Electron Microscopy Aberration Correction Using Beam-Image Shift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current electron microscopy methods struggle with inefficient data collection due to uncalibrated image aberrations, which require repetitive mechanical translation and high expertise, limiting the throughput and quality of three-dimensional structure determination.
Innovation Solution
A method involving beam-image shift and kernel canonical correlation analysis (KCCA) to calibrate and correct non-axial aberrations, using internal reference grids with known properties, and applying deconvolution techniques to predict and correct aberrations across the imaged area.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If repetitive mechanical translation is used to observe different parts of the sample, then image aberrations are minimized locally, but data collection efficiency deteriorates
Solution Approach 1:
The patent replaces repetitive mechanical translation with computational methods. By using kernel canonical correlation analysis (KCCA) to calibrate and correct non-axial aberrations across the entire imaged area, the system eliminates the need for mechanical repositioning while maintaining image quality throughout the field of view.
Solution Approach 2:
The aberration correction method provides universal correction across the entire imaged area rather than being limited to a small local region. This allows different parts of the sample to be observed simultaneously with corrected aberrations, enabling efficient data collection without mechanical translation.
2Measurement precision
If microscope alignment procedures are used to minimize aberrations, then image quality is improved, but operation complexity increases
Solution Approach 1:
The system performs self-calibration using kernel canonical correlation analysis on the acquired images themselves. The method automatically determines aberration parameters and applies corrections without requiring manual alignment procedures, making the system self-sufficient and easier to operate.
Solution Approach 2:
The patent changes the approach from manual parameter adjustment to automated parameter determination. By using KCCA to automatically calculate aberration parameters from the image data, the system eliminates complex manual alignment procedures while achieving superior correction results.
3Measurement precision
If deconvolution with multiple coefficients is used to correct aberrations, then correction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary determination of optimal deconvolution coefficients using kernel canonical correlation analysis on calibration data. By pre-calculating the coefficients that maximize the moment of the intensity distribution, the system simplifies the actual correction process while maintaining high accuracy.
Solution Approach 2:
The method uses feedback through moment maximization to determine optimal deconvolution coefficients. By iteratively adjusting coefficients to maximize the moment of the intensity distribution, the system automatically converges to the optimal correction parameters, balancing accuracy and computational efficiency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods for correcting one or more image aberrations in an electron microscopy image, including cryo-EM images, are provided. The method includes obtaining a plurality of electron microscope (EM) images of an internal reference grid sample having one or more known properties, the plurality of electron microscope images obtained for a plurality of optical conditions and for a plurality of coordinated beam-image shifts. The method may also include, among other features, determining an aberration correction function that predicts aberrations for every point in the imaged area using kernel canonical correlation analysis (KCCA).