Sophisticated computation in the process of development at Carnegie Mellon University and elsewhere will be instrumental in creating a new bio-imaging camera to produce quality images deeper under the skin.

Conventional imaging, such as ultrasound and magnetic resonance imaging, penetrate deep into the body but don’t capture optical imaging, which is used in measuring the number of white blood cells or a skin cancer.

However, current optical bio-imaging through tissue enables radiographers to see only a few millimeters below the skin—that’s because when light passes through the skin, it is scattered, making it virtually impossible to form a quality image.

“The cells in the skin bend the light. The imaging under the skin can only do so much. The light will give some image, but you get a lot of blur and stray light,” explains Aswin Sankaranarayanan, an assistant professor of electrical and computer engineering at Carnegie Mellon University, one of the researchers on the project.

The computation technique being developed, called “computational scatterography” aims to use algorithms to make sense of the light to fix the blur and stray light problems in the images. This will also enable radiographers to take images much deeper beneath the skin.

“For example, you take a photo of a football game, but the people move so you have a blur. What if you change the camera in a small way to capture more information? The goal is not to capture a sharper image [by itself] but the information with [the use of] algorithms puts it together and makes a higher quality image,” he says.

Computational illumination and imaging approaches developed at Carnegie Mellon University produce images of different body parts that show features up to two millimeters deep. Researchers expect new techniques will enable images of structures at least 10 times deeper.
Computational illumination and imaging approaches developed at Carnegie Mellon University produce images of different body parts that show features up to two millimeters deep. Researchers expect new techniques will enable images of structures at least 10 times deeper.

The computation will involve more than one algorithm. For instance, one algorithm will unbend the light so that it won’t scatter. Another would be used to separate the light to capture just the light of interest, such as from a particular layer of skin. There will be additional algorithms designed for different imaging to identify profusion or cancer, he says.

The effort, sponsored by the National Science Foundation, is led by Rice University and also includes researchers from Harvard, MIT and Cornell. The five-year, $10 million grant was announced in February.

Since then, the investigators have been working to identify the steps to take, the applications needed and the technology to bring to the table. One current experiment examines how light scatters through fog, snow and rain, which has yielded lessons that can be applied to this project.

“The long term objective is to create more complicated optical designs to view 10 millimeters beneath the skin and attack problems like white blood cell counts without a needle,” says Sankaranarayanan. The imaging cameras could also be used in remote monitoring and elsewhere.

Sankaranarayanan hopes that the team will make objective progress in the new technology in four years. “The beauty of this project is that there’s a bunch of very smart people trying to solve this problem,” he says.

Register or login for access to this item and much more

All Health Data Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access