Deep learning can discern and enhance microscopic details in smartphone photos, with the potential to transform applications in global health, telemedicine, and medical diagnostics, according to researchers at the Samueli School of Engineering of the University of California, Los Angeles.
By enhancing resolution and color details, the method allows smartphone images to approach the quality of images from laboratory-grade microscopes. This could deliver high-quality medical diagnostics to resource-poor regions without access to high-end diagnostic technologies.
The method uses an attachment placed over the smartphone lens to increase the resolution and visibility of image details. Artificial intelligence reproduces the resolution and color details needed for laboratory analysis.
The researchers captured images of medical diagnostics with a standard laboratory-grade microscope, and then with a smartphone using the 3D-printed microscope attachment. Then they fed the corresponding images into a computer system that "learns" to improve the mobile phone images, using deep-learning–based computer code.
From UCLA Newsroom
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found