A newly emerging technology holds promise for imaging and diagnosing skin-cancer tissues earlier and more accurately than ever — and Stevens Institute of Technology is at the forefront.
"Skin cancer is the most common and fastest-growing of all cancer types, with approximately 3.5 million new cases and billions of dollars of treatment cost in the U.S. occurring annually," explains Stevens electrical and computer engineering professor Negar Tavassolian. "It is usually diagnosed through visual inspection by a dermatologist, but visual inspection is subjective and can be susceptible to error."
Tavassolian, who trained at Sharif University of Technology, McGill University and Georgia Tech and did postdoctoral work at MIT, says new imaging technologies can help doctors improve the odds of early detection. She is the recipient of a recent National Science Foundation (NSF) CAREER award to develop just such an innovative medical application of millimeter-wave technology.
"Early detection of skin cancers is critical, and millimeter-wave technologies and devices have now evolved to the point where low-cost, in-depth views of skin are on the horizon," she notes. "Most of the problems, such as safety and power supply, have been solved.
"Now we propose to solve a key remaining challenge: high, useful resolution of the images."
Achieving ultra-wide bandwidths by splitting channels
Microwave-band (frequency 1 GHz to 30 GHz) imaging technology is currently used to diagnose breast cancer, lung cancer, stroke and other diseases, but millimeter-wave technology (frequency 30 GHz to 300 GHz) is relatively new and underdeployed — largely confined to military and security applications such as body-scanning in airports as well as a few industrial uses. The radars that aid automobile collision avoidance and automatic braking systems utilize millimeter waves, for instance.
Since millimeter waves can't penetrate deeply into the body, they can't be used for purposes such as imaging internal organs.
But millimeter-wave imaging is cheaper, safer, less power-intensive and much more portable than other types of body imaging, making it especially attractive for potential medical uses where superficial imaging is the goal.
Tavassolian's innovation takes the technology to a new, more powerful level. By splitting millimeter-wave bandwidths into subcarriers (channels), then processing and recombining the slices, detailed medical images can be created for proactive diagnostic purposes.
"This has not been done before, to our knowledge, in medical imaging," she explains.
During the first stages of their research, Tavassolian and graduate assistant Amir Mirbeik are testing a dielectric probe inserted into skin tissue to determine how accurately it detects contrasts and electrical differences that can help differentiate between healthy and tumorous skin.
"Because forming tumors contain higher water content than healthy skin, this contrast should be viewable on a millimeter-wave-created image," Tavassolian says.
Her team is also designing specially configured sub-band antennas, each micro-fabricated in Stevens' own clean room and each uniquely tuned to the unique bandwidth at which it will operate. These antennas will transmit signals and record backscattered responses during the experimental imaging process.
Next, Tavassolian will begin receiving excised skin samples — of normal, healthy skin and also of confirmed tumors — several times weekly from Massachusetts General Hospital in Boston and Hackensack University Medical Center in New Jersey. Those samples will be transported to Stevens frozen, imaged over a period of a few hours each, then compared against each other.
A low-cost, portable imaging system for medical centers
If the system proves the concept is viable, Tavassolian and her team will then begin developing hardware and software prototypes for a portable, low-cost imaging system that can be deployed in medical centers. Tavassolian will also create a public display on the project for New Jersey's largest interactive science museum, Liberty Science Center in Jersey City, and develop new Stevens curriculum offerings in the biomedical applications of electromagnetics.
"It's all about greater impact," she says of her decision to focus on medical applications during her career as an electrical engineer. "I have always gravitated toward medical problems."
In addition to her NSF-supported work on imaging, Tavassolian performs additional Stevens research on radio frequency and microwave technologies, bio-electromagnetics and micro-electromechanical systems with biomedical applications, including a project to develop a heart-rate and blood-pressure monitoring system utilizing acoustic signals, radar, and accelerometer data.