
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Related Articles