Rapid alignment of nanotomography data using joint iterative reconstruction and reprojection

Area of Science:

• Nanoscale imaging
• 3D reconstruction
• Tomography

Background:

• Advancements in nanoscale imaging necessitate precise alignment of projection images in x-ray and electron tomography.
• Current rotation stage limitations pose significant challenges for accurate 3D image reconstruction at the nanoscale.

Purpose of the Study:

• To develop a rapid post-acquisition alignment method for projection images in nanotomography.
• To enhance the quality of three-dimensional images obtained from nanoscale tomography datasets.

Main Methods:

• A novel iterative refinement procedure for joint estimation of alignment errors and the object.
• Application of the method to both simulated and experimental x-ray and electron nanotomography data.

Main Results:

• Demonstrated a thousand-fold reduction in residual alignment error with simulated data.
• Achieved comparable reconstructed image error levels in less than half the iterations compared to existing methods.
• Successfully applied the alignment approach to experimental nanotomography datasets.

Conclusions:

• The proposed post-acquisition alignment method effectively addresses limitations in nanoscale tomography.
• This technique enables high-quality 3D image reconstruction from challenging projection datasets.