Related Experiment Video
Updated: Jun 16, 2025
07:16
Three-Dimensional Imaging of Tumor-Bearing Tissue Using the Iterative Bleaching Extends Multiplexity Approach
Published on: April 25, 2025
139
Fourier-Based 3D Multistage Transformer for Aberration Correction in Multicellular Specimens
Thayer Alshaabi1,2, Daniel E Milkie1, Gaoxiang Liu2
1Howard Hughes Medical Institute, Ashburn, VA.
Arxiv
|June 5, 2025
Summary
Adaptive Optical Vision Fourier Transformer (AOViFT) uses machine learning to correct optical aberrations in microscopy. This AI approach enhances imaging quality and reduces hardware complexity for biological samples.
Area of Science:
- Microscopy
- Optical Engineering
- Computational Biology
Background:
- Optical aberrations significantly degrade resolution and contrast in high-resolution tissue imaging.
- Conventional adaptive optics (AO) systems are complex, costly, and slow for large-scale aberration correction.
Purpose of the Study:
- To introduce AOViFT (Adaptive Optical Vision Fourier Transformer), a novel machine learning framework for aberration sensing and correction.
- To demonstrate AOViFT's ability to restore diffraction-limited performance with reduced computational resources.
Main Methods:
- Development of a 3D multistage Vision Transformer operating on Fourier domain embeddings.
- Implementation of AOViFT for inferring optical aberrations in microscopy data.
- Validation using live gene-edited zebrafish embryos.
Main Results:
- AOViFT achieved aberration correction with significantly lower computational cost, training time, and memory footprint than conventional methods.
- The framework successfully corrected spatially varying aberrations in zebrafish embryos.
- Correction was demonstrated using both a deformable mirror and post-acquisition deconvolution.
Conclusions:
- AOViFT offers an efficient, AI-driven solution for correcting optical aberrations in microscopy.
- The framework eliminates the need for wavefront sensing hardware, simplifying experimental workflows.
- AOViFT lowers technical barriers, enabling high-resolution volumetric microscopy across diverse biological samples.
