Over a million developers have joined DZone.

How Deep Learning Can Help Us Identify Cancer Cells

Deep learning lends an eye in the war against cancer. Combining a specialized microscope with a deep neural net automates the detection of individual cancer cells.

· Big Data Zone

Read this eGuide to discover the fundamental differences between iPaaS and dPaaS and how the innovative approach of dPaaS gets to the heart of today’s most pressing integration problems, brought to you in partnership with Liaison.

Tackling cancer remains notoriously difficult, not to mention expensive.  A team from UCLA have developed a clever technique for identifying cancer cells much faster than is currently the case.

The method, which was documented in a recent study, takes a different stance to the current labeling approach that can fall down due to inaccuracy.

Instead, the new method captures an image of the cells without harming them.  It’s capable of detecting 16 different physical characteristics, such as its biomass, granularity and size.

Smarter Microscopes

The method takes advantage of a couple of UCLA inventions, including a photonic time stretch microscope and a deep learning algorithm that is capable of identifying cancer cells with high levels of accuracy.

The team believe that the microscope is just one of the possible applications of the technology.  It works in a similar way to a flash on a camera, and takes pictures of the blood cells using laser bursts.

Of course, such a process is done in nanoseconds so would ordinarily be much too fast to be digitized by normal equipment.  The new microscope is capable of overcoming this by using smart optics that both enhance the clarity of the images and slow them down long enough to be detected.  The deep learning algorithm then goes to work on identifying the cancerous cells from their healthy peers.

“Each frame is slowed down in time and optically amplified so it can be digitized,” the researchers say. “This lets us perform fast cell imaging that the artificial intelligence component can distinguish.”

Working in the Dark

The approach is also clever in the way it works with light.  Ordinarily, the light generated by taking such pictures would destroy the cells, but the photonic time stretch approach allows pictures to be taken with low-level illumination.

It’s certainly a fascinating approach, and the team believe that it could result in a much more data driven approach to diagnosis.

Check out the video below for more information.

Discover the unprecedented possibilities and challenges, created by today’s fast paced data climate and why your current integration solution is not enough, brought to you in partnership with Liaison

neural networks,deep learning,cancer

Published at DZone with permission of Adi Gaskell, DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

The best of DZone straight to your inbox.

Please provide a valid email address.

Thanks for subscribing!

Awesome! Check your inbox to verify your email so you can start receiving the latest in tech news and resources.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}