Ultrasound Imaging Adopts Optical Tools
Using principles from blending light patterns, researchers double the resolution of sound images.
Optical scientists have long known how enlightening interference patterns can be, in disciplines ranging from X-ray crystallography to telescope lens polishing. They result from the meeting of two wave forms. Throw two similar pebbles into a pond near each other, and the ripples from each will interfere with the other. That interference is often dubbed “the moiré effect” after its similarity to a type of silk fabric.
Biomedical engineers at the University of California, Davis have developed a way to improve resolution in ultrasonography by capitalizing on interference between sound waves. While ultrasound is widely used because of its safety, it suffers from poor resolution, artifacts, poor contrast and low signal-to-noise ratios. Standard ultrasound also has difficulty resolving features that are smaller than about half the wavelength of the sound used -- a restriction called the diffraction limit.
A currently available technique called structured illumination microscopy has improved resolution in optical imaging, and the UC Davis researchers adapted it for sound waves.
"When a diffraction limited pattern is superimposed on a target that has frequency components beyond the diffraction limit,” said lead author Tali Ilovitsh, “the subsequent interference pattern downshifts the high frequency components to below diffraction limit frequencies that are then captured by the transducer."
This works as a proxy for a larger aperture in the image receiver and thus sharpens the image. However, the optical technique is only useful at shallow tissue depths down to about 100 microns, said Hari Shroff, a senior investigator at the National Institute of Bioimaging and Biomedical Engineering in Bethesda, Maryland, who specializes in optical imaging techniques and was not involved in the study. Ultrasound typically images much deeper tissues, but because sound and light waves behave similarly, the optical diffraction limit workaround can also be used for acoustic signals.
The UC Davis team has taken advantage of the programmable capability of modern ultrasound machines, which allows the sonographer to design very specific sound patterns. The researchers created an algorithm for an “optically inspired beam shape,” said Ilovitsh. “With the transducer we transmit a wave form that will generate the pattern we want at the specific location of the target we want to image. Then we capture the echoes with the transducer and do post-processing to recover features that have higher resolution.”
The researchers used a pattern with five foci. Shifting the pattern in time and space returned five images to the transducer, and when the resulting data was assembled by software, normally out-of-focus tissue features were visible. The experimental targets included a wire, a multipurpose ultrasound imaging phantom, and a rat’s paw.
Shroff called the technique “intriguing,” but questioned whether a doubling of resolution represents a significant improvement. “The real question for the impact will be how much a factor of two” improves image quality, he added. “In optical microscopy it’s incredible how much you can learn with just a bit more,” he said, so “it’ll be interesting to see what they do with it.”
The acoustic method has two big advantages: It can be used with existing programmable ultrasound machines with the right adjustments, and it produces images in real time. “It is simple and effective,” said Kai Melde, who leads an acoustic holography project at the Max Planck Institute for Intelligent Systems in Stuttgart, Germany, and was not involved in the study.
“It’s exciting because it’s an extension of some of the interesting work that’s been done in optics leading to super-resolution techniques,” said principal investigator Kathy Ferrara. “It’s a new way to implement ultrasound.” The study authors foresee their method assisting in diagnosis of many medical conditions, including arthritis, cardiac abnormalities, and early detection of small malignant lesions and their borders. They published their findings in January in the journal Communications Biology.