Data-driven microscopy allows for automated context-specific acquisition of high-fidelity image data.

Area of Science:

• Biophysics
• Cell Biology
• Microscopy Techniques

Background:

• Light microscopy offers subcellular resolution but struggles with population-wide characterization.
• Existing methods lack high-fidelity imaging integrated with population context.

Purpose of the Study:

• To introduce a general framework, data-driven microscopy (DDM), for high-fidelity imaging.
• To enable real-time population-wide object characterization for context-aware imaging.
• To enhance microscopy data quality and reduce human bias.

Main Methods:

• DDM framework combines data-independent and data-dependent imaging steps.
• Utilizes real-time population-wide object characterization.
• Developed plugins for high-content screening and live adaptive microscopy.

Main Results:

• Demonstrated DDM's effectiveness in cell migration and infection studies.
• Captured rare and common events with high precision and resolution.
• Showcased improved high-content screening capabilities.

Conclusions:

• DDM provides a data-driven approach to enhance microscopy fidelity.
• Reduces bias and increases reproducibility in single-cell analysis.
• Integrates single-cell data within the broader population context for improved interpretation.