'Like a panorama, no stitching required': POSTECH develops integrated photoacoustic-ultrasound imaging system

A research team led by Kim Cheol-hong — a professor in the departments of electrical engineering, IT convergence and mechanical engineering, and the Graduate School of Convergence at POSTECH (Pohang University of Science and Technology) — has developed a three-dimensional integrated photoacoustic-ultrasound imaging system capable of capturing wide areas of the body, such as the hands and feet, in a single scan. The work was carried out in collaboration with the ultrasound probe research team at Siemens Healthineers.

Combined photoacoustic-ultrasound imaging is a diagnostic technology that visualizes internal body structures and vascular conditions simultaneously. It merges photoacoustic technology, which detects faint sounds generated by blood vessels, with ultrasound technology, which maps internal structures by emitting sound waves and reading the reflected signals. The challenge has long been that imaging wider areas requires equipment that grows exponentially more expensive and complex — forcing clinicians to capture multiple images in segments and stitch them together, a process prone to distortion.

The research team addressed this by developing a technique called digital multiplexing, which rapidly switches individual transducer elements on and off in sequential segments. Using a probe equipped with 768 elements — and an FPGA semiconductor chip that precisely controls the sequence and timing of those activations — the system maintains image clarity and resolution while capturing areas that previously required multiple passes, all in a single scan with no stitching.

Using the system, the research team scanned the palm, back of the hand, forearm and foot in a single pass, producing sharp three-dimensional maps of fine blood vessels. Conventional handheld ultrasound probes carry roughly 128 to 192 elements and can image a width of only about 38 millimeters at a time; the new probe, with its 768 elements, extends that coverage to 154 millimeters in a single scan. The team also fired multiple wavelengths of light to calculate blood oxygen levels numerically and verified the accuracy of those readings against measurements from a clinical pulse oximeter.

Until now, clinicians had to accept a degree of error inherent in capturing images in segments and stitching them together. If adopted in clinical settings, the technology would yield accurate diagnostic information in a single pass. It is expected to be particularly valuable for diabetic patients who require detailed monitoring of blood circulation, as well as for patients whose recovery of blood flow to burned or wounded tissue needs to be tracked over time.

"This system can serve as a core diagnostic platform for the early detection and prognosis management of metabolic and vascular diseases — such as peripheral artery disease and diabetic foot — that require real-time monitoring of blood flow and functional tissue changes across wide areas," Kim said. "We plan to continue research aimed at validating clinical efficacy across diverse patient groups on a large scale, and at optimizing real-time image processing speeds, with an eye toward practical application."

The findings were published July 8 in Laser & Photonics Reviews, an international journal in the field of optics.