Since Roentgen’s accidental discovery of X-rays in 1895, clinicians have adopted many techniques to look inside the human body without surgery or other invasive procedure. After more than 100 years, medical imaging is an essential part of modern medical diagnosis. A range of techniques, including X-ray imaging, magnetic resonance imaging, ultrasound imaging, and nuclear medicine imaging have shown their value. One technique, ultrasound imaging, is widely used because it is non-invasive, low-cost, safe and convenient. However, Professor Richard Prager believes it has not yet reached its full potential.
Professor Richard Prager is head of the Engineering Department of the University of Cambridge. He recently spoke to us to about ultrasound imaging and how the Centre is developing new approaches. He discussed the project he is involved in with Dr Nghia Nguyen, "Multi-modality and Hybrid 3-D Ultrasound/Photoacoustic Imaging System", which aims to design and develop new reconstruction methods for the next generation of medical ultrasound imaging systems.
During his PhD in Cambridge in the 1980s, he focused on automatic speech recognition using neural network algorithms. Recognizing the potential of these techniques in a range of pattern recognition tasks, he applied them to medicine. Since then he has focused on ultrasound imaging research, improving imaging resolution from an engineering perspective. His work on the development of the freehand 3D ultrasound systems Stradx and Stradwin showed some of the potential for using 3D ultrasound in clinical diagnosis and treatment planning. In recognition of his work on 3D ultrasound and advancing engineering education, the Royal Academy of Engineering elected him to its Fellowship in 2017.
The new project has two parts. The first aims to encourage the adoption of 3-dimensional ultrasound imaging. The second part explores the potential of a new imaging technique based on “photoacoustics”.
3-Dimensional Ultrasound Imaging
This part of the project takes a technique previously developed in Cambridge and assesses its use in a clinical setting in collaboration with Vinno, a Jiangsu-based manufacturer of clinical ultrasound equipment. The new beamforming algorithm, developed by Dr Nguyen and Prof. Prager, provides a software technique that improves the resolution and contrast of lower cost ultrasound scanning equipment, enabling more patients to benefit from ultrasound diagnosis.
In general, current medical ultrasound scanners create a 2-dimensional image, a virtual slice through the human body. 3D clinical ultrasound means the clinician can see the whole of a patient’s liver or other organ, giving new insights and hence better, faster diagnosis. Other imaging techniques such as X-ray computed tomography (CT) and magnetic resonance imaging (MRI) already allow this, but they require more expensive equipment and with CT there is more risk to the patient because of the use of ionizing radiation. 3D ultrasound imaging combines the advantages of ultrasound imaging- cheap, safe, fast and portable- with the diagnostic power of more expensive approaches. Many people could benefit from this improvement, for example with examinations taking place in local community clinics rather than large (and expensive) hospitals. With appropriate training of staff, clinics in less developed regions and countries could also perform the examinations. Combining smart software, such as the “coherent pixel-based beamforming algorithm”, with reliable, low-cost hardware is a useful first step.
Hybrid photoacoustic imaging
Photoacoustic imaging uses complex signal processing techniques to analyze the tiny sound created when a cell absorbs a short laser pulse. The sound is so quiet that it needs a complex model of how the sound travels through the human body to interpret. Working in collaboration with local universities, the project will simulate this new type of medical image scanner to explore the potential of the approach.
Professor Prager described the goal of a hybrid approach combining the advantages of photoacoustic imaging and ultrasound imaging. Traditional ultrasound imaging is based on echo scanning, using the reflection or scattering of ultrasound. Ultrasound can travel long distances in soft biological tissues. This allows highly detailed images of anatomical structures inside the body. In other words, they can give very accurate information on the shape, size and location of objects. However, the approach cannot provide other vital information such as biochemical characteristics of the tissue or the oxygen level in the blood. Optical techniques can give this information but are limited to tissue near the skin unless invasive procedures are used.
The hybrid approach aims to give a picture combining structure, motion and biochemical composition in a single test. Professor Prager described the potential. "The hybrid photoacoustic/ultrasound imaging system being researched in the Centre may be able to provide deeper penetration or higher resolution than other functional imaging approaches; it may provide high-resolution anatomical/functional images at the same time, and show changes in the mechanical properties of tissues in the body”.
"The transducers for ultrasound and photoacoustic imaging are also easy to integrate, which reduces the cost of combining these two technologies. The hybrid approach could be used for cancer detection and neuroscience, for example."
There is currently a great deal of speculation that Artificial Intelligence, and in particular Machine Learning, could make a significant contribution to medicine. Professor Prager commented on the opportunities: some developments could profoundly change clinical practice, but this depends on the complex relationships between physicians and technology. Technology will support physicians rather than replace them. The physicians must understand the technology before they can trust it. It is not yet clear how this trust will develop.
Professor Prager emphasized the importance of relationships in all aspects of engineering research. A long-time advocate for collaboration, he described the essence of a productive relationship between scientific research and industrial innovation--communication, trust and time. "Trust can only be built over time, so long-term relationships tend to be more reliable, and more productive."
Using the University of Cambridge’s Department of Engineering as an example, Professor Prager described the long-term partnerships with major companies such as Rolls-Royce and Mitsubishi Heavy Industries. Such partnerships enables researchers to work on urgent problems faced by industry, and to engage in long-term fundamental research; "Effective industry-university interaction can be built around academic research that has both immediate application value and long term benefits. The two can be coupled."
The Centre’s ultrasound project follows this proven approach. In cooperation with local industry, the project will address the immediate needs to improve existing medical equipment and, in the longer term, to develop new approaches to clinical imaging in collaboration with Nanjing universities.
Commercial products using these new technologies could both improve clinical care and provide economic benefits to the region. Only by engaging with clinical practice can new technologies deliver clinical benefits. "From my experience, research groups that maintain good relationships with industry and other users can usually do high-quality research," Professor Prager concluded.
Early results from the project are promising. Initial investigations in collaboration with Nanjing University into the hybrid photoacoustic-ultrasonic imaging technique have been published in Optics Letters, an authoritative journal in optical research. Algorithms developed by Prager and Nguyen are in trials with ultrasound equipment manufacturer Vinno Technology (Suzhou) Ltd.
As well as working with Dr Nguyen on this research project, Professor Richard Prager has been deeply involved in the Centre’s management since its creation. He expressed his confidence in the Centre’s approach. “Cambridge and Nanjing have a relationship built on trust. It is the foundation of the Centre’s programme of high quality scientific research and industrial innovation”.