Motion-corrected Noise2Noise denoising for structural analysis of cells and tissues with (cryo-)FIB/SEM.

Motion correction enables Noise2Noise denoising of (cryo-)FIB/SEM volumes with slice thickness over 10 nm, improving 3D ultrastructural analysis.

• Why it matters: Low signal-to-noise ratio in (cryo-)FIB/SEM hampers interpretation of nanoscale cellular and tissue structures, limiting insights from volume data.
• What they did: A method was developed that splits volumes into even and odd slices, applies motion correction to estimate structural differences, and trains Noise2Noise models on corrected subvolumes.
• The result: The approach, validated on representative volumes, highlights the critical role of motion correction for effective denoising, enabling clearer structural analysis at nanoscale.