Deep Learning Reconstruction – a Game Changer in CT Imaging

Canon Medical’s Advanced intelligent Clear-IQ Engine (AiCE) is the world's first deep learning reconstruction software that produces high-quality CT images with ‘Patient dose up to 90% below UK National Diagnostic Reference Levels’. VISIONS spoke to Dr. Richard Hawkins, Consultant Radiologist at the Mid Cheshire Hospitals NHS Foundation Trust, UK, about its benefits in daily practice.
Deep Learning Reconstruction (DLR) innovaton, built into an Aquilion ONE / GENESIS Edition CT scanner is delivering a ‘new era’ of patient imaging at Mid Cheshire Hospitals NHS Foundation Trust, UK.

Patient dose reductions of up to 90% below National Diagnostic Reference Levels (NDRLs) have been achieved at Leighton Hospital in Crewe by integrating into its new Canon CT scanner a Deep Learning Reconstruction AI algorithm, Advanced intelligent Clear-IQ Engine (AiCE) powered by Altivity, Canon Medical’s new AI brand. The low doses have even been achieved when examining traditionally difficult to image patients such as severely ill patients with their arms by their sides, patients unable to hold their breath and bariatric patients.

“It enables phenomenal patient dose reduction, up to 90% below the National Diagnostic Reference Levels.”

Dr. Richard Hawkins, Consultant Radiologist,
Mid Cheshire Hospitals NHS Foundation Trust, United Kingdom.
“Advanced Deep Learning of clinical image reconstruction using AiCE heralds a new era in CT. It enables phenomenal patient dose reduction, up to 90% below the National Diagnostic Reference Levels, at the same time as providing extremely high-quality clinical images and all in a rapid timeframe suitable for everyday clinical use.
This goes far beyond model-based iterative reconstruction on CT and as Canon Medical was first to innovate in this area, it offers the most mature system of this kind,” states Dr. Richard Hawkins, Consultant Radiologist at Mid Cheshire Hospitals NHS Foundation Trust.

He continues, “The clinical images generated using AiCE are much more natural and acceptable in appearance to radiologists reporting on cases.
Previously, with model-based iterative reconstruction, the images looked as if they had been painted with watercolors. This is a great improvement – once you see them you’ll never look back. As a department we have always been very proactive when it comes to keeping patient dose down and our experiences of using the system every day for inpatients and outpatients has exceeded our expectations. AiCE is a game changer for radiology.”

“Advancements in Artificial Intelligence to further the capacity and capabilities of radiology are very exciting. It isn’t theory or pilot studies, it is real and being used in the UK by busy NHS hospitals to power the improvement in patient care, speed-up processes and empower clinical confidence,” states Mark Thomas, CT Modality Manager at Canon Medical. “AiCE is trained using a deep learning algorithm to differentiate ‘noise’ from true signal, reducing distortions, preserving edges and maintaining details in image outputs at the same time as achieving lower doses than ever seen before in routine CT imaging.” //
Leighton Hospital, part of Mid Cheshire Hospitals NHS Foundation Trust, UK, is one of the first NHS hospitals using Advanced intelligent Clear-IQ Engine (AiCE), a Deep Learning Reconstruction AI algorithm on its Aquilion ONE / GENESIS Edition CT scanner. [Picture taken pre-COVID-19]

Pictured L to R: (Rear) Tamzin Culverhouse, Medical Imaging Assistant; Alex Finnie, Senior Radiographer; Matt Simpson, Consultant Radiologist; Barnaby Harrison, Account Manager at Canon Medical Systems UK. (Front) Justin Edwards, Advanced Radiographer Practitioner; Sophie Vaux, Senior Radiographer; Dr. Richard Hawkins, Consultant Radiologist; and Mark Thompson, Medical Imaging Assistant.
Interested in the scientific paper:
The future of CT: deep learning reconstruction by C.M. McLeavy, M.H. Chunara, R.J. Gravell, A. Rauf, A. Cushnie, C. Staley Talbot, and R.M. Hawkins? Scan the QR code:
sou
DISCLAIMER: The clinical results described in this paper are the experience of the author. Results may vary due to clinical setting, patient presentation and other factors.