Sometimes scientific progress comes in the form of the discovery of something completely new. Other times, progress comes down to doing something better, faster, or easier. Caltech's improved photoacoustic imaging technology PACTER simplifies operating procedures, achieves three-dimensional imaging, and reduces operating complexity, marking a major advancement in the field of medical imaging.
The latest research from the laboratory of Leehom Wang, Bren Professor of Medical Engineering and Electrical Engineering at Caltech, falls into the latter category. In a paper published in the journal Nature Biomedical Engineering, Leehom Wang and postdoctoral scholar Zhang Yide show how they simplified and improved an imaging technique they first announced in 2020.
The technology, a photoacoustic imaging technique called PATER (Photoacoustic Topography via Polarity Relay), is a specialty of Wang Jianmin's research group.
Improvements in photoacoustic imaging technology
In photoacoustic imaging, laser pulses enter tissue and are absorbed by tissue molecules, causing the molecules to vibrate. Each vibrating molecule is a source of ultrasound waves, which can be used to image internal structures in a manner similar to ultrasound imaging.
However, photoacoustic imaging is technically challenging because it produces all imaging information in a short time. To capture this information, early versions of Wang's photoacoustic imaging technology required arrays of hundreds of sensors (transducers) pressed against the surface of the tissue being imaged, making the technology complex and expensive.
Wang and Zhang reduced the number of sensors needed by using a device called a "Maiji relay," which slows the flow of information (in the form of vibrations) into the sensor. As previous reports on PATER explained:
In computing, there are two main methods of data transfer: serial and parallel. In serial transmission, data is sent over a communication channel in a single data stream. In parallel transmission, multiple data are sent simultaneously through multiple communication channels.
These two methods of communication are roughly similar to how cash registers are used in stores. Serial communication is like a cash register. Everyone is in the same line and sees the same cashier. Parallel communication is like having several cash registers, each with one line.
The system Wang designed with 512 sensors is similar to a store with many cash registers. All sensors work simultaneously, and each sensor receives partial data from the ultrasonic vibrations generated by the laser pulses.
Because the ultrasonic vibrations emitted by the system are generated in a short period of time, a single sensor would be overwhelmed if all the data were to be collected in such a short period of time. This is where the Magellanic relay comes in.
As Leehom Wang describes it, an ergodic repeater is a cavity that allows sound to reverberate around it. As ultrasonic vibrations pass through the ergodic repeater, they are stretched in time. Going back to the cash register analogy, this would be like having another employee assist a single cashier by telling customers to walk around the store until the cashier is ready to handle them, so the cashier isn't scrambling.
PACTER: next steps
The latest version of this technology, called PACTER (PhotoacousticComputedTomographyThroughanErgodicRelay), goes a step further, allowing the system to operate with a single sensor and, through the use of software, collect as much data as 6,400 sensors.
Wang, who is also the Andrew and Peggy Cherng Chair in Medical Engineering Leadership and executive officer of Medical Engineering, said PACTER improves on PATER in two other ways.
One of the improvements is that PACTER can generate three-dimensional images, while PATER can only generate two-dimensional images. This is due to the development of improved software.
"The transition to 3D imaging has greatly increased data requirements. The challenge we face is how to transmit the massively increased data through a single sensor," Zhang said. "We found the solution by changing our approach. Instead of directly employing computationally intensive methods to reconstruct 3D images from single sensor data, we first expanded one sensor into thousands of virtual sensors. This idea simplifies the process of 3D image reconstruction, bringing it closer to our traditional methods of photoacoustic imaging."
Second, unlike PATER, PACTER does not require calibration every time it is used.
"With PATER, we have to calibrate it every time we use it, and that's unrealistic. We get rid of this one-time calibration every time we use it," Wang said. Calibration is required because when the system fires laser pulses into tissue, the pulse's "echo" bounces off the transducer, making it unable to sense direct ultrasound information. PACTER solves this problem by adding a delay line to the system. The delay line forces the echo to take a longer physical path on its way back to the transducer so that it reaches the transducer after receiving the direct ultrasound information.
The paper describing the work, "Ultrafast longitudinal imaging of hemodynamics with single-shot volume photoacoustic tomography using single-element detectors," appears in the Nov. 30 issue of Nature Biomedical Engineering. Co-authors of the paper include: Hu Peng (Ph.D. 23 years), a former medical engineering graduate student; Li Lei (Ph.D. 19 years), a former medical engineering postdoctoral fellow; Cao Rui, a medical engineering postdoctoral fellow; Anjul Khadria, a former medical engineering postdoctoral fellow; Konstantin Maslov, a former Caltech staff scientist; Tong Xin, a medical engineering graduate student; and Zeng Yushun, Jiang Laiming, and Zhou Qifa from the University of Southern California.
Research funding was provided by the National Institutes of Health.
Compiled source: ScitechDaily