Filter2Noise: a framework for interpretable and zero-shot low-dose CT image denoising

Key Points

• The aim is to develop a low-dose CT image denoising framework that is both interpretable and user-controlled.
• Developed the Filter2Noise (F2N) framework for image denoising.
• Evaluated performance across different CT scanners and imaging protocols.
• Ensured the framework requires no retraining for different environments.
• Achieved competitive denoising performance.
• Maintained complete interpretability of the denoising process.
• Enabled user control for radiologists in adjusting parameters.