Canon Medical has created a new level of cardiac imaging with Precise IQ Engine (PIQE) by leveraging the power of the Aquilion ONE / PRISM Edition CT scanner‘s whole-heart, single- rotation cardiac coverage, and enhancing it with the Super Resolution benefits from the Aquilion Precision Ultra-High Resolution (UHR) CT scanner.

PIQE builds from the foundation of Advanced intelligent Clear-IQ Engine (AiCE), the industry’s first Deep Learning Reconstruction algorithm for CT, and enhances cardiac image quality by using Deep Learning to bring the benefits of the Aquilion Precision Ultra-High Resolution CT scanner to the Aquilion ONE / PRISM Edition.

Powered by Altivity, PIQE takes deep intelligence to the next level for better cardiac imaging, with significantly reduced noise and dose, and improved contrast-to-noise ratio (CNR), all with no loss in low contrast detectability relative to conventional hybrid iterative reconstruction. The combination of improved spatial resolution, while maintaining low contrast detectability (LCD) helps improve visualisation of stents and coronary plaque, both calcified and non-calcified, providing a more confident diagnosis of coronary artery disease.

The PIQE Deep Learning Reconstruction algorithm is trained using data acquired on the Aquilion Precision CT system, which features UHR 0.25 mm to double high contrast signal definition. PIQE features a next generation, three-dimensional neural network that has been trained to identify and preserve signal features, both in-plane and longitudinally, throughout the heart. The algorithm is trained on high-quality cardiac cases that have been acquired on clinically operating, Aquilion Precision systems. PIQE’s three-dimensional learning also helps ensure continuity in reconstruction of small, longitudinal vessels, which are often obscured by conventional reconstruction algorithms.

The National Institutes of Health (NIH) in Maryland, USA - the world’s largest biomedical research facility - has been collaborating with Canon Medical for the last 12 years to improve CT technology and image quality. Dr. Marcus Chen, MD, Director of Cardiothoracic Imaging at the NIH worked closely with Canon Medical in the development of PIQE.

“With PIQE, it’s as if you are wearing glasses for the first time,” remarked Dr. Chen. “I can now see very fine detail. I can see small areas of calcification. I can see small stent struts.”

“What impresses me most about PIQE is its image quality. It has incredible sharpness and detail,” added Dr. Chen. “With PIQE, we are now able to see individual stent struts, whereas, on the traditional imaging, they just look blurred. With PIQE, we are now able to better visualise inside a stent and make a diagnosis whether or not a patient has in-stent stenosis. And the amount of blooming from calcified areas is minimised with PIQE reconstruction.”

The NIH considers the new tool to be so beneficial that it has integrated PIQE fully into the standard workflow for all cardiac examinations.

“It is completely integrated into our workflow,” said Dr. Chen. “The speed of reconstruction by PIQE allows for a smooth diagnostic process. And as I am able to see smaller structures more clearly, I know that I am making better diagnoses and delivering better healthcare for the patient.”

Aquilion ONE / PRISM Edition
Canon Medical’s Aquilion ONE / PRISM Edition offers One Beat Cardiac imaging: covering the whole heart in a single rotation via 320 detector rows of 0.5 mm thickness. The ability to image the heart in one 0.275 second rotation prevents misregistration and excludes artifacts caused by stitching or beat-to-beat variation.

Aquilion Precision
Canon Medical’s Aquilion Precision CT system, introduced in 2017, was engineered from the ground up to create Ultra-High Resolution CT images. The system features 1792 detector elements per row, double the number of a conventional system, resulting in twice the intrinsic in-plane spatial resolution of a conventional CT detector. In addition, each of the 160 detector rows along the z-direction is 0.25 mm thick, half that of a standard CT detector.

Dr. Kazuo Awai is Dean of the School of Medicine and Professor and Chairman Department of Diagnostic Radiology, at Hiroshima University, Japan. He has collaborated with Canon Medical Systems over many years on the development of many of its cutting-edge technologies, including Forward Projected Model-based Iterative Reconstruction SoluTion (FIRST), AiCE, Spectral, and now PIQE. Hiroshima University is a key collaboration site for the development of PIQE and has been the first field test site for the new technology.

“PIQE images use a high-resolution CT image as the target image for training, so it is a normal CT scan, but it is very unique in that it provides an image with very high resolution,” he remarked. “For example, stents placed in the coronary arteries of the heart can be visualised very clearly, and the calcium blooming artifact on the coronary arteries is also considerably reduced. From this, I personally expect that the ability to depict fat in soft plaque will be improved in the future.”

“What I have seen is that PIQE images have a very natural texture, and the noise appears to be greatly reduced. Noise, especially low-frequency noise, is strongly suppressed. Therefore, the image quality looks very natural,” added Dr. Awai. “When low-frequency noise is improved, the ability to detect lesions improves. So, in that sense, I think this technology will increase diagnostic ability with CT. It has the potential to improve not only spatial resolution but also contrast resolution.”

“In CT, the enemy of high resolution is noise,” said Dr. Zhou Yu, Director of CT R&D at Canon Medical Research, USA. “To minimise patient dose, we often have to implement a denoising algorithm that will also sacrifice spatial resolution. As a result, we rarely get the maximum resolution that the system can provide. With deep learning reconstruction, we have redefined this resolution and noise trade-off. In PIQE, conventional denoising has been replaced with deep learning denoising, allowing the inherent resolution of the system to be expressed.”

“Most people in the field today use 2D neural networks.
The challenge with 2D is that it is hard to differentiate small features, like small vessels, from noise in 2D. With a 3D neural network, the network has additional information from adjacent slices to differentiate these small features versus noise,” explained Dr. Yu.

With all of the technical ingredients put together, it is also important to incorporate clinical knowledge into our development.

“In the end, it is the clinician who reads the image and makes the diagnosis,” said Dr. Yu. “To do that, we work closely with clinicians to understand the clinical use case, the image quality preference, and to fine-tune our training target and training method to maximise the image quality.”

The development team are excited about PIQE technology and look forward to seeing it being used in the field.

“We see great potential for PIQE to improve the quality of cardiac imaging by reducing blooming artifacts from calcium and stents, improving visualisation of small vessels, and reducing noise,” added Dr. Yu. //