cGAN QPI-DHM
Quantitative phase imaging using a digital holographic microscope and a generative adversarial network
The conventional quantitative phase reconstruction in off-axis Digital Holographic Microscopy (DHM) relies on computational processing. Regardless the implementation, any DHM computational processing leads to large processing times, hampering the use of DHM to video-rate renderings of dynamic biological processes. In this work, a conditional generative adversarial network (cGAN) for robust and fast quantitative phase imaging in DHM is reported. The reconstructed phase images provided by the GAN model present stable background levels, enhancing the visualization of the specimens for different experimental conditions. The proposed learning-based method has been trained and validated using human red blood cells. After proper training, the proposed GAN yields to a computationally efficient method, reconstructing DHM images 7X faster than conventional computational approaches.
Downloads
Feel free to download the weights yielded during the training stage for the proposed cGAN model click here
Funding and Acknowledgments
*This project was partially funded by the Vicerrectoría de Descubrimiento y Creación at Universidad EAFIT (Colombia), and the National Science Foundation (NSF) (grant number 2042563).
*C. Trujillo thanks the supercomputing resources made available by the Centro de Computación Científica Apolo at Universidad EAFIT (http://www.eafit.edu.co/apolo (accessed on 20 November 2021)) to conduct the research reported in this scientific product. R. Castaneda and A. Doblas acknowledge the support from the University of Memphis and the Herff College of Engineering.
Citation
If using this information for publication, please kindly cite the following paper:
R. Castaneda, C. Trujillo, and A. Doblas “Video-rate quantitative phase imaging using a digital holo-graphic microscope and a generative adversarial network.,” Sensors 21(23), 8021 (2021). https://doi.org/10.3390/s21238021
Support or Contact
| Researcher | Google Scholar | webSite | |
|---|---|---|---|
| Raul Castaneda | rcstdqnt@memphis.edu | RaulGoogle | RaulResearch |
| Carlos Trujillo | catrujilla@eafit.edu.co | CarlosGoogle | CarlosEAFIT |
| Ana Doblas | adoblas@memphis.edu | AnaGoogle | AnaResearch |
Please visit OIRL website for learning more about our different projects
Please visit EAFIT website for learning more about our different projects