Overcoming the field-of-view to diameter trade-off in microendoscopy via computational optrode-array microscopy

Area of Science:

• Plant Biology
• Microscopy
• Biophotonics

Background:

• High-resolution deep tissue microscopy is vital for plant biology.
• Conventional imaging optics face a trade-off between field-of-view (FOV) and probe diameter due to aberrations.
• Implanted probes offer a solution but are limited by FOV-to-diameter ratios.

Purpose of the Study:

• To overcome the FOV limitations in deep tissue microscopy.
• To develop a novel imaging technique using microfabricated non-imaging probes (optrodes) and machine learning.
• To enable large FOV, high-resolution imaging in plant tissues.

Main Methods:

• Utilized microfabricated non-imaging probes (optrodes).
• Combined optrodes with a trained machine-learning algorithm to reconstruct images.
• Employed an array of optrodes (1x2) for parallel imaging.
• Demonstrated imaging on fluorescent beads, stained plant stem sections, and stained living stems.

Main Results:

• Achieved FOV of 1x to 5x the probe diameter using single optrodes and ML.
• Demonstrated real-time video imaging of fluorescent beads at 30 FPS.
• Successfully imaged stained plant stem sections and living stems with large FOV.
• Showcased enhanced FOV by using multiple optrodes in parallel.

Conclusions:

• Microfabricated non-imaging probes combined with machine learning significantly expand FOV in deep tissue microscopy.
• This approach overcomes fundamental optical limitations of conventional imaging.
• Paves the way for fast, high-resolution deep tissue imaging in plant biology and beyond.