A fast fiducial marker tracking model for fully automatic alignment in electron tomography.

Area of Science:

• Cryo-electron microscopy
• Structural biology
• Image processing

Background:

• Fiducial marker tracking is critical for accurate automatic alignment in subtomogram averaging.
• Current methods face challenges in achieving fully automatic and effective fiducial marker tracking.

Purpose of the Study:

• To develop a robust and efficient scheme for automatic fiducial marker tracking.
• To theoretically relate projection and tracking models for improved alignment.
• To enhance the quality of automatic alignment in electron microscopy.

Main Methods:

• Theoretical analysis of affine transformation deviation for fiducial marker alignment.
• Development of a Gaussian mixture model-based algorithm for accelerated fiducial marker tracking.
• Implementation of a divide-and-conquer strategy to address lens distortions.

Main Results:

• Theoretical upper bound established for transformation deviation.
• Demonstrated effectiveness and efficiency of the Gaussian mixture model algorithm.
• Real-world experiments validate theoretical bounds and algorithmic performance.

Conclusions:

• The proposed scheme provides a reliable and efficient solution for automatic fiducial marker tracking.
• This work facilitates fully automatic alignment for large-scale electron microscopy datasets.
• The developed algorithm significantly improves the quality of subtomogram averaging.