Medical Imaging Turns To Oft-Neglected Part Of Light Spectrum

The slice of the electromagnetic spectrum sandwiched between the infrared and microwave regions used to be known as the “terahertz gap” because scientists lacked the tools to access it. Not anymore.

“In the last several years, terahertz spectroscopy has really found its footing,” says Yale University’s Charles A. Schmuttenmaer, one of the field’s pioneers.

Previously, he explains, the difficulty in generating and detecting terahertz light restricted scientists’ ability to explore ways to use the light to study matter. Now, thanks to advances in instrumentation, commercial terahertz equipment has become widely available, and that has spurred development of terahertz imaging for cancer diagnosis and other biomedical applications.

Before these recent developments, only a limited number of scientists worked with light in the terahertz region (1 THz = 10¹² Hz), also known as far-infrared radiation. For decades, they used this portion of the electromagnetic spectrum and the nearby microwave region to interrogate molecules for studies in spectroscopy, astronomy, and other areas.

The recent advent of affordable, fast-laser-based systems for generating and detecting THz light has led to nearly exponential growth in the number of research papers in the THz field, says Emma Pickwell-MacPherson. Those papers detail “a plethora of potential applications,” says Pickwell-MacPherson, an engineering professor at the Chinese University of Hong Kong and a specialist in THz imaging of biological tissue.

THz light interacts with matter differently than other types of radiation. Unlike the familiar effects of infrared light, for example, which include inducing bending and stretching motions of water and other molecules, THz light causes collective motions of groups of water molecules—water cluster dances, in effect.

Water absorbs THz light quite strongly. That property means that THz spectroscopy is highly sensitive to changes in the water content of materials. And because skin and other tissues are full of water, that sensitivity can be exploited for biomedical imaging applications.

For example, as Pickwell-MacPherson explains, healthy tissues and tumors differ in terms of how they interact with THz light based on differences in their water and blood concentrations, tissue structures, and other factors that alter their optical properties. So if a THz beam illuminates a patch of skin containing normal and cancerous regions, those differences can be discerned by analyzing the light reflected by the skin. Because water absorbs THz light so strongly, a tumor can be detected even when the differences between it and healthy tissue are subtle.

“That sensitivity can be a blessing and a curse,” Pickwell-MacPherson says, because any small change will strongly affect the signal, even one caused by an experimental artifact. As a result, researchers must take pains to control humidity, temperature, and other variables. She acknowledges that the sensitivity to experimental conditions has led to some debate about the practicality of this imaging strategy, which has slowed its adoption.

Still, THz-based imaging offers several advantages over traditional medical imaging techniques. Those benefits are driving scientists in academia and industry to push on with development despite the limitations. The biggest driver is safety. Unlike X-rays, which are used in various ways to image tumors, THz light does not ionize biological material and thus bypasses numerous health concerns.

In addition, as Joo-Hiuk Son, a physics professor at South Korea’s University of Seoul, points out, THz imaging has an advantage over magnetic resonance imaging. MRI excels at imaging deep tumors but struggles with ones at the surface. In contrast, THz light cannot penetrate deeply into soft tissue. Therefore it may be better suited than MRI to image carcinomas, a large category of common types of cancer, which includes most breast cancers that develop just below the skin and malignancies or outer layer of organ tissue.

Terahertz Spectroscopy 101

Light with submillimeter wavelengths and a frequency range of roughly 0.1 to 10 THz, or 3 to 300 cm^–1, is known as ­terahertz radiation. It can penetrate plastics, paper, and textiles, but it is absorbed strongly by water, making it a sensitive probe of biological tissue. Unlike the relatively high-frequency stretching and bending motions that infrared light induces in individual water molecules, THz light causes groups of ­water molecules to coalesce and disassemble repeatedly.

A number of studies conducted around 2005 by researchers at TeraView, a THz instrument maker in Cambridge, England, demonstrated the benefits of THz imaging techniques in detecting human breast cancer. One study, conducted on tissue samples removed during surgery, showed that THz imaging was sensitive enough to pinpoint tumors that did not contain calcium deposits. Noncalcified tumors are easy to miss because they do not show up in X-ray scans and often cannot be detected in manual breast exams.

The researchers also showed that THz techniques can distinguish the small but significant differences in the refractive index and THz absorption properties of adipose, fibrous, and tumor tissue in breasts. This distinction could aid oncologists in determining the most suitable treatment option (Opt. Express 2009, DOI: 10.1364/oe.17.012444).

Another potential advantage of THz imaging compared with standard medical imaging methods is the relatively small size of THz instruments. Unlike MRI systems and X-ray and computed tomography (CT) scanners, THz imaging systems can be small and portable and therefore used directly in the operating room, says Sir Michael Pepper, chief scientific director of TeraView, which is building and testing such devices.

