What is it about?
We used a convolutional neuronal network to see transparent samples, which are typically not seen by a bright field microscope. The great thing is that for training we use simulated data, but it produces results from real observations.
Featured Image
Photo by Dynamic Wang on Unsplash
Why is it important?
Our approach enables phase imaging based on a traditional microscope without additional hardware. It means that the advantages of phase imaging (as stain-free examination, and dry mass estimation) are now available for a wider audience and might help with medical studies (e.g., neural activity and cancer).
Perspectives
It is just the start of a long journey of trustworthy AI-assisted phase microscopy where simulated data is exploited to work with the real data.
Igor Shevkunov
Tampereen yliopisto
Read the Original
This page is a summary of: A deep learning-based concept for quantitative phase imaging upgrade of bright-field microscope, Applied Physics Letters, January 2024, American Institute of Physics,
DOI: 10.1063/5.0180986.
You can read the full text:
Contributors
The following have contributed to this page
