Neural phase microscopy with metasurface optics for real-time and nanoscale quantitative phase imaging

Area of Science:

• Optics and Photonics
• Biomedical Imaging
• Artificial Intelligence

Background:

• Quantitative phase imaging (QPI) is crucial for non-invasive analysis in various scientific fields.
• Conventional QPI systems are often bulky, slow, and require complex setups, limiting real-time applications.
• Existing phase-retrieval software can suffer from low image quality and environmental sensitivity.

Purpose of the Study:

• To introduce a compact, fast, and high-resolution QPI platform.
• To overcome the limitations of conventional QPI hardware and software.
• To enable real-time, portable, and precise phase imaging.

Main Methods:

• Integration of nanophotonic metasurfaces to simplify optical architecture and enable single-shot acquisition.
• Development of physics-informed artificial intelligence (AI) models for aberration correction and phase restoration.
• Implementation of a compact optical design reducing form factor and complexity.

Main Results:

• Achieved nanoscale resolution better than 840 nm.
• Operated at a high frame rate of 74 Hz.
• Demonstrated real-time quantitative phase information restoration with corrected optical aberrations.

Conclusions:

• The combined nanophotonic hardware and AI algorithms significantly advance QPI technology.
• The developed platform offers a path towards portable, precise, and real-time phase imaging.
• This innovation addresses key challenges in current QPI systems, enhancing its applicability.