TeraView’s portable instruments feature a handheld probe that’s connected via optical fibers and electrical cables to a small central unit that houses an ultrafast laser. Similar probes, including ones based on Raman spectroscopy and mass spectrometry, are also being developed to identify the borders of cancerous tissue during surgery.

Pepper points out that delineating this boundary is critical for a surgery’s success. For example, in breast cancer cases, surgeons try to remove as little healthy tissue as possible to ensure a patient’s quality of life is as high as possible after the procedure. But more than 25% of breast cancer patients need to undergo follow-up surgery because post-surgery pathology tests show that surgeons don’t excise all of the tumor the first time.

The objective is to reduce the number of follow-up surgeries by using the handheld THz scanner to more precisely identify the border between healthy and cancerous tissue, Pepper says. Improved precision would improve clinical outcomes and reduce costs.

Vincent Wallace, a physicist at the University of Western Australia, together with TeraView researchers and other scientists is test-driving the portable THz scanner in a clinical trial at Guy’s Hospital in London. The researchers used the portable scanner to analyze 46 freshly excised breast tissue samples from 30 patients. By comparing the results of the THz analyses to those of standard pathology methods, they conclude that the technique accurately recognizes and distinguishes between benign and malignant tissue in an ex vivo setting (Eur. J. Surg. Oncol. 2015, DOI: 10.1016/j.ejso.2015.08.036).

“We are continuing to develop classification algorithms to differentiate between tumor and normal tissues from freshly ­excised samples,” Wallace says. Pepper adds that the goal is to collect enough data to build confidence that the technology will not give false-negative results.

“False positives, one can live with,” Pepper says. “But a false negative, meaning incorrectly identifying a cancerous region as healthy, that could be deadly.” After the development work has been completed, the team plans to move the technology into the operating room for tests during surgeries, as well as to extend the trials to a hospital in Perth, Western Australia.

Arnie D. Purushotham is a surgeon and professor of breast cancer at King’s College London who is familiar with the tests involving the handheld THz device. He notes that using the device to scan a patient during surgery is more challenging than using it to analyze excised tissues. He says, “if after refining the device, it can identify with a high level of reliability the presence or absence of cancerous cells at the edges of a specimen during removal of cancerous tissue, then there is a huge likelihood it will be applicable to patient care.”

Researchers have also been making inroads using THz imaging to analyze other types of malignancies. At the University of Seoul, for example, Son and coworkers showed that the technique can readily distinguish healthy mouth tissue from oral cancer, pinpointing the locations, shapes, and sizes of tumors. The study also showed that contrast is sharper when THz imaging is conducted at –20 °C instead of 20 °C. The team attributed the sharp contrast to differences in the structures of normal and diseased cells at cold temperatures, not the effects of water, which were reduced by freezing the samples (Biomed. Opt. Express 2013, DOI: 10.1364/boe.4.001413).

In a related study, Son and coworkers showed that THz imaging can also be used to detect cancer in freshly excised brain tissue. The THz imaging data closely match results of MRI and visual analyses. According to Son, those findings futher support surgeons’ use of a compact THz imaging system to determine tumor boundaries during an operation (Biomed. Opt. Express 2014, DOI: 10.1364/boe.5.002837).

Despite these advances, THz imaging for biomedical applications still faces big challenges. Perhaps the biggest one is the strong attenuation of THz light by water. Because water strongly absorbs THz light, a THz beam can’t penetrate deeply into biological tissue.

One possible way around that problem would be to replace some of the water in a tissue sample with a material that is a weaker THz absorber. In a proof-of-concept study, Son and coworkers showed that glycerol might do the job. It is biocompatible, absorbed by tissue, and has a THz absorption coefficient that is smaller than water’s. The study showed that infusing glycerol into a sample doubled the intensity of the THz signal compared with an untreated sample (Opt. Express 2013, DOI: 10.1364/oe.21.021299).

The team also tested an approach to boost contrast between healthy and cancerous tissues. High image contrast is another requirement for extending the reach of THz imaging. Specifically, the researchers showed that by functionalizing gold nanoparticles with antibodies that bind to cancer cells, the particles collected around tumors and strongly reflected THz light, selectively enhancing those tissues’ contrast.

THz imaging is a fertile area of science that’s growing quickly. It’s also one that seems to attract conservative practitioners. None of the scientists contacted by C&EN describe the bioimaging tool as the best thing since sliced bread. Pickwell-MacPherson, for example, is “cautiously optimistic” about the technique’s future.

Wallace takes a similar position. “The field of medical imaging using THz technology is very much in its infancy, and there is a great deal of work yet to be done to show its true potential.”

But the potential payoff could be big. “If there is even a 1% chance that I could get this technique to work well,” Pickwell-MacPherson says, “and thereby reduce the death rate due to breast cancer or skin cancer, then I think it’s worth doing.”