Dynamic autofocus for continuous-scanning time-delay-and-integration image acquisition in automated microscopy.

Area of Science:

• Biomedical Imaging
• Microscopy Technology
• Computational Imaging

Background:

• Efficient image cytometry requires rapid acquisition and analysis of large image datasets (20 gigapixels).
• High-speed acquisition is crucial for numerous biomedical applications.
• Continuous-motion time-delay-and-integrate (TDI) scanning offers speed but is limited by autofocus challenges.

Purpose of the Study:

• To develop a dynamic autofocus system for high-resolution image cytometry during continuous scanning.
• To overcome the limitations of implementing high-resolution autofocus simultaneously with TDI scanning.

Main Methods:

• A "volume camera" with nine fiber optic imaging conduits capturing images from multiple focal planes.
• Mixed analog-digital processing circuits measuring high spatial frequencies for focus index calculation.
• Software for real-time analysis of focus data streams and closed-loop feedback control.

Main Results:

• The system achieves autofocus updates at 56 Hz, enabling image acquisition at 0.3x0.3 micrometer resolution and 2.3 mm/s stage speed.
• High-performance closed-loop image-content-based autofocus was demonstrated in continuous scanning for the first time.
• Successful testing in phase contrast and fluorescence imaging confirmed system efficacy.

Conclusions:

• The developed dynamic autofocus system significantly enhances image cytometry speed and resolution.
• This technology paves the way for broader adoption of high-speed microscopy in biomedical research.
• The system represents a significant advancement in real-time autofocus for continuous scanning microscopy.