Ultrasound imaging is widely used in clinical practice for its low cost, high resolution, high frame rate and portability. <br />In the last decade, the introduction of ultrafast imaging has enabled new acquisition modes such as shear wave elastography, 3D imaging systems have enabled volume acquisition but at much lower framerates. <br /><br />We believe ultrafast 4D ultrasound imaging is key issue for the development of new diagnostic applications in the next decades for the benefit of patients.
If a fully sampled matrix probe with high channel count approach remains the gold standard technique for 3D imaging in research and ultrafast 4D ultrasound imaging, it however, has several limitations for clinical applications:<br />- The full 2D matrix array and the 1024 channels scanner are expensive to build and the market has not yet been clearly identified to justify the extra-cost of this solution over conventional ultrafast scanner such as the Aixplorer ™ having only 256 channels. <br />- Scaling to larger number of elements is technically difficult. For number of channels higher than 1024 (32x32), it is technically very challenging to build the array and inherently connecting the full transducer population to the system. The cable is therefore becoming a limiting factor for portability and ergonomics of the probe. <br />- Increasing the transducer frequency to higher than 8 MHz is another technical challenge to take up since the width of individual elements of probe is reversely proportional. <br /><br />For all these reasons, Ultrafast4D project is proposing an alternative approach suitable to address the technical and economical challenges while maintaining our goal of ultrafast volume rate at high quality.
In coherent compounding as used in ultrafast imaging, plane waves or diverging waves form the natural basis to enforce synthetic focusing everywhere in the image. Since plane waves or diverging waves are so well suited to the focusing problem, a strong reduction in the number of transmitters is achievable and high frame rate is possible. In this context, we believe that the geometry of row-column transducers is particularly well suited to plane waves and could yield both good imaging quality and high volume rate provided that ultrafast scanner is used together with the coherent compounding technique.
The aim of the project is to combine plane wave coherent compounding with the row-column transducer geometry. Row-column transducers can only by addressed by full rows or full columns instead of individual elements which dramatically reduce the output cables and electronic channels from N² to 2N. They are thus perfectly suited to emit and receive plane waves or cylindrical waves either vertically or horizontally.
If row-column geometry has been studied for conventional focusing at low frame rate, synthetic aperture at even lower frame rate, individual elements addressing based on CMUT (Capacity Micromachined Ultrasonic Transducer) non-linearities, and plane wave emission with a slow row-by-row reception scheme, it has never been applied to its full potential by combining coherent plane wave compounding together with full-parallel receive using an ultrafast scanner.
The aim of the project is to investigate this well suited imaging scheme with a strong focus on in vivo applications.
We demonstrated the advantage and limitations of the proposed scheme compared to fully populated matrix.
Theory, simulations and emulation using a fully populated matrix were performed to help design a prototype and adapted ultrafast sequences.
The first prototype was built and tested in vitro and in vivo allowing to perform 3D ultrasound and Power Doppler reconstruction with good quality.
The remaining of the projet is dedicated to integration on the Aixplorer platform of two prototypes and further in vitro and in vivo investigations of the different probe prototypes.
Ultrafast 4D ultrasound associated with affordable conventional ultrafast scanner has a huge clinical potential. We can already list clinical applications that could benefit from this approach:
- In cardiology , real time 4D Doppler of all heart chambers in a single cardiac cycle. 3D imaging of this highly dynamic organ at ultrafast framerate will bring echocardiography to the next level.
- In radiology, we expect that the combination of Doppler, shear wave elastography in full 3D on a conventional ultrafast scanner could change the way screening and monitoring is currently done. Ultrasound in radiology would be able to better compete with MRI and CT with high quality quantitative volumetric data for a fraction of the cost.
A paper on the proposed scheme on an emulated Row-Column array has been published in Physics In Medicine & Biology. First in vitro and in vivo images with the concept were demonstrated.
4D in vivo ultrafast ultrasound imaging using a row-column addressed matrix and coherently-compounded orthogonal plane waves.
Flesch, M., M. Pernot, J. Provost, G. Ferin, A. Nguyen-Dinh, M. Tanter, and T. Deffieux. Physics in Medicine and Biology 62, no. 11 (2017) : 4571–4588.
Ultrasound imaging dates back to the seventies where it was introduced as a real-time imaging modality into clinics. Since then, ultrasound imaging has become a major imaging modality, mostly due to its real-time capabilities but also due to its versatility, portability and low cost compared to other imaging modalities. Ultrasound imaging is mainly used for screening, diagnosis and monitoring in many fields such as radiology cardiology, gynecology, gastroenterology, obstetrics, neonatology, neurology, ophthalmology, angiology.
As a result of increasing life expectancy and declining fertility rates, the proportion of the population aged over 60 years is growing faster than any other age group. Consequently, the pressure to maintain the health and quality of life in ageing populations will continue to be a major challenge for healthcare systems, a challenge where ultrasound imaging as still a very important role to play.
In the last decade, the Langevin Institute and SuperSonic Imagine have been at the core of the ultrafast ultrasound imaging revolution with the development of ultrafast scanners with hundreds of acquisition channels. In the new architecture, most of the imaging logic is implemented directly in software rather than in hardware. The flexibility and high quantity of acquired data available has enabled new imaging strategies based on synthetic focusing of the ultrasound field such as coherent plane wave compounding. This fundamental shift in the way of creating ultrasonic images has allowed the development of new imaging modalities which can reach thousands of frames per second with high quality. Ultrafast imaging is the key to drastically improve sensitivity of several imaging modes such as Doppler imaging but also enables the quantification of new highly promising features such as tissue elasticity based on shear wave imaging.
Recently the introduction of 3D ultrasound imaging on high-end scanner has opened the possibilities to reduce operator dependence of the images by acquiring a full volume directly. 3D ultrasound imaging has seen a lot of developments in the last decade with new technology such as electronic beamforming integrated in the probe handle allowing real time framerate of dozens of volume per second, still far below ultrafast volumerate or required for Doppler imaging. Despite this limitation, 3D imaging sparkled strong interest in cardiology research where it is crucial to follow the heart mechanics in 3D.
We believe that the combination of ultrafast and volumetric imaging will be the key for the development of new diagnostic applications in the next decades. As such, we recently demonstrated in vivo 3D shear wave elastography and 4D ultrasensitive Doppler thanks to our unique 4D ultrafast prototype with 1024 channels and matrix array built by Vermon.
With the Ultrafast4D project, we believe that those breakthroughs can be combined in a single solution, ready to move clinical imaging even further : new highly promising quantitative features, no compromise high frame rate volume imaging, affordable package.
The main objective of the Ultrafast4D project is to allow similar 4D imaging capabilities using available commercial ultrafast scanners (128-256 channels) thanks to smarter probe designs and novel imaging sequence schemes for a mucher lower cost.
The project will be mainly focused on low cost 4D ultrafast cardiac imaging - a key differentiator for the involved companies - from prototype fabrication to in vivo cardiac volumic ultrafast acquisitions on healthy volunteers. The project also aims to explore other preclinical applications at higher frequencies for vascular imaging and preclinical functional ultrasound.
Monsieur Thomas Deffieux (Inserm U979 - Institut Langevin)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Inserm Inserm U979 - Institut Langevin
SSI SUPER SONIC IMAGINE
Help of the ANR 725,079 euros
Beginning and duration of the scientific project: December 2015 - 36 Months