Canon Medical Forges New Ground with Deep Convolutional Neural Network Image Reconstruction for CT

Vienna, Austria, 27 February 2019 – Building on its advanced imaging technologies Canon Medical Systems introduces AiCE, a Deep Convolutional Neural Network (DCNN) image reconstruction technology for CT. AiCE, Advanced Intelligent Clear-IQ Engine, uses Deep Learning technology to differentiate signal from noise so that it removes noise while it preserves true signal.

With the AiCE Deep Learning approach, the DCNN is trained in the factory using perfect High-quality target data from real patient datasets. This patient data is extensively processed with advanced MBIR, Model Based Iterative Reconstruction, which provides optimal image quality and improved spatial resolution for groundbreaking results.

Following training and validation, the AiCE DCNN is then implemented into the CT scanner that allows for reconstruction speeds fast enough for busy clinical environments.

“AiCE quickly produces stunning CT images that are exceptionally detailed and with the low-noise properties that benefits all our patients for a faster and even better diagnosis”, said Henk Zomer, Senior Manager of the CT Business Unit at Canon Medical Systems Europe B.V.

Canon Medical is showcasing its AiCE technology at this year’s ECR in Vienna, February 28 – March 03, 2019 (Expo X3).