Imaging Diagnoses Patients Just Below the Surface

R&D DIGEST

A theory on the generation of high-frequency pulses that penetrate a short distance beneath a surface could lead to high-quality imaging technology. Called nonlinear constructive interference, the technique could help detect skin cancer and find flaws below tooth enamel. The method produces powerful microwave radiation without the ionizing damage that is generated by x-rays.

Ehasan Afshari calls his work a new phenomenon. “If you have a non­linear medium—in this case, it's a surface—and you launch two waves, when they hit each other at a certain angle, they could combine in a nonlinear fashion,” he says. “The amplitude of the [new] wave could be more than the summation of the two waves.” At the same time, the frequency content of the waves could increase.

Afshari, an assistant professor of electrical and computer engineering at Cornell University (Ithaca, NY), conducted the work with Harish Bhat, an assistant professor of mathematics with the University of California, Merced.

The team took lower-frequency signal sources (60 GHz) and showed that it's possible to combine them to generate power at more than 200 GHz. This kind of wave interaction and two-dimensional nonlinear surface is something new, says Afshari.

After publishing a paper on the technique, the researchers tested a prototype at a lower frequency and a lower scale, with good results. Afshari anticipates testing the device at a higher frequency and a higher power soon. He says it's possible that they'll be able to generate signals at frequencies up to around 1.16 THz, and have the first device developed within a year.

The main medical uses for the technology would involve scanning and imaging with a high resolution and a low penetration depth. “Without using anything else, we could detect [skin] cancer,” says Afshari. “In dental applications, we could see beneath the surface of the tooth and scan different layers with very high resolution.”

The method is compatible with a complementary metal oxide silicon chip, so there isn't a need to create a complex device. This means it would be small, cheap, and light. “The technique is compatible with existing technology, which is commercially available on a silicon substrate,” says Afshari. “The other advantage is that several types of digital signal processing can be integrated into the silicon.”

The system could be engineered to control penetration depth, allowing signals to push a couple of millimeters deep while providing fine resolution. The heart of it would be on a silicon chip that measures a few millimeters in size, and the cost for mass production would be a few dollars per chip. To apply it to medical applications, Afshari says the system would need packaging and other changes that would increase the cost, but not in thousands of dollars. “At this point,” says Afshari, “I'm very interested in working with a [medical device] company to further develop this.”

Afshari is working with his students to fabricate the prototype on silicon, which is the main challenge. Because his training is in circuit design and physics, Afshari wants to work with someone who understands the medical applications.

The National Science Foundation supported part of the research. Details about the new technique were published in the May issue of the journal Physical Review E.

Copyright ©2008 Medical Device & Diagnostic Industry